4 15 672 Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Alvin L. Young Collection on Agent Orange Description An account of the resource <p style="margin-top: -1em; line-height: 1.2em;">The Alvin L. Young Collection on Agent Orange comprises 120 linear feet and spans the late 1800s to 2005; however, the bulk of the coverage is from the 1960s to the 1980s and there are many undated items. The collection was donated to Special Collections of the National Agricultural Library in 1985 by Dr. Alvin L. Young (1942- ). Dr. Young developed the collection as he conducted extensive research on the military defoliant Agent Orange. The collection is in good condition and includes letters, memoranda, books, reports, press releases, journal and newspaper clippings, field logs and notebooks, newsletters, maps, booklets and pamphlets, photographs, memorabilia, and audiotapes of an interview with Dr. Young.</p> <p>For more about this collection, <a href="/exhibits/speccoll/exhibits/show/alvin-l--young-collection-on-a">view the Agent Orange Exhibit.</a></p> Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Box The box containing the original item. 184 Folder The folder containing the original item. 5360 Series The series number of the original item. Series VIII Subseries I Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Description An account of the resource <strong>Corporate Author: </strong>Bureau of National Affairs, Inc. Source A related resource from which the described resource is derived Current Report Date A point or period of time associated with an event in the lifecycle of the resource 1981 Title A name given to the resource Agent Orange: Veterans Ask High Court to Review Ruling Barring Federal Common Law Suit ao_seriesVIII https://www.nal.usda.gov/exhibits/speccoll/files/original/03f4f6a939b2cbd7cf19388bd69dbdd7.pdf ba849b6568e90db8c91b8414bd9e2b43 PDF Text Text Item D Number 05464 D Author Corporate Author United States Environmental Protection Agency (EPA), Report/Article Tltb Ambient Water Quality Criteria for 2,3,7,8Tetrachlorodibenzo-p-dioxin Journal/Book Tttte Year 1984 Month/Day February Color ° Number of Images 239 Doscrljton Notes EPA Friday, March 15, 2002 440/5-34-007 Page 5464 of 5571 �United States Environmental Protection Agency Office of Water Regulations and Standards Washington, DC 20460 EPA 440/5-84-007 February 1984 Water Ambient Water Quality Criteria for 2, 3, 7,8 - Tetrachlorodibenzo - p - dioxin �AMBIENT WATER QUALITY CRITERIA POR 2,3,7,8-TETRACHLORODIBENZO-P-DIOXIN Prepared By U.S. ENVIRONMENTAL PROTECTION AGENCY Office of Water Regulations and Standards Criteria and Standards Division Washington, D.C. Office of Research and Development Environmental Criteria and Assessment Office Cincinnati, Ohio Carcinogen Assessment Group Washington, D.C. Reproductive Effects Assessment Group Washington, D.C. Environmental Research Laboratories Corvallis, Oregon Duluth, Minnesota Gulf Breeze, Florida Narragansett, Rhode Island �DISCLAIMER This report has been reviewed by the Environmental Criteria and Assessment Office, U.S. Environmental Protection Agency, and approved for publication. Mention of trade names or commercial products does not consititute endorsement or recommendation for use. 11 �FOREWORD Section 304 (a)(l) of the Clean Water Act of 1977 (P.L. 95-217), requires the Administrator of the Environmental Protection Agency to publish criteria for water quality accurately reflecting the latest scientific knowledge on the kind and extent of all Identifiable effects on health and welfare which may be expected from the presence of pollutants 1n any body of water, Including groundwater. Proposed water quality criteria for the 65 toxic pollutants listed under section 307 (a)(l) of the Clean Water Act were developed and a notice of their availability was published for public comment on March 15, 1979 (44 FR 15926), July 25, 1979 (44 FR 43660), and October 1, 1979 (44 FR 56628). This document 1s a revision of those proposed criteria based upon a consideration of comments received from other Federal Agencies, State agencies, special Interest groups, and Individual scientists. The criteria contained In this document replace any previously published EPA criteria for the 65 pollutants. This criterion document 1s also published 1n satisfaction of paragraph 11 of the Settlement Agreement 1n Natural Resources Defense Council, et al. vs. train. 8 ERC 2120 (D.D.C. 1976), modified, 12 ERC 1833 (D.D.C. 1979). The term "water quality criteria" Is used In two sections of the Clean Water Act, section 304 (a)(l) and section 303 (c)(2). The term has a different program Impact In each section. In section 304, the term represents a non-regulatory, scientific assessment of ecological effects. The criteria presented 1n this publication are such scientific assessments. Such water quality criteria associated with specific stream uses when adopted as State water quality standards under section 303 become enforceable maximum acceptable levels of a pollutant 1n ambient waters. The water quality criteria adopted 1n the State water quality standards could have the same numerical limits as the criteria developed under section 304. However, In many situations States may want to adjust water quality criteria developed under section 304 to reflect local environmental conditions and human exposure patterns before Incorporation Into water quality standards. It Is not until their adoption as part of the State water quality standards that the criteria become regulatory. Guidelines to assist the States In the modification of criteria presented In this document, 1n the development of water quality standards, and 1n other water-related programs of this Agency, have been developed by EPA. STEVEN SCHATZOW Director Office of Water Regulations and Standards 111 �ACKNOWLEDGEMENTS Aquatic Toxicology: David J. Hansen (reviewer) Environmental Research Laboratory, Narragansett U.S. Environmental Protection Agency Charles E. Stephan (author) Environmental Research Laboratory, Duluth U.S. Environmental Protection Agency Gary A. Chapman Environmental Research Laboratory, CorvalUs U.S. Environmental Protection Agency Mammalian Toxlclty and Human Health Effects*: Debdas Mukerjee (document manager) Environmental Criteria and Assessment Office, Cincinnati U.S. Environmental Protection Agency K. Diane Courtney Roy Albert Institute of Environmental Medicine New York University Medical Center Frederick Coulston Coulston International Corporation Health and Effects Research Laboratory, Research Triangle Park U.S. Environmental Protection Agency Michael L. Dourson Environmental Criteria and Assessment Office, Cincinnati U.S. Environmental Protection Agency Donald G. Barnes Office of Pesticides and Toxic Substances U.S. Environmental Protection Agency David Firestone Food and Drug Administration Steven P. Bayard Carcinogen Assessment Group U.S. Environmental Protection Agency S. GaratUnl Institute d1 Recerche Farmacologlc "Mario Negrl" Milan, Italy David L. Bayllss Carcinogen Assessment Group U.S. Environmental Protection Agency Richard Grelssmer Oak Ridge National Laboratory D1pak K. Basu Syracuse Research Corporation Bernard H. Haberman Carcinogen Assessment Group U.S. Environmental Protection Agency Randall J.F. Bruins Environmental Criteria and Assessment Office, Cincinnati U.S. Environmental Protection Agency Lennart Hardell University Hospital Umea, Sweden *An additional 60 participants from EPA's headquarters, Research Triangle Park, Cincinnati, and regional offices and 135 observers from Industries, academla, environmental groups and news media also attended the meeting at which this chapter was reviewed. 1v �Robert Hapless Environmental Monitoring Systems Laboratory U.S. Environmental Protection Agency Charles H. Nauman Exposure Assessment Group U.S. Environmental Protection Agency Michael W. Neal Syracuse Research Corporation Rolf Hartung University of Michigan AUstalr W.M. Hay University of Leeds, U.K. James Olsen School of Medicine State University of New York Charallnggayya Hlremath Carcinogen Assessment Group U.S. Environmental Protection Agency F. Pocchlarl Institute Superlore dl Sanlta V1ale Reglna, Rome, Italy Otto Hutzlnger University of Amsterdam The Netherlands Shane Que Hee R.D. Klmbrough Centers for Disease Control C. Rappe Richard J. Koclba Dow Chemical Company Sheila L. Rosenthal Reproductive Effects Assessment Group U.S. Environmental Protection Agency University of Cincinnati Medical Center University of Umea, Sweden Marvin Legator University of Texas Medical Branch Steven H. Safe Texas A&M University - Ruth L1l1s Mt. Sinai School of Medicine Marvin Schnelderman Environmental Law Institute Prab D. Lotllkar Temple University School of Medicine Ellen SUbergeld Environmental Defense Fund Fumlo Matsumura Michigan State University E. McConnell National Institute of Environmental Health Sciences W.P. McNulty Oregon Regional Primate Research Center Robert Miller National Cancer Institute David Stalling Columbia National Fisheries Research Laboratory Jerry F. Stara Environmental Criteria and Assessment Office, Cincinnati U.S. Environmental Protection Agency Lewis Thlbodeaux University of Arkansas Thomas Tlernan Wright State University Ralph Nash U.S. Department of Agriculture Technical Support Services Staff: Mann, E. Durden, C.A. Cooper Clerical Staff: J.A. Olsen, B.L. Zwayer, P.A. Daunt, K.S. N.C. Bauer, S.J. Faehr, T. Highland, L.A. Schwaegerle �TABLE OF CONTENTS Page INTRODUCTION A-l Physical Properties Cocontamlnants of 2,3,7,8-TCDD In Chlorinated Products Synthesis Chemistry Analytical Methods for TCOO Summary of Health Effects References AQUATIC TOXICOLOGY A-l A-2 A-3 A-3 A-4 A-8 A-9 B-l Introduction Acute Toxlclty to Aquatic Animals Chronic Toxldty to Aquatic Animals Toxlclty to Aquatic Plants B1oaccumulat1on Other Data Unused Data Summary B-l B-l B-2 B-3 B-3 B-6 B-6 B-9 National Criteria B-10 References B-ll MAMMALIAN TOXICOLOGY AND HUMAN HEALTH EFFECTS C-l EXPOSURE C-l Water and Soil Related. Ingestlon from Food Inhalation Dermal C-l C-6 C-15 C-17 PHARMACOKINETICS C-18 Absorption Distribution Metabolism Excretion fc EFFECTS C-18 C-22 C-28 C-31 C-36 Acute, Subacute and Chronic Toxlclty Synerglsm and/or Antagonism Teratogen1c1ty MutagenlcHy Carclnogenldty C-36 C-69 C-72 C-102 C-115 v1 �TABLE OF CONTENTS Page CRITERION FORMULATION C-176 Existing Guidelines and Standards Current Levels of Exposure Special Groups at Risk Basis and Derivation of Criterion C-176 C-176 C-177 C-178 Estimates by Others of Carcinogenic Potency and Criteria C-182 REFERENCES C-185 APPENDIX C-241 �LIST OF TABLES No. 1. Title Predicted Bloconcentratlon Factors for 2,3,7,8-TCOO Based on Estimated and Measured Values of the OctanolWater Partition Coefficient Page B-4 2. Other Data on Effects of 2,3,7,8-TCDD on Aquatic Organisms. . B-7 1. Levels of 2,3,7,8-TCDD 1n F1sh and Shellfish C-10 2. Percentage of 2,3,7,8-TCDD 1n the Liver 24 Hours After Oral Administration of 0.5 ml of Various Formulations Containing TCDD C-20 3. Tissue Distribution of 2,3,7,8-TCDO C-23 4. Elimination of 2,3,7,8-TCOO C-33 5. Lethality of 2,3,7,8-TCDO Following Acute Exposure C-37 6. Toxic Responses Following Exposure to 2,3,7,8-TCDD: Species Differences C-42 Hepatocellular Fatty Change Observed 1n Rats Following Subchronlc Exposure to 2,3,7,8-TCDD C-56 Effects of Chronic Exposure to 2,3,7,8-TCDD In Laboratory Rodents C-59 Studies on the Potential Teratogenlc Effects of 2,3,7,8-TCDD-Contamlnated 2,4,5-T C-74 7. 8. 9. 10. Studies on the Potential Teratogenlc Effect of 2,3,7,8-TCDO 11. 12. 13. 14. 15. C-79 The Results of MutagenlcHy Assays for 2,3,7,8-TCOO 1n Salmonella typhlmurlum C-103 Distribution of Tumor Types In Two Case-Control Studies of Soft-Tissue Sarcoma C-122 Exposure Frequencies 1n Two Case-Control Studies of Soft-Tissue Sarcoma C-123 Relative Risks of Soft-Tissue Sarcoma 1n Relation to Exposure to Phenoxyacetlc Adds and Chlorophenols In Two Case-Control Studies C-125 Distribution of Hlstologlcal Types of Soft-Tissue Sarcomas C-130 �LIST OF TABLES No. 16. TUIe Page Midland County Soft and Connective Tissue Cancer Deaths 1960-1981 C-139 17. Other Occupations (Minus Forestry/Agriculture) C-145 18. Other Occupations (Minus Forestry/Agriculture/ Woodworkers) C-146 Analysis of Stomach Cancer Mortality In a Group of West German Factory Workers Exposed to 2,3,7,8-TCDO C-150 19. 20. Reanalysls of Stomach Cancer Mortality 1n a Group of West German Factory Workers Exposed to 2,3,7,8-TCDO 21. Stomach Cancer Mortality In Three Studies of Workers Exposed to Phenoxyacetlc Add Herbicides and/or 2,3,7,8-TCDD 22. C-153 C-155 Incidence of Primary Tumors 1n Female Swiss-Webster Mice by Dermal Application of 2,3,7,8-TCDD or 2,3,7,8-TCDD Following DMBA 23. 24. 25. 26. 27. C-160 Incidence of Primary Tumors In Male Swiss-Webster Mice by Dermal Application of 2,3,7,8-TCDD or 2,3,7,8-TCDD following DMBA C-161 Summary of Neoplastlc Changes After 2,3,7,8-TCDO 1n Rats C-163 Summary of Neoplastlc Lesions Produced by 2,3,7,8-TCDD 1n Sprague-Dawley Rats, Spartan Substraln, that are Statistically Significant In at Least One Sex C-166 2,3,7,8-TCDD Oral Rat Study by Dr. Kodba, with Dr. Squire's Review (8/15/80) Female Sprague-Dawley Rats Spartan Substraln (2 years) C-167 2,3,7,8-TCDD Oral Rat Study by Dr. Kodba, with Dr. Squire's review (8/15/80) Male Sprague-Dawley Rats - Spartan Substraln (2 years) . . . C-168 28. Incidence of Primary Tumors In Male Osborne-Mendel Rats . . . C-170 29. Incidence of Primary Tumors In Female Osborne-Mendel Rats . . C-172 30. Incidence of Primary Tumors 1n Female B6CF1 Mice C-173 31. Incidence of Primary Tumors 1n Male B6CF1 Mice C-174 32. Summary of Human Potency Estimates for 2,3,7,8-TCDD C-242 1x �CRITERIA DOCUMENT 2,3,7,8-TETRACHLORODIBENZO-P-DIOXIN CRITERIA Aquatic Life Not enough data are available concerning the effects of 2,3,7,8-TCDD on aquatic life and Its uses to allow derivation of national criteria. The available Information Indicates that acute values for some freshwater animal species are >1.0 pg/8,; some chronic values are <0.01 pg/8., and the chronic value for rainbow trout 1s <0.001 yg/S.. Because exposures of some species of fishes to 0.01 pg/a for <6 days resulted In substantial mortality several weeks later, derivation of aquatic life criteria for 2,3,7,8-TCDD may require special consideration. Predicted bloconcentratlon factors (BCFs) for 2,3,7,8-TCDD range from 3000-900,000, but the available measured BCFs range from 390-13,000. If the BCF Is 5000, concentrations >0.00001 pg/9, should result 1n concentrations in edible freshwater and saltwater fish and shellfish that exceed levels Identified In a U.S. FOA health advisory. If the BCF 1s >5000 or 1f uptake In a field situation 1s greater than that 1n laboratory tests, the value of 0.00001 pg/a, will be too high. Human Health For the maximum protection of human health from the potential carcinogenic effects due to 2,3,7,8-TCDD exposure through 1ngest1on of contaminated water and contaminated aquatic organisms, the ambient water concentration should be zero. This criterion 1s based on the non-threshold assumption for 2,3,7,8-TCDD. However, zero may not be an attainable level at this time. �Therefore, the levels that may result 1n an Increase of cancer risk over the lifetime are recommended estimated criteria respectively. at are 10~5, 10~* 1.3xlO~7, and 10~7. 1.3xlO~8 and The corresponding 1.3xlO~9 yg/l. If the above estimates are made for consumption of aquatic organisms only, excluding consumption of water, the levels are 1.4xlO~7, 1.4xl(Te and 1.4xlO~* yg/l, respectively. for consumption of water only, the levels 2.2xlO~9 pg/S., respectively. xl If these estimates are made are 2.2x10"*, 2.2xl(T7 and �INTRODUCTION The major source of 2,3,7,8-tetrachlorod1benzo-p_-d1ox1n (2,3,7,8-TCDO) (CAS Number 1746-01-6) appears to be as a contaminant formed during the production of 2,4,5-tMchlorophenol (2,4,5-TCP) from 1,2,4,5-tetrachlorobenzene (Mllnes, 1971; K1mm1g and Schulz, 1957; Firestone et al.. 1972). 2,4,5-TCP 1s the major chemical feedstock 1n the production of several herbicides Including 2,4,5-trlchlorophenoxyacetlc add Sllvex. (2,4,5-T), 2,4,5-T esters and Each of these chemicals may contain 2,3,7,8-TCOD as a contaminant (Buser and Bosshardt, 1974; Courtney et al., 1970; Edmunds et al., 1973; ZHko and Choi, 1971). formed during It has also been reported that 2,3,7,8-TCDO may be the pyrolysls of chlorinated phenols (Buu-Ho1 et al., 1971a,b), chlorinated benzenes (Buser, 1979) and polychlorlnated dlphenyl ethers (Llndahl et al., 1980), and thus can also be emitted by municipal Incinerators (Rappe et al., 1983a; Lustenhouwer et al., 1980; Olle et al., 1982, 1983). There 1s no clear evidence that 2,3,7,8-TCOD 1s a typical contaminant 1n the herbicide 2,4-d1chlorophenoxyacet1c add (2,4-0) (Woolson et al., 1972; Henshaw et al., 1975; Cochrane et al., 1981). Physical Properties 2,3,7,8-TCDD 1s a symmetrical, nearly planar molecule with the empirical formula C,nH.C1.0_. The four chlorine 12 4 4 2 from one another (Poland and Glover, 1973). atoms are Indistinguishable 2,3,7,8-TCDD 1s a white crys- talline solid with a melting point range of 302-305°C (Sparschu et al., 1971; Elvldge, 1971) and has a molecular weight of 321.9. The vapor pressure of this compound Is estimated to be 10~6 mm of Hg (0.1 mPa) at 1 atmosphere and 25°C (Mabey et al., 1981). The Henry's constant has been estimated to be 2.1xlO"3 atmosphere m3 mol"1 (Mabey et al., 1981). 2,3,7,8-TCDD 1s Upophlllc, exhibiting some solubility 1n fats, oils and A-l �other relatively nonpolar solvents, and Is only slightly soluble 1n water (0.2 yg/l) (Grummet and Stehl, 1973; Morris, 1981). The solubility of 2,3,7,8-TCOO 1n various organic solvents Is given below (Grummet and Stehl, 1973): Solvent Solubility (ppm) lard oil benzene o-dlchlorobenzene chloroform acetone n-octanol methanol 44 570 1400 370 110 50 10 The partition coefficient of 2,3,7,8-TCDD 1n a water:hexane system has been reported to be 1000 (Matsumura and Benezet, 1973). The octanol/water partition coefficient (K ) has been calculated by the methods of Hansch and Leo (1979) and has been experimentally measured. Calculated values for log K range from 6.84-7.28, and a measured value of 6.15 has been reported (see Section B, B1oaccumulat1on). Cocontamlnants of 2.3.7.8-TCDD 1n Chlorinated Products 2,3,7,8-TCDD Is only one of many trace contaminants found In some chlorinated Industrial products Including a few chlorinated phenols, a few chlorinated phenoxy adds (especially the herbicides 2,4,5-T and Sllvex) and hexachlorophene. Among the other trace contaminants found 1n these products are members of the polychloMnated d1benzo-p_-d1ox1ns (PCDOs), polychlorlnated dlbenzofurans (PCDFs), polychlorlnated dlphenylethers (PCDPEs), polychlorinated phenoxyphenols (PCPPs), polychlorlnated blphenyls (RGBs) and polychlorlnated benzenes (PCBz). Of these, some possess properties that make them difficult to separate analytically from the 2,3,7,8-TCDD Isomer (Klmbrough, 1974; U.S. EPA. 1980; Bumb et a!., 1980; Rappe et al., 1983b). A-2 �Syntheses 2,3.7,8-TCDD has been synthesized by several methods In moderate yield (e.g., reaction of dlchlorocatechol salts with o-chlorobenzene by refluxlng In alkaline dlmethylsulfoxlde; chlorlnatlon of d1benzo-p_-d1ox1n In the presence of ferric chloride and Iodine; UV Irradiation of PCODs of high chlorine content; Ullman reaction of chlorinated phenolates at 180-400°C; pyrolysls of chlorinated phenolates and chlorinated phenols; heating 1,2,4-trlchloro5-n1trobenzene and 4,5-d1chlorocatechol In the presence of base). These processes have been reviewed 1n U.S. EPA (1980). Chemistry 2,3,7,8-TCDD 1s considered to be relatively stable toward heat, acids and alkalies. It begins to decdmpose at 500°C, and at a temperature of 800°C, virtually complete degradation occurs within 21 seconds (Stehl et al., 1973). From a theoretical equation for thermal dissociation constant K = 1015'5 exp(-80,000/RT) sec"1 K = dissociation constant, R = universal gas constant, T = temperature for tetrachlorod1benzo-p_-d1ox1ns formulated by Staub and Tsang (1983), the 99.99% gas phase dloxln dissociation at 727°C will require about 15.4 minutes. The same equation predicts a 99.99% decomposition of tetrachloro• d1benzo-p_-d1ox1ns 1n 0.3 seconds at 977°C. molecule (FanelU et al., 1978). Gamma radiation degrades the 2,3,7,8-TCDD can be perchlorInated (Hutzlnger et al., 1972). 2,3,7,8-TCDD 1s transformed very slowly 1n aquatic systems. Of the four transformation processes (photoreactlon, blotransformatlon, hydrolysis and radical oxidation) that control the fate of a chemical In aquatic media only the first two processes are throught to effect the transformation of 2,3,7,8-TCDD (Matsumura et al., 1983). In organic solvents, 2,3,7,8-TCDO undergoes reductive photodechlorlnatlon at wavelengths <320 nm (Crosby et A-3 �al., 1971; Libert! et al., 1978). In aqueous solution hydroxylatlve dechlorlnatlon probably occurs, although this has not been seen. L1bert1 et al. (1978) showed that 2,3,7,8-TCDD spread over silica gel, aluminum, glass, ceramic tile and marble 1n the absence of an organic solvent showed various decomposition rates on UV Irradiation. Little decomposition occurred on glass or marble, but substantial degradation occurred on silica gel and aluminum. Also, 1:1 ethyl oleate/xylene was found to be a satisfactory H * donor. PUmmer et al. (1973) reported that a 2,3,7,8-TCDO suspension 1n distilled water remained unchanged when Irradiated with a sunlamp. Simi- larly, a thin dry film of 2,3,7,8-TCDD on a glass plate or 2,3,7,8-TCDD on dry and wet soils showed negligible photodegradatlon after Irradiation with sunlamps (Crosby et al., 1971). In contrast, 2,3,7,8-TCDD 1n methanol solution, or a benzene solution of 2,3,7,8TCDD In water 1n the presence of a surfactant underwent substantial photodegradatlon under sun lamp or sunlight Irradiation (PUmmer et al., 1973; Crosby et al., 1971). NestMck et al. (1980) experimentally determined the photolytlc half-life of 2,3,7,8-TCOD 1n n-hexa-decane under sunlamp Irradiation to be -57 minutes. The surfactant, l-hexyldecylpyr1o1n1um chloride, sensitized the photodecomposltlon In aqueous solution (Botre et al., 1978). The evolution of 2,3,7,8-TCDD from more highly chlorinated PCDDs on UV Irradiation from sunlight Is unlikely since dechlorlnatlon 1n organic solvents and In the presence of artificial UV sources occurs preferentially at the 2,3,7,8-pos1t1ons (Buser and Rappe, 1978; Nestrlck et al., 1980). Analytical Methods for TCDD Host of the current analytical methods used for the Identification and quantHatlon of 2,3,7,8-TCDD are based upon gas chromatography/mass spectrometry (GC/MS). This method provides both high sensitivity (detection at A-4 �low-ppt levels) and required selectivity (Crummett and Stehl, 1973; "Mernan et al., 1975; Taylor et al., 1975; Buser and Bosshardt, 1976; Buser, 1977; U.S. EPA, 1980; Tlernan, 1983). Unfortunately, the GC/MS method Is expen- sive, time consuming and difficult. Elaborate quality control and quality assurance of analytical methods are necessary. Rad1o1mmunoassay and elec- tron capture-GC have also been developed; both were essentially screening techniques (Karasek and Onuska, 1982). Sampling Method -- Two types of sampling methods can be used for collecting aqueous samples for TCDD. In the first method, no preconcentratlon of the samples during collection 1s made. Grab samples are collected 1n clean amber glass bottles of 1 8. or 1 quart capacity fitted with screw caps lined with clean Teflon or aluminum foil (U.S. EPA, 1982). The sample containers must be kept refrigerated at 4°C and protected from light during collection and shipment of grab and composite samples. All samples must be extracted within 7 days and completely analyzed within 40 days of extraction (U.S. EPA, 1982). The second method 1s the preconcentratlve method of sample collection (010omen1co et al., 1980). In t.ils method, 2-20 8, of water are allowed to pass through a 12 cm long x 1.5 cm Internal diameter XAD-2 column. The XAD-2 columns containing the polychlorlnated dloxlns should be protected from light and kept at 4°C during transportation and storage. Analysis — An appropriate volume of water (depending on the desired detection limit) with added Internal standard of either 13C^ 2 or 37 C14 2,3,7,8-TCDD In the amount of 2.5-25 ng (Harless et al., 1980; U.S. EPA, 1982) can be extracted with hexane (D1Domen1co et al., 1980), dlchloromethane (U.S. EPA, 1982; Harless et al., 1980) or petroleum ether A-5 �(Van Ness et al., 1980). Judging from the recovery data (U.S. EPA, 1982; DIDomenlco et al., 1980; Harless et al., 1980), dlchloromethane appears to be a better solvent. The extract containing TCDDs can be cleaned up by acid and base wash (Harless et al., 1980; U.S. EPA, 1982; Van Ness et al., 1980) and by subsequent liquid chromatography with an alumina column (Harless et al., 1980; Van Ness et al., 1980); however, U.S. EPA (1982) recommends another cleanup step using silica gel liquid chromatography, which may be necessary for wastewater but may be unnecessary for drinking water and clean surface water samples. The final separation and analysis Is performed by low resolution GC-HRMS (Van Ness et al., 1980; Harless et al., 1980) or high resolution GC-HRMS or LRMS (U.S. EPA, 1982). The U.S. EPA (1982) method derived from the method of Buser and Rappe (1980) seems to be an appropriate one because H recommends using a 50 m Sllar IOC capillary column that resolves 2,3,7.8TCl)0 from Us other Uomers. This same column can resolve 1,2,3,7,8-pentaCDO from other penta-CDDs, and 1,2,3,6,7,8-, 1,2,3,7,8,9- and 1,2,3,4,7,8hexa-CDDs from other hexa-CDDs (Rappe et al., 1983a). Other suitable col- umns Include SP-2330, SP-2340 and DB-5 (Tlornan, 1983). Harless et al. (1980) reported that TCDO 1n water can be accurately determined down to a concentration of 0.03 ppt. However, for determination of <1 ppt, rigorous measures must be taken to avoid the possibility of sample contamination during collection, storage, transportation or analysis. Gas Chromatography/Mass Spectrometry (GC/MS): The mass spectrometral pattern of 2,3,7,8-TCDO Is very similar to the spectra of other tetrachlorod1benzo-p_-d1ox1ns. Since other compounds (e.g., certain polychlorlnated blphenyls) present 1n the sample extract can also give rise to mass spectral A-6 �Ions at the same nominal masses as TCOOs (m/e 320 and m/e 322), two approaches are being used to Increase specificity (U.S. EPA, 1980). The first approach of applying high resolution mass spectrometry (M/&M >9000) to Increase the selectivity makes use of the small difference In the "exact" masses of TCDOs (C H Cl 0 having an "exact" 321.8936) compared with compounds of similar molecular weight. mass of Application of the optimum chromatographlc conditions and columns to maximize the resolution of compounds 1s necessary before the MS step. The second approach to avoid the problem of Interferences from closely related compounds Is to make use of low-resolution mass spectrometry Incorporated with a more selective separation step such as capillary column GC (Buser and Rappe, 1980; Rappe et al., 1983b) or high performance liquid chromatography followed by GC (NestMck et al., 1979). The former method can be used for all PCCOs and PCDFs; the latter method Is selective In characterizing only the TCDDs (Bumb et al., 1980). The following criteria have been outlined by Harless et al. (1980) for confirmation of 2,3,7,8-TCDD residues: 1. Correct GC/MS retention time for 2,3,7,8-TCDD. 2. Correct Isotope ratio for the molecular Ions 320 and 322. 3. Correct simultaneous response for the jnolecular Ions 320, 322 and 328. 4. Correct responses for the co-Injection of sample fortified with 37 C1-TCDO and 2,3,7,8-TCDD standard. 5. Response of molecular Ions 320 and 322 must be >2.5 times the noise level. Supplemental criteria that Harless et al. (1980) suggested for highly contaminated extracts are: 1. COC1 loss Indicative of TCDD structure 2. GC/MS peak-matching analysis of molecular Ions 320 and 322 1n real time to confirm the 2,3,7,8-TCDD elemental composition. A-7 �Summary of Health Effects 2,3,7,8-TCDD 1s one of the most toxic substances known. It exhibits a delayed biological response 1n many species and 1s highly lethal at low doses to aquatic organisms, birds and mammals. It has been shown to be acnegenlc, fetotoxlc, teratogenlc, mutagenlc (In a limited number of mutagenldty tests) and carcinogenic, and affects the Immune responses 1n mammals. These "findings, In conjunction with the wide distribution of contaminated products and Us extreme stability In the environment, lead to the conclusion that 2,3,7,8-TCDD represents a potential hazard to both aquatic and terrestrial life, and makes 2,3,7,8-TCDD one of the major concerns for public health. A-8 �REFERENCES Botre, C., A. Memoll and F. Alhalque. 1978. decomposltlon 1n aqueous solutions. TCDD solubH1zat1on and photo- Environ. Sc1. Technol. 12: 335-336. Bumb, R.R., W.B. Grummet, S.S. Cutle, et al. 1980. fire: A source of chlorinated dloxlns. Science. 210: Buser, H.R. 1977. Trace chemistries of 385. Determination of 2,3,7,8-tetrachlorod1benzo-p_-d1ox1n 1n environmental samples by high-resolution gas chromatography and low-resolution mass spectrometry. Anal. Chem. 49: 918. Buser, H.R. 1979. Formation of polychlorlnated dlbenzofurans (PCDFs) and d1benzo-p_-d1oxins (PCDOs) from the pyrolysls of chlorobenzenes. phere. Chemos- 8: 415-424. Buser, H.R. and H.P. Bosshardt. 1974. Determination of 2,3,7,8-tetra- chlorod1benzo-l,4-d1ox1n at parts per billion levels 1n technical grade 2,4,5-tr1chlorobenozyacet1c acid In 2,4,5-T alkyl ester and 2,4,5-T amlne salt herbicide formulations by quadruple mass fragmentography. J. Chromatogr. 90: 71. Buser, H.R. and H.P. Bosshardt. 1976. Determination of polychlorlnated d1benzo-p_-d1ox1ns and dlbenzofurans 1n commercial pentachlorophenols by combined GC-MS. J. Assoc. Off. Anal. Chem. 59: 562-569. A-9 �Buser, H.R. and C. Rappe. 1978. Identification of substitution patterns 1n polychlorlnated d1benzo-p_-d1ox1ns (PCODs) by mass phere. spectrometry. Chemos- 7: 199-211. Buser, H.R. and C. Rappe. 1980. High-resolution gas chromatography of the 22 tetrachlorod1benzo-p_-d1ox1n Isomers. Anal. Chem. 52: 2257-2262. Buu-Ho1, N.P., et al. 1971a. Tetrachlorod1benzod1ox1n - Intimations of cardnogenlcHy. C.R. Hebol. Seanc. Acad. Scl., Paris. 272: 1447. Buu-Ho1, N.P., et al. 1971b. Comptes rendus des seances de 1. Acad. Sc1. Ser. D. 273: 708. Cochrane, W.P., J. Singh, W. Miles and B. Wakeford. 1981. Determination of chlorinated d1benzo-p_-d1ox1n contaminants 1n 2,4-D products by gas chromatography-mass spectrometrlc techniques. J. Chromatog. 217: 289-299. Courtney, K.D., D.W. Gaylor, M.D. Hogan, M.L. Falk, R.R. Bates and I. Mitchell. 1970. Teratogenlc evaluation of 2,4,5-T. Science. Crosby, O.G., A.S. Wong, J.R. PUmmer and E.A. Woolson. posltlon of chlorinated d1benzo-p_-d1ox1ns. Science. Crummet, W.B. and R.H. Stehl. 1971. 168: 864-866. Photodecom- 173: 748-749. 1973. Determination of chlorinated dlbenzo- p-dloxlns and dlbenzofurans 1n various materials. 5: 15-25. A-10 Environ. Health Perspect. �D1Domen1co, A., V. Sllano, G. V1v1ano and G. Zapponl. 1980. Accidental release of 2,3,7,8-tetrachlorod1benzo-p_-d1ox1n (TCOD) at Seveso, Italy. I. Sensitivity and specificity of analytical procedures adopted for TCDD assay. Ecotoxlcol. Environ. Safety. 4(3): Edmunds, J.W., et al. 1973. 283-297. Determination of 2,3,7,8-tetrachlorod1benzo-p- dloxln In 2,4,5-trlchlorophenoxyacetlc add and 2,4,5-T alkyl ester herbi- cides. Pestle. Scl. 4: 101. Elvldge, D.A. 1971. The gas chromatographlc determination of 2,3,7,8- tetrachlorod1benzo-p_-d1ox1n 1n 2,4,5-tr1chlorophenoxyacet1c add ("2,4,5-T"), 2,4,5-T ethylhexyl ester, formulations of 2,4,5-T esters, and 2,4,5-tMchlorophenol. Analyst. 99: 721. FanelH, R., C. Chlabrando, M. Salmona, S. Gratt1n1 and P.G. Caldera. 1978. Degradation of 2,3,7,8-tetrachlorod1benzo-p_-d1ox1n 1n organic solvents by gamma ray Irradiation. Experlentla. 34: 1126-1127. Firestone, 0., J. Ress, N.L. Brown, R.P. Barron and J.N. Damlco. 1972. Determination of polychlorod1benzo-p_-d1ox1ns and .related compounds In commercial chlorophenols. J. Assoc. Off. Anal. Chem. 55(1): 85-92. Hansch, C. and A.L. Leo. 1979. Analysis 1n Chemistry and Biology. SubstHuent Constants for Correlation John Wiley and Sons, Inc., NY. p. 18-47. A-ll �Harless, R.L., E.G. Oswald, M.K. Wilkinson, et al. 1980. Sample prepara- tion and gas chromatography-mass spectrometry determination of 2,3,7,8tetrachlorod1benzo-p_-d1ox1n. Anal. Chem. 52(8): 1239-1245. Henshaw, B., et al. 1975. Gas-I1qu1d chromatography and gas-Hqu1d com- bined with mass spectrometry of a butyl ester formulation of (2,4-d1chlorophenoxy) acetic add. J. Chromatogr. 106: 33. Hutzlnger, 0., S. Safe and V. ZHko. 1972. Analysis of chlorinated aro- matic hydrocarbons by exhaustive chlorlnatlon: Qualitative and structural aspects of the perchloro-derlvatlves of blphenyl naphthalene, terphenyl, dlbenzofuran, dlbenzo-dloxln and DDE. Intern. J. Environ. Anal. Chem. 2: 95-106. Karasek, F.W. and F.I. Onuska. 1982. Trace analysis of the dloxlns. Anal. Chem. 54: 309A-326A. Klmbrough, R.D. 1974. The to.:1c1ty of polychloMnated polycycllc chemicals and related compounds. Cr1t. Rev. Toxlcol. 2: 445. K1mm1g, J. and K.H. Schulz. 1957. BerufUche akne (sog. chlorakne) durch chlorlerte aromatlsche zykllsche ather. Dermatologlca. 115: 540. (Ger.) L1bert1, A., D. Brocco, I. Allegr1n1 and G. Bertonl. 1978. Field photo- degradation of TCDD by ultraviolet radiations. In: D1ox1n: Tox1colog1cal and Chemical Aspects, F. Cattabenl et al., Ed. SP Medical and Scientific Books, NY. p. 195-200. A-12 �Llndahl, R., C. Rappe and H.R. Buser. 1980. Formation of polychlorlnated d1benzo-p_-d1ox1ns (PCOOs) from the pyrolysls of polychlorlnated dlphenyl ethers. Chemosphere. 9: 351-361. Lustenhouwer, J.W.A., K. OHe and 0. Hutzlnger. 1980. Chlorinated dlbenzop-dtoxlns and related compounds In Incinerator effluents: A review of measurements and mechanisms of formation. Chemosphere. Mabey, W.R., J.H. Smith, R.T. Podoll, et al. data for organic priority pollutants. 9: 501-522. 1981. Aquatic fate processes Monitoring and Data Support Dlv., OWRS, Washington, DC. EPA 440/4-81-014. p. 107-108. Matsumura, F. and H.3. Benezet. mlcroblal degradation of 1973. Studies on the bloaccumulatlon and 2,3,7,8-tetrachlorod1benzo-p_-d1ox1n. Environ. Health Perspect. 5: 253-258. Matsumura, F., J. Quensen and G. Tsushlmoto. of TCDD In a model ecosystem. 1983. Mlcroblal degradation In: Human and Environmental Risks of Chlori- nated D1ox1ns and Related Compounds. R.E. Tucker, A.L. Young and A.P. Gray, Ed. Plenum Press, NY. p. 191-219. Mllnes, M.H. 1971. Formation of 2,3,7,8-tetrachlorod1benzo-p_-d1ox1n by thermal decomposition of sodium 2,4,5-trlchlorophenate. Nature. Nestrlck, T., et al. 1979. 232: 395. Synthesis and Identification of the 22 tetra- chlorod1benzo-p_-d1ox1n Isomers by high performance liquid chromatography and gas chromatography. Anal. Chem. 51: 2273. A-13 �NestMck, T.J., L.L. Lamparskl and D.I. Townsend. 1980. Identification of tetrachlorod1benzo-2-d1ox1n Isomers at the 1 ng level by photolytlc degradation and pattern recognition techniques. Morris, L.A. 1981. The movement, herbicides and TCDD In the forest. Anal. Chem. 52: 1865-1875. persistence and fate of the phenoxy Res. Rev. 80: 66-135. 011e, K., J.W.A. Lustenhouwer and 0. Hutzlnger. d1benzo-p_-d1ox1ns and related compounds 1982. PolychlorInated 1n Incinerator effluents. In: Chlorinated 01ox1ns and Related Compounds: Impact on the Environment, 0. Hutzlnger et al., Ed. Pergamon Press, NY. p. 227-244. OHe, K., H.V.O. Berg and 0. Hutzlnger. 1983. Formation and fate of PCDD from combustion processes. Chemosphere. 12: 627-636. PUmmer, J.R., U.I. KUngeblel, O.G. Crosby and A.S. Wong. Chlorod1ox1ns...Origin and Fate. Adv. Chem. Ser. 120. Poland, A. and E. Glover. 1973. Inducers of 4-am1nolevul1n1c add lase. 1973. In: p. 44. Chlorinated d1benzo-p_-d1ox1ns: Potent synthetase and aryl hydrocarbon hydroxy- II. A study of the structure activity relationship. Mol. Pharmacol. 9: 736. A-14 �Rappe, C., S. Marklund, P.-A. Bergqvlst and M. Hansson. 1983a. PolychloMnated dloxlns, dlbenzofurans and other polychlorlnated polynuclear aromatlcs formed during Incineration and PCB fires. In: Chlorinated Dloxlns and Olbenzofurans In the Total Environment, Vol. 1, L.H. Keith et al., Ed. Butterworth Publishers. Rappe, C., S. Marklund, M. Nygren and A. Gar1. 1983b. Parameters for Identification and confirmation In trace analyses of polychlorlnated dloxlns and dlbenzofurans. In: Chlorinated 01ox1ns and Olbenzofurans 1n the Total Environment, Vol. 1, L.H. Keith et al., Ed. Butterworth Publishers. Sparschu, G.L., Jr., F.L. Dunn, Jr., V.K. Rowe, Jr. 1971. Study of the teratcrgenlclty of 2,3,7,8-tetrachlorod1benzo-£-d1ox1n 1n the rat. Food Cosmet. Toxlcol. 9: 405-412. Staub, W.M. and W. Tsang. 1983. Physical and chemical properties of dloxlns 1n relation to their disposal. In: Human and Environmental Risks of Chlorinated Dloxlns and Related Compounds. R. Tucker, A.L. Young and A.P. Gray, Ed. Plenum Publishing Corp., NY. p. 731-748. Stehl, R.H., et al. 1973. The stability of pentachlorophenol and chlori- nated dloxlns to sunlight, heat, and combustion. Adv. Chem. Ser. 120. Am. Chem. Soc., Washington, DC. Taylor, M.L., et al. 1975. Determination of tetrachlorodlbenzop-dloxlns 1n chemical and environmental matrices. Proceed. 23rd Ann. Conf. Mass Spectro- metry Allied Topics, p. 337. A-15 �Tiernan, T.O., et al. 1975. Measurement of tetrachlorod1benzo-p_-d1ox1ns 1n USAF herbicide stocks and In environmental samples. 6th Ann. Symp. Environ. Res., Edgewood Arsenal, MD. Tiernan, 1.0. 1983. Analytical chemistry of polychloMnated d1benzo-p- dloxlns and dlbenzofurans: A review of the current status, in: Chlorinated Dloxlns and Dlbenzofurans 1n the Total Environment, Vol. 1. L.H. Keith et al., Cd. But UT worth Publishers, p. 211-237. U.S. EPA. 1980. Dloxlns. UKI , U.S. H'A, Cincinnati, OH. EPA 600/2-80-197. U.S. EPA. 1982. Test Method: 2,3,7,8-Tetrachlorod1benzo-p_-d1ox1n-method 613. Storet No. 34675. Environ. Monitor. Support Lab., Cincinnati, OH. Van Ness, G.F., 3.G. Solch, M.I. laylor and T.O. llernan. l^RO. letra- chlorod1benzo-p_-d1ox1ns 1n chemical wastes, aqueous effluents and soils. Chemosphere. 9: 553-563. Woolson, E.A., R.F. Thomas and P.D.J. Ensor. benzo-p_-d1ox1n content 1972. Survey of polychlorodl- 1n selected pesticides. J. Agrlc. Food Chem. 20(2): 350-354. ZHko, V. and P.M.K. Choi. 1971. PCB and other Industrial halogenated hydrocarbons 1n the environment. FRB Tech. Rep. 272: 25. A-16 �Aquatic Toxicology* Introduction Most of the available data related to effects of 2,3,7,8-tetrachlorodlbenzo-p-d1ox1n (2,3,7,8-TCDD) on aquatic life have been generated by Norrls and co-workers, Isensee and co-workers, Hatsumura and co-workers, and Helder. Much of the available Information Is from studies Involving freshwater microcosms. Although such studies are Intended to provide Information on fate of a test material, some data concerning effects on aquatic life are also obtained. No tests have been conducted using saltwater organisms. The last literature search for Information that could be used 1n this chapter was conducted In November, 1983. Acute Toxldty to Aquatic Animals Although the data available concerning 2,3,7,8-TCDD do not allow calculation of an acute value for any species, some useful Information does exist. Data published by Miller et al. (1973) and Norrls and MUler (1974) Indicate that the 96-hour LC? s for a worm, Paranals sp., a snail, Physa sp., and larvae of the mosquito, Aedes aeqyptl. would be >0.2 whereas those for the coho salmon, Qncorhynchus klsutch. and the guppy, PoeclHa retlculata. would be >1 and >10 vg/1. respectively. Based on microcosm studies In which concentrations 1n water were measured at 2-day Intervals, the 96-hour LC,. for flngerllng channel punctatus. would be >0.24 vg/a, whereas catfish, Ictalurus those for Daphnla magna and a *An understanding of the Guidelines for Deriving Numerical National Water Quality Criteria for the Protection of Aquatic Life and Its Uses (Stephan et al., 1983) Is necessary 1n order to understand the following text and tables. B-l �snail, Physa sp., would be >1.3 pg/a {Isensee and Jones, 1975; Isensee, 1978). Yocklm et al. (1978) did not observe acute toxldty to D. maqna. a snail, Helosoma sp., or the mosqultof 1sh, Gambusla af finis, exposed for over 96 hours to a measured concentration of 0.0024-0.0042 yg/fc. Helder (1980, 1981, 1982a) found that the 96-hour LC5()s for embryos of northern pike, Esox ludus. and embryos and yolk-sac fry of rainbow trout, Sal mo qalrdnerl. would be >0.01 vg/8.; the 96-hour trout would be >0.1 vg/8.. LC? for juvenile rainbow Although no 48- or 96-hour LC5Qs or EC5Qs can be calculated, the available data Indicate that those for the coho salmon, guppy, 0. maqna. and a snail, Physa sp., are >1.0 pg/i. Chronic Toxldty to Aquatic Animals No standard chronic toxldty tests have been conducted on 2,3,7,8-TCOD with aquatic animals, but several exposures that have been conducted for other purposes do provide some Information concerning chronic toxldty. Because Miller et al. (1973) used static long-term exposures, no conclusions can be drawn concerning chronic toxldty from their exposures of A. aeqyptl or a snail, Physa sp., but It can be concluded that 0.2 yg/s. would cause chronic toxldty to a worm, Paranals sp. A 96-hour exposure to an Initial concentration of 0.0056 yg/a, resulted 1n 55% mortality among coho salmon within 60 days (Miller et al., 1973, 1979); thus 0.0056 yg/8. would cause chronic toxldty to this species. S1m1lar1ly, 0.1 yg/8. would cause chronic toxldty to the guppy, because exposure to 0.1 yg/a, for 5 days killed all Individuals within 40 days (Nords and Miller, 1974). In microcosms 1n which the concentrations of 2,3,7,8-TCDO were measured at 2-day Intervals, both D. maqna and a snail, Physa sp., reproduced at 1.3 yg/a, (Isensee and Jones, 1975; Isensee, 1978). Exposure to a measured concentration of 0.0024-0.0042 yg/J, killed all exposed mosqultoflsh and channel B-2 �catfish within 20 days (Yocklm et al., 1978). Based on effects caused by 96-hour exposures, trout 1980, to and 0.01 0.001 yg/S, would cause yg/8, would 1981, 1982a). 0.1 yg/a. vg chronically affect northern pike (Helder, Branson et al. (1983) reported that a 6-hour exposure adversely Apparently 0.001 chronic toxlclty to rainbow affected rainbow trout after 64-139 days. of 2,3,7,8-TCDD/J. would cause unacceptable chronic toxlclty to rainbow trout and 0.01 yg/a. would be chronically toxic to coho salmon, mosquHoflsh, channel catfish and northern pike; 1.3 yg/a. may not be chronically toxic to D_. magna or a snail, Physa sp. Toxlclty to Aquatic Plants The few data available on the toxlclty of 2,3,7,8-TCDO to aquatic plants are also from microcosm studies. The alga, Oedoqonlum cardlacum, and the duckweed, Lemna minor, were not affected by 30-day exposures to 1.3 yg/S, and 0.71 yg/t, respectively (Isensee and Jones, 1975; Isensee, 1978). Yocklm et al. (1978) did not observe any adverse effects on 0. cardlacum exposed to a measured concentration of 0.0024-0.0042 yg/1 for 32 days. B1oaccumulat1on Several equations have been developed for predicting the steady-state bloconcentratlon factor (BCF) for an organic compound from Us octanol-water partition coefficient (Kenaga and Goring, 1980; Velth et al., 1980; Velth and Koslan, 1983). Several estimated values (Leo, 1979; Mabey et al., 1982; Neely, 1983) and one measured value (Neely, 1979, 1983; Kenaga, 1980; Branson, 1983) have been reported for the octanol-water partition coefficient for 2,3,7,8-TCDO. Use of various equations with four available values for the partition coefficient, K , results 1n the predicted BCFs shown 1n Table 1. The predicted BCFs range from 3000-68,000 using the measured value of the partition coefficient and from 7000-900,000 using the calculated values. B-3 �TABLE 1 Predicted Bloconcentratlon Factors for 2,3,7,8-TCOO Based on Estimated and Measured Values of the Octanol-Water Partition Coefficient (Kow) log Kow Equation Reference 6.15a log BCF = 0.542 log Kow + 0.124 OD * e,f log BCF = 0.76 log Kow - 0.23 Estimated Measured 6.84b 7.14C 7.28d 2,870 6.780 9,860 11,700 f 27,800 93,000 157,000 201,000 log BCF 0.79 log Kow - 0.40 = g 28,700 101,000 174,000 224.000 log BCF = 0.635 log Kow + 0.7285 e 43,000 118,000 183,000 225,000 log BCF = 0.85 log Kow - 0.70 f,g g 33,700 130,000 234,000 308.000 67.800 332,000 663,000 915,000 BCF = 0 .048 Kow Branson, 1983 Mabey et al., 1982 c Neely, 1983 d Leo, 1979 e Kenaga and Goring, 1980 f Ve1th et al., 1980 9 Ve1th and Koslan, 1983 b �Several measured BCFs have been reported for 2,3,7,8-TCDD. Using microcosm studies In which the concentrations 1n water were measured at 2-day Intervals for 30-33 days, Isensee and Jones (1975) and Isensee (1978) obtained BCFs of 390-13,000 for the alga, 0. cardlacum. a snail, Physa sp., and D_. magna. In a separate 32-day microcosm study 1n which the measured concentrations of 2,3,7,8-TCDD ranged from 0.0024-0.0042 ygA, BCFs for 0. cardlacum. Physa sp., and D_. maqna ranged from 660-7070 from the seventh day to the end of the test. In a different kind of test channel catfish were held 1n a cage In a discharge plume 1n a river for 28 days. Four 24-hour composite water sam- ples were analyzed for 2,3,7,8-TCDD. A whole-body BCF of 2,000 was obtained (U.S. EPA, 1983; Thomas, 1983). In a laboratory bloconcentratlon test rainbow trout were exposed 139-day to 0.107 depuration period. yg/a for 6 hours and followed through a The resulting projected steady-state BCF was 5450 1f growth dilution was not taken Into account, and 9270 If growth dilution was taken Into account. These values are for the whole body; the concentration of 2,3,7,8-TCDD 1n muscle was about one-half that 1n the whole body (Branson et al., 1983). Corbet et al. (1983) conducted bloconcentratlon tests on 1,3,6,8-TCDD with the fathead minnow, Plmephales promelas. and rainbow trout, using a 4-day uptake phase and 48-day depuration phase. All results were based on radioactivity measurements, because no confirmatory analyses were performed. The projected steady-state BCFs were 1061 with the fathead minnow and 469 with the rainbow trout. The authors concluded the environmental behavior of 1,3,6,8-TCDD 1s quite different from that of 2,3,7,8-TCDD, based on a 10- to 15-fold difference 1n measured clearance rate constants and the fact that the projected BCF for 1,3,6,8-TCDD was much less than a predicted BCF for 2,3,7,8-TCDO. B-5 �Information on maximum permissible tissue concentrations Is available from two sources. Hawkes and Morris (1977) found that feeding activity and growth decreased and fin erosion and liver pathology Increased when a portion of the diet fed to young rainbow trout for 105 days contained 2.3 mg of 2,3,7,8-TCDDAg of food. These effects were not observed when a portion of the diet contained 0.0023 mg/kg. A diet containing 2.3 mg of 2,3,7.8- TCDO/kg 1s obviously unacceptable for rainbow trout, but It Is not known how low the average concentration In the diet would have to be to prevent unacceptable effects on survival, growth and reproduction. The U.S. FDA Issued a human health advisory on fish containing 0.000050 mg of 2,3,7,8-TCOD/kg; the FOA believed there was little cause for concern If the average concentration 1n fish was <0.000025 mg/kg (Hayes, 1981). Other Data Because delayed effects had been observed 1n tests on 2,3,7,8-TCOO with mammals, several studies were conducted to determine whether delayed effects also occurred with fishes. that lasted <6 days caused In tests with a number of species, exposures substantial mortality several weeks later (Table 2). The bullfrog wds less sensitive to Injected 2,3,7,8-TCOO than many mammalian species (Beatty et al. 1976). Unused Data Publications, such as those by Lamparskl et al. (1979), Nlemann et al. (1983) and Ryan et al. (1983), that only dealt with analytical methodology for measuring 2,3,7,8-TCDD 1n aquatic life were not used. Publications by Baughman and Meselson (1973), Young et al. (1975), Harrison et al. (1979), Harless and Lewis (1982), O'Keefe et al. (1983), Harrison and Crews (1983), B-6 �TABLE 2 Other Data on Effects of 2,3,7,8-TCDD on Aquatic Organisms Species Duration Effect Result Reference (vg/D Coho salmon, Oncorhynchus klsutch 96 hours 50% dead 1n 56 more days 0.0056 Miller et al., 1973, 1979 Rainbow trout (embryo), Salmo galrdnerl 96 hours Some mortality In 24 weeks 0.01 Helder, 1981 Rainbow trout (yolk-sac fry), Salmo galrdnerl 96 hours All dead 1n 24 weeks 0.01 Helder, 1981, 1982a Rainbow trout (yolk-sac fry), Salmo galrdnerl 96 hours Growth retarded for 23 weeks 0.001 Helder, 1981, 1982a Rainbow trout (juvenile), Salmo galrdnerl 16 hours/day for ,4 days All dead 1n 27 days 0.1 Helder, 1981, 1982a Rainbow trout (juvenile), Salmo galrdnerl 16 hours/day for 4 days Growth reduced for 10 wc>eks 0.01 Helder, 1981, 1982a Rainbow trout, Salmo galrdnerl 6 hours F1n rot, hemorrhaglng and death after 64 days 0.1 Branson et al., 1983 Northern pike (embryo), Esox lucius 96 hours Nearly all dead 1n 23 days 0.01 Helder, 1980, 1982a Northern pike (embryo), Esox lucius 96 hours Slight reduction 1n growth up to 21 days 0.0001 Helder, 1980 CD —I �TABI £ 2 (cont.) Species Duration Effect Result Reference (pg/i) Guppy, PoeclHa retlculata 120 hours Killed 18% 10 NorMs and Miller, 1974 Guppy, Poedlla retlculata 120 hours All dead 1n 17 more days 0.1 NorMs and Miller, 1974 Guppy. PoeclHa retlculata 24 hours 10% dead 1n 41 more days 0.01 Miller et al., 1979 Bullfrog (tadpole), Rana catesbelana 50 days No deaths 1000 vg/kg d.p.) Beatty et al., 1976 Bullfrog (adult), Rana catesbelana 35 days No deaths 500 vg/kg d.p.) Beatty et al. , 1976 00 l.p. = 1ntraper1toneal �Harless et al. (1983) and Stalling et al. (1983) reported concentrations 1n aquatic organisms but did not report enough data on the concentrations 1n water to allow calculation of a BCF. Botre et al. (1978) and Ward and Matsumura (1978) only dealt with fate of 2,3,7,8-TCDD and presented no data on effects on aquatic life. The bloconcentratlon tests of Matsumura and Benezet (1973) and Matsumura (1977) were static and usually lasted for only a few days. The concentra- tion of 2,3,7,8-TCDD 1n water was not measured adequately by Tsushlmoto et al. (1982). Isensee and Jones (1975) reported BCFs based on dry weight; fortunately Isensee (1978) reported results of the same tests based on wet weight. Helder et al. (1982) exposed rainbow trout to fly ash and an extract of fly ash and no conclusions can be drawn concerning effects of 2,3,7,8-TCDD on aquatic life. The data of Zullel and Benecke (1978) were not used because the test species was not Identified well enough to allow a determination of whether It 1s resident 1n North America. Young et al. (1976, 1978), EsposHo et al. (1980), Helder (1982b), Kenaga and Norrls (1983) and a report of the National Research Council of Canada (1981) only contained data published elsewhere. Summary The data that are available concerning the effects of 2,3,7,8-TCDD on aquatic organisms and their uses do not allow the calculation of an acute or chronic toxldty value for any freshwater animal species. Data available from various studies do Indicate, however, that the acute values for several freshwater species are >1.0 yg/i. Similar data Indicate that the chronic value for rainbow trout 1s <0.001 yg/8. and that chronic values for several other species are <0.01 yg/a. the two plant species exposed to 1.3 B-9 Effects were not observed on �Estimates of the bloconcentratlon factor for 2,3,7,8-TCDO range from 3000-900,000. Measured BCFs have been reported for a variety of species and range from 390-13,000. The U.S. FDA Issued a health advisory for fish containing more than 0.000050 mg of 2,3,7,8-TCDD/kg. A concentration of 2.3 mg of 2,3,7,8-TCDD/kg 1n a portion of the diet affected rainbow trout. Exposures of <6 days resulted 1n deaths among four species of fishes several weeks later. No tests have been conducted on 2,3,7,8-TCDD with saltwater species. National Criteria Not enough data are available concerning the effects of 2,3,7,8-TCDD on aquatic life and Us uses to allow derivation of national criteria. The available Information Indicates that acute values for some freshwater animal species are >1.0 yg/S,; some chronic values are <0.01 yg/l, and the chronic value for rainbow trout Is <0.001 yg/fc. Because exposures of some species of fishes to 0.01 vg/3, for <& days resulted In substantial mortality several weeks later, derivation of aquatic life criteria for 2,3,7,8-TCDD may require special consideration. Predicted bloconcentratlon factors (BCFs) for 2,3,7,8-TCDO range from 3000-900,000, but the available measured BCFs range from 390-13,000. If the BCF Is 5000, concentrations >0.00001 yg/a. should result In concentrations In edible freshwater and saltwater fish and shellfish that exceed levels Identified 1n a U.S. FDA health advisory. If the BCF 1s >5000 or 1f uptake In a field situation 1s greater than that In laboratory tests, the value of 0.00001 yg/a. will be too high. B-10 �REFERENCES Baughman, R. and M. Meselson. 1973. An anlytlcal method for detecting TCDO (dioxin): Levels of TCDD in samples from Vietnam. Environ. Health Perspect. 5: 27-35. Beatty, P.M., H.A. Holscher and R.A. Neal. 1976. ToxIcHy of 2,3,7,8- tetrachlorodibenzo-p-dioxin In larval and adult forms of Rana catesbelana. Bull. Environ. Contam. Toxlcol. 16: 578-581. Botre, C., A. Memoll and F. Alhalque. 1978. TCOD solubilizatlon and photo- decomposition 1n aqueous solutions. Environ. Sc1. Technol. 12: 335-336. Branson, O.R. 1983. Letter to C.E. Stephan. Dow Chemical U.S.A., Midland, MI. September 20. Branson, D.R., I.T. Takahashi, W.M. Parker and G.E. Blau. 1983. Bioconcen- tratlon kinetics of 2,3,7,8-tetrachlorodlbenzo-p-dioxin in rainbow trout. Abstract. Dow Chemical U.S.A., Midland, MI. September 28. Corbet, R.L., D.C.G. Muir and G.R.B. Webster." 1983. Fate of 1,3,6,8- T COO In an outdoor aquatic system. Chemosphere. 12: 523-527. Esposito, M.P., T.O. Tlernan and PB82-136847. NTIS, Springfield, VA. B-ll F.E. Oryden. 1980. Oioxins. �Harless, R.L. and R.G. Lewis. 1982. Quantitative determination of 2,3,7,8tetrachlorodlbenzo-p-dloxln residues by gas chromatography/mass spectrometry. IJK Chlorinated 01ox1ns and Related Compounds, 0. Hutzlnger, R.W. Fre1, E. Merlan and F. Pocch1ar1, Ed. Pergamon Press, NY. p. 25-35. Harless, R.L., R.G. Lewis, A.E. Oupuy and D.D. McDanlel. 1983. Analysis for 2,3,7,8-tetrachlorod1benzo-p-d1ox1n residues 1n environmental samples, in: Human and Environmental Risks of Chlorinated D1ox1ns and Related ^ Compounds, R.E. Tucker, A.L. Young and A.P. Gray, Ed. Plenum Press, NY. p. 161-171. Harrison, D.D. and R.C. Crews. 1983. A field study of soil and biological specimens from a herbicide storage and aerial-test staging site following long-term contamination with TCDD. ITK Human and Environmental Risks of Chlorinated D1ox1ns and Related Compounds, R.E. Tucker, A.L. Young and A.P. Gray, Ed. Plenum Press, NY. p. 323-339. Harrison, O.D., C.I. Miller and R.C. Crews. 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCDD) near 1979. Residual levels of herbicide storage loading areas at Eglln AFB, Florida. AFATL-TR-79-20 or ADA078819. and NTIS, Springfield, VA. Hawkes, C.L. and L.A. Norrls. 1977. Chronic oral toxldty of 2,3,7,8- tetrachlorod1benzo-p-d1ox1n (TCDD) to rainbow trout. Trans. Am. Fish. Soc. 106: 641-645. Hayes, A.H., Jr. 1981. Letter to W.G. Mllllken. U.S. FDA, Rockvllle, MD. August 26. B-12 �Helder, T. 1980. Effects of 2,3,7,8-tetrachlorod1benzo-p-d1oxln (TCDO) on early life stages of the pike (Esox Indus L.). Scl. Total Environ. 14: 255-264. Helder, T. 1981. Effects of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCDD) on early life stages of rainbow trout (Salmo galrdnerl. Richardson). ogy. Toxicol- 19: 101-112. Helder, T. 1982a. Effects of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCDO) on early life stages of two fresh-water fish species, jn; Chlorinated Dloxlns and Related Compounds, 0. Hutzlnger, R.W. Fre1, E. Merlan and F. Pocch1ar1, Ed. Pergamon Press, NY. p. 455-462. Helder, T. fish. 1982b. Effects of TCOD on early life stages of fresh water In.: Principles for the Interpretation of the Results of Testing Procedures 1n Ecotoxlcology. EUR 7549. Commission of the European Communities on Environment and Quality of Life, Luxembourg, Belgium, p. 465-471. Helder, T., E. Stutterhelm and K. Ol1e. 1982. The toxlclty and toxic potential of fly ash from municipal Incinerators assessed by means of a fish early life stage test. Chemosphere. 11: 965-972. Isensee, A.R. 1978. B1oaccumulat1on of 2,3,7,8-tetrachlorod1benzo-para- dloxln. Ecol. Bull. (Stockholm). 27: 255-262. Isensee, A.R. and G.E. Jones. 1975. Distribution of 2,3,7,8-tetrachloro- d1benzo-p-d1ox1n (TCDO) 1n aquatic model ecosystem. 9: 668-672. B-13 Environ. Sd. Technol. �Kenaga, E.E, 1980. Correlation of bloconcentratlon factors of chemicals In aquatic and terrestrial organisms with their physical and chemical properties. Environ. Sc1. Techno!. 14: 553-556. Kenaga, E.E. and C.A.I. Goring. 1980. Relationship between water solubil- ity, soil sorptlon, octanol-water partitioning, and concentrations of chemicals 1n biota. In.: Aquatic Toxicology, J.G. Eaton, P.R. ParMsh and A.C. HendMcks, Ed. ASTM STP 707. Am. Soc. Test. Materials, Philadelphia, PA. p. 78-115. Kenaga, E.E. and L.A. NorMs. 1983. Environmental toxldty of 1COD. in: Human and Environmental Risks of Chlorinated Dloxlns and Related Compounds, R.E. Tucker, A.L. Young and A.P. Gray, Ed. Plenum Press, NY. p. 277-299. Lamparskl, L.L., T.J. Nestrlck and R.H. Stehl. 1979. Determination of part-per-tr1ll1on concentrations of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n In fish. Anal. Chem. 51: 1453-145". Leo, A.J. 1979. Letter to C. Stephan. Pomona College, Claremont, CA, May 11. Mabey, W.R., J.H. Smith, R.T. Podoll, et al. 1982. Aquatic fate processes data for organic priority pollutants. EPA 440/4-81-014. U.S. EPA MonitorIng and Data Support Division, Washington, DC. p. 107. Matsumura, F. 1977. Absorption, accumulation, and elimination of pesti- cides by aquatic organisms. Environ. Sc1. Res. B-14 10: 77-105. �Matsumura, F. and H.J. Benezet. mlcroblal degradation Health Perspect. of 1973. Studies on the bloaccumulatlon and 2,3,7,8-tetrachlorod1benzo-p-d1ox1n. 5: 253-258. Miller, R.A., L.A. Morris and C.L. Hawkes. 1973. ToxIcHy of 2,3,7,8- tetrachlorodlbenzo-p-dloxln (TCDD) In aquatic organisms. Perspect. Environ. Environ. Health 5: 177-186. MUler, R.A., L.A. Morris and B.R. Loper. 1979. The response of coho salmon and gupples to 2,3,7,8-tetrachlorod1benzo-p-d1oxln (TCDD) In water. Trans. Am. Fish. Soc. 108: 401-407. National Research Council of Canada. 1981. PolychloMnated dlbenzo-p- dloxlns: Criteria for their effects on man and his environment. NRCC No. 18574. Ottawa, Canada. Neely, W.B. 1979. Estimating rate constants for the uptake and clearance of chemicals by fish. Environ. Scl. Technol. 13: 1506-1510. Neely, W.B. 1983. Dow Chemical U.S.A., Midland, Letter to C.E. Stephan. MI, June 10. Nlemann, R.A., W.C. Brumley, D. Firestone and J.A. Sphon. 1983. Analysis of fish for 2,3,7,8-tetrachlorod1benzo-p-d1ox1n by electron capture capillary gas chromatography. Anal, Chem. 55: 1497-1504. B-15 �Morris, L.A. and R.A. Miller. 1974. The toxlclty of 2,3,7,8-tetrachloro- dlbenzo-p-dloxln (TCOD) In gupples (Poecllla retlculatus Peters). Bull. Environ. Contam. Toxlcol. 12: 76-80. O'Keefe, P., C. Meyer, D. Milker, et al. 1983. Analysis of 2,3,7,8-tetra- chlorodlbenzo-p-dloxln In Great Lakes fish. Chemosphere. 12: 325-332. Ryan, J.J., J.C. PHon, H.B.S. Conacher and D. Firestone. 1983. Inter- « laboratory study on determination of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n 1n fish. 3. Assoc. Off. Anal. Chem. 66: 700-707. Stalling, O.L., L.M. Smith, J.O. Petty, et al. 1983. Residues of poly- chlorinated d1benzo-p-d1ox1ns and dlbenzofurans 1n Laurentlan Great Lakes fish. In: Human and Environmental Risks of Chlorinated D1ox1ns and Related Compounds, R.E. Tucker, A.L. Young and A.P. Gray, Ed. Plenum Press, NY. p. 221-240. Stephan, C.E., D.I. Mount, D.J. Hansen, J.H. Gentile, G.A. Chapman and H.A. Brungs. 1983. Guidelines for deriving numerical national water quality criteria for the protection of aquatic life and Us uses. U.S. EPA, Duluth, MN, July 5. Thomas, N.A. 1983. Memorandum to C.E. Stephan, U.S. EPA, Duluth, MN, July 22. Tsushlmoto, G., F. Matsumura and R. Sago. 1982. Fate of 2,3,7,8-tetra- chlorod1benzo-p-d1ox1n (TCDD) 1n an outdoor pond and 1n model aquatic ecosystems. Environ. Toxlcol. Chem. 1: 61-68. B-16 �U.S. EPA. 1983. Dow Chemical Company - Midland Plant Wastewater Characterization Study. Preliminary summary of results. U.S. EPA Region V, Environ. Services D1v., Eastern District Office. March 28. Velth, G.D. and P. Koslan. 1983. Estimating bloconcentratlon potential from octanol/water partition coefficients. In: Physical Behavior of PCBs 1n the Great Lakes, 0. Mackay, S. Paterson, S.J. Elsenrelch and M.S. Simmons, Ed. Ann Arbor Science, Ann Arbor, MI. p. 269-282. Velth, G.D., K.J. Macek, S.R. Petrocelll and J. Carroll. 1980. An evalua- tion of using partition coefficients and water solubility to estimate bioconcentration factors for organic chemicals In fish. ogy, J.G. Eaton, P.R. Parrlsh and A.C. Hendrlcks, Ed. In: Aquatic ToxicolASTM STP 707. Am. Soc. Test. Materials, Philadelphia, PA. p. 116-129. Ward, C.T. and F. Matsumura. dloxln (TCDD) In a model 1978. Fate of 2,3,7,8-tetrachlorod1benzo-paquatic environment. Arch. Environ. Contam. Toxlcol. 7: 349-357. Yocklm, R.S., A.R. Isensee and G.E. Jones. 1978. Distribution and tox1c1ty of TCDD and 2,4,5-T In an aquatic model ecosystem. Chemosphere. 7: 215-220. Young, A.L., C.E. Thalken and W.E. Ward. 1975. Studies of the ecological Impact of repetitive aerial applications of herbicides on the ecosystem of test area C-52A, Eglln AFB, Florida. AFATL-TR-75-142 or AD-A032773. NTIS, Springfield, VA. B-17 �Young, A.L., C.E. Thalken, E.L. Arnold, J.M. Cupello and L.G. Cockerham. 1976. Fate of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCDD) In the environ- ment: Summary and decontamination recommendations. USAFA-TR-76-18 or ADA033491. NTIS, Springfield, VA. Young, A.L., 3.A. Calcagnl, C.E. Thalken and J.W. Tremblay. 1978. The toxicology, environmental fate, and human risk of herbicide orange and Us associated dloxln. OEHL-TR-78-92 or AD-A062143. NTIS, Springfield, VA. Zullel, N. and G. Benecke. 1978. the toxldty of polychlor Inated Application of a new bloassay to screen blphenyls on blue-green algae. Environ. Contam. Toxlcol. 20: 786-792. B-18 Bull. �Mammalian Toxicology and Human Health Effects EXPOSURE Hater and Soil Related The amount of human exposure of 2,3,7,8-TCDD attributable to drinking water alone cannot be readily determined. A National Academy of Sciences (NAS) document states that 2,3,7,8-TCDO has never been detected 1n drinking water using methods with limits of detection In the parts per trillion (ppt) range {NAS, 1977). The two most likely sources of 2,3,7,8-TCDD contamination are discharge of contaminated Industrial effluents, and washouts from contaminated disposal sites. However, even after contamination 2,3,7,8-TCDO should remain strongly sorbed to sediments and biota {Isensee and Jones, 1975). In one study, >90% of 2,3,7,8-TCDD In the aquatic media was present In the sorbed state (Ward and Matsumura, 1978). The possibility of 2,3,7,8TCDD leaching Into the groundwater appears remote. Helling (1971), Kearney et al. (1972) and Helling et al. (1973) found that 2,3,7,8-TCDD tended to remain on or near the surface of the soil. The mobility of 2,3,7,8-TCDD In five different soil types was examined by Kearney et al. (1973). They found that decreasing mobility of 2,3,7,8-TCDD was associated with Increasing organic content of the soil. Based on this observation, and the fact that dloxlns were relatively Immobile In all soils tested, they concluded that underground water supplies probably would not be contaminated with 2,3,7,8TCDD. Similar conclusions were made by Matsumura and Benezet (1973) who hypothesized that any movement In the soil environment would most likely occur via horizontal transfer of soil and dust particles. Nash and Beall (1980) conducted studies on the fate of 2,3,7,8-TCDD In a mlcroagroecosystem and found that 80% of the applied 2,3,7,8-TCDD remained C-l �In the upper 2 cm of soil. Trace amounts of 2,3,7,8-TCDD detected at depths of 8-15 cm suggested that some movement of dloxln Into the soil had occurred. Analysis of water leachate samples showed no detectable 2,3,7,8-TCOO following two applications (days 0 and 35) of SHvex containing 44 ppb 2,3,7,8-TCDD. However, similar analyses of Teachable samples taken 42 days after a third application of Sllvex containing 7500 ppb 2,3,7,8-TCDD Indicated a maximum concentration of 0.05-0.06 ppt of 2,3,7,8-TCDD. The downward vertical migration of 2,3,7,8-TCDD Into the first 1.5 cm of soil was reported around Seveso, Italy (DIDomenlco et al., 1980a,b). The monitoring of Seveso soil 1 year after the accident showed that the highest 2,3,7,8-TCDD levels were not present In the topmost soil layer (0.5 cm), but very often In the second (0.5-1.0 cm) or third (1.0-1.5 cm) layers. In view of the low water solubility of 2,3,7,8-TCDD, probable explanations of this vertical migration could be saturation of sorptlon sites 1n soil, solvatlon of 2,3,7,8-TCDD by organic solvents (NRCC, 1981), or blotlc mixing by earthworms or other soil Invertebrates. Nevertheless, both studies support the view that 2,3,7,8-TCDD does not migrate readily 1n soils. The photodecomposltlon of 2,3,7,8-TCDD on wet or dry soil under artificial and natural sunlight was studied by Crosby et al. (1971). The photodecomposition was found to be negligible 1n soils. Similarly, Pllmmer et al. (1973a,b) determined that photodecomposltlon of 2,3,7,8-TCDD on soils was too slow to be detected. In a later experiment, Pllmmer (1978) found that although 2,3,7,8-TCDD decomposed significantly on a precoated silica plate (-22%) 1n 8 hours of sunlight Irradiation, practically no decomposition of 2,3,7,8-TCOD was observed from 2,3,7,8-TCDD sorbed on soil under similar conditions. C-2 �The photodegradatlon of 2,3,7,8-TCOO 1n combination with other pesticide mixtures was studied by Crosby and Wong (1977). When Agent Orange containing 15 ppm of 2,3,7,8-TCDD was applied on the surface of glass plates (5 mg/cm2), rubber plant (Hevea brasH1ens1s) (6.7 mg/cma), and on the surface of sieved Sacramento loam soil (10 mg/cm2) and exposed to sunlight, 2,3,7,8-TCDD was found to photodecompose. The loss of 2,3,7,8-TCDD 1n 6 hours was >50% from the glass plates, -100% from the surface of leaves and -10% from the surface of soil. The rapid photolysis of 2,3,7,8-TCDD from these surfaces Indicates that the herbicide formulation provided a hydrogen donor which probably allowed the photolysis to occur. The authors attributed the slower photolysis of 2,3,7,8-TCDD In soil to a shading effect by the soil particles. The overall half-life of 2,3,7,8-TCDD In soil was first reported to be 1-3 years (Kearney et al., 1972). Studies performed by the U.S. A1r Force suggested that the half-life of this chemical In soils under relatively dry conditions (Utah test area) was -330 days. In more moist soils and under warm conditions (Florida test area), the half-life was -190 days. This 1s consistent with the blodegradatlon half-life of -0.5 year for 2,3,7,8-TCDD determined from the soil In rural Missouri after the accidental spraying of TCDD-contam1nated oil (IARC, 1977). Important factor also. However, physical removal may be an More recent data (Young, 1983; Wlpf and Schmld, 1983) Indicate that the half-life Is closer to 10 years. The half-life of 2,3,7,8-TCDD following an accidental 2,3,7,8-TCDD release from a trlchlorophenol manufacturing plant at Seveso, Italy, was studied by D1Domen1co et al. (1980a). The disappearance of 2,3,7,8-TCDD from the topmost soil layer after 1 year was speculated to be due to photodegradation, volatilization, or vertical movement through the soil. These C-3 �Investigators estimated the first half-life of 2,3,7,8-TCDD In soil at the time of Us release to be 5 months. One month after release, the next 2,3,7,8-TCDO half-life was estimated to be 1 year, whereas 17 months later It was estimated to be >10 years. labeled TCOO adsorbed It has recently been shown that radio- to soil becomes progressively more resistant to extraction (Phlllppl et al., 1981; Huetter and Phlllppl, 1982) and, therefore, the persistence of 2,3,7,8-TCDD residues In aged soil 1s probably greater as well. f 2,3,7,8-TCDD exhibits relatively strong resistance to mlcroblal biodegradation. Only 5 of -100 mlcroblal strains that have the ability to degrade persistent pesticides show slight ability to degrade 2,3,7,8-TCDD (U.S. EPA, 1980d).- Ward and Matsumura (1977) reported that the half-life of 2,3,7,8-TCDD In sediment-containing Wisconsin lake waters was 550-590 days. In lake water alone, -70% of the 2,3,7,8-TCDD remained after 589 days. Using an outdoor pond as a model aquatic ecosystem, Tsushlmoto et al. (1982) and Matsumura et al. (1983) estimated the apparent half-life of 2,3,7,8-TCOD to be ~1 year. Although blodegradatlon may have been responsible for part of the degradation, other Investigators (Huetter and Phlllppl, 1982) have reported the virtually complete lack of blodegradabUHy of 2,3,7,8-TCDD. The blodegradatlon half-life of 2,3,7,8-TCDD can be estimated from the theoretical rate constant values based on relative rates of transformation reported In the literature or on structure-activity analogy values given by Mabey et al. (1981). Assuming the blotransformatlon rate constant of IxlO"10 ms. cell"1 hr"1 (Mabey microorganisms et al., 1981) and the concentration of capable of degrading TCDD as 5xl05 cell ml"1 (Burns et al., 1981), the half-life of blodegradatlon 1s estimated to be >1 year. C-4 �2,3,7,8-TCOD on dry and wet soils showed negligible photodegradatlon after Irradiation with sunlamps (Crosby et a!., 1971). In order to explain the longer half-life of 2,3,7,8-TCDD 1n a model laboratory ecosystem than 1n an outdoor pond, Matsumura et al. (1983) and Tsushlmoto et al. (1982) speculated photolysis as the most likely cause. In the outdoor environment where the Intensity of sunlight was higher compared to the laboratory experiments, algae-mediated photosens1t1zat1on of 2,3,7,8-TCOD may cause some photodecomposltlon of this compound. From the available Information, 1t Is difficult to predict the fate of 2,3,7,8-TCDD 1n aquatic media under environmental photolytlc conditions. In the presence of hydrogen atom donating sub- strate(s) In surface waters, photolysis may be a significant fate process. Although several Investigators Implicated volatilization as one of the major reasons for the observed disappearance of 2,3,7,8-TCDD from aqueous solution during mlcroblal studies, little quantitative Information regarding the volatilization of 2,3,7,8-TCDD from 2,3,7,8-TCDD may undergo some water-mediated aquatic media Is available. evaporation 1n aquatic media (Matsumura et al., 1983). A transport model to estimate TCDD volatilization from a cooling pond on an Industrial site on the basis of measured concen- trations In the pond bottom sediment and pond surface area led to an estimated rate of 15-16 mg/year (Thlbodeaux, 1983). Using the formulas of Llss and Slater (1974), a vapor pressure value of *~10~6 torr (0.1 m Pa) and a solubility value of 6.2xlO~10 mole/1, NRCC (1981) calculated the vola- tilization half-life for 2,3,7,8-TCDD to be 6 minutes from water of 1 cm depth and 10 hours from water of 1 m depth. Evaporation half-life 1s directly proportional to water depth and Inversely proportional to mass transfer coefficient (Thlbodeaux, 1979). The limitations of the L1ss-Slater theory to predict the rate of volatilization have been discussed In the NRCC C-5 �(1981) document. The Llss-Slater model does not consider terrestrial matrices (suspended solids, sediments, biota, etc.) normally encountered 1n natural surface water. A computerized EXAMS model, considering sorptlon of TCDD on the suspended and bottom sediments and otherwise employing the L1ss-Slater model, gave the result that may account for the 100% of the fraction lost due to volatilization under the most favorable conditions (NRCC, 1981). The volatilization half-life for 2,3,7,8-TCDO has been estimated to be 5.5 and 12 years from pond and lake water, respectively. However, It should be remembered that these are estimated values and no experimental confirmation of these values 1s yet available. Ingestlon from Food The occurrence of 2,3,7,8-TCDD 1n food could result from (1) contamination of plant crops with 2,3,7,8-TCDD as a result of using herbicides such as SHvex and 2,4,5-T (for weed control); (2) consumption by livestock of 2,3,7,8-TCDD-contam1nated forage; or (3) magnification of residues through the food chain. Conceivably, 2,3,7,8-TCDD could also be deposited on food crops after being formed during certain combustion processes (NRCC, 1981). Galston (1979) has speculated that under certain conditions 2,3,7,8-TCDD might enter the human body from a 2,4,5-T-treated food chain and might accumulate 1n the fat and be secreted 1n the milk. Studies with either the seeds or the mature plants of soybeans or oats showed that 2,3,7,8-TCDD was neither absorbed by the seeds after spraying nor taken up from the soil Into the mature plants (Isensee and Jones, 1971; Matsumura and Benezet, 1973). However, young plants accumulated up to 40 ppb of 2,3,7,8-TCDD (Isensee and Jones, 1971). From the analysis of several parts of fruit trees and kitchen-garden plants such as carrots, onions, potatoes and narcissuses collected from the contaminated (400-1000 vq/m2 of 2,3,7,8-TCDD In soil) C-6 �Seveso area In Italy, Cocuccl et al. (1979) concluded that 2,3,7,8-TCDD 1s translocated from soil to the aerial parts of the plants, probably through the conductive vessels. This study further suggested that the plants may eliminate 2,3,7,8-TCDD by an unknown mechanism within 4-10 months after transplantation 1n unpolluted soils. However, the study of Cocucd et al. (1979) contradicts the Investigations of W1pf et al. (1982) In which vegetation samples analyzed from the Seveso area from 1976 through 1979 strongly suggested that the contamination 1n vegetation was from local dust and not from plant uptake. Unlike the Seveso Incident where release of 2,3,7,8-TCDD Into the environment took place, normal use of herbicides containing 2,3,7,8-TCDD Impurity may not cause detectable 2,3,7,8-TCDD contamination of the crop. Jensen et al. (1983) analyzed rice grain from fields 1n Arkansas, Louisiana and Texas after application of 2,4,5-T (containing 0.4 ppm 2,3,7,8-TCDD) at a maximum rate of 2.25 Ibs/acre. No 2,3,7,8-TCDD residues (detection limit 2-10 ppt) were found In these rice grains nor were any found In 30 samples of rice purchased In retail stores throughout the United States. Contamina- tion of fruits, vegetables or grains In the United States with 2,3,7,8-TCDD has never been Investigated. The presence of polychlorlnated dloxlns In the fat of cattle that had grazed on pasture treated with 2,4,5-T has been reported (U.S. EPA, 1980d). The levels of 2,3,7,8-TCDD ranged from 4-70 ppt. Other Investigators have failed to detect 2,3,7,8-TCDD (detection limit 1 ppt) 1n fats of cattle grazing on pasture or rangeland treated with normal applications of 2,4,5-T (Kocher et al., 1978). C-7 �Bovine milk collected after the accident In the Seveso area was analyzed by FanelH et al. (1980). The concentration of 2,3,7,8-TCDD was found to vary from none detected (detection limit <40 ppt) to as high as 7.9 ppb. Other Investigators have failed to detect 2,3,7,8-TCOO (detection limit 1 ppt) In surveillance samples of milk (after normal application of 2,4,5-T on pasture) from the states of Oklahoma, Arkansas and Missouri, or quarantined milk In the state of Michigan (Lamparskl et al., 1978; Mahle et al., 1977). Firestone et al. (1979) fed pentachlorophenol containing several dloxlns (not 2,3,7,8-TCDD) to lactatlng cows for 70 days. The concentration factor for 1,2,3,6,7,8-hexachlorod1benzo-p_-d1ox1n, the dloxln of highest concentration, In milk fat was -2.4 times relative to Us concentration In the diet. The analysis of human milk and urine for 2,3,7,8-TCOD has been reported. A study of 103 samples of breast milk from mothers living 1n sprayed areas In the United States revealed no 2,3,7,8-TCDD at a detection limit of 1-4 ppt (U.S. EPA, 1980d). About 6 of the 9 human milk samples analyzed by Langhorst and Shadoff (1980) may have contained 2,3,7,8-TCDD at levels slightly higher than the detection limits (0.2-0.7 ppt). Because of the lack of validation of the precision and accuracy of data, however, It was concluded that 2,3,7,8-TCDD was not present. Bumb et al. (1980) analyzed charcoal-broiled steak In order to detect any 2,3,7,8-TCDD formed as a result of the broiling process. TCOO was detected (detection limit 1-10 ppt). C-8 No 2,3,7,8- �2,3,7,8-TCDD has been reported In several species of commercial and noncommercial fish In several rivers and lakes In the United States and Canada. The levels of 2,3,7,8-TCDD In fish and shellfish as determined by various authors are given 1n Table 1. In some cases the values listed are means or composites of more than one organism, species or location. Values for Individual analyses as high as 695 ppt In fish tissue have been reported (Harless and Lewis, 1982). The efficiency of various extraction and cleanup procedures for 2,3,7,8-TCDD analysis 1n fish has been discussed by Brumley et al. (1981). Results of analyses shown In Table 1 Indicate that the 2,3,7,8-TCDD levels 1n fish and shellfish depend not only on the sites from which they are collected, but also on the type of species collected. For example, fish and shellfish collected from Bayon Meto/Arkansas River, THtabawassee/Sag1naw River, Saglnaw Bay, Lake Ontario, Lake Huron, and Cayuga Creek showed higher levels of 2,3,7,8-TCDD than those collected from Lake Erie, Lake Michigan, Lake Superior and the Atlantic Ocean. In addition, certain types of aquatic species that are bottom feeders, have high fat content or are carnivorous, such as catfish, carp, trout and salmon, showed higher levels of 2,3,7,8-TCDD than bass, bullhead or suckers. The Influence of 2,3,7,8- TCDD levels 1n surrounding water on the bloconcentratlon of fish species 1s not known. Cordle (1983) estimated the potential maximum human dally Intake of 2,3,7,8-TCDD for residents of the Great Lakes region who regularly consume Great Lakes fish. WHhln the population subset consuming these species (-17 million Individuals), daily consumption of fish tissue was 15.7 g at the 90th percentlle, and 36.8 g at the 99th percentlle. Within the smaller subset consuming pike (number of individuals not specified), dally consumption at the 99th percentlle was 83.95 g. C-9 For hypothetical mean 2,3,7,8-TCOD �TABLE 1 Levels of 2,3,7,8-TCDD 1n F1sh and Shellfish Type/Section of F1sh Sampling Site Concentration3 (PPt) Reference Edible fleshb Bayou Heto/Arkansas River 480 HHchum et al., 1980 Catfish Bayou Heto/Arkansas River ND (7)-50 HUchum et al.. 1980 Buffalo Bayou Heto/Arkansas River ND (7-13) HUchum et al., 1980 Predatorb Bayou Heto/Arkansas River 15-230 HHchum et al.. 1980 Bottom feeder'' Bayou Heto/Arkansas River 77 HUchum et al., 1980 Lake Ontario/Lake Erie/ Wei land Canal ND (<2) Josephson, 1983 Eel, Smelt and Catfish/muscle Lake Ontario/Lake Erie/ Wei land Canal 2-39 Josephson, 1983 Crayfish Bergholtz Creek, Love Canal 3.7 Smith et al., 1983b Catfish, Bass and Walleyed pike 2,4,5-T contaminated watershed 1n Arkansas and Texas; TUtabawassee and Saglnaw Rivers ND (5-10) Shadoff et al., 1977; U.S. EPA, 1980d; Buser and Rappe, 1980 Lake trout/whole body Lake Ontario 51-107 O'Keefe et al., 1983 Chinook salmon/skinless fillet Lake Ontario 26-39 O'Keefe et al., 1983 Coho salmon/skinless fillet Lake Ontario 20-26 O'Keefe et al., 1983 Rainbow trout/skinless fillet Lake Ontario 17-32 O'Keefe et al., 1983 <•> Rock bass/muscle w j o �TABLE 1 (cont.) Type/Section of F1sh Sampling Site Concentration3 (ppt) Reference O'Keefe et al., 1983 Brown trout/skinless fillet White perch/skinless fillet Lake Ontario White sucker/skinless fillet Lake Ontario Smallmouth bass/skinless fillet Lake Ontario 5.9 O'Keefe et al.. 1983 Brown bullhead/skinless fillet o Lake Ontario Lake Ontario 3.6 O'Keefe et al.. 1983 Carp, Goldfish/skinless fillet Cayuga Creek 87 O'Keefe et al.. 1983 Northern pike/skinless fillet Cayuga Creek 32 O'Keefe et al., 1983 Pumpkin seed/skinless fillet Cayuga Creek 31 O'Keefe et al., 1983 Rock bass/skinless fillet Cayuga Creek 12 O'Keefe et al., 1983 Coho salmon/skinless fillet Lake Erie 1.4-O.5 O'Keefe et al., 1983 Walleye p1ke/sk1nless fillet Lake Erie 2.6 O'Keefe et al.. 1983 Smallmouth bass/skinless fillet Lake Erie 1.6-<2.4 O'Keefe et al., 1983 Carp, Goldfish/skinless fillet Lake Erie ND (2.6) O'Keefe et al., 1983 Lake trout/whole body Lake Huron 21 O'Keefe et al., 1983 Carp/skinless fillet Lake Huron 26 O'Keefe et al., 1983 Channel Catfish/skinless fillet Lake Huron 20 O'Keefe et al., 1983 8-162 17-26 ND (3.2J-10 O'Keefe et al.. 1983 O'Keefe et al.. 1983 I �TABLE 1 (cont.) Type/Section of F1sh Sampling Site Concentration 3 (ppt) Reference Sucker/skinless fillet Lake Huron Yellow Perch/skinless fillet Lake Huron ND (8.7) O'Keefe et al., 1983 Coho salmon/skinless fillet Lake Michigan ND (3.8) O'Keefe et al., 1983 Rainbow trout/skinless fillet Lake Superior 1.0 O'Keefe et al., 1983 Perch, Sucker Saglnaw Bay Catfish Saglnaw Bay 14-37 Nlemann et al., 1983C Carp Saglnaw Bay 23-47 Nlemann et al., 1983C Catfish Bayou Meto/Arkansas River ND (3.8) Nlemann et al., 1983C Bottom feeders*5 Bayou Meto/Arkansas River ND (6.7)-12 Nlemann et al., 1983C Lake trout Lake Ontario 34-54 Nlemann et al., 1983C Rainbow trout Lake Ontario 43 Nlemann et al., 1983C Ocean haddock Atlantic Ocean Carp Lake Huron Channel catfish THtabawassee, Saglnaw and Grand Rivers 28-695 Harless and Lewis, 1982 Carp THtabawassee, Saglnaw and Grand Rivers ND (7)-153 Harless and Lewis, 1982 25 ND (3.8)-25 ND (4.6) 3-28 O'Keefe et al., 1983 Nlemann et al., 1983C Nlemann et al., 1983C Stalling et al., 1983 �TABLE 1 (cont.) Type/Section of F1sh Sampling SHe Concentration 3 (ppt) Reference Yellow perch THtabawassee and Saglnaw Rivers NO (5)-20 Harless and Lewis, 1982 Smallmouth bass Grand River 7-8 Harless and Lewis, 1982 Sucker THtabawassee River and Saglnaw Bay ND (4)-21 Harless and Lewis, 1982 Lake trout Lake Michigan NO (5) Harless and Lewis, 1982 Lake trout/whole body Lake Ontario at Burlington, Canada 61.2 Ryan et al., 1983 Rainbow trout/whole body Lake Ontario at Toronto Harbor, Canada 32.3 Ryan et al., 1983 Lake trout/whole body Lake Huron at Burnt Island, Canada 30.4 Ryan et al., 1983 Ocean haddock/fillet East Coast, Canada NO (1-10) Ryan et al., 1983 o i When not detected, the detection limit 1s Indicated within the parentheses. Organisms not further Identified 1n the report. C 0nly the GC/MS results of these authors are Included 1n tabulation These are the mean concentrations 1n samples showing detectable levels of 2,3,7,8-TCDO. ND = Not detected �residue levels of 25-100 ppt, estimates of dally Intake would thus range from 0.39-8.4 ng 2,3,7,8-TCDD/day. As shown 1n Table 1, tissue residues for species 1n certain areas do fall within this range. In order to derive an ambient water quality criterion for the protection of human health fom the harmful effects of 2,3,7,8-TCDD, 1t 1s necessary to estimate the average level of exposure of the U.S. population which would result from a particular concentration of 2,3,7,8-TCDD In ambient fresh or estuaMne waters (45 FR 79348). Data from a recent survey on fish and ^ shellfish consumption 1n the United States were analyzed by SRI International (U.S. EPA, 1980a). The results were used to estimate that the per capita consumption of freshwater and estuarlne fish and shellfish In the United States Is 6.5 g/day (Stephan, 1980). A bloconcentratlon factor (BCF) relates the concentration of a chemical In aquatic species to the concentration 1n water. Several regression equations can be used to estimate a BCF value for 2,3,7,8-TCDD from Its octanol-water partition coefficient (K ). Using three calculated values ow of K , the regression equations estimate the BCF 1n the range of 7000-900,000. Using the only measured K value, the regression equations predict a BCF for 2,3,7,8-TCDD 1n the range of 3000-68,000. The available measured BCFs, however, range from 390-13,000. The sources of the theo- retical and experimental BCF values cited here can be found 1n detail 1n Section B. Until further Information 1s available, the U.S. EPA's best current estimate for the BCF of 2,3,7,8-TCDD In aquatic organisms 1s 5000. Thus a BCF of 5000 will be used 1n the "Criterion Formulation" section to estimate the human exposure to 2,3,7,8-TCDD which would result from consumption of aquatic organisms taken from 2,3,7,8-TCDD-contamlnated waters. If the BCF Is actually >5000 or 1f uptake In a field situation 1s greater than 1n laboratory tests, human exposure will be underestimated. C-14 �Inhalation No data pertaining to the Inhalation exposure of 2,3,7,8-TCDD were found. However, the spraying of older formulations of 2,4,5-T containing 2,3,7,8-TCOO Impurity may lead to a concomitant exposure to 2,3,7,8-TCDD. Exposure could be through spray drift and through the vapor phase. From mlcroagroecosystem chamber and field studies, Nash and Beall (1980) determined the atmospheric concentration of 2,3,7,8-TCDD at various times after the application of emulsified and granular Sllvex (1.3-2.0 kg/ha Sllvex) containing 44 ppb and 7.5 ppm 2,3,7,8-TCDD Impurity, respectively. Using trltlated 2,3,7,8-TCDD, these authors found that atmospheric concentrations of 2,3,7,8-TCOD vary not only with the number of days elapsed after application (lower concentration at longer time period), but also with formulation (granular form gave lower concentration than emulslflable concentrate), and the 2,3,7,8-TCDD Impurity present 1n Sllvex (higher Impurity levels produced higher atmospheric concentrations). Depending on these variables, the atmo- spheric concentration 1n mlcroagroecosystem (10~1S g/ma), was found to vary from chambers, expressed 0.09 fg/m3 (granular 1n fg/m3 Sllvex applied at 1.3 kg/ha, concentration measured 35 days after application) to 79,800 fg/m3 (emulsified Sllvex applied at 2.0 kg/ha, concentration measured during application). Air filter samples collected from Elizabeth, NJ, after an Industrial fire on April 22, 1980, were analyzed for 2,3,7,8-TCDD by Harvan et al. (1981). Of the nine samples analyzed by coll1s1on-1nduced-d1ssodat1on mass-analyzed 1on kinetic energy spectrometry by these authors, one con- tained 20 pg of 2,3,7,8-TCDD, four contained <9 pg, and. four others probably contained 5-12 pg of unspecified TCDD Isomer which was not the 2,3,7,8-TCDO Isomer. C-15 �The atmospheric concentrations of 2,3,7,8-TCDO near two hazardous waste sites have been monitored. In one study, U.S. EPA (1982) failed to detect any 2,3,7,8-TCDD In the atmosphere (detection limit 1-20 ppt) at the Love Canal, NY, area. In another study of a waste disposal site near Jackson- ville, AR, Thlbodeaux (1983) reported an average concentration of 1100 ppt of 2,3,7,8-TCDO 1n two air partlculate samples collected near the disposal site. The levels of 2,3,7,8-TCDD 1n atmospheric dust was monitored 1n the Seveso, Italy, area between 1977 and 1979. The concentrations of 2,3,7,8- TCDD were found to be In the range of 0.06-2.1 ppb of dust using dustfall jars, and 0.17-0.50 ppb of dust by high volume sampling (D1Domen1co et al., 1980c). Another Incineration. source of atmospheric emission of polychlorlnated dloxlns Is The concentrations of TCDDs 1n fly ash from municipal Incin- erators have been studied by several authors (Elceman et al., 1979, 1980; NestMck et al., 1982; Karasek et al., 1982; Bumb et al., 1980; Buser and Bosshardt, 1978; Tlernan et al., 1982; Taylor et al., 1983). Isomer known The TCDD to be the most toxic (e.g., 2,3,7,8-TCDD) was either not detected or detected at a low level. The quantities emitted In Incinerators vary, probably because of differing efficiencies, and since few municipal Incinerators have been reliably characterized for PCDD/PCDF emissions over extended time Intervals, the data base 1s still Inadequate. Whereas Bumb et al. (1980) and Buser and Rappe (1980) detected 0.4 ng/g of 2,3,7,8-TCDD In the fly ash from a United States municipal Incinerator, the U.S. EPA concluded that emissions from five municipal waste combustors did not present a public health hazard for residents living In the Immediate vicinity (CEQ, 1981). 2,3,7,8-TCDD has been detected In the emissions of some municipal C-16 �waste Incinerators 1n Europe (Glzzl et al., 1982; Benfenatl et al., 1983; Taylor et al., 1983; OHe et al., 1982, 1983; Lustenhouwer et al., 1980; Barnes, 1983). For an Industrial boiler In the United States where penta- chlorophenol (PCP) was known to have been burned, Rappe et al. (1983) reported ~5 ppm PCDDs In the baghouse and bottom ash. However, >90% of the PCDDs were lower chlorinated congeners than octa-CDD, the expected dlmeMzatlon product of PCP. Among the large number of Isomers found, only a small amount of 2,3,7,8-TCDD could be quantified. Analyses of soot samples from a transformer fire 1n Blnghamton, NY, 1n February 1981, revealed that 2,3,7,8-TCDD (0.6 ppm) and 1,2,3,7,8-penta-CDD were the dominating Isomers of the PCDDs formed (Buser and Rappe, 1983; Rappe et al., 1983). The origin of the polychlorlnated dloxlns was probably the chlorobenzenes In the transformer oil (Buser, 1979). Analyses of wipe tests from a garage adjacent to this accident site did reveal the presence of polychlorlnated d1benzo-p_-d1ox1ns prior to the cleaning of the garage. Following the clean-up, no contamination was found (Tlernan et al., 1982; Tlernan, 1983). Dermal Dermal exposure to 2,3,7,8-TCDD 1s likely to be most significant during the spraying of 2,4,5-T. Lavy et al. (1980) determined the exposure levels of applicators spraying 2,4,5-T (ESTERON 245) during typical applications 1n a forest. The average dermal exposure to 2,4,5-T was estimated to be 0.6 mg/kg bw. If the 2,3,7,8-TCDD content 1n 2,4,5-T Is assumed to be <0.1 ppm and the absorption rate Is assumed to be the same, an exposure of 0.6 mg/kg of 2,4,5-T will correspond to <60 pg/kg bw of 2,3,7,8-TCDD for dermal exposure. Lavy et al. (1980) found a slightly lower level of 2,3,7,8-TCDD concentration (-12.5% lower) than the predicted value. No 2,3,7,8-TCDD was detected 1n any of the urine samples (detection limit 1.7 ng/i). C-17 �PHARHACOKINETICS The pharmacoklnetlcs of 2,3,7,8-TCOD has been Investigated 1n a number of laboratory animals, and there are several recent reviews on this subject (Neal et al., 1982; Gas1ew1cz et al., 1983a; Olson et al., 1983). This section will examine our current understanding of the absorption, distribution, metabolism and excretion of 2,3,7,8-TCDD 1n various mammalian species. Absorption The dermal and gastrointestinal absorption of 2,3,7,8-TCDD have been ^ Investigated 1n several species. No studies are available on the pharmaco- klnetlcs of 2,3,7,8-TCDD through the Inhalation route of exposure. Absorption From the Gastrointestinal Tract Experimentally, 2,3,7,8-TCOD 1s generally administered In the diet or by gavage 1n an oil vehicle. In Sprague-Dawley rats given a single oral dose of 1.0 wg [14C]2,3,7,8-TCDD/kg of bw, absorption from the Intestinal tract was estimated at -83% (Rose et al., 1976). WHh repeated oral dosing at 1.0 yg/kg/day (5 days/week x 7 weeks), absorption was observed to be \ approximately that observed for the single oral dose. With a much larger single oral dose, 50 vg/kg bw, -7054 of the dose was absorbed by SpragueDawley rats (Piper et al., 1973). In these studies, the chemical was administered by gavage 1n acetone:corn oil (1:25 or 1:9). One study 1n the guinea pig reported that -50% of a single oral dose (quantity not mentioned) of 2,3,7,8-TCDD 1n acetone:corn oil was absorbed (Nolan et al., 1979). The gastrointestinal absorption of 2,3,7,8-TCDD was also examined 1n the hamster, the species most resistant to the acute toxldty of this toxin (Olson et al., 1980a). Olson et al. (1980b) administered hamsters a single, sublethal, oral dose of [1,6-9H]-2,3,7,8-TCDD 1n olive oil (650 and reported that 74% of the dose was absorbed. C-18 yg/kg) When 2,3,7,8-TCDD was �administered to rats 1n the diet at 7 or 20 ppb (0.5 or 1.4 ^g/kg/day) for 42 days, 50-60% of the consumed dose was absorbed (Fries and Marrow, 1975). These findings Indicate that over a wide range of doses and under these experimental conditions, 2,3,7,8-TCDO 1s generally well absorbed from the gastrointestinal tract of the three species that have been examined. Contact with 2,3,7,8-TCDD 1n the environment would most often Involve exposure to a complex mixture containing the toxin, as opposed to the above experimental situation, where 2,3,7,8-TCDD was administered 1n the diet or through an oil vehicle. The Influence of dose and vehicle or adsorbent on gastrointestinal absorption has been Investigated 1n rats by Polger and Schlatter (1980), using hepatic concentrations 24 hours after dosing as an Indicator of the amount absorbed. They found a linear relationship between ng 2,3,7,8-TCDD administered 1n 50% ethanol (for doses of 12-280 ng, equivalent to 0.06-1.4 vg/kg) and the percentage of the dose 1n hepatic tissues (36.7-51.5%). At the next higher dose of 1070 ng, however, the percentage fell off to about 42%. Their results regarding the Influence of vehicle or adsorbent on gastrointestinal absorption have been summarized 1n Table 2. Administration of 2,3,7,8-TCDD 1n an aqueous suspension of soil resulted In a decrease In the hepatic levels of 2,3,7,8-TCDO as compared with hepatic levels resulting from administration of 2,3,7,8-TCDD 1n 50% ethanol. The extent of the decrease was directly proportional to the length of time the 2,3,7,8-TCDD had been In contact with the soil. When 2,3,7,8-TCDD was mixed In an aqueous suspension of activated carbon, absorption was almost totally eliminated (<0.07% of the dose 1n hepatic tissues). C-19 �TABLE 2 Percentage of 2,3,7,8-TCDD 1n the Liver of Rats 24 Hours After Oral Administration of 0.5 ml of Various Formulations Containing TCDD* Formulation 5054 Ethanol TCDD Dose (ng) No. of Animals Percentage of Dose 1n the Liver 7 36.7 ± 1.2 12.7, 22.9 17 24.1 ± 4.8 21.2, 22.7 10 16.0 ± 2.2 14.7 Aqueous suspension of soil (37%, w/w) that had been 1n contact with TCDD for: 10-15 hours 8 days Aqueous suspension of activated carbon (2554, w/w) 14.7 6 *Source: Polger and Schlatter, 1980 C-20 <0.07 �Ph1l1pp1 et al. (1981) and Mutter and Ph1l1pp1 (1982) have shown that radlolabeled 2,3,7,8-TCDD becomes progressively more resistant with time to extraction from soil. Similarly, the feeding of fly ash, which contains PCDOs, to rats 1n the diet for 19 days resulted In considerably lower hepatic levels of PCOOs than did the feeding of an extract of the fly ash at comparable dietary concentrations of PCDDs (van den Berg et al., 1983). The PCOOs were tentatively Identified as 2,3,7,8-TCDO, 1,2,3,7,8-PeCDO, 1,2,3,6,7,8-HxCDD and 1,2,3,7,8,9-HxCOO and the difference 1n hepatic levels noted between fly ash-treated and extract-treated rats was greater for the more highly chlorinated Isomers than H was for 2,3,7,8-TCDD. These results Indicate the Importance of the formulation or vehicle containing the tox1n(s) on the relative bloavallabllHy of 2,3,7,8-TCDO, PeCDD and HxCDDs following oral exposure. Information on the absorption of 2,3,7,8-TCDD through the skin 1s found only 1n a study by Polger and Schlatter (1980). The authors administered 26 ng 2,3,7,8-TCDO In 50 y9. methanol to the skin of six rats. hours, the liver contained 14.8^2.6% of the dose. After 24 By comparing to the hepatic levels obtained after oral administration 1n 50% ethanol (1n the same study), the amount absorbed from a dermal application can be estimated at -4054 of the amount absorbed from an equivalent oral dose. This comparison assumes that hepatic levels are valid estimates of the amount absorbed from both oral and dermal routes and that absorption from methanol 1s equivalent to absorption from 50% ethanol. As compared with dermal application 1n methanol, dermal application of 2,3,7,8-TCDD to rats In vaseline or polyethylene glycol reduced the percentage of the dose 1n hepatic tissue to 1.4 and 9.3%, respectively, but had no observable effect on the dose of 2,3,7,8- TCDO required to Induce skin lesions (~1 yg/ear) 1n the rabbit ear assay. C-21 �Application of 2,3,7,8-TCDD 1n a soil/water paste decreased hepatic 2,3,7,8TCDD to -2% of the administered dose and Increased the amount required to produce skin Application lesions to 2-3 vg In rats and In an activated carbon/water rabbits, respectively. paste essentially eliminated absorption, as measured by percent of dose In the liver, and Increased the amount of 2,3,7,8-TCDD required to produce skin lesions to -160 yg. These results suggest that the dermal absorption and acnegenlc potency of 2,3,7,8TCDD are dependent on the formulation (vehicle or adsorbent) containing the toxin. Distribution The tissue distribution of 2,3,7,8-TCDD In a number of species 1s summarized 1n Table 3. From these data It 1s apparent that 2,3,7,8-TCOO dis- tributes preferentially to the liver and adipose tissue of most species that have been examined. Piper et al. (1973) used a single oral dose of [14C]2,3,7,8-TCDD to study distribution and excretion 1n male SpragueDawley rats. Most of the radioactivity (53.2%) was excreted via the feces, but the urine and expired air accounted for 13.2 and 3.2%, respectively. Analysis of the tissues after 3 days showed liver and adipose tissue to contain the highest percent of the dose per gram of tissue, with 3.18 and 2.60%, respectively. Rose et al. (1976) also examined the distribution of [1*C]2,3,7,8-TCDD 1n the rat. Twenty-two days after a single oral dose of 1.0 yg/kg, liver and adipose tissue had retained most of the 14 C activity, with 1.26 and 1.25% of the label retained per gram of tissue, respectively. With repeated oral doses, the activity was again localized mainly 1n the liver and adipose tissue, but the liver had five times as much radioactivity as did the fat. C-22 �TABLE 3 Tissue Distribution of 2,3,7,8-TCDD Species Route of Administration Tissues with the Highest Concentration of 2,3,7,8-TCDD References Rat oral liver Fries and Marrow, 1975 Rat oral liver > fat Rose et al., 1976 Rat oral liver > fat Piper et al., 1973 Rat oral liver > fat Kodba et al., 1978 Rat oral liver > fat Allen et al., 1975 Rat 1.D. liver > fat Van Miller et al., 1976 Mouse oral liver > fat > kidney > lung Manara et al., 1982 Mouse 1.p. liver > fat > kidney > lung > spleen Manara et al., 1982 Rhesus monkey l.p. fat > skin > liver > adrenal = thymus Van Miller et al., 1976 Golden Syrian hamster 1.p. or oral liver > fat Olson et al., 1980a Guinea pig oral fat > liver > adrenals > thymus > skin Nolan et al., 1979 fat > liver > skin > adrenals Gas1ew1cz and Neal, 1979 Guinea pig NA = Not applicable �With the single oral dose, no radioactivity was detected In either the urine or expired air. Indicating that most 1f not all of the elimination of 2,3,7,8-TCDD and/or oral doses, the a Us metabolites was through the feces. With repeated *C activity was also excreted primarily through the feces, but significant amounts were found In the urine, especially of the female rats. Male rats given 1.0 yg/kg/day of 2,3,7,8-TCDD for 7 weeks excreted an average of 3.1% of the cumulative dose 1n the urine while the female rats excreted an average of 12.554 In the urine (Rose et al., 1976). Fries and Marrow (1975) have also reported evidence of sex differences 1n tissue distribution 1n rats. During 42 days of administration of 2,3,7,8- TCDD, -85% of the total body residue of male rats was located 1n the liver, while 70% of the total body residue of female rats was located In this organ. Studies performed by Van Miller et al. (1976) on rhesus monkeys and rats using single 1.p. doses of trltlated 2,3,7,8-TCDD (400 yg/kg bw) showed that while rats had over 40% of the 2,3,7,8-TCDD 1n the liver 7 days after dosing, the monkeys had only about 10% 1n the same organ at that time. In two strains of mice, the liver contained -35% of an administered dose of 2,3,7,8-TCDD 1 day after oral or 1.p. administration (Manara et al., 1982). The liver was also found to be the major site of accumulation of 2,3,7,8TCDD In the hamster, with 20% of the dose localized 1n the liver (5.3% of dose/g liver) at 3 days following a sublethal dose of 650 yg 3H-2,3,7,8TCDD/kg (Olson et al., 1980a). In all three species, 1-22 days after single-dose oral or 1.p. administration, levels of 2,3,7,8-TCDD 1n adipose tissue were generally slightly lower than levels 1n the liver, and were considerably higher than concentrations 1n other tissues (Piper et al., 1973; Rose et al., 1976; Van Miller et al., 1976; Olson et al., 1980a; Manara et al., 1982), Including the thymus (Rose et al., 1976; 1976; Olson et al., 1980a). C-24 Van Miller et al., �Koclba et al. (1978) found that female rats maintained on a dally dietary 2,3,7,8-TCDD Intake of 0.1 vg/kg/day for 2 years had an average 2,3,7,8-TCDD content of 8100 ppt 1n fat and 24,000 ppt 1n the liver. Rats given 0.01 yg/kg/day had an average of 1700 ppt of 2,3,7,8-TCDD 1n the fat and 5100 ppt In the liver. For both of these dally dosages the 11ver:body fat ratio of 2,3,7,8-TCOD was 3:1. At the lowest dose level of 0.001 yg/kg/day, both fat and liver contained an average of 540 ppt 2,3,7,8TCDD. Koclba et al. (1976) presented evidence that steady-state had been reached after <13 weeks of feeding of 2,3,7,8-TCDD. McNulty et al. (1982) reported that 2 years after administration of a single oral dose of 1 yg/kg of 2,3,7,8-TCDD to an adult rhesus macaque monkey, tissue levels of the compound were 100 ppt 1n adipose tissue and 15 ppt 1n liver. These results Indicate that prolonged retention of 2,3,7,8TCDD may occur In this species. The tissue distribution of 2,3,7,8-TCDD 1n the guinea pig appears to be similar to the monkey, with the highest concentration of the toxin being found 1n adipose tissue (Gaslewlcz and Neal, 1979; Nolan et al., 1979). The Interspecles difference 1n the tissue dis- tribution of 2,3,7,8-TCDD may be related to the relative adipose tissue content of a given species and/or the affinity of 2,3,7,8-TCDD for the hepatic mlcrosomal fraction; however, the significance of these differences remains 1n doubt. For example, the hepatotoxldty of 2,3,7,8-TCDD In a given species does not appear to be related to the hepatic concentration of the toxin (Neal et al., 1982). 2,3,7,8-TCDD has been demonstrated to be teratogenlc and fetotoxlc 1n the rat (see Teratogenldty section); the ability of 2,3,7,8-TCDD to gain access to the developing fetus of Fischer 344 rats following a single oral dose of [14C]2,3,7,8-TCDD was Investigated by Moore et al. (1976). C-25 They �found low concentrations of 2,3,7,8-TCDD 1n the fetus at gestation days 14, 18 and 21. The radioactivity appeared to be evenly distributed throughout the fetus on days 14 and 18; however, Increased levels of radioactivity were detected In fetal liver on day 21. Nau and Bass (1981) (more recently reported by Nau et al., 1982) Investigated the fetal uptake of 2,3,7,8-TCDD In NMRI mice following oral, l.p. or s.c. administration of the compound at dose levels of 5, 12.5 or 25 vg/kg 1n dlmethylsulfoxlde (DMSO):corn oil or acetone-.corn oil. The chemical was usually administered as a single dose 2 days prior to sacrifice. All three modes of administration produced similar maternal and embryonic or fetal levels of 2,3,7,8-TCDO at 5 and 12.5 vg/kg. At 25 vg/kg, higher maternal and fetal tissue levels were obtained with s.c. administration, and much higher levels were obtained with 1.p. administration, than were obtained with oral administration. Embryonic 2,3,7,8-TCDD concentrations were maximal on gestatlonal days 9 and 10; however, low levels were found 1n the embryo and fetus between gestatlonal days 11 and 18. This sharp decrease In 2,3,7,8-TCDD concentration coincides with placentatlon. 2,3,7,8-TCDD concentrations In the placenta were an order of magnitude greater than In the fetus 'Itself. The affinity of fetal liver for 2,3,7,8-TCDD was relatively low, as compared to maternal liver; however, 2,3,7,8-TCDD levels 1n fetal livers were 2-4 times higher than the levels 1n other fetal organs. An attempt was made to correlate 2,3,7,8-TCDO levels 1n the fetuses with the observed Incidence of cleft palate, but no clear relationship was observed. Autorad1ograph1c studies of tissue localization following 1.v. administration of [i«C]2,3,7,8-TCDO In DMSO to three strains of mice Indicated that the liver had the highest concentration and longest retention of radioactivity 1n the body, followed by the nasal mucosa (Appelgren et al., 1983). C-26 �In pregnant mice, the concentration of radioactivity 1n the fetuses was lower than In the dams, but a similar, selective labelling of the liver and the nasal mucosa was seen In the fetuses at day 17 of gestation. In the adult animals, labelling of the adrenal cortex was about equal to that of the liver at 1 hour after dosing, but thereafter was much lower than In the liver. Labelling of the thymus, lymph nodes, bone marrow and prostate were low at all observation times (I.e., 5 minutes to 61 days after Injection). Very few data are available on the tissue distribution of 2,3,7,8-TCDO 1n humans. Facchettl et al. (1980) reported tissue concentration of 2,3,7,8-TCDD at levels of 1-2 ng/g In liver and <0.1 ng/g 1n thyroid, brain, lung, kidney and blood 1n a woman who died 7 months after potential exposure to 2,3,7,8-TCDD from the Seveso accident. This pattern of 2,3,7,8-TCDD distribution, however, may not be representative for humans since the woman at the time of death had an adenocarclnoma (which was not considered related to the accident) Involving the pancreas, liver and lung. In addition Young et al. (1983) reported preliminary results of the analysis of adipose tissue from soldiers exposed to Agent Orange. Two analyses were performed, one using the exact mass of 321.8936 and the other the signal profile at masses 321.8936 and 319.8965. Three groups were studied consisting of 20 veterans claiming health problems related to Agent Orange exposure, 3 A1r Force officers with known heavy exposure to Agent Orange during disposal operations, and 10 control veterans with no known herbicide exposure. In the first group, 10 of the 20 had measurable levels of 2,3,7,8-TCDD (5 with 5-7 ppt, 3 with 9-13 ppt and 1 with 23 and 35 ppt and another with 63 and 99 ppt). In the second group only two officers had measurable 2,3,7,8-TCDD levels and these did not exceed 3 ppt. In the 10 control veterans, 4 had 2,3,7,8-TCDD levels between 7 and 14 ppt. Levels of C-27 �2,3,7,8-TCDD 1n adipose tissue did not appear to be associated In this study with 111 health or any particular symptom. However, H was considered that Information on background levels of 2,3,7,8-TCDD In adipose tissue was too limited to draw any firm conclusions. Metabolism Vlnopal and Caslda (1973) found no evidence of water soluble metabolites of 2,3,7,8-TCDD following Incubation with mammalian liver mlcrosomes or l.p. Injection Into mice. In the same experiment, only unmetabollzed 2,3,7,8- TCDD was extractable from mouse liver 11-20 days after treatment. Van Miller et al. (1976) claimed that the slow elimination of 2,3,7,8-TCDD they observed In both rats and monkeys after l.p. Injections suggested that 2,3,7,8-TCDD was not readily metabolized. Metabolites of 2,3,7,8-TCDD have been detected In the bile and urine of Syrian Golden hamsters after single oral or 1.p. doses (Olson et al., 1980a) and 1n the bile of dogs following repeated direct Introduction of the chemical Into the duodenal lumen (Polger et al., 1982a). Polger and Schlatter (1979), Ramsey et al. (1979) and Ramsey et al. (1982) demonstrated biliary excretion of several [14C]2,3,7,8-TCDD by rats after repeated oral dosing. metabolites of The metabolites were tentatively Identified as glucuronldes of hydroxylated 2,3,7,8-TCDD. The amounts of metabolites found were small, Indicating that 2,3,7,8-TCDD 1s only slowly metabolized In the liver. Previous work by Piper et al. (1973) using single oral doses of 2,3,7,8-TCDD concluded that, since small amounts of radioactivity were found 1n the urine and expired air of male rats during the first 10 days, metabolic alteration or breakdown must occur. by Rose et al. (1976) using oral doses stated that while the The study 14 C activity In the rat livers appeared to be present as unchanged 2,3,7,8-TCDD, a C-28 �significant amount of radioactivity found 1n the feces appeared to come from substances other than 2,3,7,8-TCOO; the excretion of 14 C 1n the urine also Indicated that metabolism had occurred. Polger et al. (1982b) Investigated the toxlclty of 2,3,7,8-TCDD metabolites by administering extracts of bile from 2,3,7,8-TCDO-treated dogs to male guinea pigs In single oral doses equivalent to 0.6, 6.0 and 60 yg/kg of parent compound. Other groups of guinea pigs received bile extract from untreated dogs or 2,3,7,8-TCDD Itself. A comparison of the mortality data at 5 weeks after dosing Indicated that the acute toxldty of 2,3,7,8-TCDD to guinea pigs was at least 100 times higher than was the acute toxlclty of Us metabolites. More recently, Olson et al. (1983) reported that all of the radioactivity 1n urine and bile from 14 C-2,3,7,8-TCDD-treated rats, hamsters and guinea pigs corresponded to metabolites of 2,3,7,8-TCDD. The enzymatic hydrolysis of the 2,3,7,8-TCDD metabolites from the rat and hamster altered the chromatographlc profile of the metabolites, Indicating the presence of glucuronlde conjugates In bile and sulfate conjugates Blttner, 1983). 1n urine (Olson and The apparent absence of these metabolites In extracts of hamster and rat liver suggest that once formed, the metabolites of 2,3,7,8TCOD are readily excreted (Olson et al., 1980a; Rose et al., 1976). These results also Indicate that urinary and biliary elimination of 2,3,7,8-TCDD 1s dependent upon metabolism of the toxin. Although urine and bile appear to be free of unmetabollzed TCDD, data from the hamster and rat Indicate that from 10 to 40% of the 2,3,7,8-TCDD-deMved radioactivity 1n feces represents unchanged 2,3,7,8-TCDD (Olson et al., 1983; Olson and BHtner, 1983). The dally presence of unchanged 2,3,7,8-TCDD 1n feces and absence In bile suggests that direct Intestinal elimination may be the C-29 Us �source for the fecal excretion of 2,3,7,8-TCDD. This finding demonstrates that the half-life for elimination of 2,3,7,8-TCDO may not directly reflect the in vivo rate of 2,3,7,8-TCOD metabolism 1n a given animal. Nevertheless, the metabolism of 2,3,7,8-TCDO does 1n part regulate Us elimination or relative persistence 1n a given animal. Several metabolites of 2,3,7,8-TCDD have recently been Identified. Sawahata et al. (1982) Investigated the in vitro metabolism of 2,3,7,8-TCDD 1n Isolated rat hepatocytes. The major product was deconjugated with B-g1ucur«n1dase, deMvatlzed with dlazomethane, and separated Into two compounds by high performance liquid chromatography (HPLC). These metabolites were subsequently Identified as l-hydroxy-2,3,7,8-TCDO and 8-hydroxy-2,3,7tr1chlorod1benzo-p_-d1ox1n. Polger et al. (1982b) Identified six metabolites In the bile of dogs that were given a lethal dose of [3H]2,3,7,8-TCDD. The major metabolite was 1,3,7,8-tetrachloro-2-hydroxyd1benzo-p_-d1ox1n; 2,3,7,8tr1chlor-3-hydroxyd1benzo-p_-d1ox1n and 1,2-d1chloro-4,5-hydroxybenzene were also Identified as minor metabolites. The structures of the three remaining metabolites were not determined; however, two appeared to be trlchlorohydroxyd1benzo-£-d1ox1ns and the third was apparently a chlorinated 2-hydroxyd1phenyl ether. Data on the metabolism of 2,3,7,8-TCDO suggests that reactive epoxlde Intermediates may be formed. Poland and Glover (1979) have Investigated the i§ vivo binding of [1,6-3H]-2,3,7,8-TCDD derived radioactivity to rat hepatic macromolecules. They found maximum levels equivalent to 60 pmol 2,3,7,8-TCDO/mole of amlno adds 1n protein, 12 pmol 2,3,7,8-TCOD/mole of nucleotlde In rRNA, and 6 pmol of 2,3,7,8-TCDD/mole of nucleotlde In DNA. This corresponds to one 2,3,7,8-TCOD-DNA adduct/35 cells. Poland and Glover (1979) suggest that 1t 1s unlikely that 2,3,7,8-TCDD-1nduced oncogenesls 1s through a mechanism of covalent binding to DNA and somatic mutation. C-30 �Further studies In other species, possibly with [14C]-2,3,7,8-TCOD, are needed to confirm these results and assess the relationship between covalent binding and the short and long-term toxldty of 2,3,7,8-TCDD. Isolated rat hepatocytes In suspension have been used as an \n vitro system for assessing 2,3,7,8-TCDD metabolism under various conditions (Olson et al., 1981). Data Indicate that the rate of 2,3,7,8-TCDD metabolism In rat hepatocytes correlates directly with drug Induced changes 1n hepatic cytochrome P-450 monooxygenase activity, suggesting that 2,3,7,8-TCDD 1s metabolized by this enzyme (Neal et al., 1982). Pretreatment of rats with 2,3,7,8-TCDD has been shown to enhance the rate of 2,3,7,8-TCDD metabolism 1n Isolated hepatocytes, demonstrating that 2,3,7,8-TCDD can Induce Us own rate of metabolism. Beatty et al (1978) also found a correlation between hepatic mixed-function oxldase (MFO) activity and the toxldty of 2,3,7,8TCDD 1n rats. In both naturally occurring age and sex-related differences In MFO activity, and following administration of Inducers and Inhibitors of MFO enzyme systems, hepatic MFO activity was directly correlated with the 20-day LD5Q. Olson and BHtner (1983) reported that the rate of 2,3,7,8-TCDO metabolite formation Jjn vitro was higher In hepatocytes from the hamster than In hepatocytes from the rat. Qualitative evaluation of J_n vivo and 1_n vitro metabolites by HPLC also suggested significant Interspedes variability. The authors suggested that such differences 1n metabolism may partially explain the differences 1n toxldty among spedes. Excretion The following discussion assumes process. that elimination Is a first order WHh the exception of the guinea pig, which may follow zero order kinetics (Gaslewlcz and Neal, 1979), elimination data yield a straight line C-31 �on a semi logarithmic plot. Indicating that elimination Is a single, first order process. H1les and Bruce (1976) have pointed out that the studies of Allen et al. (1975) and Piper et al. (1973) can be Interpreted equally well by either zero or first order kinetics. The majority of the data, however, seem to support the assumption of a first order elimination process. The excretion of 2,3,7,8-TCDD and Us metabolites has been Investigated 1n a number of species. Table 4 summarizes results on the elimination of 2,3,7,8-TCDO-deMved radioactivity, following a single exposure to 3H- or [14C]-2,3,7,8-TCDD. These studies show that 2,3,7,8-TCDD was slowly excreted from the bodies of all species tested, with a half-life In the body of 10-43 days. In the Syrian Golden hamster, the least sensitive mammalian species to the acute toxldty of 2,3,7,8-TCDD, excretion occurred readily through both the urine (3554 of administered dose, 41% of total excreted radioactivity) and feces (50% of the administered dose, 59% of total excreted radioactivity) (Olson et al., 1980b; Gaslewlcz et al., 1983a). The high levels found 1n the urine of Infant monkeys were probably due to the Incomplete separation of urine and feces (Van HUler et al., 1976). In all the other species tested so far, excretion occurred mainly through the feces (80-100% of total urinary and fecal radioactivity) with only minor amounts of 2,3,7,8-TCDD metabolites found 1n the urine (Piper et al., 1973; Allen et al., 1975; Rose et al., 1976; Gaslewlcz and Neal, 1979). Only Piper et al. (1973) reported the excretion of metabolites In the expired air. During 21 days following administration of a single oral dose of [14C]2,3,7,8-TCOD to rats, 3.2% of the administered radioactivity (4.6% of the excreted radioactivity) was recovered 1n the expired air. C-32 �TABLE 4 Elimination of 2,3,7,8-TCDD Species Single Treatment vg/kg (route) Half-Life for Elimination (days) Relative % of TCOO-Oer1ved Radioactivity Reference Feces Urine 6.0 Gas1ew1cz and Neal, 1979 Guinea pig 94.0 1.45 (oral) 22 - 43 NT NT Nolan et al., 1979 Rat 03 30.2 + 5.8 Guinea pig c-> 2 (1.p.) 1.0 (oral) 31+6 >99 <1 Rose et al., 1976 20.0 Piper et al., 1973 Rat 50 (oral) 17.4 + 5.6 80.0 Rat 50 (oral) 21.3 + 2.9 95.5 4.5 Allen et al., 1975 Rat 400 (i.p.) NT 91.0 9.0 Van Miller et al., 1976 Monkey (adult) 400 (I.p.) NT 78.0 22.0 Van Miller et al., 1976 Monkey (Infant) 400 (1.p.) NT 39.0 61.0 Van Miller et al., 1976 Mouse C57BL/65 DBA/2J B6D2F-,/J* 10 (1.p.) 10 (I.p.) 10 (1.p.) 11.0 + 1.2 24.4 f 1.0 12.6 + 0.8 72.0 54.0 72.0 28.0 46.0 28.0 Gas1ew1cz et al., 1983a,b Gas1ew1cz et al., 1983a,b Gas1ew1cz et al., 1983a,b Hamster 650 (1.p.) 10.8 + 2.4 59.0 41.0 Olson et al., 1980a Hamster 650 (oral) 15.0 + 2.5 NT NT Olson et al., 1980a *OffspMng of C57BL/6J and DBA/2J which are heterozygous at the Ah locus NT = Not tested �Rose et al. (1976) Investigated the elimination of [1*C]2,3,7,8-TCDD In rats given repeated oral doses of 0.01, 0.1 or 1.0 yg/kg/day Monday through Friday for 7 weeks, or a single dose of 1.0 v9/kg. In the singledose study, no **C was excreted 1n the urine or expired air; 1n the repeated-dose study, however, 3-18X of the cumulative dose was excreted 1n the urine by 7 weeks. This study Indicated that steady-state concentrations will be reached In the bodies of rats 1n -13 weeks. The rate constant defining the approach to steady-state concentrations was Independent of the dosage of 2,3,7,8-TCOO over the range studied. This Is consistent with the observations of Fries and Marrow (1975) who found that the total retention In the bodies of rats was proportional to total Intake. When rats were maintained on a diet containing either 7 or 20 ppb 2,3,7,8-TCOO, the amount of 2,3,7,8-TCDD retained In the body was 5.5 times the dally Intake of 2,3,7,8-TCDD at 14 days, 7.5 times the dally Intake at 28 days, and 10.0 times the dally Intake at 42 days. The data 1n Table 4 suggest some Interspedes differences 1n the halflife for elimination (t 1/2) of 2,3,7,8-TCOO. In the hamster, the least sensitive species to the acute toxldty of 2,3,7,8-TCDD, a mean t 1/2 of 10.8 days was observed (Olson et al., 1980a,b), and 1n the guinea pig, the most sensitive species to the acute toxldty of 2,3,7,8-TCDD, the mean t 1/2 was 30.2 days (Gas1ew1cz and Neal, 1979). The observed Interspedes differences 1n the t 1/2 of 2,3,7,8-TCDD may 1n part be related to the relative sensitivity of a given species to the acute toxldty of 2,3,7,8-TCOD. The Intraspedes differences 1n the t 1/2 of 2,3,7,8-TCOO 1n three mouse strains may be due to the finding that the DBA/2J strain possesses greater adipose tissue stores (Gaslewlcz et al., 1983b). than ~2-fold the C57BL/6J and B6D2F1/J strains The sequestering of the UpophlUc toxin In C-34 �adipose tissue stores of the DBA/20 mouse may contribute to the greater persistence of 2,3,7,8-TCDD 1n this strain. In all of the rat studies shown In Table 4, urinary and fecal elimination were monitored for a period of only 20-22 days, and from these data H was assumed that elimination followed a single component, first order kinetic model. Recently, Olson and BHtner (1983) examined the elimination of 2,3,7,8-TCOD-derlved radioactivity 1n rats over a 35-day period following a single l.p. exposure at 1 wg aH-2,3,7,8-TCOO/kg. They observed first order kinetics for elimination, with a fast component having a t 1/2 of 7 days (representing 13% of total elimination) and a slow component having a t 1/2 of 75 days (87% of total). The second, slow component for elimination was evident only when urinary and fecal elimination were monitored for >30 days. This study suggests that 2,3,7,8-TCDD may be more persistent than earlier studies suggested. A preliminary study In the rhesus monkey Indi- cates that 2,3,7,8-TCDD may be exceptionally persistent 1n adipose tissue. McNulty et al. (1982) estimated the apparent half-life of 2,3,7,8-TCDD 1n the fat of a monkey to be ~1 year. Studies 1n the rat, guinea pig, hamster and mouse have found that all of the 2,3,7,8-TCDD derived radioactivity excreted In the urine and bile corresponds to metabolites of 2,3,7,8-TCDD (Olson et al., 1983). The apparent absence of 2,3,7,8-TCDD metabolites In liver and fat suggests that, once formed, the metabolites of 2,3,7,8-TCDD are readily excreted. Thus, urinary and biliary elimination of 2,3,7,8-TCDD Is dependent upon metabolism of the toxin. Although urine and bile appear to be free of unmetabollzed 2,3,7,8TCDD, data from the hamster and rat Indicate that a significant amount (10-40%) of unchanged 2,3,7,8-TCDD may be excreted Into the feces (Olson et al., 1983). Unmetabollzed 2,3,7,8-TCDD thus appears to enter the Intestinal C-35 �lumen by some route other than bile (direct Intestinal elimination) for a number of days following treatment. Studies In lactatlng rats have also found that unchanged 2,3,7,8-TCOD may be excreted 1n the milk of lactatlng animals (Moore et al., 1976; Luder et al., 1975). Lactation, direct Intestinal elimination, and perhaps sebum may serve as routes for excretion of 2,3,7,8-TCDO, which are not dependent upon metabolism of the toxin. These data suggest that the in vivo half-life for elimination of 2,3,7,8-TCOD may not directly reflect the rate of 2,3,7,8-TCDD metabolism 1n a given animal (Neal et al., 1982). EFFECTS Acute. Subacute and Chronic Toxldty The acute LD5Q for 2,3,7,8-TCDO Table 5. The oral LD 1n several species 1s shown 1n values range from 0.6 yg/kg bw for guinea pigs to 5051 yg/kg bw for hamsters (Schwetz et al., 1973; Vos et al., 1974; McConnell et al., 1978a,b; Henck et al., 1981; Olson et al., 1980b). The dermal LD for rabbits was 275 yg/kg of body weight (Schwetz et al., 1973); death was sometimes delayed as long as 40 days following acute exposure. Of the laboratory animals studied, the guinea pig was the most susceptible to the toxic effects of 2,3,/,8-TCDO (Schwetz et al., 1973; Gupta et al., 1973; Grelg et al., 1973). The acute toxldty has also been found to vary with the sex, age and strain of the test animal. Schwetz et al. (1973) found male Sherman rats more sensitive to 2,3,7,8-TCDD than females, while Beatty et al. (1978) found female Sprague-Dawley rats more sensitive than adult male rats. Thus, no general sex difference 1s apparent 1n the rat, perhaps due to strain differences 1n sensitivity to 2,3,7,8-TCDD. A significant sex difference was observed 1n the C57BL/10 mouse, with the oral LD5 C-36 In females being 3-fold �TABLE 5 Lethalltjr of 2,3.7,8-TCDD Following Acute Exposure Duration of Observation (yg/kg) NR 2 -8 weeks 0.6 (0.4-0.9)* time to death was 5-34 days, the 2,3,7.8-TCDD was 91X pure Schwetz et al.. 1973 gavage/corn o1lacetone (9:1) NR 2-8 weeks 2.1 (1.5-3)* time to death was 9-42 days, the 2,3,7,8-TCDD was 99X pure Schwetz et al., 1973 H/9 gavage/corn oil NR 30 days 2 median time to death was 17-20 days, marked weight loss, thymus atrophy. Intestinal hemorrhage, no porphyMa and only mild liver Injury HcConnell et al.. 1978a Rats/Sherman M/5-10 gavage/corn ollacetone (9:1 ) B 16 32 63 2-8 weeks 22 time to death was 9-27 days, the 2,3.7,8-TCDD was 91X pure Schwetz et al.. 1973 Rats/Sherman F/NR gavage/corn oilacetone (9:1) NR 2-8 weeks time to death was 13-43 days, the 2,3.7.8-TCOD was 91X pure Schwetz et al.. 1973 Rats/SpragueOawley M/6 1. p. /olive oil NR 20 days 60 t-Dso (yg/kg, mean +_ S£) adult male. 60.2 *. 7.8; weanling male, 25.2 + 1.4 Beatty et al., 1978 Rats/SpragueDawley F/6 1 .p. /olive oil NR 20 days 25 LDso (yg/kg, mean * SE) adult female, 24.6 + 2.0 Beatty et al., 1978 Monkey/rhesus F/3 gavage/corn oil 0 70 350 >35 days <70 weight loss, edema, severe thymus atrophy, loss of hair, mild liver damage McConnell et al.. 1978b H1ce/C57Bl M/14 gavage/corn oilacetone (9:1) 0 100 150 200 60 days 114 time to death In the high dose group was 15-20 days, body weight loss, edema 1n 25X of treated animals. severe thymlc and spleen atrophy. hemorrhage In the region of the eye and small Intestine, liver necrosis 1n the centrllobular region Vos et al. , 1974 Species/Strain Sex/No./ Group Guinea pigs/ Hartley M/NR gavage/corn oilacetone (9:1) Guinea pigs/ Hartley H/NR Guinea pigs/ Hartley Route/Vehicle i 03 Dose Tested (yg/kg) LD50 45 (30-66)* Comments Reference �TABLE 5 (cont.) Species/Strain Sex/Mo./ Group M1ce/C57Bl H/9 Route/Vehicle gavage/corn oil Dose Tested (Mg/kg) NR Duration of Observation 30 days 1050 (yg/kg) 283.7 Comments median tine to death was 22-25 days. dose-related body weight loss, thynlc atrophy. Increased liver weight and Reference NcConnell et al.. 1978a porphyrla. gross and historic liver alterations, subcutaneous edema, Intestinal hemorrhage HUC/C57BL/10 M/5 gavage/arachls oil H1ce/C57BL/10 F/S gavage/arachls oil N1ce/CS7BL/6J H/NR 1. p. /olive oil c~> i CO 00 85 107 135 170 213 45 days 146 85 107 135 170 213 269 330 426 536 45 days >450 NR 30 days 95X confidence Units of 111-211 132 Smith et al.. 1981 ug/kg. Host deaths occurred from 22-26 days after dosing. Signs of porphyrla, edema, hemorrhage. 1 of 4 animals died at dose of 426 ug/kg Smith et al.. 1981 Gaslewlcz et al. , 1983a.b Mlce/OBA/2J M/NR 1. p. /olive oil NR 30 days 620 M1ce/B6D2F1/J N/NR 1 .p. /olive oil NR 30 days 300 Rabbits/ MfcF/NR gavage/corn oil- NR 2-8 weeks New Zealand Rabbits/ New Zealand acetone (9:1) H&F/S 1 .p. /corn oil 32 63 126 252 500 4 weeks 115 (38-345)* NR Gaslewlcz et al.. 1983a.b BGD2Fj/J mice are the offspring of C57BL/6J and DBA/2J and are heterozygous at the Ah locus. Gaslewlcz et al. , 1983a.b time to death was 6-39 days, the Schwetz et al.. 1973 2,3,7.8-TCOD was 91X pure time to death was 6-23 days. 2-3 animals/group died In all but the low exposure group Schwetz et al.. 1973 �TABLE 5 (cont.) Species/Strain Sex/No./ Group Rabbits/ New Zealand H&F/NR dermal/acetone Hamster/ golden Syrian M/6 gavage/corn oilacetone (9:1) Hamster/ golden Syrian M&F/S-6 Hamster/ golden Syrian Route/Vehicle Dose Tested (vg/kg) 31.6 63 126 252 500 Duration of Observation 1.050 (vg/kg) Comments Reference 3 weeks 275 (H2-531) time to death was 12-22 days Schwetz et al., 1973 0 300 600 1000 3000 6000 55 days 5051 (387618,487; 95X confidence) time to death was 26-43 days, the liver and thymus appeared to be the primary target organs, only 1 death occurred 1n the 300 and 3000 tig/kg group Henck et al., 1981 1. p. /olive oil 0 500 1000 2000 3000 50 days >3000 significant dose-related decrease 1n thymus weight starting at 500 t>g/kg. only 2 deaths occurred out of 11 hamsters In the 3000 vg/kg group. Olson et al.. 1980b M/5 gavage/ollve oil 500 1000 2000 3000 50 days 1157 death generally occurred between 24 and 45 days, decrease 1n body weight above 2000 gg/kg, prollferatlve 11e1t1s with mild to severe Inflammation Olson et al.. 1980b Dogs/beagle M/2 gavage/corn oilacetone (9:1) 3000 2-8 weeks NA all animals died Schwetz et al.. 1973 Dogs/beagle f/2 gavage/corn oilacetone (9:1) 30 100 2-8 weeks NA all animals survived Schwetz et al., 1973 c~> i CO 1 0 *The number 1n parentheses appears to Indicate the range of lethal doses; however, the studies did not specify what these numbers represented. NA = Not applicable; NR = not reported �greater than that In males (Smith et al., 1981). In a study of age-related differences, Beatty et al. (1978) reported weanling male rats to have an acute ID of 25 pg/kg 1n contrast to the value of 60 pg/kg 1n adult males. Vos et al. (1974) found 0, 17 and 44% mortality 1n mice of 4, 2 and 1 months of age, respectively, following 4 weekly doses of 25 pg 2,3,7,8TCDD/kg. These limited studies suggest that young animals may be more susceptible to the acute toxlclty of 2,3,7,8-TCOO. Various strains of mice have been used to study the mechanism of action of 2,3,7,8-TCDD, based on the ability of the toxin to Induce enzymes that have been shown to segregate with a single genetic locus, the Ah locus (Poland et al., 1974, 1976a,b). The "non-responsive" strains (e.g., DBA/2J) appear to be less responsive to enzyme 1nduct1en due to an altered receptor with lower affinity for 2,3,7,8-TCOO, 1n comparison to the "responsive" strains (e.g., C57BL/6J). Gaslewlcz et al. (1983b) reported that the "responsive" C57BL/6J mice have an acute LD for 2,3,7,8-TCDD of 132 ^g/kg, compared with an LD 620 pg/kg In the "non-responsive" DBA/2J mice. An Intermediate LD of of 300 pg/kg was also reported for B6D2F./J mice, which are offspring of C578L/6J and DBA/2J (B6D2F,/J mice are heterozygous These results suggest that the acute LDg at the Ah locus). for 2,3,7,8-TCDD varies with the strain of mouse and the relative activity "responsiveness" at the Ah locus. The hepatotoxldty of 2,3,7,8-TCDD 1s well established, especially 1n rats, mice and rabbits where the hepatic lesions are particularly severe (Mllnes, 1971). Sublethal doses of 2,3,7,8-TCDD 1n rats produced signifi- cant Hver damage, characterized by fatty changes, centMlobular necrosis (Cunningham and Williams, 1972), megalocytosls, and unusual numbers of multlnucleated giant hepatocytes (Gupta et al., 1973). A single dose of 0.1 pg/kg In rats produced Increased liver weights (Harris et al., 1973). C-40 �In rats given single doses of 5 and 25 yg/kg 2,3,7,8-TCOO, Fowler et al. (1973) reported extensive proliferation of the smooth and rough endoplasmlc retlculum, especially near the bile ducts. Twenty-eight days after dosing, the electron micrographs of the livers were Indistinguishable from controls. Similar results were observed by Jones and Butler (1974) and Jones and Grelg (1975). In mice, exposure to 1-10 vg/kg 2,3,7,8-TCDD/day produced liver damage as Indicated by elevated SGOT, SGPT, serum LDH, alkaline phosphatase and blllrubln levels (Z1nkl et al., 1973). A number of toxic responses have been observed following exposure to 2,3,7,8-TCDO and these have been summarized for a number of species 1n Table 6. 2,3,7,8-TCOO tox1c1ty exhibits marked Interspecles variability, with some responses being highly species specific and confined to one or a few species. Loss of body weight or reduced weight gain and thymlc atrophy are the most consistent toxic responses of 2,3,7,8-TCDD exposure 1n various species, with the latter being one of the most sensitive Indicators of toxldty. In general, the toxlcologlc pattern observed with 2,3,7,8-TCOO Is not unique; It also occurs with certain halogenated dlbenzofurans, chlorinated blphenyls, naphthalenes, and bromlnated dloxins (McConnell, 1980). An extended period was observed between treatment and death. During this period the animals had poor weight gain or loss of weight and appeared to be "wasting away". At death, loss 1n body weight was reported to be as great as 50% for some species (McConnell, 1980). In female Wlstar rats Intubated with 2,3,7,8-TCDO at a dose of 100 yg/kg, the weight loss was blphaslc (Courtney et al., 1978). The Initial weight loss occurred rapidly during the first 7-10 days after treatment and was associated with decreased food and water consumption. This Initial phase of weight loss was reversed with the resumption of normal food Intake for 4 or 5 days, only to be C-41 �TABLE 6 Toxic Responses Following Exposure to 2,3,7,8-TCDD: Species Differences3 Monkey Guinea P1g Cowb Rat Mouse RabMt b Ch1ckenb Hamster Hyperplasla and/or metaplasia Gastric mucus Intestinal mucosa Urinary tract B1le duct and/or gall bladder Lung: focal alveolar Skin rs i Hypoplasla, Atrophy, or Necrosis Thymus Bone marrow Testicle Other Liver lesions PorphyMa Edema 0 0 0 0 0 0 0 + 0 0 0 0 0 ft + References: Monkey (McConnell et al., 1978a; Norback and Allen, 1973; Allen et al. t 1977); Guinea pig (McConnell et al., 1978a; McConnell, 1980; Moore et al., 1979; Turner and Collins, 1983); Cow (McConnell, 1980); Rat (McConnell, 1980; Kodba et al., 1978, 1979); Mouse (Schwetz et al., 1973; McConnell et al., 1978a; Vos et al., 1973); Rabbit (K1mm1g and Schultz, 1957; Schwetz et al., 1973; Vos and Beems, 1971); Chicken (Schwetz et al., 1973; Norback and Allen, 1973; Allen and Lallch, 1962; Vos and Koeman, 1970); Hamster (Olson et al., 1980b; Henck et al., 1981) Responses followed exposure to 2,3,7,8-TCDD or structurally related chlorinated aromatic hydrocarbons. c Symbols: 0, lesion not observed; +, lesion observed (number of H +" denote severity); i, lesion observed to a very limited extent; blank, no evidence reported 1n literature. Skin lesions 1n cattle are observed, but they differ from the skin lesions observed In other species. Source: Adapted from Poland and Knutson, 1982 �followed by a second, more gradual, decline 1n food and water Intake and weight until death. Providing animals with an adequately nutritious liquid diet by Intubation did not appreciably alter the pattern of weight loss nor affect survival. In contrast, Gaslewlcz et al. (1980) observed that providing rats with total parenteral nutrition would prevent some of the weight loss Induced by 2,3,7,8-TCDD; however, there was no protection from the lethal effects of 2,3,7,8-TCDD. Also, severe thymlc atrophy has been universally observed 1n all species given lethal doses of 2,3,7,8-TCDD, and since weight loss and thymlc atrophy are both associated with malnutrition, van Logten et al. (1981) Investigated the effects of dietary protein on the toxldty of 2,3,7,8-TCDD. Groups of female Fischer 344 rats administered 2,3,7,8-TCDD (20 wg/kg) and maintained on low (3.5%). normal (26%) or high (55%) protein diets maintained approximately the same body weight (gains were -0.2^3, 7_f6 and 7^3 g for each dietary group, respectively) during the subsequent 10-day period. The weight gain 1n treated animals was 10-18 g less than that In the respective control rats. Dietary protein also had no effect on preventing or enhancing the 2,3,7,8-TCDD Induced thymlc atrophy. In yet another study, Seefeld and Peterson (1983) suggest that a reduction 1n food Intake caused by 2,3,7,8-TCDD Is primarily responsible for the loss of body weight or depressed growth rate of rats. Pair-fed control rats lost weight at the same rate and to the same extent as their weight matched 2,3,7,8-TCDD-treated partners (25 or 50 vg/kg) until day 10 after treatment. At 20-35 days after treatment, the body weight of the two groups began to diverge, with the pair-fed control group having body weights that were 20-30 g higher than the corresponding 2,3,7,8-TCDD groups. They pro- pose a hypothesis that 2,3,7,8-TCDD lowers a regulated level of "set-point" for body weight control 1n the rat. The ensuing change 1n food Intake Is C-43 �thought to occur secondarily to the change 1n "set-point". Thus, the pre- cise mechanism for the 2,3,7,8-TCDD Induced weight loss remains uncertain; however. It 1s evident that weight loss Is a contributing factor to 2,3,7,8-TCOO Induced mortality and morbidity. Feeding a diet containing 7 ppb of 2,3,7,8-TCDD to rats caused an Increase of liver weight while unexpectedly, 20 ppb caused less of a liver weight gain. After the feeding of 2,3,7,8-TCDD was discontinued, recovery was greater 1n the 7 ppb groups (Fries and Marow, 1975). The hepatotoxlclty of 2,3,7,8-TCDD was most severe In rats, mice and rabbits (Vos and Beems, 1971; Gupta et al., 1973; Schwetz et al., 1973; Vos et al., 1974). 2,3,7,8TCDD-1nduced liver alterations 1n the guinea pig and hamster were generally limited to the responses accompanying liver hypertrophy (Turner and Collins, 1983; Olson et al., 1980b). Limited steatosls, focal necrosis, and cyto- plasmic hyalln-Uke bodies were also observed 1n the guinea pig (Turner and Collins, 1983). Comparative studies Indicate that the guinea pig and hamster were the least sensitive to 2,3,7,8-TCDD-1nduced hepatotoxlclty, which 1s In contrast to the 5000-fold difference in the acute LD for 2,3,7,8-TCDD 1n these species. 2,3,7,8-TCDD affects porphyMn metabolism and causes significantly elevated excretion of porphyrlns and i-am1nolevu!1n1c add section). Goldstein et al. (1978) showed synthetase, a rate-limiting enzyme (see Metabolism that a-am1nolevu!1n1c add In porphyrln synthesis, was slightly Increased (2-fold) In male C57B1 mice given 4 weekly doses of 2,3,7,8-TCDD at 25 yg/kg. 2000-fold. This dose of 2,3,7,8-TCDD Increased liver prophyrln levels Catabollsm of porphyrln by uroporphyrlnogen decarboxylase also appeared to be decreased 1n 2,3,7,8-TCDD-treated mice. (UD) Smith et al. (1981) reported a decrease 1n UD activity from -25 to 7 nmoles/hr/g liver In C-44 �male and female C57B1 mice 3 weeks after a single oral exposure to 2,3,7,8TCDD at a dose of 75 yg/kg. No effect of 2,3,7,8-TCOD on UD activity was observed In DBA/2 mice, which were Insensitive to the Induction of prophyrla. A time course of changes In UO activity with length of time after exposure to 2,3,7,8-TCDD Indicated a steady decline In activity starting 3 days after exposure to 2,3,7,8-TCOD, which continued until day 21 when the study was terminated. Sweeney and Jones (1978) reported similar results after 5 weekly doses of 2,3,7,8-TCDD at 25 ug/kg. In this study, the UD activity declined -48% 1n C58B1 mice and only 4% In DBA/2 mice. factors besides the Increase 1n a-am1nolevul1nlc add decrease Other synthetase and the In UD activity may also participate In the dramatic Increase In liver porphyrln In mice, associated with exposure to near lethal doses of 2,3,7,8-TCDD. A number of biochemical studies have resulted from the observation that 2,3,7,8-TCDD produces fatty livers and a resulting Increase In total hepatic Upld content In several species. rat produced an Increase A sublethal dose of 2,3,7,8-TCDD In the In trlglycerldes and free fatty adds and a decrease 1n sterol esters, while a lethal dose Increased cholesterol esters and free fatty adds (Albro et al., 1978). Poll et al. (1980) treated rats with a single 1.p. Injection of 2,3,7,8-TCDD at doses of 2.5, 5, 10 and 20 vg/kg. At day 21 after treatment there was a dose-related Increase 1n total plasma cholesterol and high density Upoproteln cholesterol, while no change was observed 1n trlglycerldes or very low and low density Upoprotelns (VLDL and LDL, respectively). At a dose of 20 yg/kg the maximum Increase In HDL cholesterol and total cholesterol occurred 30 days after treatment, and a significant elevation was still present at 60 days after treatment when the study was terminated. C-45 Slight changes In the apoprotelns �of HDL From 2,3,7,8-TCDD rats and control rats were Indicative of new apoproteln synthesis. Although the Increase In HDL cholesterol may be 1n response to eliminating excess llplds, the exact function has not been clearly shown. In contrast to rats, male Hartley strain guinea pigs given a single 1.p. Injection of 2,3,7,8-TCDD at a dose of 2 v9/kg had Increased hyperl1p1dema characterized by Increases 1n VLDL and LDL (Swift et al., 1981). In animals sacrificed 7 days after exposure to 2,3,7,8-TCDD, there was an Increase In total serum Upld, cholesterol esters, tMglycerldes and phosphollplds when comparison was made to pair-fed, weight-paired or ad_ libitum fed control groups. Serum-free fatty acids were not changed quantitatively; however, some qualitative changes occurred, reflecting an Increase 1n the types of fatty acids which were abundant 1n the adipose tissue of guinea pigs. Analysis of Upoprotelns revealed a 19-fold Increase In VLOL and a 4-fold Increase 1n LDL, with no change observed 1n the levels of HDL. The VLDL was also qualitatively different 1n the 2,3,7,8-TCDD treated animals, containing less cholesterol ester and an altered C apoproteln. The Importance of these qualitative changes 1s unclear. The hyper!1p1dem1a may result from the 2,3,7,8-TCDD Induced mobilization of free fatty adds, which are then used 1n the synthesis of VLDL and are subsequently formed Into LDL. The relationship of the changes In serum Upld levels to the mechanism of 2,3,7,8TCDD toxlclty needs further study. Gupta et al. (1973) reported slight to moderate thymlc atrophy 1n guinea pigs after 8 weekly oral doses of 0.2 vg/kg. The thymlc atrophy was characterized by a decrease 1n the number of cortical thymocytes, reduction In size of the thymlc lobules, and the absence of a demarcation between cortex and medulla. There was a relative depletion of lymphold cells 1n the C-46 �spleen and the lymph nodes. observed In addition, moderate thymlc atrophy was In rats after 31 dally oral doses of 1 yg/kg 2,3,7,8-TCDD. Thymlc atrophy has also been noted In monkeys (Norback and Allen, 1973). In later studies 2,3,7,8-TCDD was found to suppress cell-mediated Immune function 1n young rats without affecting humoral Immune function. Suppression of T-cell function was selective In that "helper" cell function was not suppressed (Faith and Moore, 1977). Recently, the effects of 2,3,7,8-TCDD on thymus Involution In rats were found not to Involve the adrenal or pituitary glands and were not prevented by treatment with growth hormone (van Logten et al., 1980). Increased susceptibility to Salmonella Infection was found In mice treated Intragastrlcally with 2,3,7,8-TCDD at doses between 1 and 20 yg/kg bw once weekly for 4 weeks. Such Increased susceptibility after 2,3,7,8- TCDD administration was not seen with Herpes virus Infection (Thlgpen et al., 1975). Thymus atrophy with consequent suppression 1n cell-mediated Immunity as measured by several parameters was found by Vos et al. (1978) 1n mice after various doses of 2,3,7,8-TCDD up to 50 vg/kg bw. were dose related. The effects Juvenile and adult mice treated with 2,3,7,8-TCDD 1n their feed at 10 and 100 ppm displayed several dose related changes, Including depression In total serum protein, gamma globulin and albumin. Primary and secondary antibody responses to both tetanus toxold and sheep erythrocytes were also reduced, as well as resistance to challenge with either Salmonella typhlmurlum or Usterla monocytogenes (H1nsd1ll et al., 1980). Neonatal B6C3F, mice, exposed to prenatal (maternal dosing on day 14 of gestation) and postnatal (days 1, 7 and 14 after birth) doses of 0, 1.0, 5.0 or 15.0 v9/kg 2,3,7,8-TCDD were studied for Immunotoxlc effects and host susceptibility (Luster et al., 1980). C-47 In the bone marrow, hypocellularlty �and depressed macrophages-granulocyte progenitor cells and plurlpotent stem cells were associated with 2,3,7,8-TCDD exposure at the 5.0 and 15.0 pg/kg dose levels. Host susceptibility to L. monocytogenes was tested 1n the 2,3,7,8-TCDD-exposed neonates. and PYB6-tumor cells Death occurred 1n 73% and 40% of the L.. monocytoqenes Inoculated (1.2xl06 viable organisms) mice 1n the 5.0 and 1.0 yg/kg dose groups, respectively, compared controls. to 28% of Tumor development occurred In 44, 60 and 22% of the neonates Inoculated with 5x10* tumor cells from the 5.0 vg 2,3,7,8-TCDO/kg, 1.0 vg 2,3,7,8-TCDO/kg and control groups, respectively. While thymlc atrophy may be one of the most sensitive Indicators of experimental exposure to 2,3,7,8-TCDD, animals given a lethal dose of 2,3,7,8-TCDD do not appear to die from Infections, nor does a germ-free environment protect them from death (Grelg et al., 1973). In the Gupta et al. (1973) study, rats also showed degenerative changes In the renal collecting tubules, degenerative changes of the thyroid follicles, necrosis and ulceratlon of the glandular stomach and hemorrhage Into the adrenals. This latter set of findings generally occurred at higher dose levels than the minimum dose needed to provide thymlc atrophy or liver enlargement. More recent studies on the effects of 2,3,7,8-TCDD on renal functions have been carried out. Analzl and Cohen (1978) reported an In- crease 1n the renal secretion of phenolsulfonphthaleln (PCP) and a significant decrease 1n glomerular filtration rate (GFR) compared with controls 1n rats treated with 10 yg/kg (l.p.) of 2,3,7,8-TCDO. These authors attrib- uted these effects to the toxldty of 2,3,7,8-TCDD on glomerular structures. However, other reports concluded that 2,3,7,8-TCDD causes no specific functional lesions In the kidney, rather that the effects on renal functions reflect a general toxicosis (Pegg et al., 1976). C-48 �Pronounced dermal effects with 2,3,7,8-TCDD treatment have been reported for rabbits by Schwetz et al. (1973). Mllnes (1971) observed chloracne 1n rabbits after a single oral dose of 1 yg/kg. sensitive Dogs, though apparently less to 2,3,7,8-TCDD than rabbits (lethal effects) following oral administration, have exhibited hair loss (Schwetz et al., 1973). PcConnell et al. (1978b) found facial alopecia with acne-Uke eruptions, blepharitis, weight loss and anemia In rhesus monkeys after single oral doses of 70 or 350 yg/kg 2,3,7,8-TCDD. Allen et al. (1977) observed loss of facial hair and eyelashes, accentuated hair follicles, dry scaly skin and gastric mucosal dysplasla 1n eight female rhesus monkeys fed a diet containing 500 ppt 2,3,7,8-TCDO for up to 9 months. from severe pancytopenla. Eventually 5 of the 8 monkeys died In humans, the most characteristic and frequently observed lesion produced by 2,3,7,8-TCDD and other chlorinated aromatic hydrocarbons Is chloracne (Crow, 1981; Taylor, 1979). This lesion consists of hyperplasla and hyperkeratosls of the InterfolUcular epidermis, hyperkeratosls of the hair follicle, especially at the Infundlbulum, and squamous metaplasia of the sebaceous glands which form keratlnaceous comedones and cysts (Klmbrough, 1974). A number of studies have been directed toward evaluating the mechanlsm(s) for the toxldty of 2,3,7,8-TCDD. Such studies will ultimately provide a better estimate of man's relative sensitivity to 2,3,7,8-TCDD and other compounds having a similar mode of action. studies may be able to explain the marked Specifically, these Interspedes differences 1n relative sensitivity to 2,3,7,8-TCDO, and thus help establish man's relative sensitivity. These studies may also some day provide for the better treatment of human exposure to these toxins. C-49 �Pharmacogenetlc studies have played an Important role In understanding the biologic and toxic effects of drugs and xenoblotlcs. Nebert and co- workers have shown that carcinogenic polycycllc aromatic hydrocarbons (PAHs) Induce the cytochrome P-450-dependent monooxygenase, aryl hydrocarbon hydroxylase (AHH), In certain responsive strains of mice (e.g., C57BL/6J, BALBc, C3HF/He), whereas this PAH Induction activity Is minimal or nonexistent 1n non-responsive strains (DBA/2J) (Nebert, 1979, 1982; Nebert and Jensen, 1979; Nebert et al., 1981, 1983). The gene complex responsible for the Induction of AHH and several other enzymes has been designated the Ah locus, which comprises regulatory, structural and possible temporal genes. Extensive studies on genetically Inbred responsive and non-responsive mice (and their backcrosses) Indicate that these differences a-re related to the Ah regulatory gene and Us gene product, the Ah. cytosollc receptor protein. This receptor protein Interacts with PAH Ugands and the resultant PAH:Ah receptor complex translocates Into the nucleus and presumably Initiates the Induction of AHH via a process comparable to that proposed for the steroid hormones. Since the carcinogenic and toxic effects of PAHs are dependent on their oxldatlve metabolism to reactive electrophlUc forms, 1t 1s not surprising that the A|i receptor plays an Important role 1n mediating their toxldty and cardnogenlclty (KouM, 1976; Kourl et al., 1974; Benedict et al., 1973; Shum et al., 1979; Thomas et al., 1973; Legraverend et al., 1980; Robinson et al., 1975; Mattlson and Thorgelrsson, 1979). Responsive mice are more susceptible to the toxic (Inflammation, fetotoxldty, primordial oocyte depletion) and carcinogenic effects of PAH at organs/tissues In direct contact with the applied chemical; In contrast, non-responsive mice are more susceptible to the tumorlgenlc effects of PAHs at tissue/organ sites remote C-50 �from the Initial site of exposure to the PAHs. These differences 1n susceptibility are due to several factors Including AHH-med1ated toxlcatlon and detoxlcatlon. Genetic studies also support the role of the Ah_ receptor In mediating the toxic and biologic effects of 2,3,7,8-TCDD. Initial studies by Poland and coworkers (Poland et al., 1974, 1983; Poland and Glover, 1975; Nebert et al., 1975; Poland and Knutson, 1982) demonstrated that the mlcrosomal AHHIndudng activity of 2,3,7,8-TCOO and 3-methylcholanthrene (HC) In several genetically Inbred mice strains were similar. 2,3,7,8-TCOD Induced AHH In several Like MC and related PAHs, responsive mouse strains (e.g., C57BL/6J); In contrast to MC, 2,3,7,8-TCOO Induced mlcrosomal AHH In the DBA/2J non-responsive mice; however, the ED for this biologic response was significantly higher than values reported for the responsive mice. In genetic crosses between responsive C57BL/6 and non-responsive DBA/2 mice H was also shown for both MC and 2,3,7,8-TCDO that the trait of responsiveness Is Inherited In a simple autosomal dominant mode (Poland and Knutson, 1982). It has been suggested that the observed differences 1n the activities of MC and 2,3,7,8-TCDD are related to their relative Ah receptor affinities (Poland and Knutson, 1982) and pharamcoklnetlc and metabolic factors, which would more rapidly diminish the "available" concentrations of MC due to metabolism and excretion. Several studies with 2,3,7,8-TCDD 1n genetically Inbred mice support the receptor mediated hypothesis. ase, The Induction of UDP-glucuranosyl transfer- DT dlaphorase, a-am1nolevu!1n1c add, T-aldehyde dehydrogenase glutath1one-S-transferase 8, and chollne klnase by 2,3,7,8-TCDD or MC 1n genetically Inbred mice has also been shown to segregate with the Ah_ locus (Beatty and Neal, 1976a; Owens, 1977; Klrsch et al., 1975; OletMch et al., C-51 �1978; Ishldate et al., 1980; Poland and Glover, 1973). Toxicology studies with genetically Inbred mice confirm the role of the Ah locus 1n mediating several toxic effects Including porphyrla, Immunotoxlclty, a wasting syndrome, thymlc atrophy and cleft palate formation (Jones and Sweeney, 1980; Poland and Glover, 1980; Courtney and Moore, 1971; Vecchl et al., 1983). Poland et al. (1982) have also linked the tumor-promoting activity of 2,3,7,8-TCDD 1n hairless mice to the cytosollc receptor. ln_ vitro studies with XB cells 1n culture also support the role of receptor 1n mediating a dose-related cell keratlnlzatlon by 2,3,7,8-TCDD which resembles some of the characteristics of chloracne (Knutson and Poland, 1980). This cell line Is also responsive to AHH Induction and contains a cytosollc receptor binding protein. Although the murlne Ah_ receptor has not been characterized, several studies confirm that a protein with high affinity for MC and 2,3,7,8-TCDD 1s present 1n low concentrations In the hepatic (-30-50 fmolar) and extrahepatic tissues of responsive C57BL/6J mice (Greenlce and Poland, 1979; Okey et al., 1979, 1980; Poland et al., 1976b; Mason and Okey, 1982; GaslewUz and Neal, 1982; Okey and Vella, 1982; Okey, 1983; Nebert et al., 1983). Although the Ah receptor has not been detected 1n the cytosol of DBA/2J mice, after the administration of radlolabeled 2,3,7,8-TCDD to these mice, some of the radlolabel 1s detected 1n the nuclei of the non-responsive mice. Moreover, the sedimentation characteristics of the [3H]-2,3,7,8-TCOD: nuclear protein complex 1n D8A/2J mice are similar to those observed with the bound Ah cytosollc receptor protein In C57BL/6J mice using a sucrose density gradient centrlfugatlon separation technique (Okey, 1983). Several reports have also demonstrated that the cytosollc Ah receptor protein migrates Into the nucleus of the cell only after binding with 2,3,7,8-TCDD C-52 �(Greenlee and Poland, 1979; Okey et al.. 1979, 1980) and this parallels the observations noted for the Interactions between steroids and their receptor proteins. It should be noted, however, that the binding affinity and concentration of the cytosol receptor for 2,3,7,8-TCDD In liver from guinea pig, rat, C57BL/6 mouse, rabbit and hamster are very similar despite a 5000-fold difference In LDou for 2,3,7,8-TCDD between the guinea pig and hamster rn (Poland and Knutson, 1982; Gaslewlcz et al., 1983a). Thus the affinity and concentration of hepatic cytosol receptors does not alone explain the profound Interspedes variability In sensitivity to TCDD. In a subchronlc study, Koclba et al. (1976) fed rats 0, 0.001, 0.01, 0.1 or 1.0 yg 2,3,7,8-TCDO/kg of body weight by gavage for 5 days/week for 13 weeks. The dosing at 1.0 yg/kg/day caused decreased body weights, damage, thymlc some mortality, lethargy, liver pathology, biochemical atrophy, decreased evidence of liver lymphatic tissues, disturbances of porphyrln metabolism and slight alterations 1n the hematopoletlc system. There was also evidence of mild adverse effects on the male and female reproductive systems. The effects on the reproductive system Included decreased size of the testls and secondary sex organs 1n 2 of 5 males and uteri In 4 of 5 females. The 0.01 yg/kg/day level was considered by the authors to be the no-observed-adverse-effect yg/kg/day level was the no-observed-effect level (NOAEL) and the 0.001 level (NOEL) for this treatment regimen. The dietary administration of 2,3,7,8-TCDO to rats at dose levels equivalent to 0, 0.001, 0.01 or 0.1 yg/kg/day for three generations (Murray et al., 1979) produced effects on liver, thymus and reproduction (discussed In C-53 �the Teratogenlclty section) at 0.01 and 0.1 yg/kg/day. authors, the 0.001 According to the yg/kg/day exposure was a NOAEL (however, equivocal effects were noted In some generations at this dose). Liver toxldty was the only effect of treatment observed by hlstologlc examination of Osborne-Hendel rats and B6C3F mice administered 2,3,7,8- TCDO for 13 weeks In a preliminary subchronlc toxlclty study designed to define an acceptable dose for a chronic toxldty study (U.S. DHHS, 1980b). The anMals 1n groups of 10 males and 10 females were administered the compound 1n corn olV.acetone (9:1) twice a week at doses for rats of 0.0, 0.5, 1, 2, 4 and 8 yg/kg/week, and for mice at doses of 0.0, 1, 2, 5, 10 and 20 yg/kg/week. Deaths occurred at the two high dose levels In rats, with 4 females 1n the 8 yg/kg/week and 1 1n the 4 yg/kg/week group dying, while only 2 male rats 1n the 4 yg/kg/week group died. Deaths were accompanied by severe toxic hepatitis. Hepatic lesions were observed In all other rats examined 1n groups administered 1-8 yg/kg/week; however, not all animals 1n each group were submitted to necropsy. Normal liver histology was observed In the two male rats examined from the low dose groups and only threshold toxic effects occurred In the low dose female rats. Similar effects of treatment were observed In mice, with a single death occurring In each sex at the high exposure level along with reports of hepatic lesions on hlstologlc examination. In contrast to rats, female mice were less sensitive to the hepatotoxlc effect of 2,3,7,8-TCDD than were the male mice. Hepatic lesions were observed 1n all dose groups of male mice, while the 1 and 2 yg/kg/week dose groups of female mice had normal livers. Although the group sizes were small, making conclusions tenuous, 1t appeared that sex differences 1n the sensitivity to the toxic effects of 2,3,7,8-TCDD occurred, and that the more sensitive sex may vary with species tested. C-54 �In a more extensive subchronlc study In rats, King and Roesler (1974) followed the development of toxldty by a series of Interim sacrifices during 28 weeks of exposure to 2,3,7,8-TCOD and a 12-week, post-treatment recovery period. Groups of 35 male and 35 female Sprague-Dawley rats were Intubated twice weekly with 2,3,7,8-TCDD 1n corn o1l:acetone (9:1) at doses of 0.0, 0.1 and 1.0 yg/kg/week. No treatment-related deaths occurred; however, 3 animals from each group of each sex were killed after 2, 4, 8 and 16 weeks, and 10 animals of each sex were killed after 28 weeks of treatment. In addition, 3 rats of each sex were killed 4 and 12 weeks after termination of exposure. Animals were monitored for gross changes during the study and were examined for gross and hlstologlc changes at necropsy. Besides a dose-related decrease In body weight gain 1n male rats and a decrease In body weight gain In the high dose female rats, the only effect of exposure to 2,3,7,8-TCDD was hlstologlc changes In the liver. Liver pathology was normal 1n all treated groups up through the Interim kill at 10 weeks. Fatty changes 1n the liver were considered the most Important observation and the data Is summarized In Table 7. The fatty changes ranged from single large I1p1d droplets In a few centMlobular heptocytes to I1p1d droplets 1n all centrllobular hepatocytes with extension Into the mldzonal hepatocytes. No clear dose-response pattern was observed In this study; however, 1t did appear that the severity of fatty changes was greater 1n female rats. During the recovery period fatty changes progressively de- creased In severity, but were still present In some treated animals 12 weeks after cessation of exposure. Other hlstologlc changes observed 1n the liver of a small number of animals, predominantly In the animals killed at 28 weeks, Included single cell or very small areas of necrosis, Increased nuclear size, subtle distortion of liver architecture, and hyperchromatlc C-55 �TABLE 7 Hepatocellular Fatty Change Observed 1n Rats a Following Subchronlc Exposure to 2,3,7,8-TCOD 28 Weeks 4-Weeks Recovery 12-Weeks Recovery Treatment Groupb Ho Ha N NS NS 2/3 1/3 NS NS NS 3/3 NS NS NS NS 2/3 1/3 NS NS NS 2/3 NS NS 1/3 NS 1/3 1/3 1/3 NS 2/3 NS NS 1/3 NS N S Male Control 7/10 3/10 Males at 0.1 yg/kg 2/10 1/10 1/10 5/10 1/10 Males at 1.0 yg/kg 0/10 1/10 2/10 NS 5/10 2/10 NS H1 Ho Ha N H1 Ho Ma o I en en Female Control 10/10 NS NS NS NS 3/3 NS NS NS NS 3/3 NS NS NS NS Females at 0.1 yg/kg 4/9 4/9 1/9 NS NS 3/3 NS NS NS NS 3/3 NS NS NS NS Females at 1.0 yg/kg 1/10 4/10 4/10 1/10 NS 1/3 1/3 NS 1/3 NS 1/4 3/4 NS NS NS a Source: King, and Roesler, 1974 ''Animals were treated twice weekly by gavage with 2,3,7,8-TCDD dissolved 1n corn o1l:acetone (9:1) N S M1 Mo Ma NS = = = = = None Slight: random hepatocyte containing a solitary, large I1p1d droplet-equivocal Mild: several centrllobular hepatocytes contain I1p1d Moderate: most centrllobular hepatocytes contain I1p1d Marked: all centrllobular and some midzonal hepatocytes contain I1p1d Not specified �nuclei. All of these lesions were considered to be slight or mild, and less toxlcologlcally relevant than the fatty changes. The data on hepatic steatosls Indicated that the liver was a sensitive organ to the toxic effect of 2,3,7,8-TCDD, and although some recovery occurred after termination of treatment, the recovery process was slow and not complete by the time the study was terminated. The recovery time was also demonstrated to be long In a subchronlc study by Goldstein et al. (1982) of 2,3,7,8-TCDD-1nduced porphyrla. Groups of eight female Sprague-Oawley rats were given 2,3,7,8-TCDD 1n corn oil'.acetone (7:1) weekly by gavage for 16 weeks at doses of 0.0, 0.01, 0.1 or 10.0 yg/kg/week and killed 1 week after the last treatment. Additional groups of rats received doses of 0.0 or 1.0 yg/kg/week for 16 weeks and were allowed to recover for 6 months. The high dose level was lethal to all animals within 12 weeks, while the only other gross sign of toxldty was a decrease In body weight gain In the group receiver;> 1.0 yg/kg/week. After 16 weeks of exposure to 2,3,7,8-TCDD, liver porphyrlns were elevated -1000-fold In 7 of 8 animals receiving 1.0 yg/kg/week, but only 1 of 8 animals In the 0.1 yg/kg/week group had elevated porphyrln levels. effect was observed 1n the low dose animals. No After a 6-month recovery period the porphyrln level 1n animals exposed to 1 yg/kg/week was still 100-fold higher than values 1n the control group. A similar pattern was observed for urinary excretion of uroporphyrln. The rate limiting enzyme 1n heme synthesis, d-amlnolevullnlc add synthetase, was also elevated at both the time of termination of treatment and at the end of the recovery period; however, other enzymes that were Increased after 10 weeks of treatment, cytochrome P-450, aryl hydrocarbon hydroxylase, and glucuronyl transferase, returned to near normal levels by 6 months. C-57 It was clear that a �6-month recovery period from subchronlc exposure to 2,3,7,8-TCOD at a dose of 1.0 yg/kg/week was not sufficient for complete reversal of 2,3,7,8TCOD-1nduced porphyrla. In rats. Increased urinary porphyMn was also observed after subchronlc exposure to 2,3,7,8-TCOD (Cantonl et al., 1981). Female CO rats were orally administered weekly doses of 2,3,7,8-TCDO at levels of 0.01, 0.1 and 1.0 wg/kg/for 45 weeks. The Initial Increase was observed 1n the high dose group at 3 months, and 1n the other two groups at 4 months, after the start ^ of exposure. Not only did the absolute amount of porphyMn Increase, but the relative distribution also changed to compounds containing more carboxyl groups. Only 1n the high dose group did the livers, at the terminal necropsy, show signs of excess prophyrln under examination by ultraviolet light. The toxic effects, other than neoplasla, of long-term exposure to 2,3,7,8-TCDO have been studied 1n rats, mice and monkeys. The primary purpose of many of the studies In rodents was to assess the cardnogenlclty of 2,3,7,8-TCDD. (These effects are discussed In detail 1n the Carclnogenesls section.) The observation of non-neoplast1c systemic toxic effects 1n these studies was often limited, and observations were made near the end of the natural Hfespan when conditions associated with aging may have obscured some effects produced by 2,3,7,8-TCDD. Table 8 summarizes the toxic effects of chronic exposure to 2,3,7,8-TCDD and provides Information on the exposure levels which result 1n the observed effects. Human health effects related to excessive exposures to 2,3,7,8-TCDD have been noted 1n several Instances. However, 1n many of these cases H Is difficult to quantify the exposure to 2,3,7,8-TCDD leading to the observed symptoms. Most of the exposures occurred 1n relation to the manufacture of C-58 �TABLE 8 Effects of Chronic Exposure to 2.3.7.8-TCDD In Laboratory Rodents Species/ Strain ' Dose Sex/No. Rat/Sprague- H/10 Daw ley Treatment Schedule Duration of Study Parameters Honltored Effects of Treatment Reference 0.0 ppt NA 95 weeks survival 40* survived until 95 weeks, the first death occurred at week 68 Van Killer et al.. 1977a H/10 1 ppt continuous In diet for 78 weeks 95 weeks survival BOX survived until 95 weeks, the first death occurred at week 86 Same as above H/10 5 ppt continuous In diet for 78 weeks 95 weeks survival 60% survived until 95 weeks, the first death occurred at week 33 Same as above H/10 50 ppt continuous In diet for 78 weeks 95 weeks survival 60X survived until 95 weeks, the first death occurred at week 69 Same as above H/10 500 ppt continuous In diet for 78 weeks 95 weeks survival SOX survived until 95 weeks, the first death occurred at week 17 Same as above H/10 1000 and 5000 ppt continuous In diet for 78 weeks 95 weeks survival No animals survived until 95 weeks, the first death occurred at week 31 Same as above H/10 50,000. 500.000 and continuous In diet 95 weeks survival No animals survived until 95 weeks. the first deaths occurred at weeks 2 and 3 Same as above -2193 ppt (0.1 vg/kg/day) continuous In diet for 2 years 2 years extensive hlstopathology, heroatology, urine analyses, and clinical chemistry Cumulative mortality Increased (F); Koclba et al.. Body weight gain decreased (H.F); 1978. 1979 Red blood cell count decreased (H.F); Packed cell volume decreased (H.F); Hemoglobin decreased (H,F); Retlculocytes Increased (H.F); White blood cell count decreased (F); SGPT Increased (F); 6-Glutarayl transferase Increased (F); Alkaline phosphatase Increased (F); Urinary coproporphyrln Increased (F); Urinary uroporphyrln Increased (F); Urinary 4amlnolevullnlc acid Increased hepatic degeneration Increased (H.F) en 10 Rat/Sprague- H8J750&50 Dawley �TABLE B (cont.) Species/ Strain Sex/No. Rat/SpragueDawley M8.F/50&50 Dose -208 ppt (0.01 pg/kg/day) Treatment Schedule Duration of Study Parameters Monitored Effects of Treatment Reference continuous In diet for 2 years 2 years extensive hlstopathology, hematology, urine analyses and clinical chemistry Urinary coproporphyMn Increased (F); Koclba et al., Urinary uroprophyrln Increased (F); 1978. 1979 Hepatic degeneration Increased (H.F) extensive hlstopathology, urine analyses and clinical chemistry No differences from values obtained from control animals Same as above (0.001 yg/kg/day) continuous 1n diet for 2 years 2 years Dawley Rat/Osborne- M&F/75S.75 Hendel 0.0 Pg/kg/week NA 106 weeks extensive hlstopathology Toxic hepatitis; 0/74 (M). 0/75 (F) U.S. DHHS. 1980b Rat/Osborne- H&F/50&50 Mendel 0.5 pg/kg/week administered by gavage biweekly for 104 weeks 107 weeks extensive Mstopathology Toxic hepatitis; 14/50 (M). 32/50 (F) Same as above Rat/Osborne- M8.F/50J.50 Mendel 0.05 ng/kg/week administered by gavage biweekly for 104 weeks 107 weeks extensive hlstopathology The Incidence of toxic hepatitis was not elevated 0/5C (M). I/SO (F) Same as above Rat/Osborne- H&F/50&50 Mendel 0.01 administered by gavage biweekly for 104 weeks 107 weeks extensive hlstopathology The Incidence of toxic hepatitis was not elevated 1/50 (M). 0/50 (F) Same as above M1ce/B6C3Fl M&F/75&75 0.0 wg/kg/week NA 105-106 weeks extensive hlstopathology Toxic hepatitis; 1/73 (M), 0/73 (F) Same as above M1ce/B6C3Fl M&F/50&50 0.5 vg/kg/week (M) 2.0 yg/kg/week (F) administered by gavage biweekly for 104 weeks 107 weeks extensive hlstopathology Toxic hepatitis; 44/50 (M), 34/47 (F) Same as above M1ce/B6C3Fl M&F/50&50 0.05 Pg/kg/week (M) 0.2 pg/kg/week (F) administered by gavage biweekly for 104 weeks 107 weeks extensive hlstopathology The Incidence of toxic hepatitis was not elevated 3/49 (M), 2/48 (F) Same as above M1ce/B6C3Fl M&F/50&50 0.01 ng/kg/week (M) 0.04 yg/kg/week (F) administered by gavage biweekly for 104 weeks 107 weeks extensive hlstopathology The Incidence of toxic hepatitis was not elevated 5/44 (M). 1/50 (F) Same as above Rat/Sprague- H&F/SO&SO -22 ppt o o Mg/kg/week �TABLE 8 ( c o n t . ) Species/ Strain Sex/No. Dose Treatment Schedule Duration of Study Parameters Monitored Effects of Treatment Reference Mice/Swiss N/38 0.0 yg/kg/week NA S88 days histology on all organs Dermatitis and amyloldosls; 0/38 Toth et al.. 1978. 1979 Mice/Swiss M/44 0.007 yg/kg/week administered by gavage weekly for 1 year 649 days histology on all organs Dermatitis and amyloldosls; 5/44 Same as above Mice/Swiss M/44 0.7 yg/kg/week administered by gavage weekly for 1 year 633 days histology on all organs Oermatltls-and amyloldosls; 10/44 Same as above Mice/Swiss N/43 7.0 yg/kg/week administered by gavage weekly for 1 year 424 days histology on all organs Early mortality, dermatitis and amyloldosls; 17/43 Same as above NA » Not applicable o i �2,4,5-trlchlorophenol or 2,4,5-tr1chlorophenoxyacet1c add. The best known among these may be the release of 2,3,7,8-TCDD due to the 1976 explosion of a malfunctioning 2,4,5-trlchlorophenol reactor 1978). In Seveso, Italy (Carrerl, The area closest to the plant received exposures of up to 5000 vg/m2 of soil; more remote areas received 1-75 vg/m2, and within these areas 12 cases of chloracne developed (Regglanl, 1980). Other symptoms Included acute dermatitis. All but the most severe cases of chloracne recovered within 26 months. No neurological, visceral or Immune effects were noted 1n these reports. Pocchlarl et al. (1979) reported a more extensive study of the Seveso Incident. These authors showed 75 cases of chloracne due to 2,3,7,8-TCDD exposure; 15 of those cases were severe or very severe. After 18-24 months, 19 cases had fully recovered while 1 case was still listed as severe. A subsequent survey 1n children by these authors uncovered an additional 137 cases of mild to serious chloracne. As Indicated In the Pocchlarl et al. (1979) report, signs of liver damage were also found after the Seveso Incident 1n Italy. Raised serum transamlnase and f-glutamyl transferase levels were found 1n -20% of the people living 1n or near the area of greatest deposition of 2,3,7,8-TCDD (Bert et al., 1976; Hay, 1976). In addition, some neurological effects were also noted among the exposed people. Immuno- loglcal and cytogenetlc Investigations yielded normal results. The most commonly reported symptom related to 2,3,7,8-TCDD exposure 1n man has been chloracne (Bauer et al., 1961; Klmmlg and Schulz, 1957; Schulz, 1957; Firestone, 1977; Ougols and Colomb, 1956, 1957; Dugols et al., 1958; Blelberg et al., 1964; Oliver, 1975; Poland et al., 1971; Klmbrough, 1980). The acneform lesions of the skin may develop a few weeks after the exposure and may persist for over 1 year following the cessation of exposure. Other C-62 �skin problems which have been reported Include hyperplgmentatlon, hlrsutlsm, Increased skin Fragility and veslcobullar eruptions on exposed areas of the skin. Host cases of chloracne have been reported for Industrial workers. Bauer et al. (1961) reported that 31 cases of chloracne occurred within a few months after a factory manufacturing 2,4,5-T near Hamburg, Germany, Instituted a change 1n Us Industrial process. On this occasion, the Inves- tigators were able to show that the chloracne was not caused by purified 2,4,5-T, but that 1t was attributable to 2,3,7,8-TCDO which was a contaminant 1n the technical grade 2,4,5-T. A similar outbreak of chloracne affected 60 workers at a 2,4,5-T plant 1n Michigan 1n 1964 (Firestone, 1977). In addition to chloracne, some, occupational exposures to 2,3,7,8-TCOD have provided evidence for liver damage and for neurological effects. Hyperllpemla and hypercholesterolemla were reported (1n addition to chloracne) for 17 workers at a plant producing trlchlorophenol 1n France (Ougols and Colomb, 1956, 1957; Dugols et al., 1958). Among 29 subjects who exhibited chloracne after working 1n a plant which produced 2,4-D and 2,4,5-T 1n Newark, New Jersey, 11 had Increased excretion of uroporphyrlns. Three of these were diagnosed as porphyMa cutanea tarda, and two of these also exhibited elevated serum glutamlc-oxalacetlc transamlnase levels (Blelberg et al., 1964; Firestone, 1977). Pazderova-Vejlupkova et al. (1981) reported that 80 workers developed chloracne, nausea, fatigue and weakness engaged 1n the lower extremities while 1n the production of 2,4,5-sodlum trlchlorophenoxyacetate and trlchlorophenoxyacetate butylester 1n a plant located In Prague, Czechoslovakia. Prominent clinical symptoms among 55 of the 80 workers Included hypercholesterolemla, hyperl1pem1a and hyperphosphoHpemla, Increased plasma C-63 �alpha and gamma globulins, and decreased plasma albumin. Porphyrla cutanea tarda was observed In 11 of the 55 workers tested. In some cases. Illness subsided while other cases became more severe during a 3 to 4-year Follow-up period. Long-term pathological symptoms (remaining evident 5 years after exposure) Include deviations 1n Upld metabolism, abnormal glucose tolerance, and high urinary excretion of uroporphyrlns. Polyneuropathy, usually of the lower extremities, occurred during the period of Illness and the symptoms remained after 4 years. Singer et al. (1982) also Indicated a decrease In nerve conduction velocities of sura! nerves 1n workers exposed to phenoxy acid herbicides (average exposure, 7 years) when compared to a similar group of non-exposed workers (40.3 m/sec 1n exposed vs. 42.8 m/sec 1n non-exposed, p=0.02). Although the causative agent was not known, dloxln contaminants are suggested. Poland et al. (1971) re-examined all of the employees of the Newark factory 1n 1969, after the level of 2,3,7,8-TCDD 1n the trlchlorophenol had been reduced from 10-25 mg/kg to <1 mg/kg. Chloracne was still found In 13 of 73 workers. A number of employees exhibited hyperplgmentatlon or facial hypertrlchosls. No definite diagnoses of porphyrla were made, and only one worker had a mild case of uroporphyrlnurla. The authors suggested that chloracne and porphyrla cutanea tarda are essentially Independent syndromes. In the Newark workers, Poland et al. (1971) noted that serum cholesterol was elevated 1n 10% of the cases and serum lactic dehydrogenase was elevated In 29% of the cases. Seven persons (10% of the workers) had a white blood cell count of <5000/mm3. In addition, -30% of the workers reported suf- fering from gastrointestinal symptoms (nausea, vomiting, diarrhea, abdominal pains, blood In the stools); -10% of the workers had other symptoms, such as weakness of the lower extremities, headaches and decreased auditory acuity. C-64 �Some hypomanla was observed, with the degree of hypomanla (as measured on the Minnesota Multlphaslc Personality Inventory Hypomanlc Scale) showing an association with the severity of chloracne. Among the workers exposed 1n Hamburg, many also showed clinical signs of systemic toxldty, mainly muscular weakness, loss of appetite and weight, sleep disturbances, orthostatlc Impairment. hypotension, abdominal pain and liver Most of the workers presented psychopathologlcal changes that were Interpreted to be a specific syndrome (Bauer et al., 1961; K1mm1g and Schulz, 1957; Schulz, 1957). Teleglna and Blkbulatova (1970) reported that the production of 2,4,5-T 1n the USSR began In 1964. At a specific site, 128 workers showed skin lesions, and among 83 examined, 69 had chloracne. Many, especially those with severe skin lesions, also presented evidence of liver Impairment. In addition, 18 workers had a neurasthenic syndrome. Similar findings were described by Jlrasek et al. (1973, 1974), who reported 76 cases of chloracne following exposure to 2,3,7,8-TCOD between 1965 and 1968 In a plant 1n Czechoslovakia which produced 2,4,5-T and pentachlorophenol. Fifty-five of these cases were followed medically for over 5 years; some had symptoms of porphyrla cutanea tarda, uroporphyrlnurla, abnormal liver tests (elevated blllrubln levels, Increased SGOT or SGPT activities, enlargement. and elevated bromsulphthaleln clearance times) and liver The majority of the patients also suffered from neurasthenia and a depressive syndrome. In 17 persons, signs of a peripheral neuropathy, especially 1n the lower extremities, were confirmed by electromyographlc examinations. More than half of the patients showed raised blood levels of cholesterol and total I1p1ds. C-65 �Three scientists were poisoned In the course of an experimental preparation of 2,3,7,8-TCDD made by heating potassium trlchlorophenate (Oliver, 1975). Two scientists developed typical chloracne 6 and 8 weeks after the exposure. Delayed symptoms, possibly due to 2,3,7,8-TCDD, developed ~2 years after the Initial exposure 1n two of the scientists. These symptoms were personality changes, Including loss of energy and drive, Impairment of vision, taste and muscular coordination, disturbance of sleep, gastrointestinal symptoms and hlrsutlsm. Hypercholesterolemla (1n excess of 300 mg/100 ma,) occurred In all three patients. In November 1953, an accident occurred at a factory In Ludwlgshafen, Federal Republic of Germany, during the manufacture of trlchlorophenol (Goldman, 1972; Hofmann, 1957). As a consequence, 53 workers were affected by chloracne, 42 1n a severe form. The son of one of the workers developed chloracne following contact with his father's clothes. Twenty-one of the 42 workers with severe chloracne showed signs of damage to Internal organs or disturbances of the nervous system. The most relevant features were poly- neuritis, sensory Impairment and liver damage. In a follow-up study of these workers, Thelss and Goldmann (1977) reported that of the 53 workers exposed to 2,3,7,8-TCDD In the 1953 accident, 22 were still working at the factory, 16 had retired and 15 had died (6 while still employed at the factory and 9 while 1n retirement). Of the 22 workers still employed, 2 still had acne of the face and scrotum, 1 had paralysis of the left leg and 1 had permanent loss of hearing. The remainIng workers were well, except for scars left by the chloracne. Of the 15 deaths, 7 were from cardiovascular disease, 2 of which were myocardlal Infarcts, 1 due to mitral stenosis, 4 from cancer, 2 from suicide, 1 from necrotlc pancreatitis and 1 from esophageal hemorrhage. C-66 The 16 men who had �retired and were still alive were well. No abortions or miscarriages were reported 1n the wives of exposed workers still employed at the factory. 2,3,7,8-TCDD has also been Identified as the cause of an outbreak of poisoning In humans, horses and other animals In 1971 (Carter et al., 1975; Klmbrough et al., 1977). Exposure was related to the spraying of waste oil contaminated with 2,3,7,8-TCDD on riding arenas for dust control. The most severe effects occurred 1n a 6-year-old girl who played In the arena soil. Her symptoms Included headache, nosebleedlng, diarrhea, lethargy, hemorrhag1c cystitis and focal pyelonephritis. Three other children and one adult who were frequently In the arena complained of skin lesions. In two of the children, the described lesions were similar to chloracne. Exposure to 2,3,7,8-TCDD and other dloxlns occurring as contaminants In • Agent Orange have been associated with many health effects reported 1n veterans and residents of Vietnam. Symptoms Include numbness of extremi- ties, skin rashes and Irritation, liver dysfunction, weakness, loss of sex drive, and psychological changes (Holden, 1979). The toxic effects attributed to 2,3,7,8-TCDD exposure were studied over a 10-month period In a group of 78 Vietnam veterans who claimed to have been exposed to Agent Orange (Bogen, 1979). Symptoms reported by the veterans Included gastrointestinal complaints (anorexia, nausea, diarrhea, constipation, abdominal pain), joint pain and stiffness, and neurological complaints (numbness, dizziness, headaches, depression and bouts of violent rage). It Is mentioned that these patients had previously been chronically 111 and had frequent Infections and allergies. This study was apparently based on personal evaluations of health 1n a survey-type format. No control group was used for comparison and no clinical or medical evaluations of health were made. Most of these complaints were non-specific, Judgmental and occur commonly 1n the general public. C-67 �In an effort to evaluate the toxic effects attributed to 2,3,7,8-TCDD as a contaminant of Agent Orange, Stevens (1981) established a minimum toxic dose of 2,3,7,8-TCDD and determined the amount of this contaminant to which veterans may have been exposed during Agent Orange spraying. Based on studies 1n which rhesus monkeys were fed small amounts of dietary 2,3,7,8-TCDD and analogy with human data on the minimum toxic dose of 2,3,7,8-tetrachlorod1benzo-p_-furan (TCDF), the cumulative minimum toxic dose of 2,3,7,8TCDD In man was calculated to be 0.1 yg/kg. Based on application rates (4.1 g Agent Orange/m2) and 2,3,7,8-TCDD concentration In the herbicide (2 ppm), the average concentration of TCDD on sprayed surfaces of Vietnam was ~8 yg/m2. Based on a comparison of the development of toxic systems 1n humans exposed to 2,3,7,8-TCDD 1n the Seveso accident and a child exposed In an Eastern Missouri horse arena, the measured environmental levels of 2,3,7,8-TCDD, and estimates of the absorbed dose necessary to produce these symptoms, the author calculated an average Intake transfer factor (ratio of absorbed compound to 2,3,7,8-TCDD. environmentally available compound) of 1:2050 for Assuming this absorption-to-exposure ratio and even assuming that a soldier was directly sprayed (exposed to 8 yg/m2) for each day of his 1 year service 1n Vietnam, his cumulative Intake would be only 1.4 yg or 0.02 yg/kg of 2,3,7,8-TCDD. Based on these calculations, Stevens (1981) reports that 5 years of direct dally contact with Agent Orange would be necessary to reach a toxic level of 2,3,7,8-TCDD and feels that claims of Illness caused by 2,3,7,8-TCDD 1n Agent Orange are without merit. Exception Is made, however, for certain workers (forest exposed to 2,4,5-T and 2,3,7,8-TCDD for many years. C-68 Industries) who could be �Synerglsm and/or Antagonism 2,3,7,8-TCDD elicits diverse toxic and biologic effects and therefore can be expected to alter the activities of other chemicals. many compounds, Including 2,3,7,8-TCDO, which Induce For example, drug-metabolizing enzymes or act as cancer promoters, can greatly Influence the activity of other carcinogens and toxins. This type of Interaction can result In nonadditive effects which could be called synerglsm or antagonism. However, when the mechanism of action of the Interacting chemical Is different (such as Initiators and promoters of carclnogenesls), the Interaction effects should be called modulation. of a carcinogen. Thus a cancer promoter modulates the effects An example of a synerglstlc or antagonistic effect would occur when two chemicals that elicit the same toxic effect are coadmlnlstered and the resultant magnitude of the toxic response 1s non-additive. It 1s clear that 2,3,7,8-TCOD Interacts with chemicals via modulation and synerglsm/ antagonism as Indicated below. Results from several cocardnogenlclty studies appear to give only limited support to the modulating effect of 2,3,7,8-TCDO. The DBA/2N mouse strain, which responds only weakly to the sarcomatogenlc action of MC, becomes susceptible after treatment with 2,3,7,8-TCDD {Kourl, 1976; Kourl and Nebert, 1977). As an extension of this % study, Kourl et al. (1978) demonstrated 1n two Inbred strains of mice, C57BL/6Cum and DBA/2Cum, that administration of 2,3,7,8-TCDD simultaneously with MC appears to enhance the carcinogenic response. The authors concluded that their results suggest that 2,3,7,8-TCDD acts as a cocarclnogen, possibly as an Inducer of AHH at the site of Inoculation. In contrast, when mice were painted with 1 pg 2,3,7,8-TCDD prior to 7,l2-d1methylbenz(a)anthracene (OMBA) Initiation and 12-0-tetradecanoylphorbol-13-acetate (TPA) promotion, tumor formation was C-69 �inhibited (Berry et al.t 1979). The greatest degree of Inhibition (94%) was seen when the 2,3,7,8-TCDD was applied 5 days prior to Initiation by DMBA; when pretreatment was at 3 days and 1 day, the Inhibition was 86 and 3X, respectively. 2,3,7,8-TCDD did not promote the cardnogenlcHy of DMBA In a two-stage skin tumor 1 genes 1s assay with GDI female mice (Berry et al., 1979). 2,3,7,8-TCDD was applied twice weekly for 30 weeks at 0.1 wg 1n acetone to both DMBA treated and untreated control rats. No paplllomas were observed In either group. The recent results from the NCI cardnogenesls testing program (U.S. DHHS, 1980a) Indicate that In female Swiss-Webster mice the Incidence of flbrosarcoma In the Integumentary system 1s higher when 2,3,7,8-TCDD 1s applied alone or following DMBA application than In the control. The Incidence between the two experimental groups 1s comparable. This report has been discussed 1n detail In the carclnogenlclty section of this document. In order to test the potential of 2,3,7,8-TCOO as a promoter of hepatocarclnogenesls, rats which had received a single 10 mg/kg dose of dlethylnltrosamlne (DEN) following partial hepatectomy were given 2,3,7,8-TCDD (0.14 and 1.4 pg/kg s.c. once every 2 weeks) for 7 months. Animals which received (a) only a single Initiating dose of DEN after partial hepatectomy and no further treatment, or (b) 2,3,7,8-TCDD alone with no Initiating dose of DEN exhibited relatively few enzyme-altered fod and no hepatocellular carcinomas. However, animals Initiated with DEN and then given 2,3,7,8-TCDD had a marked Increase 1n enzyme-altered foci. At the higher dose of 2,3,7,8-TCDD, hepatocellular carcinomas were present In 5 of 7 rats. By means of three different enzyme markers used to evaluate the phenotypes of the enzyme-altered foci, a distinct phenotype heterogeneity of the fod was C-70 �noted with a shift towards phenotypes exhibiting a greater deviation from normal liver when 2,3,7,8-TCDD was given following DEN-partlal hepatectomy. QuantHatlon of the numbers of enzyme-altered fod was performed by relating measurements made from two-dimensional tissue sections to the number of foci per unit volume of liver using relationships established In the field of stereology. The total volume of the liver occupied by the enzyme-altered foci, but not their number, Increased with the dose of 2,3,7,8-TCDD administered following DEN-partlal hepatectomy. These studies demonstrate that 2,3,7,8-TCDD Is a potent promoting agent for hepatocardnogenesls (PHot et al., 1980). D1G1ovann1 et al. (1979) noted an Inhibitory effect of 2,3,7,8-TCDD on mice Initiated with benzo(a)pyrene [B(a)P] and promoted with TPA. Again, the greatest Inhibition of skin tumor formation (65%) was seen when 2,3,7,8TCDD was applied 5 days prior to B(a)P Initiation. Inhibition was 57 and 13% at 3 and 1 days pretreatment, respectively. D1G1ovann1 et al. (1979) and Berry et al. (1979) suggested that this antlcardnogenlc effect was related to the ability of 2,3,7,8-TCDD to Induce monooxygenase systems of the skin. 2,3,7,8-TCDD pretreatment has been observed to modify the effects of barbltuates and other xenoblotlcs (Grelg, 1972). Adult male Porten rats were given a single oral dose of 200 yg 2,3,7,8-TCDD/kg bw 1-3 days preceding treatment with 100 mg/kg zoxazolamlne hydrochlorlde or 150 mg/kg hexabarbltone sodium. 2,3,7,8-TCDD pretreatment resulted In a 54% decrease In the duration of the paralysis Induced by zoxazolamlne and a 2-fold Increase In the sleeping time produced by hexabarbltone. C-71 �The synerglstlc or antagonistic not been well documented. effects of chemical combinations have A recent report compares the Immunotoxlclty of 2,3,7,8-TCDD, 2,3,7,8-TCDF, and 2,3,7,8-TCDF plus 2,3,7,8-TCOO (coadmlnlstered) (Rlzzardlnl et al., 1983). Seven days after administration of 1.2 vg/kg of 2,3,7,8-TCDD to C578L/6J mice, sheep red blood cells were Injected by l.p. administration and plaque-forming cells (RFC) 1n the spleen were counted 5 days later. 2,3,7,8-TCDD Inhibited antibody production by 80%. In a parallel study, a dose of 2,3,7,8-TCDF was administered (10 yg/kg) and no significant Immunotoxlc effects were observed. 2,3,7,8-TCDO (1.2 yg/kg) plus 2,3,7,8-TCDF Coadmlnlstratlon of (10 yg/kg) resulted 1n 5054 reduction 1n antibody production and demonstrates a significant antagonistic effect by 2,3,7,8-TCDF. Coadmlnlstratlon of these two Isostereomers resulted In antagonistic effects with respect to the Induction of hepatic mlcrosomal cytochrome P-450 and 7-ethoxycoumar1n 0-deethylase. Sweeney et al. (1979) found that Iron deficiency protected mice against the development of hepatocellular damage (Including porphyrla) normally caused by 2,3,7,8-TCDD exposure. In contrast, the teratogenlc and fetoxlc data reported by Neubert and Dlllmann (1972) and Courtney and coworkers (see Teratogen'cHy section) suggests that Coadmlnlstratlon of phenoxy herbicides and 2,3,7,8-TCDD may also result 1n synerglstlc effects. Teratogenlclty Courtney et al. (1970a,b) were the first to report that 2,4,5-T was capable of causing teratogenlc effects 1n rats and mice. In these studies, rats and two strains of mice were exposed s.c. or orally to 2,4,5-T containing 30 ppm 2,3,7,8-TCDD. The mixture was teratogenlc and fetotoxlc to mice at >46.4 mg/kg. Rats were more sensitive, exhibiting fetotoxlc responses at 10 mg/kg for this 2,4,5-T/2,3,7,8-TCDD mixture. C-72 Since this Initial report, �research has focused on determining the role of 2,3,7,8-TCOD contamination 1n eliciting the teratogenlc response. These studies are summarized 1n Table 9. Neubert and Dlllmann (1972) conducted a detailed study to determine the significance of 2,3,7,8-TCOD contamination. three 2,4,5-T samples: These Investigators assayed a highly purified sample containing <0.02 ppm 2,3,7,8-TCOD (referred to as Sample A), a purified sample Identical to that used by Roll (1971) that contained 0.05+0.02 ppm 2,3,7,8-TCDO (Sample 8), and a commercial sample containing an undetermined quantity of 2,3,7,8-TCDD (Sample C). All three samples Induced cleft palates 1n NMRI mice at sufficiently high doses. In terms of the number of fetuses with cleft palate/the total number of fetuses, the dose/response pattern observed by Neubert and Dnimann (1972) was similar to that observed by Roll (1971) using a similar grade of 2,4,5-T. In addition to the three 2,4,5-T samples, Neubert and Dlllmann (1972) also assayed a sample of 2,3,7,8-TCDD alone and 1n various combinations with the highly purified sample of 2,4,5-T. This approach allows at least partial quantification of the significance of 2,3,7,8-TCDD contamination 1n 2,4,5-T-1nduced cleft palates. When the Utter 1s used as the basic experimental unit, the Incidences of cleft palate (number of Utters with cleft palate/total numbers of litters) versus the dose can be plotted on log dose/problt response paper, correcting for central response using Abbott's equation. According to this method, the ED for cleft palate Induction are: 2,3,7,8-TCDO: 2,4,5-T (Sample A): 2,4,5-T (Sample B): C-73 4.6 vg/kg bw 115 mg/kg bw 46 mg/kg bw (by eye-fH) �TABLE 9 Studies on the Potential Teratogenlc Effects of 2.3.7.8-TCDD-Contamlnated 2,4.5-T Species/Strain HIce/NMRI HIce/NHRI Vehicle Rape-seed oil Rape-seed oil Form of 2.4.5-T acid acid TCDO Level <0.02 ppm (Sample A) 0.05+0.02 ppn Dally Dose Treat- Obserment vation Maternal Response Day Days 8, 15. 30. 45. 6-15 Fetal Response Reference 18 No toxic effects; decreased maternal weight at doses of 90 mg/kg and greater Significant Increases In the Neubert and Incidence of cleft palates Dlllmann, 1972 at doses above 30 mg/kg (see text for additional details). Significantly decreased (p<0.005) fetal weight at all dose levels. 60. 90 and 120 mg,Ag 30. 60 and 90 mg/kg 6-15 18 No toxic effects; decreased maternal weight at 90 mg/kg Increases In the Incidence Same as above of cleft palate at 60 and 90 mg/kg; significantly decreased (p<0.005) fetal weight at all dose levels (Sample 8) HIce/NHRI Rape-seed oil acid NR (Sample C) 90 mg/kg 6-15 18 No toxic effects but decreased maternal weight Increase In the Incidence Same as above of cleft palate; significant (p<0.005) decrease In fetal weight HIce/NMRI Rape-seed oil butyl ester NR 12 and 17 mg/kg 6-15 18 No toxic effects Significant decrease In fetal weight but no effect on mortality; Increase In the frequency of cleft palate similar to that seen with acid (see text) H1ce/NHRI NR acid 0.05*0.02 ppm 20. 35. 60, 90 and 130 mg/kg 6-15 NR Toxic effects observed at 90 and 130 mg/kg Increases In the percentage Roll. 1971 of resorptlons and/or dead fetuses at 90 and 130 mg/kg; Increases In the Incidence of cleft palate and retardation of skeletal development at 35 mg/kg and above Mice/CD-I Corn o1l:acetone (9:1) acid <0.05 ppm 115 mg/kg 10-15 18 No significant effect on weight gain or I1ver-tobody weight ratios No effect on fetal mortality Courtney. 1977 or fetal weight but an Increase In the Incidence of cleft palate Same as above �TABLE 9 (cont.) Species/Strain Vehicle Form of 2.4.S-T TCOD Level Dally Dose Treat- Obserment vation Maternal Response Days Day Fetal Response Reference M1ce/C57BL/6 add 30 ppm 46.4 and 113 rag/kg 6-14 18 NR Significant (p<0.01) InCourtney creases In the Incidence of et al., 1970a.b cleft palate In the high dose group and cystic kidney In both dose groups; Increased fetal mortality also observed In the high dose group MIce/AKR o i Honey: water (1:1) Honey:water (1:1) acid 30 ppm 113 mg/kg 6-15 19 Increase 1n llverto-body weight ratio Significant (p<0.05) Increases In the Incidence of cleft palate and fetal mortality Same as above Rat/SpragueGavage/hydroxyDawley (groups propyl-methylof 25 rats) cellulose acid 0.5 ppm 1. 3, 6, 12 or 24 mg/kg/ day 6-15 20 No effect on body weight and no observable signs of toxlclty A slight but statistically significant (p<0.05) decrease In Implantations and litter size In lowest dose group only; no frank teratogenlc effects based on a detailed examination of the control and 24 mg/kg dose group; the only effect noted was an Increase In the Incidence of 5th partially ossified sternebrae Emerson et al.. 1970. 1971 This appears to be a full publication of the abstract summary by Thompson et al.. 1971 Rat/Wlstar Gavage/aqueous gelatin or corn oil acid <0.5 mg/kg 25. 50, 100 or 150 mg/kg/ day 6-15 22 Some maternal mortality and decreased body weight gain at 150 mg/kg; no signs of toxlclty at 100 mg/kg or below At 100 or 150 mg/kg decreased fetal weight. Increased fetal mortality and an Increase 1n the Incidence of skeletal anomalies; no significant effect at the two lower dose levels Khera and McKlnley, 1972; Khera et al.. 1971 Rat/Ulstar Gavage/aqueous gelatin or corn oil butyl ester <0.5 mg/kg 50 or ISO mg/kg/day 6-15 22 NR No significant effect on fetal mortality, fetal weight, or the Incidence of anomalies Khera and McKlnley. 1971; Khera et al.. 1971 �TABLE 9 (cont.) Species/Strain Vehicle Form of 2.4.5-T TCOO Level Dally Dose Treat- Obserment vatlon Maternal Response Days Day Fetal Response Reference Rat/Hoitzman Gavage/1:! solu- add tlon of honey and water 30 ppm 4.6. 10.0 and 10-15 46.4 mg/kg/day 20 NR Significant (p<0.01) InCourtney creases 1n fetal mortality et al.. 1970a.b at the two higher dose levels; dose related Increases In the percent of abnormal fetuses per litter; a high Incidence of cystic kidneys In treated groups Rat/CD Gavage/15X sucrose solution add 0.5 ppm 10.0, 21.5. 46.4 and 80.0 mg/kg/day 6-15 20 Reduced maternal weight gain at the two higher dose levels (p<0.05) and Increased I1ver-to-body weight ratio at the highest dose level (p<0.05) Increase 1n the Incidence of Courtney and ktdney anomalies, but no Moore, 1971 Increase In cleft palate Rat/Strain not specified Gavage/cnethocel acid 0.5 ppm 50 mg/kg 6-15 NS No effect on mortality or body weight gain No significant effect on fetal mortality or fetal weight; a significant (p<0.05) Increase In the Incidence of delayed ossification Sparschu Rat/Strain not specified Gavage/methocel Same as above o i acid Syrian hamster/ Gavage/acetone. add Hesocrlcetus corn oil and careuratus boxymethyl cellulose In ratio of 1:5.8:10 NS = Not specified; NR = not reported et al.. 1971a 0.5 ppm 100 mg/kg 6-10 NS Increased mortality and decreased body weight gain Increase 1n the Incidence of delayed ossification and poorly ossified or malallgned sternabrae (p<0.05) <0.1-4.5 ppm 20. 40, 80 and 100 mg/kg 6-10 14 NS Collins et al., Dose-related Increases 1n 1971 fetal mortality, gastrointestinal hemorrhages and fetal abnormalities; see text for discussion of effect of TCDD level on development �If the assumption were made that all teratogenlc activity 1n the 2,4,5-T samples were attributable to 2,3,7,8-TCDO contamination, the expected ED™ for Samples A and B would be 230,000 mg/kg (0.0046 mg/kg x 0.02 ppnT1) and 92,000 mg/kg observed ED (0.0046 mg/kg x 0.05 ppnf1), respectively. Since the was lower by a factor of over 1000, 2,3,7,8-TCDD appears not to be the sole factor 1n 2,4,5-T-1nduced cleft palate. The nature of possible Interaction between 2,4,5-T and 2,3,7,8-TCDD 1s more difficult to define. Based on assays of five mixtures of 2,3,7,8-TCDD and the highly purified 2,4,5-T, Neubert and DUlmann (1972) noted a greater than additive effect on the Induction of cleft palates. A similar conclu- sion can be reached If one assumes that Sample A was a "totally pure" sample of 2,4,5-T. According to the assumption simple similar action (Finney, 1971) and by treating Sample B as a mixture of 2,3,7,8-TCDD and 2,4,5-T, the expected ED for Sample B would be 119.8 mg/kg. The observed value of 46 mg/kg again suggests a greater than additive effect. A more detailed statistical analysis of these data, however, would be required to support the assumptions of simple similar action or Independent joint action that are Implicit In these analyses. Furthermore, the Inability to define precisely the levels of 2,3,7,8-TCDD 1n the 2,4,5-T samples and the possible « significance of other contaminants would preclude an unequivocal Interpretation of the results of the analysis. Nevertheless, three of the studies summarized 1n Table 6 {Neubert and DUlmann, 1972; Roll, 1971; Courtney, 1977) have demonstrated the Induction of cleft palate In mice by using 2,4,5-T samples containing 2,3,7,8-TCDD levels of 0.05^0.02 ppm or less. Although 2,3,7,8-TCDD contamination Is un- doubtedly a factor In the teratogenlc activity of 2,3,7,8-TCDD contaminated 2,4,5-T, the above analysis suggests that 2,3,7,8-TCDD contamination 1s not C-77 �the sole factor, and that some teratogenlc activity must be attributed to 2,4,5-T Itself or other contaminants 1n 2,4,5-T. Effects on reproductive success and fertility have also been studied 1n four groups of C57BL/6 male mice (25 animals/group) following oral Ingestlon of mixtures of 2,4-D, 2,4,5-T and 2,3,7,8-TCOD. Dally doses of -40 mg/kg 2,4-D, 40 mg/kg 2,4,5-T and 2.4 wg/kg 2,3,7,8-TCDD 1n Group II; 20 mg/kg 2,4-0, 20 mg/kg 2,4,5-T and 1.2 vg/kg 2,3,7,8-TCDD 1n Group III; and 40 mg/kg 2,4-0, 40 mg/kg 2,4,5-T and 0.16 animals were given. vg/kg 2,3,7,8-TCDD 1n Group IV Vehicle control animals 1n Group I had corn oil added to the feed. At the conclusion of an 8-week dosing period, treated animals were mated to untreated virgin females. No significant decrease In fertility, reproduction and germ cell toxldty were noted. Survival of offspring and the development of the newborns apparently were not affected (Lamb et al., 1980). Courtney and Moore (1971) tested a purified sample of 2,3,7,8-TCDD for teratogenlc potential. A summary of this study and others assessing the teratogenlc potential of purified 2,3,7,8-TCDD are presented 1n Table 10. CD-I, DBA/20 and C57B1/6J mice were given s.c. Injections of 2,3,7,8-TCDD at 1 or 3 vg/kg/day on days 6-15 of gestation 1n the study by Courtney and Moore (1971). This dose regime did not result 1n maternal toxldty, although an Increase 1n the maternal I1ver/bw ratio was observed In DBA/2J and C57B1/6J mice. 2,3,7,8-TCDD had no measurable effect on fetal mortality; however, anatomical abnormalities were observed 1n all strains and at all dose levels, with C57B1/6J being the most sensitive strain. The abnormalities observed were cleft palate and unspecified kidney anomalies. C-78 �TABLE 10 Studies on the Potential Teratogenlc Effect of 2,3.7.8-TCDD Dally Dose Treatment Days Observation Day Maternal Response Fetal Response 19a Increased liver/body weight ratio fetoddal. cleft palate, cystic kidney Courtney et al.. 1970b Increased liver/body weight ratio cleft palate, kidney anomalies Courtney and Moore, 18a none reported cleft palate, kidney anomalies Moore et al., 1973 7-16 18° Increased liver/ body weight ratio cleft palate, hydronephrotlc kidneys, hydrocephalus, open eyes, edema, petechlae Courtney. 1976 0.001. 0.01. 0.1. 1.0. 3.0 v9/kg 6-15 18a none reported cleft palate, dilated renal pelvis Smith et al., 1976 0.3. 3.0. 4.5. 6-15 18 no effect observed fetoddal at the high dose, cleft palate at doses at or above 5 vg/kg Neubert and Dlllmann, Species/Strain Vehicle House/C57BL/6 Mouse/AKR DMSO or honey : water (1:1) 21.5, 46.4. Mouse/CD-I Nouse/OBA/2J Mouse/C57Bl/6J DMSO 0.5. 1. 3 vg/kg Mouse/C57Bl/6 Acetone: corn oil (1:9) 1, 3 vg/kg 10-13 or 10 Mouse/CD-I DMSO or corn oil 25. 50. 100. Mouse/CF-1 corn oil: acetone (98:2) House/NHRI rape-seed oil 9.0 vg/kg 6-14 or 9-17 113.0 mg/kg 200. 400 vg/kg 6-15 17a or 18 o I ID Rat/CD Rat/SpragueDawley DMSO corn oil/ acetone 0, 0.5, 2.0 v9/kg 6-15. 9 and 10, I3 an(l 1* corn oil/ anlsole 0.0, 0.125. 0.25. 1. 2, 4. 8, 16 vg/kg 6-15 1972 none reported kidney malformations at both dose levels Courtney and Moore, 1971 20a vaginal hemorrhage at 2.0 and 8.0 vg/kg Intestinal hemorrhage at 0.125 and 0.5 vg/kg. fetal death at higher doses, subcutaneous edema Sparschu et al.. 1971b 22 maternal toxlclty observed at or above 1 ng/kg Increased fetal death observed at or above 1 ng/kg. subcutaneous edema and hemorrhages In the 0.25-2 groups Khera and Ruddlck. 1973 and 8.0 vg/kg Rat/Ulstar 1971 20a or 0. 0.03. 6-15 0.125. 0.5, 2.0 Reference �TABLE 10 (cont.) Species/Strain Vehicle Dally Dose Rat/Sprague- corn oil/ 0.1. 0.5. Dawley acetone Treatment Days Observation Day 2.0 vg/kg 1-3 21 diet decrease 1n body weight gain In the high dose group (9:1) Rat/SpragueDawley Maternal Response 0.001. 0.01 and 0.1 pg/kgc throughout gestation postparturition low fertility at 0.01 and 0.1 iig/kg decreased body weight at 0.01 and 0.1 v9/kg dilated renal pelvis Fetal Response Reference decreased fetal weight In the 0.5 and 2 ng/kg group Glavlnl et al.. 1982a low survival at 0.01 and 0.1 tig/kg, decreased body weight at 0.01. slight dilated renal pelvis at 0.001 Murray et al.. 1979 pg/kg 1n the F] but not succeeding generations Rabbit/ New Zealand corn oil/ acetone (9:1) 0.0. 0.1, 0.25. 0.5 and 1 yg/kg o i 28 maternal toxlclty at doses of 0.25 pg/kg and above 'First day of gestation designated day zero o 6-15 ''First day of gestation designated day one CD c The high dose level (0.1 pg/kg/day) was discontinued due to very low fertility 1n adults Increases In extra Mbs and total soft tissue anomalies Glavlnl et al.. 1982b �Moore et al. (1973) treated pregnant C57B1/6 mice with an oral dose of 2,3,7,8-TCOD at 1 or 3 ng/kg/day on days 10-13 of gestation, or 1 yg/kg on day 10 of gestation. At the high dose level, the average Incidence of cleft palate was 55.4%. Kidney anomalies (hydronephrosls) were observed on an average of 95.154 of the fetuses/Utter, with 83.1% having bilateral kidney anomalies. When the dose was decreased to 1 yg/kg/day, the average Incidence of cleft palate dropped to 1.9%; however, the Incidence of kidney anomalies remained relatively high, with an average Incidence of 58.9%. On the average, bilateral kidney anomalies occurred 1n 36.3% of the fetuses/ Utter. A single dose of 1 vg/kg on day 10 of gestation produced kidney anomalies 1n 34.3% of the fetuses; however, no cleft palates were observed. When C57B1/6 mice were treated with 1 vg/kg on day 10 of gestation and were then allowed to Utter, the detection of kidney lesions on postnatal day 14 was found to depend 2,3,7,8-TCDD-treated mother. largely on whether the pups nursed on a When pups from a 2,3,7,8-TCOO-treated mother nursed on control mice, kidney anomalies were found In only 1/14 Utters. In contrast, when pups from control mothers nursed on 2,3,7,8-TCDD-treated mice, kidney anomalies were observed In 4/14 litters. In the pups exposed to 2,3,7,8-TCDO both jji utero and during the postnatal period, kidney anomalies were observed 1n 5/7 litters. Kidney anomalies observed following Ui utero exposure or exposure through the milk were similar, and these kidney anomalies may not be considered a purely teratogenlc response. Neubert et al. (1973) reviewed what was known of the embryotoxlc effects of 2,3,7,8-TCDD 1n mammalian species. Also reported were their own studies and previous work (Neubert and Dlllmann, 1972) using NMRI mice, 1n which cleft palate was observed to be a common abnormality; however, no kidney C-81 �anomalies were reported. Neubert and Dlllmann (1972) administered 2,3,7,8TCDD by gavage to 20 female mice on days 6 through 15 of gestation at doses of 0.3, 3.0, 4.5 and 9.0 vg/kg. At day 18 of gestation, extensive reabsorptlon was observed 1n the high dose group with 6/9 Utters totally resorbed. In the few surviving fetuses, there was an 81% Incidence of cleft palate. At lower doses, there were 9 and 3% Incidences at doses of 4.5 and 3.0 vg/kg, respectively, and no cleft palates were observed In 138 fetuses examined In the 0.3 vg/kg group. Fetal mortality was Increased at the 9.0 vg/kg dose If animals were treated only on days 9 through 13; however, the Incidence of cleft palate remained high at a frequency of 60%. In a series of experiments to determine the time of gestation at which 2,3,7,8-TCDD was effective 1n Inducing cleft palate, mice were treated for a single day between days 7 and 13 of gestation with 2,3,7,8-TCDD at a dose of 45 vg/kg. A maximum number of Inducted cleft palates occurred when animals were treated on either day 8 or 11 of gestation, while exposure to 2,3,7,8TCOD after day 13 of gestation produced no cleft palates In the fetuses. Courtney (1976) compared the teratogenlc potential of 2,3,7,8-TCDD administered orally with 2,3,7,8-TCDD administered s.c. CD-I mice were dosed with 2,3,7,8-TCDD on days 7 through 16 of gestation at levels of 25, 50, 100, 200 or 400 vg/kg/day; the 400 vg/kg dose was not used 1n animals treated by s.c. Injection. Doses of 200 or 400 vg/kg/day produced vaginal bleeding and high rates of abortion. A dose of 100 vg/kg/day was fetotoxlc, resulting 1n decreased fetal weight and survival. Anatomic abnormalities were observed at all dose levels, with cleft palate and hydronephrotlc kidneys being most common. Other abnormalities observed Included hydrocephalus, open eye, edema and petechlae. Subcutaneous administration of 2,3,7,8-TCDD produced a greater teratogenlc response at a lower dose than C-82 �oral administration, with abnormalities observed 1n 87% of the fetuses following s.c. administration and 42% after oral administration of a dose of 25 yg/kg/day. The effects of 2,3,7,8-TCDD on the Incidence of fetal anomalies were also studied by Smith et al. (1976) In CF-1 mice. The mice were given 0.001-3.0 yg 2,3,7,8-TCDD/kg/day by gavage from day 6 through 15 of gestation. The Incidence of cleft palate was found to be significantly Increased In 1.0 and 3.0 yg/kg/day dose groups, and the Incidence of kidney anomalies was significantly Increased at 3.0 yg/kg/day. There were no observable teratogenlc effects In the study at 0.1 yg/kg/day; however, some were noted at lower dose levels, although not statistically significantly elevated. Poland and Glover (1980) compared cleft palate formation by 2,3,7,8-TCDD 1n the responsive C57BL/6J, the non-responsive B6D2F /J strains of mice. DBA/2J and the hybrid Female mice were mated with male mice of the same genetic strain and on day 10 of pregnancy the pregnant mice were given a single s.c. dose of 3.0, 10.0 or 30.0 yg/kg of 2,3,7,8-TCDD dissolved 1n p-dloxane or the solvent (control) alone (0.4 ma/kg). On day 18 the animals were killed and the number of cleft palates and resorbed fetuses was determined. At doses of 3.0 and 10.0 yg/kg of 2,3,7,8-TCDD, cleft palates (3% Incidence among live fetuses) were only observed 1n the C57BL/60 mice at the higher dose level. At a dose of 30 yg/kg, the Incidence of cleft palates among live fetuses for the C57BL/6J, B6D2F../J and DBA/23 mice was 54, 13 and 2%, respectively. This study also reported that cleft palate formation was significantly higher 1n several other responsive mouse strains compared with non-responsive mice. At a dose level of 30 yg/kg of C-83 �2,3,7,8-TCDD, the Incidence of cleft palates among live fetuses for the responsive C57BL/6J, A/J, BALB/cByJ and SEC/1REJ mice was 54, 73, 65 and 9554, respectively. The only responsive mouse (CBA/J) strain that was resistant to 2,3,7,8-TCOO-med1ated cleft palate was also resistant to the teratogenlc effects of cortisone. In contrast, the Incidence of cleft palates 1n the non-responsive DBA/2J, RF/J, AKR/J, SWR/J and 129/J mice was between 0-354 at the 30 vg/kg dose level. Thus the responsive mice, containing high levels of the Ah receptor, are highly susceptible to the effects of 2,3,7,8-TCOD 1n producing cleft palate, whereas the non- responsive mice, which contain low (or 0) levels of the Ah receptor protein, are resistant to this teratogenlc effect of 2,3,7,8-TCOO. These data and other results (Hassoun and Dencker, 1982) suggest that cleft palate formation elicited by 2,3,7,8-TCOD segregates with the Ah locus. In an early study, Courtney and Moore (1971) tested the teratogenlc potential of 2,3,7,8-TCDD 1n pregnant rats (CD) Injected s.c. on a dally basis with 2,3,7,8-TCDD (0.5 or 2 vg/kg) 1n dimethyl sulfoxlde on days 6 through 15, days 9 and 10, or days 13 and 14 of gestation. The only remarkable anomaly was kidney malformations In fetuses exposed to 2,3,7,8-TCDD. In the group exposed transplacentally at a dose of 0.5 yg/kg, 4/6 Utters had fetuses with kidney malformations (average number of kidney defects/ Utter was 1.8). An 11 and 3454 Incidence of kidney anomalies occurred 1n groups exposed to 2,3,7,8-TCDD on days 9 and 10, and 13 and 14, respectively. In addition, six hemorrhaglc gastrointestinal tracts were observed 1n the treated group (these data were not enumerated with respect to dose); however, this was considered a primary fetotoxlc effect of 2,3,7,8-TCDD and not a malformation. C-84 �2,3,7,8-TCDD was administered by gavage to groups (10-14 animals/group) of pregnant Sprague-Oawley rats at dose levels of 0, 0.03, 0.125, 0.5, 2.0 and 8.0 yg/kg/day on days 6 through 15 of gestation (Sparschu et a!., 1971b). No adverse teratogenlc effects were reported 1n fetuses exposed transplacentally at the 0.03 yg/kg level. At the 0.125 yg/kg level, three dead fetuses were reported, fetal weights were slightly depressed, and intestinal hemorrhage was noted In 18 of 127 examined fetuses. given doses of 0.5 vg/kg, In the group the number of viable fetuses was reduced, resorptlons were Increased, 6 dead fetuses were reported, and 36 of 99 fetuses suffered an Intestinal hemorrhage. In the 2.0 yg/kg group, only 7 live fetuses were reported (occurring In only 4/11 Intestinal hemorrhage. litters), 4 having Early and late resor-ptlons were prevalent. No live fetuses, but many early resorptlons, were reported 1n the group exposed to 8.0 yg 2,3,7,8-TCDD/kg/day. Subcutaneous edema appeared dose-related, occurring In a considerable number of fetuses from the higher dose groups. Male fetuses appeared to be more susceptible to 2,3,7,8-TCDD exposure; however, there was no significant difference In the sex ratio of live fetuses. Khera and Ruddlck (1973) tested a wide range of doses of 2,3,7,8-TCDD for teratogenlc and fetotoxlc potential. Groups of 7-15 Wlstar rats were Intubated with 2,3,7,8-TCDD at doses of 0.125, 0.25, 1, 2, 4, 8 and 16 * vg/kg on days 6 through 15 of gestation. At day 22 of gestation, there were no live fetuses 1n groups exposed to >4 vg/kg, and reduced Utter size was observed 1n the 1 and 2 yg/kg group. Unspecified maternal toxlc- 1ty was reported In all groups where there was fetal mortality. exposed In groups to 0.25-2 yg/kg, there were fetal anomalies observed as either gross or microscopic lesions consisting of subcutaneous edema of the head and neck, and hemorrhages 1n the Intestine, brain and subcutaneous tissue. C-85 �The Incidences of grossly observed lesions were 0/18, 2/11, 7/12 and 11/14 In the control, 1, 1 and 2 yg/kg dose groups, respectively (the study was conducted In two parts, and the 1 yg/kg dose was repeated). WHh regard to the other dose levels tested, the table enumerating the results had an entry of "not done". The Incidence of microscopically observed lesions for the control, 0.25, 0.5, 1, 1 and 2 yg/kg groups was 0/10, 3/10, 3/6 and 3/7, respectively. 1/33, 3/31, There were no effects of treatment observed 1n the 0.125 yg/kg group. Khera and Ruddlck (1973) also exposed dams to 2,3,7,8-TCDD at doses of 0.125, 0.25, 0.5 and 1 yg/kg on days 6 through 15 of gestation and allowed the dams to Utter and wean the pups. In this experiment, maternal toxlclty was reported 1n the 0.5 and 1 yg/kg group. At birth, there were fewer viable pups, and the pups had lower body weights In all but the 0.125 yg/kg group. At weaning on day 21 after birth, there were no surviving pups In the 1 yg/kg group, and 40% of the pups 1n the 0.5 yg/kg group did not survive. Fostering pups from dams exposed to 2,3,7,8-TCDD at 1 yg/kg onto control dams did not appreciably Increase survival, while fostering control pups onto dams exposed to 2,3,7,8-TCDD, did not Increase pup mortality. These data suggest that poor pup survival was a result from delayed toxlclty from in utero exposure to 2,3,7,8-TCDD. G1av1n1 et al. (1982a) assessed the effect of small doses of 2,3,7,8TCDD administered during the prelmplantatlon period 1n Sprague-Dawley rats. The animals. In groups of 20, were treated by gavage with 2,3,7,8-TCDD at doses of 0.0, 0.1, 0.5 and 2 yg/kg on days 1-3 of gestation. (The legends to the tables In this paper Indicated that the low dose was 0.125 yg/kg.) C-86 �At day 21 of gestation, no toxic effects were observed In the dams except for a decrease from 19.3-12.9 g In average maternal weight gain 1n the high dose animals as compared to controls. In the fetuses, weight was signifi- cantly reduced (p<0.05) In the 0.5 and 2 yg/kg groups. Malformed Utters and malformation/fetuses examined were 2, 5, 5 and 6, and 2/270, 8/260, 5/255 and 8/253, respectively, In the control 0, 0.1, 0.5 and 2 yg/kg groups; however, these Increases In the treated animals were not statistically significant. The anomalies observed were restricted to cystic kidney. This exposure to 2,3,7,8-TCDD early 1n pregnancy did not affect Implantation frequency, and the decrease 1n fetal weight was considered a result of 2,3,7,8-TCDD delayed Implantation. In a second study, Glavlnl et al. (1983) administered the same doses of 2,3,7,8-TCDD (0.0, 0.125, 0.5 and 2 yg/kg) dally to 15 female CRCD rats per group by gavage 1n corn o1l:acetone (9:1) for 2 consecutive weeks prior to mating. Females that did not become pregnant during three estrous cycles were necropsled to determine signs of toxlclty, while pregnant animals were allowed to proceed to day 21 of gestation at which time necropsies were performed with particular emphasis on reproductive organs and reproductive success. At the lowest dose tested (0.125 vg/kg), there were no overt clinical signs of toxlclty In the dams or adverse effects In any of the fetal parameters examined. maternal weight was decreased. At the 0.5 and 2 yg/kg levels, average Also, one animal 1n each of these groups did not become pregnant, although necropsy did not reveal any obvious dysfunctions. The only other overt sign of toxlclty was Ustlessness during the treatment period 1n the animals of the high-dose group. The only significant (p<0.01) fetal effect observed 1n the 0.5 yg/kg group was an Increase 1n postlmplantatlon losses from 2.9% 1n the control group to 10.2%. In the C-87 �high-dose group, there were decreases In corpora lutea and Implantations (averages of 17.654 1n control and 14.9% 1n treated animals, and 15.5% In control and 12.0% In treated animals, respectively), and Increases 1n both pre- and postlmplantatlon losses of 11.7% for controls and 19.5% (p<0.05) 1n treated animals, and 2.9% 1n control and 30.3% (p<0.001) 1n treated animals, respectively. In addition to these signs of fetal toxlclty, 9 of 10 litters 1n the high-dose group contained at least one malformed fetus as compared with 1/13, 2/13 and 2/13 1n the control, 0.125 and 0.5 yg/kg groups. The predominant fetal malformations were cystic kidney and dilated renal pelvis, which have been observed 1n other studies In which 2,3,7,8-TCOO was administered during gestation. The reproductive effects of 2,3,7,8-TCOO were also examined 1n a 3-generatlon study using Sprague-Dawley rats (Murray et al., 1979). Throughout the study, animals were continuously maintained on diets providing doses of 0, 0.001, 0.01 or 0.1 yg 2,3,7,8-TCDO/kg/day. The parental group (f ) was maintained for 90 days on the test diets prior to mating. The f rats were mated twice, producing the filial generations (f and f-ig)- Selected f,D and f_ rats were mated at -130 days of age to produce the ID f and f £ f. Utters, respectively. In later generations, the high dose O group (0.1 vg 2,3,7,8-TCDO/kg/day) was discontinued because few offspring were produced 1n this group. At the Intermediate dose (0.01 yg/kg/day), 2,3,7,8-TCOD caused lower body weights In exposed rats of both sexes (f and fp). At the low dose, no toxic effects were discerned. Fertility was greatly reduced 1n the fQ generation exposed to 0.1 yg 2,3,7,8-TCDD/kg/day. At 0.01 yg 2,3,7,8-TCDO/kg/day, was significantly (p<0.05) reduced In the f and f (of any generation) yg 2,3,7,8-TCDD/kg/day was not exposed to 0.001 C-88 rats. fertility Fertility In rats �different from that of control rats. Decreases In Utter size were noted 1n the f IA group exposed to 0.1 yg/kg/day and the f. and fJ litters t exposed at 0.01 yg/kg/day. Statistically significant decreases 1n fetal survival throughout gestation were noted in fC. and fsJ Utters of the 0.01 yg 2,3,7,8-TCDO/kg/day exposed dams. At 0.001 yg 2,3,7,8-TCOO/kg/ day, a decreased gestatlonal survival was reported for the f» Utters, but not for other generations. f and f I t\ pups exposed to 0.01 neonatal survival was noted among yg 2,3,7,8-TCDD/kg/day, but not among t f,R or f3 pups. 0.01 Decreased vg (0.001 Postnatal body weights of the f~ and f3 Utters at 2,3,7,8-TCDD/kg/day were significantly depressed. vg 2,3,7,8-TCDD/kg/day) necropsy At the low dose of 21-day-old pups revealed a statistically significant (p<0.05) Increase In dilated renal pelvis In the f generation. Subsequent generations at this dose level or any at the Intermediate dose (0.01 yg 2,3,7,8-TCDD/kg/day) did not have a significant Increase In this abnormality. Significantly decreased thymus weight and Increased liver weight were reported In the f O generation, but not In the f, generation (f,, generation data not obtained) of the Intermediate dose group. Murray et al. (1979) concluded that 2,3,7,8-TCDD Ingested at 0.01 or 0.1 yg/kg/day Impaired reproduction among rats, and NOAELs were associated with 0.001 yg 2,3,7,8-TCDD/kg/day. Nlsbet and Paxton (1982) reevaluated the primary data of Murray et al. (1979) using different statistical methods. From this revaluation It was concluded that 2,3,7,8-TCDD signifi- cantly reduced the gestatlonal Index, decreased fetal weight, and Increased I1ver-to-body weight ratios and the Incidence of dilated renal pelvis In both lower dose groups. Nlsbet and Paxton (1982) concluded that the dose of 0.001 yg/kg/day was not a NOAEL 1n this study. C-89 �A single report by Glavlnl et al. (1982b) describes the effects of exposure to 2,3,7,8-TCOD on fetal development 1n rabbits. Groups of 10-15 New Zealand rabbits were administered 2,3,7,8-TCDD by gavage at doses of 0.0, 0.1, 0.25, 0.5 and 1 wg/kg on days 6 through 15 of gestation. The dams were examined for Implantation sites, resorptlons, and live fetuses, and the fetuses were examined for malformations on day 28 of gestation. Decreased maternal weight gain and unspecified signs of maternal toxlclty occurred^ 1n dams exposed to 2,3,7,8-TCOD at doses of >0.25 yg/kg. At doses of 0.5 and 1 yg/kg, there were 2 and 4 deaths, respectively, among the dams. There were Increases 1n abortions and resorptlons at a dose of >0.25 yg/kg with no live fetuses detected 1n the high dose group. In the fetuses, the most common observation was a significant Increase In extra ribs from 33.3% 1n the controls to 82, 66.6 and 82% 1n the 0.1. 0.25 and 0.5 vg/kg dose groups. Although there was no significant Increase 1n specific soft tissue anomalies, there was an Increase from 0/87 to 3/78, 2/33 (p<0.05) and 2/28 (p<0.05) 1n total soft tissue anomalies 1n the control, 0.1, 0.25 and 0.5 yg/kg groups. The most prevalent soft tissue anomaly was hydronephrosls, which the authors point out was a common finding In rat fetuses exposed to 2,3,7,8-TCDD \n_ utero. These effects were considered to be signs of embryotoxldty rather than a teratogenlc effect. In addition to the fetotoxlc effects of prenatal exposure to 2,3,7,8TCDD, Norman et al. (1978) demonstrated that 2,3,7,8-TCDD could Induce liver mlcrosomal enzymes following in utero exposure. Pregnant New Zealand rabbits were given s.c. Injections of 2,3,7,8-TCOD at a dose of 30 nmol/kg (9.6 yg/kg) on day 24 of gestation, and the livers of newborns were exam- ined for enzyme activity within 12 hours after birth. While this treatment Increased the liver cytochrome P-450 levels In the adults ~2-fold, C-90 from �1.8-3.7 nmol/mg protein, the Increase In the newborns was ~5-fold, from 0.3-1.6 nmol/mg protein. SDS-polyacrylam1de gel electrophoresls revealed that 2,3,7,8-TCDD Induced a single form (form 6) of cytochrome P-450, and that this form was one of the two that were also Induced by 2,3,7,8-TCDD In the adult liver. The Identity of form 6 was confirmed by Immunologlc reaction and Us peptlde fingerprint. It was shown that Induction of cytochrome P-450 In newborns resulted In levels of benzo(a)pyrene hydroxylase and 7-ethoxyresoruf1n-0-deethylase activity similar to adult levels. The consequence to the newborn of these changes In the development of liver mlcrosomal enzymes has not been established. Dougherty et al. (1975) failed to find evidence for teratogenlclty or embryotoxldty In rhesus monkeys which were given on days 22-38 of gestation dally oral doses (1n gelatin capsules) of up to 10 mg/kg/day of 2,4,5-T containing 0.05 ppm 2,3,7,8-TCDD. The 2,3,7,8-TCDD dose at the highest dose level of 2,4,5-T administered (10 mg/kg/day) would correspond to 0.5 vg 2,3,7,8-TCDD/kg/day. However, 1t should be noted that palate closure In the monkey occurs on gestatlonal days 42-44 and the kidney 1s also a late developing organ. Adverse effects of exposure to 2,3,7,8-TCDD on reproductive success 1n monkeys have also been described. Schantz et al. (1979) fed a diet containing 50 ppt 2,3,7,8-TCDD to rhesus monkeys for 20 months. the study the female monkeys were bred to control males. Seven months Into There were four abortions and one stillbirth, two monkeys did not conceive even though they were mated repeatedly, and two monkeys carried their young to term. The total 2,3,7,8-TCDD Intake over the seven months was estimated by the authors to be 0.35 yg/kg, corresponding to a calculated dally dose of 0.0015 yg 2,3,7,8-TCDD/kg/day. C-91 �Allen et al. (1979) fed adult female rhesus monkeys on diets containing 50 or 500 ppt of 2,3,7,8-TCDD for 7 months. These exposure levels correspond to total doses per animal at the end of 7 months of 1.8 and 11.7 yg 2,3,7,8-TCDD, respectively. Although menstrual cycles were not affected 1n either treatment group, 5/8 animals 1n the high dose group had decreased serum estradlol and progesterone levels. low dose animals. Hormone levels were normal In the At 7 months, the females were bred to non-exposed males, and 6/8 and 3/8 females 1n the low and high dose groups, respectively, were Impregnated. Of the Impregnated animals, 4/6 and 2/3 had spontaneous abortions, while the remaining Impregnated animals had normal births. McNulty (U.S. EPA, 1980b) treated pregnant rhesus monkeys by gastric gavage to 2,3,7,8-TCDD 1n a vehicle of corn o1l:acetone solution. Grogp I animals were administered a total dosage of 5 yg/kg bw (two animals), 1 yg/kg bw (four animals) and 0.2 yg/kg bw (four animals) In nine divided doses, 3 times/week during weeks 4, 5 and 6 (days 20 through 40) after conception. Group II, consisting of 12 animals, received single doses of 1 vg/kg bw of 2,3,7,8-TCDD on days 25, 39, 35 and 40 after conception. Three animals were exposed In each of these 4 days. The vehicle control group, consisting of 11 animals, was treated with corn o1l:acetone only, on the same schedule as Group I animals. Both the females, who received the highest dose (5 yg/kg), had fetal losses. In the next lower dosed animals (1 yg/kg 1n both groups) 12 of 16 females had fetal losses; and 1n the lowest dosed animals (0.2 yg/kg 1n Group I) one abortion occurred 1n four pregnancies. females. Maternal toxIcHy was observed In many of these treated The difference 1n frequency of fetal loss between all pregnant animals given 1 yg/kg and the rate of historical abortion In the author's breeding colony was found to be significant. C-92 The author concluded that �short exposure to 1 yg/kg bw of 2,3,7,8-TCDD during early pregnancy results 1n fetal loss 1n rhesus monkeys and the results appear to be related to the adverse effects of 2,3,7,8-TCDD on the fetus (U.S. EPA, 1980b). A positive association between 2,4,5-T exposures and Increases 1n birth defects or abortions has been reported 1n human populations In Oregon (U.S. EPA, 1979), New Zealand (Hanlfy et al., 1981) and Australia (Field and Kerr, 1979). A lack of any such association has been reported In human popula- tions 1n Arkansas (Nelson et al., 1979), Hungary (Thomas, 1980), New Zealand (Dept. of Health, New Zealand, 1980; McQueen et al., 1977) and Australia (Aldred, 1978). Almost all of the reports are geographic correlation studies, and because of the uncertainties Inherent In this type of epidemlologic Investigation, as well as the difficulties 1n distinguishing the effects of 2,4,5-T from those of 2,3,7,8-TCDD contamination, none of the reportedly positive associations unequivocally Identify either 2,4,5-T or 2,3,7,8-TCDD as the causative agent. Similarly, the reportedly negative associations do not rule out 2,4,5-T or 2,3,7,8-TCDD as potential teratogens or abortlfadents 1n humans. Based on a report of a high Incidence of abortions In a small group of women living around Alsea, Oregon, who may have been exposed to the herbicide 2,4,5-T from aerial spraying (Smith, 1979), the U.S. EPA (1979) Initiated a study, often referred to as the "Alsea II study", to determine 1f spontaneous abortion rates differed between the exposed and unexposed population, If spontaneous abortion rates evidenced seasonal variation In these two groups, and 1f such seasonal variations were associated with 2,4,5-T spray application. Spontaneous Abortion Rate Index, as defined by the U.S. EPA, Is "basically the ratio of the number of hospitalized spontaneous abortions to C-93 �the number of births corresponding to the spontaneous abortions, based on the residence zip code of the women contributing to each event." Upon completion of the study, the EPA concluded that (1) the 1972-1977 Spontaneous Abortion Rate Index for the study area was significantly higher than In the Rural Control Area or the Urban area; (2) there was a statistically significant seasonal cycle In the abortion Index In each of the areas with a period of about 4 months. In particular there was an outstanding peak 1n the study area In June; and (3) there was a statistically significant correlation between-the Spontaneous Abortion Rate Index and spray patterns 1n the study area when a lag-time of 2 or 3 months was Included. The EPA concluded however, "This analysis Is a correlational analysis, and correlation does not necessarily mean causation." M1lby et al. (1980), citing three critiques of the Alsea II study that were not published 1n the open literature, state that the statistical method and basic design of the Alsea II study were sufficiently flawed to make this study of no use 1n human risk assessment. The Alsea II study has also been reviewed by a panel of epidemiologists who, In a published report of their meeting, also concluded that the basic design of the study was Inadequate to demonstrate either an effect or absence of an effect of exposure to 2,4,5-T (Coulston and Olajos, 1980). The major Inadequacies of the study were that the data collection methods were biased and would likely result In the underestimation of abortions, particularly 1n the urban area (the Incidence of abortions 1n all three groups was within the expected background rate of 8-15%); only a small portion of the area from which the exposed subjects were selected was actually sprayed with 2,4,5-T; and the study was not controlled for other factors such as age, smoking habits and alcohol consumption, which may affect the spontaneous abortion rate. Based on a new report C-94 �by SmHh (1979), the U.S. EPA Is attempting or has attempted to correlate 2,3,7,8-TCDD levels 1n the affected areas with the observed rate of abortion. No published reports have been encountered on the outcome of this effort. In the only other report encountered on a population 1n the United States, Nelson et al. (1979) noted a general Increase 1n the reported Incidence of facial cleft 1n both high and low exposure groups In Arkansas from 1948 to 1974. In this study, exposure estimates were based on average rice production In different areas of Arkansas, and the Incidence of cleft palate was determined by screening birth certificates and checking records of the Crippled Children's Services. No consistent exposure/effect correla- tions were noted, and the general Increase with time 1n the Incidence of facial clefts was attributed to better reporting procedures; however, there does not have to be a direct correspondence of malformations In human beings and experimental animals. Of the four reports available from New Zealand (Dept. of Health, New Zealand, 1980; McQueen et al., 1977; Hanlfy et al., 1981; Smith et al., 1982a), the report by the Dept. of Health 1s essentially anecdotal, Involving two women who gave birth to malformed children (one with an atrlal septal defect and a malformation of the trlcuspld valve of the heart and the other with biliary atresla). In both cases, exposure to 2,4,5-T could not be ruled out. Based on an analysis of spraying records, the time course of the pregnancies and plant damage near the women's homes, however, the Department of Health, New Zealand (1980) concluded that there was Insufficient evidence to Implicate 2,4,5-T spraying as a causative factor. Even 1f the spraying had been Implicated, a lack of Information on 2,3,7,8-TCDD levels In the spray and the absence of any monitoring data on 2,4,5-T or 2,3,7,8-TCDD would limit the usefulness of this report. C-95 �The study by McQueen et al. (1977) 1s not published 1n the open literature but Is summarized by M1lby et al. (1980). According to the summary, McQueen et al. (1977) "...examined the epidemiology of neural-tube defects In three areas In New Zealand and concluded 'there Is no evidence to Implicate 2,4,5-T as a causal factor 1n human birth defects."1 No additional details are provided. Hanlfy et al. (1981) performed an epldemlologlc study 1n Northland, New Zealand, 1n areas where spraying of 2,4,5-T was carried out by various companies for a number of years. The rate of birth defects was obtained from an examination of hospital records In seven mutually exclusive areas on a monthly basis over a period extending from 1959-1977. The rate of birth defects from 1959-1965 represented the rate for a non-exposed population since this was prior to the use of 2,4,5-T, while the Incidence of birth defects from 1972-1976 represented the rate for the exposed population. During the time of the survey there were 37,751 births, 436 stillbirths, 264 deaths shortly after birth, and 510 congenital anomalies. Three categories of birth defects, heart abnormalities, hypospadlas and eplspadlas, and talipes, had elevated rate ratios of >1 (p=0.05) In comparisons between the exposed (1972-1976) and control (1959-1965) populations. Exposure estimates were made for the seven areas and for different years using company records of aerial spraying and a model that factored 1n assumed fractional removal rates/month (this factor was assumed to be either 1.0 or 0.25). Comparisons of the rate of specific malformations with exposure demonstrated a statistically significant association between the occurrence of talipes and exposure when the fractional removal rate was assumed to be 0.25. There was, however, no statistically significant association where 1.0 was used as the fractional removal rate. C-96 �Smith et al. (1982a) Investigated the outcome of pregnancy 1n families of professional 2,4,5-T applicators and agricultural contractors Zealand. 1n New Agricultural contractors were chosen as the control population since both sprayers and contractors were of the same economic group with similar outdoor occupations. The survey was conducted by mall with 8954 of the chemical applicators responding and 83% of the agricultural contractors responding to questions asking whether they used 2,4,5-T and Us temporal relationship to reproductive histories regarding birth, miscarriages, stillbirths and congenital defects. The relative risks of congenital defects and miscarriages were 1.19 (0.58-2.4554 confidence limits) and 0.89 (0.61-1.30% confidence limits) for the wives of chemical sprayers as compared to the wives of agricultural contractors. These data Indicate that exposure of fathers and mothers (e.g., while cleaning clothes) had no effect on the outcome of pregnancy. Biases that may have affected the results, such as the age of the mother at childbirth, smoking habits and birth to Maori parents were Investigated and eliminated as possible confounders. The two reports from Australia (Aldred, 1978; Field and Kerr, 1979) also present apparently conflicting results. The report by Aldred (1978) Is not published In the open literature, but the following summary Is taken from MUby et al. (1980): "The report concluded that birth defects 1n a group of babies born 1n the [Yarram] district In 1974 and 1976 could not be attributed to exposure to 2,4,5-T or 2,4-D." Additional details that might be useful In assessing the rationale for this statement are not provided In the summary. The report by Field and Kerr (1979) plotted the Incidence of neural-tube defects (anencephaly and menlngomyelocele) 1n New South Wales, Australia, over the years 1965-1975, and the previous years usage of 2,4,5-T In all of Australia. The authors noted a decrease C-97 In the Incidence of �neural-tube defects expected on the basis of the plotted line In 1975 and 1976, when Australia Instituted monitoring of 2,4,5-T to ensure a 2,3,7,8TCOD level <0.1 ppm. The data were not tested for significance, although Field and Kerr (1979) Indicate that they consider the epldemlologlcal data on neural-tube defects to be "relatively complete," they do not comment on the Increasing Incidence of neural-tube defects with time and whether or not an Increase In the thoroughness of reporting neural-tube defects could have contributed to the apparent correlation of 2,4,5-T exposure with these defects. A visual replottlng of the data suggests that the Incidence of cleft palate correlates better wHh 2,4,5-T usage than with time. Nonethe- less, the appropriateness of correlating 2,4,5-T usage In all of Australia with the Incidence of -defects In one area of Australia 1s questionable. Thomas (1980) used an approach similar to that of Field and Kerr (1979) on data from Hungary. compared One major difference, however, 1s that Thomas (1980) the Incidence of stillbirths, cleft Up, cleft palate, splna blflda, anencephalus and cystic kidney disease In all of Hungary between 1976 and 1980 with 2,4,5-T use In 1975 1n all of Hungary. Because Hungary requires compulsory notification of malformations diagnosed from birth to age 1 year, because a relatively large percentage (55%) of the Hungarian population lives 1n rural areas where 2,4,5-T exposure may be expected to be greatest, and because annual use of 2,4,5-T 1n Hungary had risen from 46,000 kg In 1969 to 1,2000,000 kg 1n 1975, Thomas (1980) considered Hungary to be "...probably the best country 1n which to examine possible health effects of this herbicide." In any event, all Indices of birth defect rates decreased or remained stable over the period of study. In addition to contamination of 2,4,5-T being a potential source of 2,3,7,8-TCOD exposure, 2,3,7,8-TCOO 1s also an Inadvertant contaminant of C-98 �2,4,5-tMchlorophenol (TCP). Chronic exposure to 2,3,7,8-TCDO may occur during the manufacture of TCP and high level acute exposure to 2,3,7,8-TCDD has occurred after an accident 1n July, 1976 at the ICHESA TCP chemical factory In Seveso, Italy (Bonaccors! et a!., 1978). reaction used In this accident, the to produce TCP became uncontrolled, producing conditions favorable for 2,3,7,8-TCDD formation prior to venting the contents of the chemical reactor Into the atmosphere. The resulting cloud of chemicals settled over a heavily populated area. Although the amount of 2,3,7,8-TCDD released was not known, the reported cases of chloracne, a symptom of acute exposure to 2,3,7,8-TCDD, Indicated that exposure to 2,3,7,8-TCDD had occurred. Some preliminary results are available from epldemlologlc studies of reproductive events 1n the Inhabitants of Seveso, and recently a study has become available on the reproductive history of men employed In the chemical manufacturing Industry with possible chronic exposure to 2,3,7,8-TCDD (Townsend et al., 1982). Epldemlologlc studies to determine the reproductive effects In Individuals exposed to 2,3,7,8-TCDD and TCP following the accidental contamination of a populated area around Seveso, Italy, are not completed. The Incidence of spontaneous abortions occurring between March 1976 and January 1978 have been reported for Inhabitants 1n the area around Seveso by Bonaccorsl et al. (1978), Regglanl (1980) and B1sant1 et al. (1980). The spontaneous abortion rate 1n the contaminated area for the three trimesters following the accident was 13.1, 11.0 and 13.05%, which was similar to the worldwide 15-20% frequency of spontaneous abortion. Subdividing the contaminated area Into highly, moderately, and least contaminated, and examining the rates for each area Individually, also failed to demonstrate any change 1n the spontaneous abortion rate. The Incidence rates of malformations also were examined; C-99 �however, the numbers were too small for meaningful assessment. There are several reasons why these studies would not Indicate that the effect of 2,3,7,8-TCDO exposure 1n this accident had no effect on human reproduction. The authors note that there are many difficulties In Interpreting these data. The Incidence rates of spontaneous abortions and birth defects were not adequately available for the region prior to the accident as a result of suspected under-reporting. There was Inadequate reporting even after the accident due to political turmoil with regard to the management of health services. Also, an unknown number of pregnancies were surgically aborted for fear of 2,3,7,8-TCDO Induced birth defects. In a recent review of the progress of ep1dem1olog1c Investigations of the Seveso accident, Tognon! and Bonaccorsl (1982) Indicated that the data on spontaneous abortions and malformation rates still needed verification, and that these data were too preliminary to allow for conclusions. Townsend et al. (1982) Investigated the reproductive history of wives of employees potentially exposed to 2,3,7,8-TCOO during chlorophenol production. A total of 930 potentially exposed males were Identified who had worked for >1 month between January 1939 and December 1975 1n a job with potential 2,3,7,8-TCDD exposure. Exposure estimates of low, moderate and high were made by an Industrial hyglenlst primarily from job description and surface contamination data; however, the high potential exposure group was reserved for process workers during 1963-1964 when changes 1n operations resulted In a number of cases of chloracne. The control population was an equal number of male employees not Involved 1n any process that might cause exposure to 2,3,7,8-TCOO and matched for date of hire. In these groups, 586 wives were Identified and 370 agreed to participate as the exposed group, C-100 �while 345 wives In the control group agreed to participate. After Identification of the participants, a personal wives to determine pregnancy outcome. Interview was conducted with the Of the total of 737 conceptions 1n the exposed category and 1785 conceptions 1n the control category (conception which occurred 1n the exposed group prior to work records Indicating potential exposure to 2,3,7,8-TCOD were placed 1n the control group), there was no statistically significant Increase 1n spontaneous abortions, stillbirths, Infant deaths or selected congenital malformations. Sample sizes were too small to provide meaningful data 1f the populations were subdivided by extent of exposure. It was suggested that many confounding factors could account for these negative results, such as the Inappropriate selection of the populations, unidentified covarlables and Insufficient power; however, 1t was maintained that these results were consistent with animal data, which report that paternal exposure to 2,3,7,8-TCDO does not affect the conceptus. Poole (1983), 1n testimony before the House Committee on Science and Technology, described a re-analysis of the primary data used by Towsend et al. (1982). In this re-analysis, the rate of cleft palate and cleft Up were reported to be elevated by 1.9 (90% confidence Intervals of 1.0-3.6) In the years 1971-1974 for both the control and exposed groups (the comparison population was not described). At the same House Committee hearing, Houk (1983) presented data from the Birth Defect Monitoring Program of the Center for Disease Control on the yearly rate of cleft palate alone or cleft Up with or without cleft palate for births In Midland County, Michigan (the site of a chlorophenol production facility) during the years 1970-1981. The data Indicated an Increased rate for these defects of between 50 and 100% 1n the years 1971-1975, with the rate returning to expected from 1976-1981. C-101 �The observed Increase was only statistically significant 1f the rates for cleft palate alone and cleft Up with or without cleft palate were combined; however, It was the opinion of Houk (1983) that these defects should not be combined since the causal mechanism may be different. The Michigan Department of Public Health (1983a) also reported these results and, 1n addition, demonstrated that the same results occurred 1f the comparison was made with other counties In Michigan as well as with the general population of the United States. It was noted 1n this report that "runs" of Increases 1n oral ^ cleft for successive years have occurred In six other counties with no obvious potential for chemical exposure described. The Michigan Department of Public Health (1983a) Interpreted the data to Indicate that a more detailed case control study was necessary to determine 1f any common factors may exist, such as exposure to chemicals contaminated with 2,3,7,8-TCDD. Mutagenldty Short-term ^n vitro test systems have been developed to assess the biologic, toxic and genotoxlc effects of chemicals. These assays have proven to be useful Indicators of potential activity of diverse Industrial chemicals, a broad range of drugs and xenoblotlcs, carcinogens and crude environmental extracts. The most widely used short-term test system, the Ames test for bacterial mutagenesls, employs several strains of Salmonella typhlmurlum which are highly susceptible to the effects of mutagenlc chemicals. Despite the obvious utility of the Ames test and related short-term assays, their predictive capabilities (I.e., the correlation between bacterial mutagenlcIty and cardnogenlcHy) have not been fully assessed (Bartsch et al., 1982). Mutagenldty assays 1n microorganisms have been used to assess the genotoxlc effects of 2,3,7,8-TCDD; however, the results of most of these assays have Indicated little potential for mutagenlc effects (Table 11). C-102 �TABLE 11 The Results of MutagenlcUy Assays for 2.3.7,8-TCDO 1n Salmonella typhlmurlum Strains of Salmonella typhlmurlum Type of Assay Reference S-9 TA1535 TA1537 TA1538 TA1532 TA1950 TA1975 TA1978 G46 TA100 TA1531 TA1534 +/- NT NT 0 0 0 0 NT NT NT NT NT NT NT McCann. 1978 Plate Incorporation +/- NT NT 0 0 0 0 NT NT NT NT NT NT NT McCann. 1978 Plate Incorporation* +/- 0 0 0 0 0 0 0 0 0 0 0 NT NT Gilbert et al., 1980 Fluctuation test CO TA1S30 Spot test i o TA98 +/- 0 0 0 0 0 0 0 0 0 0 0 NT NT Gilbert et al., 1980 NT 0 NT NT NT t NT NT NT 0 NT QR QR Seller, 1973 0 NT 0 0 0 NT NT NT NT NT 0 NT NT Gelger and Neal , 1981 Plate Incorporation NT NT NT 0 NT NT NT NT NT NT NT NT NT Gelger and Neal , 1981 Suspension assay NT 0 NT NT NT * NT NT NT NT NT NT NT Hussaln et al.. 1972 0 NT 0 0 NT NT NT NT NT NT 0 NT NT Zelger. 1983 Spot test Plate Incorporation Suspension assay + +/- *The assay was performed under both aerobic and anaerobic conditions. NT = Not tested; QR * Questionable response �Hussaln et al. (1972) exposed Salmonella typhlmuMum histldene-dependent strains TA1530 and TA1532 1n liquid suspension to plating Into selective medium to observe 2,3,7,8-TCDD followed by reversion to prototypes. No Increase In the reversion rate was observed with strain TA1530 at exposure levels of 1 and 10 yg/ml. These exposures resulted 1n cell survivals of 90 and <1%, respectively. In strain TA1532 Increased reversion frequency was not observed resulted 1n a 0-50% at 2,3,7,8-TCDD concentrations of 2-3 yg/mJ., which decrease In survival; however, at 2,3,7,8-TCDD levels which resulted In a 9954 decrease In survival, there was an Increased number of revertant colonies/surviving cells. The dose levels were not specified. The source of the 2,3,7,8-TCDD sample studied 1n this paper was the Food and Drug Administration, and Its reported purity was 99%. Also, Seller (1973) observed a positive mutagenlc response In a spot test of 2,3,7,8-TCDD performed 1n the absence of a metabolic activation system. of the sample studied was not provided. However, the purity In tester strains G46 and TA1530, the ratio of revertants/108 cells In the treated plates divided by spontaneous revertants/108 cells was <1. In strains TA1531 and TA1534, the ratio was between 1 and 2, which was considered a "doubtful" mutagenlc response, while In strain TA1532, the ratio was >10. of the 2,3,7,8-TCDD levels tested In this assay. There was no mention The positive controls, dlethylsulfate, 2-am1nopur1ne and 2-amlnofluorene, produced ratios of 2 to 5, <1 and 5 to 10, respectively, 1n strain TA1532. In both the study by Hussaln et al. (1972) and the study by Seller (1973), 2,3,7,8-TCDD produced a positive mutagenlc response only 1n the S. typh1mur1um strain TA1532, which Is sensitive to framesnlft mutagens. Hussaln et al. (1972) also performed a mutagenldty test of 2,3,7,8-TCDD In two other m1crob1al test systems. A positive response was observed In C-104 �Escherlchla coll Sd-4 as Indicated by a reversion to streptomycin Independence. In this assay, cells were treated 1n suspension for 1 hour with 2,3,7,8-TCOD at 0.5-4 yg/msi. mutants x 10~8, as compared 10~B) occurred The greatest mutation frequency (256 to the control frequency of 2.2 mutants x at a dose level of 2 yg/mj.. The absolute number of colonies/plate was 7 for the control and 46 for the treated plate. The dose of 2 vg/mSl caused an 89X decrease In cell survival. In the second test system, the ability of 2,3,7,8-TCOD to Increase prophage Induction 1n E_. coll K-39 cells was examined. The vehicle control, DMSO, Inhibited prophage Induction as compared to the untreated controls, while the most effective dose level of 2,3,7,8-TCOD (0.5 yg/ma) resulted In an Increased prophage Induction as compared to vehicle control but not as compared to the untreated controls. Hussaln et al. (1972) concluded that 2,3,7,8-TCDD was capable of causing Increases 1n the reverse mutation rate In E_. coll Sd-4 and that 2,3,7,8-TCDO had a weak ability to Induce prophage In E_. coll K-39 cells. The studies which followed these two early reports of Hussaln et al. (1972) and Seller (1973) failed to detect mutagenlc activity of 2,3,7,8-TCDD In S. typhlmurlum. Wasson et al. (1978) reported on a personal communica- tion rrom McCann (1978) that 2,3,7,8-TCDD was Inactive In both the spot test and plate Incorporation assay with S. typhlmurlum strains TA1532, TA1535, TA1537 and TA1538. Doses and other experimental protocols were not men- tioned except that the tests were performed both with and without metabolic activation. Gilbert et al. (1980) reported that 2,3,7,8-TCDD gave "substan- tially negative results" with S. typhlmurlum strains TA98, TA100, TA1530, TA1535, TA1537, TA1538, G46, TA1532, TA1950, TA1975 and TA1978. Both the standard plate Incorporation assay and the bacterial fluctuation test were used, and both were performed with and without S-9 prepared from the livers C-105 �of Aroclor 1254 pretreated rats. In the plate Incorporation assay the test compound was tested at 1-2000 yg/plate under both aerobic and anaerobic conditions. Details were not provided for the fluctuation assay. It 1s difficult to assess possible reasons for the conflicting results between the earlier studies and these later mutagenldty assays, since Information on experimental conditions was limited 1n the negative studies. In an attempt to resolve the conflicting results and observe a mutagenlc response, Gelger and Neal (1981) tested 2,3,7,8-TCDD 1n the standard plate Incorporation assay using S-9 prepared from different sources. maximize the amount of compound tested, dloxane, a better solvent for 2,3,7,8-TCDD than the commonly employed DMSO, was used. of dloxane, the limited In order to solubility Even with the use of 2,3,7,8-TCDD allowed only 20 yg/plate to be tested, a dose which was shown to be non-toxic to the cells. The S-9 used 1n these assays was prepared from the livers of Aroclor 1254 pretreated male Sprague-Dawley rats and male Golden Syrian hamsters, and from 2,3,7,8-TCDD Induced male hamsters. In all assays at 2,3,7,8-TCDD concentrations of 0.2, 2, 5 or 20 yg/plate, and regardless of the source of the S-9, there was no observed mutagenlc response. In further attempts to duplicate the previous positive results, Gelger and Neal (1981) tested the same concentrations of 2,3,7,8-TCDD 1n strain TA1537, a more sensitive direct descendent of strain TA1532, for mutagenlc activity 1n the absence of S-9. assays Again, no Increase In the number of revertants was observed. either with or without S-9, In positive controls had predictable Increases In the number of revertant colonies. The authors concluded that 2,3,7,8-TCDD was not active under the conditions of this assay; however, testing at higher concentrations may elicit a positive response. C-106 It was �also noted that many other polychlorlnated aromatic compounds are not mutagenlc 1n the Ames test, even though there 1s positive evidence of cardnogenldty. Mutagenlc effects of 2,3,7,8-TCDD 1n yeast were observed by Bronzett! et al. (1983). Positive results for reversion and gene conversion obtained jm vitro and 1n the host-mediated assay. yielded small revertants. dose-related Increases 1n trp were The in vitro experiments convertants and 1lv An S10 metabolic activation system was required. Exposure of the yeast to 2,3,7,8-TCDD at the highest level tested {10 yg/msi) resulted In 16% survival and yielded 4-fold Increases In reversion and gene conversion. In the host-mediated assay, male mice were exposed to 25 pg of 2,3,7,8-TCDD/kg (Bronzettl et al., 1983). After 5, 10, 20 or 30 days, 0.2 mi of a yeast culture (4xl08 cells) was Instilled retroorbltally. Four hours later, the liver and kidneys were removed and the yeast cells 1n these organs were assayed for mutagenlc responses. Increases (4- to 6-fold) In reversion and gene conversion were observed 1n yeast cells obtained from the livers and kidneys. The toxic response of the animals to an exposure of 25 yg/kg was not described 1n this report. this high dose. Toxlclty should be expected at The positive results described 1n this paper may suggest that 2,3,7,8-TCDD 1s mutagenlc 1n yeast, but more definitive studies are needed before a firm conclusion can be drawn. Hay (1982) has found that 2,3,7,8-TCDD dissolved 1n DMSO transformed baby hamster kidney cells (BHK) In vitro. The dloxln Isomers 2,8-d1chloroand 1,3,7-tr1chlorod1benzo-2-d1ox1n also transformed BHK cells but the response was weak. The unchlorlnated d1benzo-2-d1ox1n and the fully chlorinated octachlorod1benzo-p_-d1ox1n were both negative In the BHK assay (I.e., C-107 �there was no cell transformation). More recently, Rogers et al. (1982) reported that 2,3,7,8-TCDD Induced mutations In the excess thymldlne, thloguanlne and methotrexate selective systems 1n L5178Y mouse lymphoma cells 1n culture. The National Toxicology Program from four assay systems: (Zleger, 1983) provided data on TCOO the S. typhlmurlum (strain TA98, TA100, TA1535 and TA1537) h1st1d1ne reversion assay, the sex-linked recessive lethal test In DrosophHa. and cytogenetlc studies (sister chromatld exchange and chromosome aberrations) 1n Chinese hamster ovary cells. obtained In all these assays. Negative results were However, these studies cannot be evaluated because the procedures used to obtain the data were not described. The solubility of 2,3,7,8-TCDD 1n water 1s only 0.2 yg/a "(Grummett and Stehl, 1973). Therefore, negative In vitro results must be viewed with caution unless precise descriptions of the preparation of each test sample are supplied (e.g., were the samples predlssolved and, If so, In what solvent). in vitro reactions of 2,3,7,8-TCDD with bacterlophage QB RNA were evaluated by Kondorosl et al. (1973). Active RNA was purified from QB phage followed by Incubation for 1 hour at 37°C with 0.0, of 2,3,7,8-TCDD. 0.2, 2.0 or 4.0 pg/ms. At all concentrations tested, 2,3,7,8-TCDD had no effect on the transfect1v1ty of QB RNA. Other compounds tested Included the alkylatlng agents methyl, ethyl and Isopropyl methanesulfonate, and dlethyl pyrocarbonate, all of which Inactivated QB RNA under the same experimental conditions. The authors suggested that 2,3,7,8-TCDD Inactivity In this assay Indicated that 2,3,7,8-TCDD was an Intercalating agent, and hence would require double stranded DNA 1n order to Interact. The data presented 1n this study, however, were Insufficient to support this conjecture. C-108 �In. vivo binding of radlolabeled 2,3,7,8-TCDD to liver macromolecules was studied 1n Sprague-Dawley rats by Poland and Glover (1979). Both male and female animals were administered [1,6-3H]2,3,7,8-TCDD by 1.p. Injection at a dose of 7.5 yg/kg. mCl/kg. This dose corresponded to a tritium level of 0.87 The animals were killed either 12, 48 and 168 hours after treat- ment, or 24 hours after treatment when the animals were pretreated with the enzyme Inducers phenobarbltal or unlabeled 2,3,7,8-TCDD. Following sacri- fice, Isolation of macromolecules, and removal of free labeled 2,3,7,8-TCDD, the amount of label bound to protein, RNA and DNA was determined. greatest non-extractable binding of labeled 2,3,7,8-TCDD occurred The to protein; however, the amount of label bound was small and only amounted to 0.03-0.IX of the total radioactivity administered. The total amount of label associated with RNA and DNA was, respectively, only 50 and 4 cpm above background. Time after exposure, sex, or prior enzyme Induction had no significant effect on 2,3,7,8-TCDD binding. As a result of the extremely low levels of radioactivity associated with RNA and DNA, H 1s uncertain whether 2,3,7,8-TCDD truly binds covalently to these macromolecules and 1f so, whether there 1s any biological significance to this low level of apparent binding. The effects of 2,3,7,8-TCDD exposure on the extent of chromosomal aberrations In the bone marrow of male rats were reported In an abstract by Green and Moreland (1975). In the Initial experiment, no Increase 1n chromosomal aberration was observed after 5 dally gavage treatments at a 2,3,7,8-TCDD dose of 10 yg/kg. In the second portion of this study, rats were exposed by a single 1.p. Injection of 2,3,7,8-TCDD at 5, 10 or 15 yg/kg or a single gavage treatment at 20 yg/kg. The animals at the two highest exposure levels were killed 24 hours post-treatment, C-109 while the �remaining animals were killed 29 days post-treatment. Again, no Increase 1n chromosomal aberrations was observed, except 1n the positive control group exposed to trlethylenemelamlne. In a later report, a small but significant Increase In chromosomal aberrations was observed 1n the bone marrow cells of male and female OsborneMendel rats (Green et al., 1977). Bone marrow cells for cytogenetlc analysis were obtained from Osborne-Mendel rats used In a range-finding study preliminary to a chronic bloassay (Green et al., 1977). The animals 1n groups of 8 males and 8 females received twice weekly Intubations of 2,3,7,8-TCDD at respective doses of 0.25, 2.0 and 4.0 pg/kg for 13 weeks. 1.0, 2.0 and 4.0, or 0.25, 0.5, Because It was not required for the range-finding study, a control group was not Included. Bone marrow cells were analysed for abnormalities and cells 1n mitosis In the animals which survived to the end of the study (4-8 animals/group). Increases In chromosomal The only significant aberrations In comparison to the low dose group were In males at 2 and 4 yg/kg and females at 4 yg/kg. The greatest Incidence observed was 4.65% of the cells with chromosomal breaks 1n the high-dose males, and this was considered only weakly positive. The weak response, as well as the lack of data from control animals and the reported difficulty of obtaining cells from the high-dose animals as a result of 2,3,7,8-TCDD toxldty, makes the conclusion from this study that 2,3,7,8TCDD produced chromosomal breaks tenuous. Czelzel and Klraly (1976) reported an Increased Incidence (p<0.001) of chromatld-type and unstable chromosome aberrations 1n the peripheral lymphocytes of workers exposed to the herbicides 2,4,5-trlchlorophenoxyethanol (2,4,5-TCPE) and Bumlnol. The 2,3,7,8-TCDO levels In the final product were <0.1 mg/kg; however, the exposure levels for Individual workers were not available. C-110 �Hulcahy (1980) reported In a letter no Increased Incidences of chromosomal aberrations 1n lymphocytes of 15 soldiers exposed to "Agent Orange". The exposure was for 6-15 months and all subjects complained of symptoms, Including skin eruptions, which they associated with "Agent Orange". The analyses were performed with lymphocytes obtained -10 years after the last exposure, and comparisons were made with eight subjects who had no history of exposure to 2,3,7,8-TCOD. Neither sister chromatld exchange nor structural aberrations Including both gaps and breaks were Increased. The authors note that the long time between exposure and analysis may have accounted for the negative results. Also, both Regg1an1 (1980) and Mottura et al. (1981) have studied Inhabitants In Seveso, Italy, exposed to 2,3,7,8-TCOO from an accident 1n a trlchlorophenol manufacturing plant. Regglanl (1980) examined 4 adults and 13 children (3-13 years) for chromosomal aberrations within 2 weeks of the accident. These 17 Individuals were examined to support claims of, and determine extent of, Injury before an Inquest Judge. Although burnllke skin lesions 1n these 17 Individuals Indicated chemical exposure, no Increase 1n chromosomal aberrations was detected. The methods of performing the analyses and the actual number of aberrations detected were not described. Similar negative results were reported In an abstract by Mottura et al. (1981). In this study, subjects were chosen from the area of heavy contamination following the accident (acute high level exposure), from the working population of the plant (chronic low level exposure) and a non-exposed control population. The number of subjects 1n each group was not enumerated. The specimens were examined by three Independent laboratories and no laboratory reported an Increase 1n chromosomal aberrations, although there was a significant difference In the reported scores between laboratories. C-lll There �was no Information In this abstract on the extent of Individual exposure or the length of time that elapsed between the accident and obtaining samples for analyses of chromosomal aberrations. DILernla et al. (1982) conducted additional studies on lymphocytes prepared 1n 1976 and 1979 from 8 persons considered acutely exposed to 2,3,7,8-TCDD 1n the Seveso accident, 8 ICMESA factory workers (considered chronically exposed), and 14 control subjects (8 had chromosomes prepared In 1976 and 6 1n 1979). Cells were examined for average number of satellite associations (SAs) (evidence for functional Mbosomal genes), both on a cell basis and for the large acrocentrlc chromosomes (D group chromosomes). There was no change In the frequency of SAs on a per cell basis In any of the groups as compared to control values, nor 1n D group chromosomes from acutely exposed subjects examined Immediately after the accident. There was, however, a decrease 1n the average frequency of SAs 1n group D chromosomes of acutely exposed subjects examined 1n 1977 and In ICMESA workers at both the 1976 and 1979 examinations. Although the biologic relevance of these observations has not yet been confirmed, Dllernla et al. (1982) observed a similar decrease 1n SAs after exposure of lymphocytes to x-1rrad1atlon. It was concluded that the decrease In SAs may have resulted from mutagenlc damage to functional nucleolar organizing regions. The potential of exposure to 2,3,7,8-TCDD to result In chromosomal damage has been studied In experimental animals and humans. Most of the studies 1n experimental animals gave no evidence that 2,3,7,8-TCDD may result 1n chromosomal aberrations; however, there 1s a report of a single positive response which was weak and little detail was provided In the report to assess the quality of the results. In the studies of humans, exposure occurred to chemicals which contain 2,3,7,8-TCDD as a contaminant. C-112 �In two of these studies Involving Individuals exposed 1n the Seveso accident, there was no observed Increase In the Incidence of chromosomal aberrations. In a third report of Individuals exposed at Seveso, there were changes observed 1n lymphocyte chromosomes from exposed workers which were suggested to have risen from mutation 1n functional nucleolar organizing regions; however, this bloassay has yet to be validated. In the only positive study, workers 1n a chemical plant were exposed to the herbicides, 2,4,5-TCPE and Bumlnol, as well as 2,3,7,8-TCOO. The participation of the herbicides 1n the resulting Increase In the workers of chromosomal aberrations cannot be excluded. At present, the data from experimental animals and humans are too limited to designate 2,3,7,8-TCOD as a clastogenlc agent. In summary, a limited number of Initial studies of the mutagenlclty of 2,3,7,8-TCDO In bacteria reported positive results 1n S. typhlmurlum strain TA1532 In the absence of a mammalian metabolic activation system (Hussaln et a!., 1972; Seller, 1973). More recent attempts to repeat these results with strain TA1532 or related strains have failed (Gelger and Neal, 1981; Gilbert et a!., 1980; HcCann, 1978). These authors have also reported no Increase In mutation rate when 2,3,7,8-TCDD was tested 1n the presence of a mammalian metabolic activation system. In other 1_n vitro assays, 2,3,7,8-TCDD has produced a positive response 1n reversion to streptomycin Independence 1n £. coll Sd-4 cells and questionable positive response with prophage Induction In £. coll K-39 cells (Hussaln et al., 1972). Also, 2,3,7,8-TCDD has been reported to be mutagenlc 1n the yeast S_. cerevlslae 1n both the In vitro assay with S-10 and the host-mediated assay (Bronzettl et al., 1983). Rogers et al. (1982) have also reported positive mutagenlclty results 1n the mouse lymphoma assay. In the E_. coll studies, the poor survival of the cells or the Interference of the vehicle solvent, DMSO, with the assay makes C-113 �the evaluation of the studies difficult. With the data available, It 1s not possible to resolve the conflicting reports on the mutagenlc potential of 2,3,7,8-TCDD. Overall, the data Indicate little potential for the Interaction of 2,3,7,8-TCDD with nucleic adds or the ability of 2,3,7,8-TCOD to produce chromosomal aberrations. Kondoros! et al. (1973) demonstrated that 2,3,7,8- TCOD did not react with RNA jin. vitro In the absence of a metabolic activation system. .In vivo studies using radlolabeled 2,3,7,8-TCDD Indicated some association of non-extractable label with RNA and DNA (Poland and Glover, 1979); however, the very low level of bound label observed suggest that the "binding" may have been merely an artifact. Similar marginal data were available on the clastogenlc effect of 2,3,7,8-TCDD. Although one \n_ vivo study 1n rats (Green and Moreland, 1975) failed to demonstrate any treatment-related chromosomal aberration, a second study by the same authors (Green et al., 1977) using a longer exposure period reported a small Increase 1n the number of aberrations. In humans exposed to 2,3,7,8-TCDD during the manufacture of 2,4,5-TCPE and Bumlnol, Czelzel and Klraly (1976) reported an Increase 1n the number of chromosomal aberrations, while no Increase was detected 1n Individuals exposed to 2,3,7,8-TCDD following an Industrial accident 1n Seveso, Italy (Regg1an1, 1980; Mottura et al., 1981). The studies of the clastogenlc effect of 2,3,7,8-TCDD were presented with little or no experimental detail to assist 1n evaluating the merits of the reports. 2,3,7,8-TCDD The data available are too limited to Indicate whether can Interact with nucleic adds or produce chromosomal aberrations. The differences among the results described above could be due to several factors, such as treatment protocols, solubility problems, purity of C-114 �the samples tested and the high toxlclty of 2,3,7,8-TCDD. This chemical may be a weak mutagen, but because It Is very toxic, the dose range for detecting a positive genetic effect may be very narrow. experimentation made. Is necessary Therefore, additional before any conclusive determination can be Suggested further testing Includes the ability of 2,3,7,8-TCDD to Induce forward mutations In mammalian cells In culture, additional yeast and bacterial studies and the sex-linked recessive lethal test 1n Drosophlla. Cardnogenlclty Ep1dem1oloq1ca1 Studies Case Reports. Observations of an unusual occurrence of relatively rare soft-tissue sarcomas were first made by Hardell (1977). Of some 87 patients seen from 1970-1976 at the Department of Oncology, University Hospital, Umea, Sweden, seven Individuals with soft-tissue sarcomas were Identified. All seven had had occupational exposure to phenoxy acids 10-20 years earlier. The tumors were 2 lelomyosarcomas, 1 Uposarcoma, 1 rhabdomyosarcoma, 1 myxofIbrosarcoma and 2 additional sarcomas of which the hlstopathology was uncertain but was probably a neurofIbrosarcoma In one and a rhabdomyosarcoma In the other. The clustering of this rare tumor type among these patients prompted the author to suggest that ep1dem1olog1cal studies be done to determine If exposure to phenoxy acids and the Impurities they contain are related to the occurrence of soft-tissue sarcomas. Zack and Susklnd (1980) reported the finding of a soft-tissue sarcoma death 1n a cohort study of workers exposed to 2,3,7,8-TCDD 1n a trlchlorophenol process accident In NHro, West Virginia. This tumor, a fibrous hlstlocytoma, was noted by the author as a rare event. This study, referred to as the N1tro study, 1s discussed later. C-115 �Cook et al. (1980) In a cohort mortality study of 61 male employees of a trlchlorophenol manufacturing area, who acquired chloracne following a 1964 Incident, noted four deaths by the end of his study period, but one of the four was a flbrosarcoma. The authors did not seem to attribute any special significance to this finding at the time. Ott et al. (1980) 1n a cohort mortality study of 204 employees exposed to 2,4,5-T during Its manufacture from 1950 to 1971, revealed no soft-tissue sarcomas among 11 deaths that had occurred by 1976. But only 1 of these 11 was a malignant neoplasm. In a discussion of the cohort studies of Zack and Susklnd, Cook, a third unpublished study by Zack (1n which a llposarcoma was found), and a fourth study by Ott et al. (1980), Honchar and Halperln (1981) noted 3 (2-.9%) soft-tissue sarcomas 1n a total of 105 deaths, compared roughly to 0.07% deaths 1n U.S. males aged 20-84 years (ICO 171, 8th Revision, 1975)* Indicating an unusual excess of such tumors. This may be somewhat of an under- estimate due to the posslbllty that some soft-tissue sarcomas may be coded to categories other than ICD 171. Separately none of the reported case studies reported a significant excess of soft-tissue sarcomas. The number of soft-tissue sarcomas noted by Honchar and Halperln was Increased by a fourth when Cook (1981a) found a malignant fibrous hlstlocytorca after a later review of the medical records from his earlier cohort study. Cook, who was familiar with the earlier three cases, went on to say that frank chloracne occurred previously 1n two cases of the 4 having a diagnosis of *Department of Health, Education, and Welfare. U.S. Public Health Service. National Center for Health Statistics of the United States, 1974. Vol. II. Mortality, Part A. C-116 �malignant fibrous hlstlocytoma. A third case diagnosed as a flbrosarcoma worked In a trlchlorophenol (TCP) process area contaminated with 2,3,7,8TCOD. This Individual exhibited facial dermatitis, but no diagnosis of chloracne was made. The last case was diagnosed as a Uposarcoma, and the Individual had been employed earlier 1n a plant producing 2,4,5-T. Cook noted that although chloracne was not reported, It could not be discounted. He also noted that all four were smokers and suggested that smokers with chloracne caused by 2,3,7,8-TCDD exposure may be subject to an Increased risk of fibrous soft-tissue sarcomas. Hardell and Eriksson (1981) discounted this hypothesis by citing that only one of Hardell's seven cases exhibited chloracne prior to the appearance of the soft-tissue saromcas, and that 1n his later case control study, he found no difference 1n smoking habits between his cases and controls. Moses and Sellkoff (1981) reported discovering a fifth soft-tissue sarcoma In a worker employed at the Monsanto Chemical Company at a time when trlchlorophenol and 2,4,5-T were being produced. He died of a retropeM- toneal neurogenlc sarcoma (malignant schwanoma) In 1980 at the age of 58. The employee, prior to his death, 1n a detailed occupational history said that he was potentially exposed to these chemicals while he was a truck driver, hauler and maintenance worker, but that he did not work 1n the production of either chemical. He was a non-smoker and did not have a history of chloracne. Johnson et al. (1981) treated a father and son with soft-tissue sarcomas (the 33-year-old son was diagnosed as having a fIbrosarcomatous mesothe- lloma, while the 53-year-old father had a Uposarcoma). Both were exposed to halogenated phenol derivatives. The author noted that 2,4-d1chlorophenol C-117 �can be a precursor of 2,4-D and 2,4,5-T. The father had had prolonged exposure prior to his disease. according to the author. The son supposedly had a shorter latency, In neither case Is the follow-up time given. Sarma and Jacops (1981) reported three cases of thoracic soft-tissue sarcoma 1n Individuals who were exposed to Agent Orange while serving 1n Vietnam. The diagnoses were fibrous hlstlocytoma, medlastlnal fIbrosarcoma, and a pleural/dlaphragmatlc lelomyosarcoma. All three served 1n areas where defoliants were used at the time. One was drenched with the material 1n one spraying. Bishop and Jones (1981) found two cases of non-Hodgk1n's lymphomas of the scalp 1n a related clinical study of 158 employees of a pentachlorophenol manufacturing plant 1n Wales. Homologues of 2,3,7,8-TCOD occurred as contaminants at up to 300 ppm at Intermediate manufacturing stages and 5 ppm 1n the final products. M1ld, moderate and severe cases of chloracne were seen 1n many employees, Including the two men who subsequently developed lymphomas. Both men worked In processes where exposure to other chemicals occurred, Including exposure to aromatic hydrocarbons. The authors reported that only 0.28 tumors of this type could be expected to occur In a group of 158 workers (ICD 200 and 202), although the basis for the computation of expected numbers 1s not stated. Olsson and Brandt (1982) noted that of 123 male patients seen at his clinic In Sweden with a recent diagnosis of non-Hodgk1n's lymphoma (NHL), 5 had cutaneous lesions as the only clinically detectable manifestation of NHL. Four of the five had repeatedly sprayed large areas with phenoxy add herbicides. In the remaining 118 NHL patients, only seven had a similar occupational exposure to phenoxy acids. significant at P<0.001. The authors reported this to be Olsson and Brandt suggested that a relationship C-118 �exists between cutaneous presentation of NHL and occupational exposure to phenoxy adds, and believed their observations were similar to those of Bishop and Jones. Adding these case studies together, the total number of workers exposed to phenoxy adds and/or chlorophenols Is small, but considering the rarity of this cancer. It Is unusual that so many cases of soft-tissue sarcomas have occurred. A Lancet editorial (Anonymous, 1982) calls this phenomenon "disturbing." It Is suggestive of an association of cancer with exposure to phenoxy adds and/or chlorophenols, and consequently with the dloxln Impurities found 1n these herbicides. Soft-Tissue Sarcomas. Soft-tissue sarcomas (STS) constitute a collec- tion of heterologous lesions that Include both malignant and non-malignant tumors. Not all of them have their origin In primordial mesenchymal cells. Some exceptions are tumors of peripheral nerves, and neuroectodermal tumors which are classified as STS, but are derived from non-mesenchymal cells. Classification, grading and staging of STSs Is difficult because of the capacity of such cells to differentiate Into many different tissues. Fairly precise hlstcgenetlc classification of such tumors Is accomplished through consideration of growth patterns and cell morphology and evaluation of Intracellular and extracellular products of tumor cells. There are a dozen distinctly different classes of mesenchymal cells that develop Into the following six well-defined tissue complexes: fibrous tissue, tendosynovlal tissue, adipose tissue, muscle, vessels and bone. STSs can be Induced 1n any of these tissue types (Hajdu, 1983). The classification of STSs for cause of death coding In the ninth and latest revision of the International Classification of Diseases (ICO, 1975) places STSs Into one of several categories. But chiefly, they fall Into "malignant neoplasms of connective C-119 �and other soft-tissue" (ICD 171). Lymphosarcomas, retroperltoneal sarcomas and extra skeletal STSs of the bone are coded elsewhere. In some Instances, 1f site 1s mentioned, H 1s coded to the site, e.g., lelomyosarcoma of the stomach (ICD 151.9), neuroflbroma of the chest wall (215.4). Questions have been raised concerning the appropriateness of lumping together malignant tumors of different sites and tumor types In order to derive risk estimates. It may not be scientifically appropriate to do so because an elevated risk cannot readily be ascribed to a particular site or type as 1s usual with most carcinogenic chemicals and substances. Unfortu- nately, with respect to STSs, tallies of deaths due to STSs of particular sites and types are not maintained separately by the vital statistics offices because of their rarity, and therefore, H 1s Impossible to derive risk estimates for particular types at given sites. Altogether, -2000 deaths/year can be attributed to STSs In the United States, most of which are coded to ICD category 171 for purposes of developing Incidence and mortality rates for this composite cause. Within ICO 171, Individual types that may be correlated with exposure cannot be Identified. A separate problem that potentially could arise from assigning STSs to multiple ICD codes 1s that Incidence and death rates due to STSs may be underestimated. Furthermore, risk estimates derived from dividing observed cases (or deaths) by expected cases (or deaths) could be biased upward. This could happen when observed STSs classified to ICD codes other than ICO 171 are lumped together while expected STSs are based upon ICD 171 only. Thus, action of this sort, especially with respect to cohort studies of Individuals exposed to dloxln-contalnlng herbicides and/or chlorophenols, C-120 �could lead to risk estimates that may be biased upward by the Inclusion of STSs In the observed category for risk estimation that should be coded to categories other than 171. Prompted by clinical observations over a 7-year period of malignant sarcomas 1n seven men with previous occupational exposure to phenoxyacetlc acid herbicides (Hardell, 1977), researchers at the Department of Oncology, University Hospital, Umea, Sweden, Initiated epldemlologlc studies to test the hypothesis of an etlologlc association (Hardell and Sandstrom, 1979). The Investigators elected to conduct case-control studies, a type of epidemiclogic research particularly well suited for rare diseases with long periods of Induction (Cole, 1979). Cases were defined as male patients with sar- comas of soft connective tissue, such as smooth muscle (lelomyosarcoma) and fat (Uposarcoma). The distribution of tumor types In the two studies 1s shown 1n Table 12. Sarcomas of harder connective tissues, such as bone and cartilage, were excluded. According to the authors, these tumors may have a different etiology and there occurred a different age-distribution 1n patients with these tumors as compared to that of STS (Hardell, 1983). Two case-control studies were conducted, the first 1n northern Sweden (referred to below as Study A), and the second In the southern part of the country (Study B). The frequencies of exposure to the substances of primary Interest are shown In Table 13. In the north, occupational exposure to phenoxyacetlc adds took place 1n both forestry and agricultural work. the south, these exposures were predominantly agricultural. In The phenoxy- acetlc adds to which exposure occurred consisted predominantly of 2,4,5-T and 2,4-D 1n both studies. Exposure to 2,4,5-T 1n the absence of 2,4-D was rarely reported 1n either study. Exposure to chlorophenols, which contain C-121 �TABLE 12 Distribution of Tumor Types 1n Two Case-Controls Studies of Soft-Tissue Sarcoma Percent of Cases Diagnosis Tissue of Origin Study Aa (n=52) Study Bb (n=110) Lelomyosarcoma Smooth muscle 30 23 Fibrous hlstlocytoma Subcutaneous connective tissue 17 25 Llposarcoma Fat tissue 14 6 Neurogenlc sarcoma Nerve tissue 10 4 Anglosarcoma Blood vessels 8 2 Hyxosarcoma Primitive connective tissue 6 8 Fibrous tissue 4 8 Other sarcomas 11 24 Total 100 100 F Ibrosarcoma a Unpub!1shed Information strom, 1979) b supplied by Hardell Er1ksson et al., 1979, 1981 C-122 to EPA (Hardell and Sand- �TABLE 13 Exposure Frequencies In Two Case-Control Studies of Soft-Tissue Sarcoma Percent Exposed Study A Substance(s) Cases (n=52) Study B Controls (n=206) Cases (n=110) Controls (n=219) Phenoxyacetlc adds only Chlorophenols only Both 23.1 11.5 1.9 6.3 2.4 0.5 12.7 10.0 0 2.3 3.6 0 Total 36.5 9.2 22.7 5.9 *Sources: Study A, Hardell and Sandstrom, 1979; Study B, Eriksson et al., 1979, 1981 C-123 �chlorinated dlbenzodloxln Impurities (Levin et al., 1976) occurred mostly 1n sawmill work and paper pulp production. both to phenoxyacetlc add Very few persons reported exposure and chlorophenols In these studies. Of the two predominant phenoxyacetlc adds, only 2,4,5-T 1s known to be contaminated with 2,3,7,8-TCDD. In Study B, a relative risk of 4.9 (90% confidence Intervals 1.6-11.1) was found In relation to exposure to phenoxyacetlc add herbicide other than 2,4,5-T (2,4-D, MCPA, mecoprop, dlchloroprop). Relative risks 1n relation to the three major categories of exposure are shown In Table 14.* Studies A and B Indicate a risk of developing STSs among workers exposed to phenoxyacetlc adds only, chlorophenols only, or phenoxyacetlc adds and/or chlorophenols several times higher than among persons not exposed to these chemicals. In each comparison, the point estimate of relative risk 1s high and unlikely to have resulted by chance alone. Since little 1s known of the etiology of STSs, the consideration of confounding 1n these studies was largely a hypothetical matter. The authors prevented the effects of age, sex, and place of residence as possible confounding factors In the selection of controls.t Because of the high cor- relation between exposure to the substances of Interest and employment In agriculture and forestry, a reasonable hypothesis could be developed that some other unknown factor present In these occupations was responsible for the elevated relative risks. *In the analyses considering phenoxyacetlc adds only and chlorophenols only, persons exposed to the other categories of substances were excluded. In Study A, the three persons exposed to both chlorophenols and phenoxyacetlc adds were Included 1n all comparisons. tControls were matched Individually to cases on the basis of these factors. Unmatched analyses are presented 1n Table 24 for the sake of simplicity. The matched-method relative risks for exposure to phenoxyacetlc adds and/or chlorphenols were 6.2 (P<0.001) 1n Study A and 5.1 (P<0.001) In Study B. C-124 �TABLE 14 Relative Risks of Soft-Tissue Sarcoma 1n Relation to Exposure to Phenoxyacetlc Adds and Chlorophenols 1n Two Case-Control Studies3 Phenoxyacetlc Acids Only Phenoxyacetlc Adds and/or Chlorophenols Chlorophenols Only Study A Study B Study A Study B Study A Study B Relative r1skb 5.3 6.8 6.6 3.3 5.7 4.7 90% Confidence 1nterval c 2.7-10.2 3.1-14.9 2.8-15.6 1.6-7.0 3.2-10.2 2.7-8. 3 Significance level^ <0.001 <0.001 <0.001 <0.005 <0.001 a Source: Study A, Hardell and Sandstrom, 1979; Study B, Eriksson et al., 1979, 1981 b Unmatched odds ratio c Test-based method of Hlettlnen, 1976 square statistic, no continuity correction, one-tailed test <0.001 �To test this hypothesis, 1t 1s possible to calculate the relative risk 1n relation to the phenoxyacetlc add exposure 1n Study B, restricting the analysis to workers within agriculture and forestry. The result Is a relative risk of 6.1 (90X confidence Interval 2.4-15.4). This finding strongly suggests that some confounding risk factor for STS distributed throughout agriculture and forestry work was not responsible for the overall Increase In risk found In relation to phenoxyacetlc add exposure. Because exposure histories were obtained by means of questionnaires and Interviews, the major potential source of bias In these studies stems from the need to rely upon the personal recollection of cases and controls for exposure histories. The published papers Indicate that the researchers paid a great deal of attention to this potential problem and state that they took all reasonable precautions to avoid 1t during the conduct of the study. In addition, the relative risk calculated by considering the agriculture and forestry workers who did not report exposure to phenoxyacetlc adds or chlorophenols and comparing them to unexposed persons 1n other occupations was 0.9 (9054 confidence Interval 0.3-2.4) 1n Study B. This suggests that a great deal of recall bias was not present (Axelson, 1980). In an update of his earlier study, Eriksson et al. (1981) obtained Information on the effects of phenoxy acids In the absence of the 1mpur1t1es--polychlor1nated dlbenzodloxlns and dlbenzofurans. The risk ratio given exposure to phenoxy adds free of polychlorlnated dlbenzodloxlns and dlbenzofurans equaled 4.2 based upon 7 out of 14 respondents who Indicated exposure to phenoxy add herbicides. When consideration was given to only phenoxy adds that contain such Impurities, the risk was 17.0. A description of the basis Tor the determination of exposure or non-exposure to dloxlns Is not well presented 1n this study. C-126 �The author concluded that exposure to phenoxy acids and chlorophenols "might con-,!' t ute a risk factor In Uu: iloviO , , ut uf .uF!-tissue sarcomas." Ihls risk relates nol. only l.o ?,4,!>-tr1chlorophenoxy acids conlnlnin.j I'oxln Impurities, but to other phenoxy adds as well. Some doubt was raised con- cerning the possible mlsclasslfIcatlons of Individuals who were exposed to phenoxy acids free of polychlorlnated d1benzod1ox1ns (I.e., 1n particular, "dlchoroprop" In the Eriksson study). In a recent communication from Hardell (1983), Eriksson recalculated his risk estimates after reclasslfylng his dlchoroprop-exposed cases and controls Into the category of probable exposure to phenoxy acids contaminated with polychlorlnated dlbenzodloxlns and removing them from the non-exposed category. His new estimates were 4.0 based upon 5 out of 8 respondents who were exposed to phenoxy adds free of contamination and 10.9 for those exposed to contaminated plionoxy a t ' ; - ^; first estimate was of only borderline significance utilizing the Mletlnen test based statistic, thus, weakening any finding that the risk of STS extends to phenoxy adds free of dloxln. In a cohort mortality study (Cook et al., 1980a) of 61 males Involved 1n a 1964 chloracne Incident, employees In a trlchlorophenol manufacturing area were found to have chloracne due to skin absorption of 2,3,7,8-TCDD. The skin lesions characterizing chloracne ranged from a few comedones on thr back of one employee (predating his entry Into the process area where exposure could occur) to severe cysts and comedones over the faces, scalps, ears, necks and backs of the remaining employees of the group. Since the main route of exposure was not through the respiratory tract, no measurements of dloxln In the air were provided by the author. On the other hand, the author did subjectively divide the cohort of 61 males Into potentially "high" vs. "low" exposure by place of work based upon dermal C-127 exposure, �although not stated. Vital status was traced from the data of the Incident through 1978. Altogether only 4 deaths were observed by the end of the follow-up, vs. 7.8 expected. Of these, 3 were cancer vs. 1.6 expected. The remaining death was hypersensitive heart disease vs. 3.8 expected. The hlstopathologlc causes of death of the three cancer victims were 1) Hbrosarcoma, 2) glloma with metastases, and 3) adenocarclnoma. The authors report that all three victims smoked a minimum of one pack of cigarettes a day for "many years." Cancer mortality 1s slightly elevated 1n this cohort despite Us relative low sensitivity, the lack of a sufficient latent period, and the presence of the healthy worker effect. utable to any particular cause. This Increased mortality was not attribAdditionally, the authors state that only one of the cancer deaths possessed "documented" evidence of chloracne, although this appears to be at variance with the definition of the cohort, which was reported by the authors to consist of males who reported to the medical department with skin conditions subsequently "diagnosed as chloracne." The authors furthermore concluded that the latency period was sufficient to "allow the Identification of a potent human carcinogen," since 1t "exceeded 14 years." Orris (1981) criticized this conclusion with a refer- ence to the Hardell and Sandstrom (1979) study 1n which the authors noted that the latent period for soft-tissue tumors may be as long as 27 years and for many, over 14 years. Cook (1981b) countered that the Hardell and Sandstrom (1979) conclusions were based upon questionable data In that the self-administered questionnaires used 1n that study provided neither valid quantitative nor qualitative estimates of exposure. Therefore, It could not be used to determine latent periods. In any case, Hueper and Conway (1964) noted that the latent period for the chemical Induction of solid malignant tumors In man exceeds 15 years and 1s probably <30 years. C-128 �Although the Hardell and Sandstrom (1979) study has some deficiencies, the Cook et al. (1980a) study provides little evidence to support the premise advanced by the authors that dloxln "cannot be considered to be a potent human carcinogen with organ or tissue specificity." There Is a distinct likelihood that the latent period for the development of STSs and related tumors due to exposure to dloxln may not have been achieved within the 14-year follow-up period specified 1n the study. Furthermore, a much larger cohort may be needed In order to detect a significantly Increased cancer risk. Smith et al. (1982b) conducted a case-control study of 102 males Identified from the New Zealand Cancer Registry as having STSs (ICO 171) between 1976 and 1980. For each case, three controls each with another form of cancer were matched by age and year of registration. The selection of cancer controls from the same registry was done to eliminate recall bias and/or Interviewer bias. The distribution of hlstological types 1n the cases 1s given In Table 15. The Interview to elicit occupational history Information was accomplished via the telephone either with the next of kin to the patlent or the patient himself 1f he was well enough. Anxiety was alleviated by the mailing of a letter prior to the Interview, the purpose of which was to Inform the person of the Intention of the Interviewer to ask some questions about his occupational history. Apparently, the questions asked were not specific enough to Identify definite exposure to phenoxy herbicides and/or chlorophenols. The authors asked only about current occupation or last occupation If retired. Compari- sons between cases and controls were accomplished by use of occupational groupings according to the Standard Classification System of New Zealand focusing on those occupational groups with a potential for exposure to C-129 �TABLE 15 Distribution of H1stolog1cal Types of Soft-Tissue Sarcomas Cell Type Number of Cases Percent Flbrosarcoma 25 24 Llposarcoma 20 20 Rhabdomyosarcoma 9 9 Lelomyosarcoma 7 7 Malignant H1st1ocytoma 6 6 Other 22 21 Unspecified 13 13 102 100 Total C-130 �phenoxy herbicides and chlorophenols. Expected cases for each major occupational classification were derived based upon the occupational distribution of the controls. The authors found no unusual excess of cases of STS In any major occupational category. In agriculture, forestry and fishing, 14 cases were observed vs. 14.0 expected. In laborers, production and transport workers, 35 cases were observed vs. 37.0 expected. A further breakdown of these two broad categories Into finer subcategorles within the major occupational categories revealed no significant excesses. The study, however, 1s not useful In assessing the risk of STS from exposure to phenoxy adds and/or chlorophenols for several reasons. First, as was pointed out by the authors, but subsequently dismissed by them as having not much of an Influence, 1s the possibility that switching from one major occupational category to another over the time period Involved for latent conditions to manifest themselves could Introduction a negative bias Into any estimates of relative risks. The latency for STS Is felt to be a minimum of 15 years (Hueper and Conway, 1964). The finding of no switching from one occupational category to another that was noted In the "first 20 Interviews" In which a change could be noted Is not necessarily Indicative of fidelity to the same Job over long periods In all 408 cases and controls. Information Identifying a switch may be lacking In those cases and controls In which a switch did occur only because the switch resulted In separation of the earlier work history from the latter. Besides the "first 20 Interviews" where a change could be noted Is not representative of the entire cohort In any case. Furthermore, the authors do not know absolutely that any of their cases and controls were exposed to phenoxy adds and/or C-131 chlorophenols since �apparently no effort was made to confirm "potential" exposures. Only dif- ferences In occupational classification were noted where "potentially" cases or controls could have had exposure to the dloxln-contalnlng herbicides. It was pointed out that the risk estimates noted do not "preclude" the possibility that an association may be found In this study when the cases and controls (or surviving kin) are Interviewed for chemical spraying at a later time. The authors themselves conclude that the preliminary study results "should not be taken as substantial evidence against the hypothesis that phenoxy herbicides and chlorophenols may cause human cancer." It should be noted that the distribution of tumor types differed considerably from the Harden and Eriksson study to the Smith study. Lelomyo- sarcomas, malignant hlstocytomas, neurogenlc sarcomas and myxosarcoma seem to predominate 1n the Hardell and Eriksson study, whereas fIbrosarcomas and Uposarcomas appear prominently 1n the Smith study. More attention should be devoted to the study of the dlsbrlbutlons of STS types 1n registry data everywhere 1n order to determine 1f such variations In the reporting of STS types are random occurrences. It 1s possible that the cancer effect of exposure to phenoxy herbicides may be narrowed to just certain types of STSs, the predominant ones In the Swedish studies. In a later study of STSs, Smith et al (1983a) conducted a case-control study of STSs 1n males that were reported to the New Zealand Cancer Registry by Public Hospitals between 1976 and 1980. The author matched one cancer control randomly chosen from the registry with each case, Initially starting with 112 of each. Controls were matched for year of registration and by date of birth ± 2 years. Inquiries were made by the authors with the hos- pital consultant, family doctor, and finally the next-of-kln or patient 1f alive. Telephone Interviews were conducted by only one Interviewer who had C-132 �no knowledge of the patients cancer history and were completed on 80 cases and 92 controls. Because some 32 potential cases (14 Ineligible) and 20 controls were excluded or lost from the study for various reasons, 1t raises a question whether control of confounding by age and year of registration was maintained In the final group of 172 cases and control Included In the analysis. Presumably the corresponding "matched" case or control to each of the 52 lost members of the total study group were not excluded. Patients were classified as having had potential exposure to phenoxyacetic acids 1f they had definite, probable or possible exposure to phenoxyacetlc add through spraying or hand contact. The actual chemical was Identified only In some Instances. The authors concluded In all remaining situations that 1f the member sprayed "gorse" and/or "blackberries" this was tantamant to potential exposure to phenoxyacetlc add. Smith calculated elevated but non-significant relative risks of exposure to phenoxyacetlc acid ranging from 1.3 In those Individuals who were "probably exposed" for a minimum of 5 days not In the previous 10 years prior to cancer registration to 1.6 In Individuals "probably exposed" for a minimum of 1 day not 1n the previous 5 years prior to cancer registration. When risk ratios were calculated after stratifying by year of birth and whether or not the patient or a relative was Interviewed, the rates Increased to 1.7 (from 1.6) 1n the latter and 1.4 (from 1.3) In the former calculation, although still nonsignificant. It would be of Interest to repeat the above calculations excluding only those with potential exposure occurring only within the 15-year period just prior to cancer registration. Furthermore, the categories of exposure "probably or definitely" exposed for >1 day or. even 5 days raises a question whether any of the cases or controls could really be said to have ever come 1n contact with enough phenoxyacetlc add C-133 to justify such a �designation. It could be that, In fact, potentially exposed Individuals 1n New Zealand have had little or no contact with the herbicide. The authors did conclude that the finding of a relative risk of 1.7 In Individuals with >1 day exposure not 1n the last 5 years cannot be entirely discounted. But then the authors state that 1f exposures of >5 days prior to 10 years before cancer registration are not Included they would expect an Increase, and since they do not see an Increase, there 1s no evidence of a "real causal link." One might question whether this 1s a suitable criterion for providing evidence of a causal association. Perhaps a more valid group for study would be one where the potential exposure was considerably longer than "5 days" and >15 years prior to Initial cancer registration. As kind of a subtle Justification for the finding of no significant risk In workers exposed In phenoxy adds, the author alludes to the fact that there are currently 500 full-time workers registered In New Zealand who do full time ground spraying and altogether some 2000 workers who were at some time professionally Involved 1n phenoxyacetlc add herbicide spraying from the air or ground with exposure "very much greater" than that of patients 1n this study. This kind of argument has appeal 1f these workers could be shown to have had their exposure sufficiently far 1n the past that latency considerations could be adequately addressed. However, the real question again remains how much real exposure did those patients 1n the study really have 10-15 years earlier, and 1n what numbers. The author remarks that H 1s surprising that he found no STS victims who had ever worked full-time 1n phenoxyacetlc add herbicide spraying. Perhaps they have not yet been ob- served for a long enough period. However, as was pointed out by the author, the findings do not support the hypothesis that exposure to phenoxyacetlc C-134 �add herbicides causes STS. But neither do they support a negative finding without better documentation regarding actual exposure and time of actual exposure. Pazderova-Vejlupkova et al. (1981) studied 80 workers Involved In the production of 2,4,5-sodlum trlchlorophenoxyacetate and butylester of tr1chlorophenoxyaceUc acid who subsequently became 2,3,7,8-TCDD during the period 1965-1968. Only 55 members of this group were followed for 10 years. 111 from exposure to The remaining 25 either refused participation or moved leaving no forwarding address. Most patients developed chloracne while 11 developed porphyrla cutanea tarda. Chief chemical signs were metabolic disturbances, pathologically elevated Uplds with abnormalities 1n the llpoproteln spectrum, and "pathological" changes 1n glucose tolerance. Other symptoms noted were biochemical deviations consistent with "a mild liver lesion," light steatosls, perlportal flbrosls or activation of Kupffer cells, or nervous system focal damage (peripheral neuron lesion In lower extremetles). Altogether six patients were reported to be deceased during this 10-year period, 2 from bronchogenlc carcinoma, 1 from cirrhosis, 1 atherosclerosis preclpue cerebl and 2 1n auto accidents. No STSs or lymphomas were found. Since there was no comparison population with which to estimate relative risk for cancer, the study must be classified at best as clinical with respect to cancer. The six deaths that occurred during the 10-year observation period In the 55 cannot be construed to be associated with exposure to the 2,4,5-T. Because of the small number of cases and the short follow-up period, nothing can be said concerning the association of exposure with cancer, especially specific types of cancer such as STS or non-Hodgk1n's lymphoma. C-135 �et al. (1982, 1983) cators Formed In 1972 studied a cohort 1926 herbicide appli- from personnel records of four Finnish employers (I.e., the Forestry Authority, Highway Authority, State Railways and a state-owned electric power company). Chlorinated phenoxyadds had been used since the 1950's 1n Finland for spraying. They constituted 2:1 mixtures of emulsified esters of 2,4-D and 2,4,5-T dissolved 1n water. Analyses from old herbicide formulations dating back to the 1960's revealed that these mixtures contained 0.1-0.9 mg/kg of 2,3,7,8-TCDO). This cohort of male workers was exposed a minimum of 2 weeks during at least one growing season through 1980 from 1955-1971. Follow-up continued 9 years for mortality but only until 1978 for morbidity. Fifteen Individuals could not be traced by 1980. Expected deaths were generated based upon cause- and age-specific national Finnish death rates for 1975. Expected cases were similarly calculated based upon national Incidence rates of 1975. By 1980, 144 deaths had occurred vs. 184.0 expected, a deficit of 22% 1n observed mortality. Only 26 cancer deaths had occurred vs. 36.5 expected, a 29% deficit. The authors separated out "natural" deaths from the total. The observed residual deaths equaled 39 while the expected deaths equaled 28.7. This excess was of borderline significance. sidered 10-year and 15-year latent periods. The authors also con- Even after 15 years, the defi- cit of deaths continued to manifest Itself both 1n categories of all causes and total cancers; 35 observed vs. 53.6 expected and 5 observed vs. 11.3 expected, respectively. Similarly, the 7-year follow-up of cancer morbidity revealed 26 cases of cancer vs. 37.2 expected. After 10 years latency, 16 cancer cases were observed vs. 20.1 expected. C-136 None of the 26 cancer deaths �or 26 cancer cases were of the STS or lymphoma type. and 0.5 lymphomas were expected.) (However, only 0.1 STS In no Instance was cancer of any site significantly elevated. The authors note that this unusual deficit of mortality and morbidity of between 70-8254 (even after 15 years from Initial exposure) 1s probably a consequence of the "healthy worker effect" 1n that only able-bodied and healthy Individuals were selected Into the Industry. The fact that the cohort was assembled In 1972 from records of persons who were exposed as early as 1955 (17 years prior) raises the likelihood that In 1972 a "survivor" population remained (45 deaths prior to 1972 were eliminated from the cohort) that was relatively healthy. Furthermore, the unusually large num- ber of not "natural" expected and observed deaths (probably accidents and external causes) occurring to this cohort Indicate a relatively youthful population was under scrutiny. The leading cause of death to persons under 35 years Is from accidents, based on national vital statistics. The authors correctly note that, because of limitations In the study material, only powerful carcinogenic effects could be detected. Risk ratios higher than 1.5 for all cancers, 4.0 for lymphomas and 10.0 for STS could be excluded based on this data set from the authors own calculations. Hore follow-up Is needed In order to provide a stable assessment of the relationship between exposure and cancer. The authors concluded that this study will allow no assessment of STS because "the number of persons having a sufficiently long latency period 1s too small." It was suggested that more valid conclusions could be made only with the passage of time. Recently, the Michigan Department of Public Health (1983b), produced an ecological study of soft and connective tissue cancer mortality rates 1n Midland and other selected Michigan counties. C-137 They found that mortality �rates for this cause were 3.8-4.0 times the national average for the periods 1960-1969 and 1970-1978, respectively, for white females In Midland. These estimates are based upon 5 deaths and 7 deaths, respectively, and are listed In Table 16. No excess risk was reported among white males, however. The Michigan Department of Health concluded that because of the occurrence of these two successive elevated rates, It Is unlikely to be a chance happening. At the same time the age-adjusted male and female cancer mortality rates for Midland were below that of the State of Michigan 1n the period 1970-1979. Midland County 1s the home of a major chemical company that produced phenoxyacetlc add herbicides until recently. The authors state that a detailed review of death certificates, hospital records, residency and occupational histories of the 20 male and female cases revealed no "commonalities" suggesting a "single causative agent" although a majority or their spouses had worked at this chemical facility. They recommend that a casecontrol study should be Instituted to evaluate possible Influences, such as lifestyle, occupation or location of residence on the risk of STS. In a separate review of the ep1dem1olog1cal evidence for STS from exposure to 2,4,5-T-conta1n1ng herbicides, the United Kingdom Ministry of Agriculture, Fisheries and Food (1983) concluded that there was no evidence vO recommend altering their earlier conclusion that formulations of phenoxy acid herbicides and related wood preservatives as "presently cleared" are safe and may continue to be used. This report too readily discounts the positive studies of Hardell and Eriksson as being biased, and H makes no reference to the later validity study by Hardell (1981) of his own work utilizing colon cancer controls (see Section on Malignant Lymphoma). In this report Hardell effectively answered these early criticisms that were reiterated by the British In their report. At the same time, the British C-138 �TABLE 16 Midland County Soft and Connective Tissue Cancer Deaths 1960-1981* Type of Malignancy Identification Primary Site Year of Death Age Type 1961 F 24 Hemanglosarcoma Face Skull and upper lobe of lung 1963 F 75 Llposarcoma Right gluteal Unknown Unknown 1964 CO <£> Sex F 51 Lelomyosarcoma Uterus Widespread 11-63 1968 F 37 Llposarcoma Spine Lungs, pelvis 1969 h 45 Flbrosarcoma Lelomyosarcoma Right thigh Uterus Lung, liver Adrenal gland and skin 1970 F 59 Kaposl sarcoma Right leg Lymph nodes 1970 F 56 Flbrosarcoma Lelomyosarcoma Right thigh Abdominal wall Spine Lung 1974 F 1 Rhabdomyosarcoma Inguinal area Unknown 1976 F 77 Llposarcoma Right thigh Buttock, lung, Mb, lymph nodes 12-74 64 Lelomyosarcoma Left knee Liver, lymph nodes, lung, bone 7-70 1978 Metastases Month and Year Diagnosed 5-58 1-66 10-68 8-68 1960 1967 8-73 �TABLE 16 (cont.) Identification Type of Malignancy Sex Age Type 1978 F 26 Rhabdomyosarcoma Rectum Lung, neck, Inguinal region 6-76 1978 F 88 Flbrosarcoma Right cheek Facial area 6-78 1979 F 27 Lelomyosarcoma Left thigh Lung 3-78 1962 M 63 Rhabdomyosarcoma Left lower leg Lung and right outer chesfwall 8-61 1967 M 77 Mesothelloma Lung Lung, peritoneum and diaphragm 6-67 1967 M 20 Rhabdomyosarcoma Pharynx PerlorbHal area and liver 1-67 1969 M 32 Llposarcoma Left arm Perineum and buttock 6-64 1971 M 76 Lelomyosarcoma Small Intestine Liver 1972 M 89 Lelomyosarcoma Retroperltonal region Hepatic system 7-72 1976 M 53 Flbrosarcoma Per1t1oneum Lung, liver 3-75 Year of Death o I Primary Site *Source: Michigan Department of Public Health, 1983b Metastases Month and Year Diagnosed 10-69 �report appears to put undue emphasis on non-positive studies that do not demonstrate a risk, although most of them have methodological limitations (I.e., low power, Insufficient latency and Inappropriate study method). In short, the British review appears to be overly optimistic about the safety of 2,4,5-T herbicides. In summary, the associations reported 1n the two Swedish soft-tissue sarcoma studies are great enough to make 1t unlikely that they have resulted entirely from random variation bias or confounding, even though the possibility cannot be dismissed that bias or confounding was present. Therefore, the studies provide a strong suggestion that phenoxyacetlc add herbicides, chlorophenols or their Impurities are carcinogenic 1n humans. Malignant Lymphoma. A separate series of clinical observations at the Department of Oncology 1n Umea, Sweden (Hardell, 1979), led the researchers to conduct a case-control study of malignant lymphoma In relation to phenoxyacetlc add, chlorophenols, and other organic compounds (Hardell et a!., 1980, 1981). Approximately 33% of the cases 1n this study were patients with Hodgkln's disease; the remainder of the lymphomas were non-Hodgk1n's forms. This study employed essentially the same methods and produced results closely comparable to these from the STS studies: 5-fold statistically significant to 6-fold relative risks In relation to phenoxyacetlc chlorophenols. adds and In addition, an elevated relative risk was found In connec- tion with exposure to organic solvents, such as benzene, trlchloroethylene, and styrene. In the published report, the methods and results were Incom- pletely documented, especially the possibility of confounding by exposure to the organic solvents. C-141 �In the update of the earlier 1980 study. Harden et al. (1981), utilizing the same basic data source, found that 36.1% of the cases had been exposed to phenoxy herbicides or chlorophenols, while only 9.6% of their controls were so exposed. The estimated relative risk was 6.0 when matching was considered and 5.3 when matching was eliminated. When cases and con- trols who were exposed to chlorophenols only were excluded, the relative risk of lymphoma from phenoxy acids alone was 4.8 (95% C.I. 2.9-8.1). On the other hand, If phenoxy adds are excluded and consideration Is given to just chlorophenols (which Includes combined exposure to phenoxy acids and chlorophenols), then the relative risk equaled 4.3 (95% C.I. 2.7-6.9). The author further subdivided this group Into "low-grade" vs. "high-grade" exposures to chlorophenols. A continuous exposure of not more than 1 week or repeated Intermittent exposures totaling not more than 1 month was classified as low-grade. The relative risk for high-grade exposure was 8.4 (95% C.I. 4.2-16.9), while that for low-grade exposure equaled 9.2 (95% C.I. 1.6-5.2). If exposure to organic solvents Is examined, given that cases and controls exposed to only phenoxy adds and/or chlorophenols were excluded except for combined exposure to organic solvents, H 1s found that highgrade and low-grade relative risks were 2.8 (95% C.I. 1.6-4.8) and 1.2 (95% C.I. 0.5-2.6), respectively. However, the author notes that exposure to phenoxy acids and high-grade organic solvents (exposure to chlorophenols excluded) produced a relative risk of 11.2 (95% C.I. 3.2-39.7) based upon a few cases and controls with exposure to both. The authors concluded that "exposure to organic solvents, chlorophenols and/or phenoxy adds constitutes a risk factor for malignant lymphoma." C-142 �This latter study Is still subject to the same methodological criticisms to which the earlier study was subjected. Chief among those Is the possi- bility of observational and/or recall bias creeping Into the responses that are elicited from self-administered questionnaires on kind and length of exposure. Secondly, confounding by exposure to potentially carcinogenic organic solvents and other agents could have had an effect, although the author assures the reader that they did not. Other research has tentatively suggested that lumberjacks may be at Increased risk of lymphoma (EdUng and Granstam, 1979). The NHro study found three deaths from cancers of the lymphatic and hematopoletlc system, against only 0.88 expected (P = 0.06, one-tailed Polsson test). The lymphoma case-control study (Harden et al., 1980, 1981) Is consistent with the two STS studies discussed above. On the other hand, the consistency could also reflect an as-yet unidentified methodologlc flaw 1n all these studies. The two Swedish case control studies on STSs and a later case control study of malignant lymphoma (Hardell et al., 1981) were subjected to a validity analysis with respect to the assessment of exposure by Hardell and Eriksson (1981). To answer the question raised regarding the recall of occupation In a forestry/agriculture job, secondary to the recall of exposure to phenoxy adds and/or chlorophenols, the cases and controls were divided Into three groups: those who worked their entire time since 1950 1n an agriculture/forestry job, those who worked some time In an agriculture/ forestry job but not exclusively, and the remainder who never worked 1n a forestry/agriculture job. The study found that the risk ratio was still 8.2 for STS 1n exclusively agriculture/forestry workers who were exposed to phenoxy acids compared to workers found In other occupations having no apparent exposure to phenoxy adds or chlorophenols. C-H3 Even when comparing �phenoxy acid and/or chlorophenol exposed agricultural/forestry workers exclusively with non-exposed agricultural/forestry workers, the risk ratio was still 7.1. This argument seems to answer effectively questions regardIng recall of occupation secondary to exposure. On the other hand, the relative risk remains 5.4 when comparing phenoxy acid and/or chlorophenol exposed workers exclusively In occupations other than agriculture/forestry with non-exposed workers 1n those same occupa- tions, thus, suggesting the presence of either recall bias or still another occupation with potential exposure to phenoxy adds and/or chlorophenols (Table 17). When woodworkers are separated out (possible exposure to chlorophenols In treatment of wood) the risk ratio becomes 9.7 (Table 18). These data suggest the presence of some recall bias. Another focus of this study was to determine 1f observational bias on the part of the Investigators could explain the significantly high risk estimates, lo answer the question, the study compared the exposure data derived from the Interviewee's returned questionnaires only with the combined Information from both the phone Interviews and questionnaires. The study found no substantial differences 1n the frequency of reporting exposure. Still a third consideration of possible bias Involves recall of exposure to phenoxy adds and/or chlorophenols because of subject knowledge of having cancer In the cases versus no knowledge of cancer In the referent population. The study chose as a referent group for the 52 STS cases (Hardell and Sandstrom, 1979) and the 169 malignant lymphomas (Hardell et al., 1981) a group of 154 colon cancer cases from the same population source and compared their exposure to phenoxy adds and/or chlorophenols by broad age groupings, and by rural vs. urban residence. C-144 �TABLE 17 Other Occupations (Minus Forestry/Agriculture)* Group Phenoxy Aclds/Chlorophenols Cases Referents Non-exposed 11 68 5 167 RR = 5.4 *Source: Hardell and EMkkson, 1981 C-145 Xs = 11.01 (P<0.01) �TABLE 18 Other Occupations (Minus Forestry/Agrlculture/Woodworkers)* Group Phenoxy Adds/Chlorophenols Non-exposed Cases 4 66 Referents 1 160 RR = 9.7 X 2 = 5.98 (P<0.05) *Source: Hardell and Erlkkson, 1981 C-H6 �Utilizing a Mantel-Haenszel rate ratio, the study found the risk of exposure to phenoxy adds remaining significantly high at 5.5 and to chlorophenols 5.4 In the STS cases compared to the colon cancer controls. Simi- larly, with the malignant lymphomas, the Identically derived risk ratios remain significantly high at 4.5 with respect to phenoxy acids and/or chlorophenol exposure 1n the cases, hence, the study concludes, no "substantial observational bias" exists. If the study Is assuming that recall bias was and Is the same as observational bias, then such a conclusion may not be entirely warranted from the comparison. Certainly, 1t appears that no recall bias existed because of subject "knowledge of having cancer" based on the authors analysis. But 1t does not rule out the possibility that recall bias can still be present In their data for other reasons. Hardell refers to an Intense "debate about phenoxy acids and their presumptive risk" In Sweden at the time the colon cancer study was conducted. But, there Is no reason to think that colon cancer victims would assume their disease was brought about from exposure to dloxln containing chemicals 1f no connection was suggested. It seems plausible that STS and/or non-Hodgkln's lymphoma patients would either learn at the time of their diagnosis that exposure to dloxln containing chemicals was the likely cause of this rare type of tumor or quickly learn from other sources, such as the news media, that exposure to herbicides containing dloxln could cause their rare form of cancer. Whereas, colon cancer victims (a rather common form of cancer) would not necessarily be led to believe that exposure to the same dloxln containing chemicals caused their disease. Hence, 1t 1s not difficult to Imagine that such unusual victims of cancer could better "remember" exposure to such chemicals than could colon cancer patients. C-147 �Therefore, although this study may explain any biases Introduced from secondary recall of occupation, observational bias Introduced from the telephone Interviewer and recall bias due to subject knowledge of cancer, 1t does not adequately answer questions of recall bias Introduced through the acquired awareness on the part of the victim of STS or non-Hodgk1n's lymphoma that his condition may have been caused by exposure to dloxln containing herbicides. Stomach Cancer. Studies of two of the oldest cohorts of workers known to have been exposed to phenoxyacetlc acid herbicides and/or 2,3,7,8-TCDD report stomach cancer mortality rates significantly higher than expected, but the results 1n each study were based on small numbers of deaths. In one study (Axelson et al., 1980), 348 Swedish railroad workers with at least 46 days of herbicide exposure between 1955 and 1972 were followed through October 1978. The workers were grouped on the basis of their primary herbicide exposures: those primarily exposed to phenoxyacetlc acids (2,4-D and 2,4,5-T) only, to amltrole (amlnotrlazole) only, and to both types of herbicides. After a 10-year latency was achieved, 3 stomach cancer deaths were observed vs. 0.71 expected (P<0.05). None were attributable to amltrol alone, but two were assigned to phenoxy adds alone while the remaining stomach cancer death occurred 1n a worker exposed to both amltrol and phenoxy adds In combination. The excess was more pronounced (3 observed vs. 0.57 expected, P<0.05) among those with early exposure (1957-1961) to phenoxy adds and/or amHrol. If persons who were exposed to just amltrol alone are excluded, thus leaving Individuals exposed to phenoxy acid alone and amltrol 1n combination, the excess 1s enhanced further (3 observed vs. 0.41 expected, P<0.01). C-148 �Axelson et al. (1980) also notes an excess 1n total "tumors" after 10 years latency as well (15 observed vs. 6.87 expected, P<0.005). This 1s pronounced In those exposed early to phenoxy acids alone (6 observed vs. 2.60 expected, P<0.01) and phenoxy acids In combination with amltrol (5 observed vs. 1.34 expected, P<0.05). Presumably, "tumors" In Sweden are analogous to malignant neoplasms In the United States. The author states that no specific type of tumor predominates and no breakdown by tumor type Is provided. The other study showing Increased stomach cancer mortality 1s the follow-up of 75 workers exposed to 2,3,7,8-TCDD during and after a 1953 runaway reaction at a trlchlorophenol manufacturing facility 1n Ludwlgshafen, Federal Republic of Germany (Thelss and Frentzel-Beyme, 1977). Two sources were used to calculate expected deaths: national mortality rates for the period 1971-1974, and 1972-1975 rates for Rhlnehessen-Palatlnate, the region In which Ludwlgshafen 1s located.* The results, shown 1n Table 19, Indicate an Increased rate of stomach cancer mortality that also Is not likely to have been due to chance alone. Two aspects of the methodology used should be noted that could have Influenced these results. First, the available report does not Include an analysis allowing for a minimum period of cancer Induction. It Is known that all three stomach cancer deaths 1n the Ludwlgshafen cohort occurred more than 10 years after Initial exposure. Employing a 10-year restriction to follow-up (as 1n the Swedish cohort study) would result In a higher relative risk estimate by reducing the number of expected deaths. *The report originally Included expected deaths using rates for the city of Ludwlgshafen, which were later shown to be Inaccurate. C-149 �TABLE 19 Analysis of Stomach Cancer Mortality In a Group of West German Factory Workers Exposed to 2,3,7,8-TCDD* Source for Expected Deaths Stomach Cancer Deaths Relative Risk Significance Level Observed Expected 3 0.559 5.4 0.02 3 0.495 6.1 0.01 Federal Republic of Germany 1971-1974 RhlnehessenPalatlnate 1972-1975 *Source: Thelss and Frentzel-Beyme, 1977 C-150 �Secondly, national and regional mortality rates from the 1970's were used to generate expected deaths to compare with observed mortality over a much longer period (1953-1977). Strong temporal trends In stomach cancer mortality In West Germany during the late 1950's and 1960's would make these expected figures Inaccurate. Without knowledge of such trends, the direc- tion and magnitude (1f any) of this possible source of bias cannot be estimated. The researchers also used an Internal control group which does not raise the second concern discussed above. This group consisted of 75 men, each matched to study group members by age and date of entry Into employment, and selected at random from a 11st of over 10,000 persons who had been Included 1n previous cohort studies by the same Investigators. No stomach cancer deaths occurred In this control group during the follow-up period. Thus, use of the Internal control groups also Indicates an excess of stomach cancers 1n the exposed workers. In an update of this earlier study, Thelss et al. (1982) continued the follow-up of his cohort through 1979 by adding 2 additional years of followup and apparently reducing the size of his cohort from 75 to 74. Altogether 21 deaths (4 more than from the earlier study) occurred vs. 18 and 19 deaths 1n the 2 matched (1 to 1) Internal comparison groups. WHh respect to can- cer deaths, the numbers were respectively 7, 5 and 5. The first control group was manually matched from the total number of persons (5500 Included In the cohort until the end of 1976) and the second, at random, by computer for some 8000 employees. In addition, 19 expected total deaths were esti- mated based on 1970-1975 mortality statistics of Rhlnehessln-Palatlnate, 18 expected deaths based on 1970-1975 mortality statistics of Ludwlgshafen, and 20 expected deaths based upon 1971-1974 mortality statistics of the Federal C-151 �Republic of Germany. Just as In the earlier study, the three stomach car- cinomas noted earlier appear to be significantly elevated Irregardless of which external comparison group 1s used (Table 20). On the other hand, one stomach cancer appeared In the randomized Internal control group. None appeared 1n the manually matched Internal control. No other elevated risks for any other cause were evident and no STSs appeared. When latency was considered only, the risk of stomach cancer remained significantly elevated after a lapse of 10 years (3 observed, 0.52 expected, P<.016) and then after a lapse of 15 years (2 observed, 0.23 expected, P<.02) based upon death rates of Rhlnehessln-Palatlnate, 1970-1975. Again, these study conclusions are limited by the small size of the study group and the very few cancer deaths noted at any particular site. Thus, It 1s Insensitive to the detection of a significantly elevated risk for most causes of cancer, especially STS and lymphomas. Although, stomach cancer 1s elevated significantly, 1t Is based only upon three deaths and since one stomach cancer death has been noted 1n an Internal control group 1n the updated version, H appears that this finding has been weakened somewhat. Furthermore, as was pointed out earlier, trends 1n stomach cancer mortality during the 1950's, 1960's and 1970's could make the comparison of stomach cancer mortality with expected deaths less valid based upon 1970-1975 rates. In summary, the evidence that phenoxyacetlc adds and/or 2,3,7,8-TCOD might Increase the risk of stomach cancer consists of two studies, each of which reports a statistically significant excess that Is based on only three stomach cancer deaths. Further follow-up of these and similar cohorts 1s warranted, but firm conclusions cannot be made on the basis of the available data. C-152 �TABLE 20 Reanalysls of Stomach Cancer Mortality 1n a Group of West German Factory Workers Exposed to 2,3,7,8-TCDD* Source for Expected Deaths Stomach Cancer Deaths Observed Federal Republic of Germany 1971-1974 Relative Risk Significance Level Expected 0.7 4.3 0.034 1970-1975 0.64 4.7 0.027 Ludwlgs-Shafen 1970-1975 0.61 4.9 0.024 Rhlnehessln- Palatlnate *Source: Thelss et al., 1982 C-153 �Four additional cohort studies have reported results that do not show Increased stomach cancer mortality rates In groups oF workers exposed to phenoxyacetlc acids and/or 2,3,7,8-TCDD. These are studies of 2,4,5-T production workers In Midland, Michigan (Ott et al., 1980), Finnish phenoxyacetlc add herbicide applicators (R11h1mak1 et al., 1978), the NHro study In which workers were exposed to 2,3,7,8-TCDD (Zack and Susklnd, 1980) and trlchlorophenol manufacturing workers (Cook et al., 1980a). As previously mentioned, the N1tro study Included a single death from STS and a weakly suggestive Increase In lymphatic and hematopoletlc system cancer mortality. The Midland study Included only one cancer death, a tumor 1n the respiratory system. In the Finnish study, hlstologlc Information on tumor types was not provided; however, there were no deaths from lymphoma. The results pertinent to stomach cancer mortality 1n the three studies are shown 1n Table 21. Neither the Midland study nor the NHro study con- tradicts the findings of the Swedish and West German Investigations previously discussed. This can be shown In two ways. First, the upper 95/4 confidence limits for the relative risk estimates from these two "negative" studies exceed even the highest point estimates of relative risk (6.1) from the two "positive" studies (see Tables 14 and 19). This Indicates that the relative risk estimates from the Midland and NHro studies, even though equal to zero, are nevertheless not significantly different from the estimates of 6.1, given the sample sizes, follow-up periods, age distribution and comparison group rates. C-154 �TABLE 21 Stomach Cancer Mortality In Three Studies of Workers Exposed to Phenoxyacetlc Add Herbicides and/or 2,3,7,8-TCDD Stomach Cancer Deaths Observed Relative Risk 95% Confidence Interval Reference Expected 0 0.14a 0 0-26.3 Ott et al ., 1980 5 6.9*.b 0.7 0.2-1.7 R11h1mak1 et al., 1978 0 0.5D 0 0-7.4 Zack and Susklnd, 1980 Estimated from total cancer expected deaths (see footnote 1n text). ^Entire follow-up period without regard for minimum time for cancer Induction (Ott et al., 1980 used a 10-year minimum Induction period). C-155 �In addition, the smallest relative risk In the Midland study (a = 0.05, <p = 0.2 one-tailed Polsson test) was 21.4 (3 observed deaths, 0.14 expected}.* Similarly, the smallest detectable relative risk In the NHro study (a = 0.05, 9 = 0.2, one-tailed Polsson test) was 10.0 (5 observed deaths, 0.5 expected). This calculation, however, was based on results for the entire follow-up period. If, as 1n the Midland study, a minimum period of cancer Induction had been employed, the expected deaths would have been fewer and the smallest reasonably detectable relative risk would have been greater. Midland This analysis of statistical power Indicates that the NHro and studies had very low probabilities of detecting the ~6-fold Increases 1n risk suggested by the Swedish and West German Investigations. Statistically, the study of Finnish herbicide applicators 1s Inconsistent with the results of the Swedish and West German cohort studies. The smallest reasonably detectable relative risk (a = 0.05, <p = 0.2, one- tailed Polsson test) was only 3.1 (11 observed deaths, 3.6 expected).t The study, therefore, appears powerful enough to detect relative risks even smaller than those seen In the Swedish and West German studies. A partial explanation for this apparent Inconsistency could lie In the fact that the *0tt et al. (1980) did not report expected deaths from stomach cancers. The figure 0.14 was obtained by multiplying the numbers of expected deaths from all cancers (2.6, allowing a 10-year minimum Induction period) by the percentage of stomach cancers among the expected deaths 1n the NHro study (0.5/9.04 = 5.554). The two studies used United States white male mortality rates and covered similar calendar years 1n follow-up (1949-1978 1n NHro and 1950-1976 1n Midland), but a similarity 1n age distributions cannot be established from the published reports. •f-The expected stomach cancer deaths were estimated 1n the same manner as for the Midland study. A proportion of 20% of all cancer deaths was applied because Finnish male mortality rates are known to be very high. C-156 �Finnish study set the minimum period of herbicide exposure for membership 1n the cohort at 10 days (2 working weeks) and noted that the "total strength of exposure has, In most cases, been a few weeks only." The Swedish study of herbicide applicators set the minimum exposure at 46 days {>! spraying season). There are also certain Inconsistencies In the data from the Finnish study which the authors note but find difficult to explain. In particular, no cancer deaths occurred during the latter part of the study period among Forestry Authority workers (1 of 4 groups Included 1n the cohort), even though 9.0 deaths were expected. This finding strongly suggests some defi- ciency In follow-up or In the source records from which vital status was determined. In summary, four cohort studies of workers exposed to phenoxyacetlc add herbicides and/or 2,3,7,8-TCOO do not report Increased risks of stomach cancer. Only one of these, however, was statistically powerful enough to be Inconsistent with the two studies that tentatively suggest an Increase 1n stomach cancer risk. The available report of this study of Finnish herbicide applicators contains methodologlc questions that require clarification. Summary of Epldemlologlcal Studies. The net result of adding together the number of workers exposed to phenoxy acids and/or chlorophenols from all case studies was an unusually high number of STSs, considering the rarity of the disease. It Is suggestive of an association of cancer with exposure to phenoxy adds and/or chlorophenols, and consequently, with the Impurities found In these herbicides, Including 2,3,7,8-TCDD. Two Swedish case-control studies report highly significant association of STS with exposure to phenoxy add and/or chlorophenols. They do not pinpoint the risk to the dloxln contaminants, however. C-157 In fact, 1n one study, �the risk was found to extend to phenoxy acids free of dloxln Impurities. In that study, the risk Increases to 17 when phenoxy acids known to contain dloxln Impurities (polychlorlnated dlbenzodloxlns and dlbenzofurans) are considered. The extent of observer bias and recall bias Introduced Into these studies by the employment of an undesirable methodology (self-administered questionnaires) Is probably not of sufficient magnitude to have produced the highly significant risks found In the studies. However, the possibility exists that these biases could have played a role 1n the determination of these risks, and consequently the data must be considered limited for the cardnogenldty of phenoxy add herbicides and/or chlorophenols 1n the absence of confirmatory studies. Later studies that did not reveal a significant excess risk of STS have severe limitations with their methodologies. These problems make these latter studies Inadequate to evaluate the risk of STSs from exposure to phenoxy acids and/or chlorophenols and, consequently, 2,3,7,8-TCDD. Therefore, the Swedish case-control studies provide limited evidence for the carc1nogen1c1ty of phenoxy adds and/or chlorophenols In humans. However, the evidence for the human carclnogenlclty for 2,3,7,8-TCDD based on the epldemlologlc studies Is only suggestive due to the difficulty of evaluating the risk of 2,3,7,8-TCDD exposure 1n the presence of the confounding effects of phenoxy adds and/or chlorophenol. Substantially weaker evidence exists Incriminating 2,4,5-T and/or 2,3,7,8-TCDD as the cause of malignant lymphoma and stomach cancer 1n humans. Studies In Animals When outbred Swiss mice were given weekly doses of 2,3,7,8-TCDD by gavage, an Increase In Hver tumors was observed (Toth et al., 1979). C-158 �Animals receiving 0.007 yg/kg/week for 1 year showed an elevated tumor Incidence over vehicle-treated mice; In the 0.7 yg/kg/week group the Increase was statistically significant (P<0.005). Mortality was sufficiently high at 7.0 yg/kg/week as to Interfere with cardnogenldty evaluation. D1G1ovann1 et al. (1977) reported a mouse skin painting study. The authors Indicated that 2,3,7,8-TCOD was a weak Initiator on the skin. A bloassay of 2,3,7,8-TCOD for possible cardnogenldty was conducted by the Illinois Institute of Technology Research, Chicago, Illinois, on a contract with the NCI Cardnogenesls Testing Program by dermal administration of the test material In Swiss-Webster mice for 104 weeks 1980a). (U.S. DHHS, Thirty male and female Swiss-Webster mice were dermally treated with an acetone suspension of 2,3,7,8-TCDO for 3 days/week for 104 weeks. Similar groups were pretreated with 1 application of 50 yg dlmethylbenzanthracene (DMBA) 1n 0.1 ma. acetone 1 week before 2,3,7,8-TCDO administration began. Female mice received 0.005 yg 2,3,7,8-TCDD/appl1cat1on, and the male mice received 0.001 yg 2,3,7,8-TCDD. As vehicle controls, 45 mice of each sex received 0.1 ml acetone 3 times/week. Thirty animals of each sex were used as untreated controls (Tables 22 and 23). Throughout the bloassay, mean body weights of the male or female groups of mice administered 2,3,7,8-TCDD, or 2,3,7,8-TCDD following DMBA, were essentially the same as those of the corresponding vehicle-control group. Mean body weights of dosed and vehicle-control groups of the females were less than those of the untreated control group throughout the study, and for the males were less than mean body weights of untreated controls during the first 80 weeks. In female mice, the Incidence of flbrosarcoma In the Integumentary system In groups dosed with 2,3,7,8-TCDD and 2,3,7,8-TCDD following DMBA was significantly higher than that 1n the corresponding controls (see Table 22). C-159 �TABLE 22 Incidence of Primary Tumors In Female Swiss-Webster Mice by Dermal Application of 2,3,7,8-TCDD or 2,3,7,8-TCDD Following DMBAa Tissue: Types of Neoplastlc Growth Vehicle Control 2,3,7,8-TCDD 2,3,7,8-TCDD plus DMBA Integumentary System; Flbrosarcoma 2/41 (554) 8/27 (30%)b 8/29 (28%)c 4/41 (10%) 1/25 (4%) 3/28 (11%) 5/41 (12%) 1/25 (4%) 3/28 (11%) 9/41 (22%) 2/25 (8%) 6/28 (21%) 14/41 (34%) 10/27 (37%) 8/29 (28%) 2/41 (5%) 3/41 (7%) 0/27 (0%) 0/27 (0%) 1/29 (3%) 1/29 (3%) Lung; Alveolar/Bronchlolar Adenoma Alveolar/Bronchlolar Carcinoma Alveolar/Bronchlolar Carcinoma or Adenoma Hematopo1et1c System; Lymphoma All Sites: Hemangloma Hemangloma or Hemanglosarcoma a Source: U.S. DHHS, 1980a bp<0.007 C P<0.010 C-160 �TABLE 23 Incidence of Primary Tumors 1n Male Swiss-Webster Mice by Dermal Application of 2,3,7,8-TCOD or 2,3,7,8-TCDD Following DMBA* Tissue: Types of Neoplastlc Growth Vehicle Control 2,3,7,8-TCDD 2,3,7,8-TCDD plus DMBA 3/42 (7%) 6/28 (21%) 6/30 (20%) 6/41 (15%) 1/28 (4%) 5/29 (17%) 1/41 (2%) 1/28 (4%) 2/29 (7%) 7/41 (17%) 2/28 (7%) 6/29 (21%) 4/42 (10%) 2/28 (7%) 5/30 (17%) 1/42 (2%) 4/28 (14%) 0/30 (0%) Integumentary System: Flbrosarcoma Lung: Alveolar/Bronchlolar Adenoma Alveolar/Bronchlolar Carcinoma Alveolar/Bronchlolar Carcinoma or Adenoma Hematopoletlc System: Lymphoma or Leukemia All Sites: Hemanglosarcoma *Source: U.S. DHHS, 1980a C-161 �It was concluded that, under the conditions of this bloassay, 2,3,7,8TCOO applied to the skin was carcinogenic for female Swiss-Webster mice, Inducing Increased Incidences of flbrosarcoma 1n the Integumentary system (U.S. OHHS, 1980a). Van Miller et al. (1977a,b) administered 0, 1, 5, 50, 1000 and 5000 ppt of 2,3,7,8-TCDD In the diet of male Sprague-Dawley rats. Higher concentra- tions (50, 500 and 1000 ppb) were also administered, but all of the rats fed at those three highest concentrations died early 1n the test. After 65 weeks, all surviving animals underwent laparotomles, and all tumors were blopsled. The rats were kept on the 2,3,7,8-TCDD diet for a total of 78 weeks and then placed on the control diet. After a total of 95 weeks all surviving animals were sacrificed and necropsled. The results of the study are summarized 1n Table 24. Although the study by Van Miller et al. (1977a,b) demonstrated the occurrence of neoplasms upon 2,3,7,8-TCDD exposure, the study has a number of shortcomings. The protocol was unusual and only a small number of animals were used. The occurrence of tumors did not follow a clear-cut dose-response relationship. Furthermore, the complete absence of tumors 1n the controls Is a highly unusual finding. Kodba et al. (1978) reported a more extensive carclnogenlclty study on 2,3,7,8-TCDD. Groups of 100 Sprague-Dawley rats (Spartan substraln, 50 males and 50 females/group) were maintained for up to 2 years on diets supplying 0.1, 0.01 or 0.001 vg 2,3,7,8-TCDD/kg/day. consisted of 86 males and 85 females. Ine control group The terminal necrospy examination was conducted at the end of 2 years of treatment. Females reclevlng 0.1 vg 2,3,7,8-TCDD/kg/day experienced a greater mortality than controls during the second half of the study. Extensive clinical chemistry data were reported as part of the study. C-162 �TABLE 24 Summary of Neoplastlc Changes After TCDD 1n Rats3 Concentration of 2.3,7.8-TCOD 1n Diet (ppt) Approximate Dally Dose No. of Animals with Neoplasms'5 No. of Neoplasms Diagnosis 0 0.0 0 0 NA 1 0.00004 0 0 NA 5 0.0001 5 6 1 ear duct carcinoma 1 lymphocytlc leukemia o I 1 adenocardnoma (kidney) cr> co 1 malignant hlstlocytoma (peritoneal) 1 anglosarcoma (skin) 1 Leydlg cell adenoma 50 0.0014 1 flbrosarcoma (muscle) 1 squamous cell tumor (skin) 1 astrocytoma (brain) 500 0.014 1 fibroma (muscle) 1 carcinoma (skin) 1 adenocardnoma (kidney) 1 scleroslng semlnoma (testes) �TABLE 24 (cont.) Concentration of 2,3,7.8-TCDD 1n Diet (ppt) 1000 Approximate Dally Dose No. of Animals with Neoplasms^ No. of Neoplasms 0.057 Diagnosis 1 cholanglocardnoma (liver) 1 anglosarcoma (skin) 1 glloblastoma (brain) 2 malignant histlocytoma (peritoneal) 5000 0.29 10 4 squamous cell tumors (lung) o 4 neoplastlc nodules (liver) 2 cholanglocarclnomas (Hver) a Source: Van Miller et al., 1977a ^10 animals per group NA = Not applicable �A portion of the data for the hlstopathologlc lesions found Is summarized 1n Table 25. The only lesions that are listed are those which were statistically different from control levels for at least one dose and In one sex. The following neoplastlc lesions were found to be Increased above con- trol levels (P<0.05): Hepatocellular hyperplastlc (neoplastic) nodules - females only Hepatocellular carcinomas - females only Stratified squamous cell carcinoma of palate or nasal turblnae - males and females Keratlnlzlng squamous carcinoma of the lung - females only Stratified squamous carcinoma of the tongue - males only Adrenal cortical ademona - males only Dr. Robert Squire, pathologist at the Johns Hopkins University Medical School and consultant to the U.S EPA Carcinogen Assessment Group (CAG), evaluated the hlstopathologlc slides from 2-year 2,3,7,8-TCOO by Kodba et al. (1978). rat feeding studies on Squire and his associates examined all livers, tongues, hard palates and nasal turblnates, and lungs available from the 2,3,7,8-TCDO study. His hlstopathologlcal findings, as well as Koclba's hlstopathologlcal evaluations, are summarized 1n Tables 26 and 27. Although there are some differences between the diagnoses of Kodba and Squire, the conclusions about the target organ for cancer Induction and the dose levels at which Induction occurred are the same whether Squire's or Koclba's diagnoses are considered. A bloassay of 2,3,7,8-TCDD for possible cardnogenldty was conducted by administering B6C3F the test material by gavage mice for 104 weeks (U.S. DHHS, 1980b). C-165 to Osborne-Mendel rats and Fifty rats and mice of each �991-3 ixaj iopxa jai)SL-| ai|i Aq I Q J ^ U O O UIQJ.J (so*0>d) l^ajsjjip AnupDUiuBis q 8/.6L ' * L P la Pqpox :aojno$ p E L 2 L qV OL 9 92 S 2 9 6 qS I 0 0 q2 9 i tL H 9 PIUO^DOUJO JLjDoaLjd 92 X3^ JOD [PU3JPP-PUIOU3PV SPdJDUPd puiouapp jpupy L 0 L 0 q2L EL 8L Efr 2 0 0 L qE L [ Q snoiupnbs payuej}s 0 fr * 8 2 Q 0 0 PUJOUPJPD AJPUJUJPW <n 9E SE £/. 2 [ 0 0 uisp[d -oau ^jpiuuipiu uB^uag 0 0 L L 9 5 Q LL 2L 82 Q 0 0 0 EL qi OL puiouapp AjpunUd ptuodiL/puiouappojqy /puiojqy snosup^noqns snjam-joumi uB t u3g L 0 0 0 puuoupjpo snoiupnbs Buiziu^pjs^-Buni (SJOUJH; a^euiqjn; LPSPU BuipntDui) qfr L 0 0 qfr 0 0 0 snoiupnbs pauUPJ^S qLL 2 0 L L 0 0 2 jpinueioiedaH q£2 q8L E 8 E 05 OS 98 OS LO'O LOO'O 0 L'O aLnpou D^s^Ld OS L'O SatPlUaj OS LO'O OS LOO'O 58 0 S3J.PW p xas 003i-8'/.'E'2 ^q paonpojj suoisan ouspidoaN jo 52 318V1 s[piuiuy jo jaqmnty App/6)|/6rt asoQ �TABLE 26 2,3,7,8-TCDD Oral Rat Study by Dr. Koclba, wHh Dr. Squire's Review (8/15/80) Female Sprague-Dawley Rats - Spartan Substraln (2 years) Dose Levels (yq/kg/day) Tissues and Diagnoses* 0 (control ) 0.001 0.01 0.1 S K S K S K 0/86 0/86 0/50 0/50 0/49 0/49 8/47 (P<10~3) 7/49 (P<10~3) 0/54 1/54 0/30 0/30 1/27 1/27 5/22 (P<10~2) 5/24 (P<10 2) Liver Neoplastlc nodules/ hepalocellular carcinomas 16/86 9/86 8/50 3/50 27/50 (P<10~4) 18/50 (P<10~3) 33/47 (P<10~8) 34/48 (P<10~12) Total combined (1.2, or above) (each 16/86 9/86 8/50 3/50 27/50 (P<10~4) 18/50 (P<10~3) 34/47 (P<10~8) 34/49 (P<10 il) Lung squamous cell carcinomas o Nasal turblnate/ hard palate squamous cell carcinomas S K animal had at least one Lumor above) *Where result 1s significantly different than the appropriate control, probability (P) of Incorrectly rejecting the null hypothesis 1s approximated 1n parentheses. S = Dr. Squire's hlstopathologlc analysis; K = Dr. Kodba's histopathologic analysis �TAUI f 71 2,3,7,8-TCDD Oral Rat Study by Dr. Kodba, with Dr. Squire's Review (8/15/80) Male Sprague-Dawley Rats - Spartan Substraln (2 years) Dose Levels (yq/kg/day) Tissues and Diagnoses* 0 (control ) S Nasal turblnate/ hard palate squamotis cell carcinomas Tongue squamous cell carcinomas Total - 1 or 2 above (each rat had at least one tumor above) 0.001 0.01 0.1 K S K S K S K 0/55 0/51 1/34 1/34 0/26 0/27 6/30 (P<10"2) 4/30 0/77 0/76 2/44 1/49 1/49 1/49 3/44 (P<0.05) 3/42 (P<10~s) 0/77 0/76 2/44 2/49 1/49 1/49 9/44 (P<10~4) 7/42 *Where result 1s significantly different than the appropriate control, probability (P) of Incorrectly rejecting the null hypothesis 1s approximated 1n parentheses. S = Dr. Squire's hlstopathologlc analysis; K = Dr. Koclba's hlstopathologlc analysis �sex were administered 2,3,7,8-TCDD suspended 1n a vehicle of 9:1 corn o1l:acetone 2 days/week for 104 weeks at doses of 0.01, 0.05 or 0.5 vg/kg/week for rats and male mice and 0.04, 0.2 or 2.0 yg/kg/week for female mice. controls. Seventy-five rats and 75 mice of each sex served as vehicle One untreated control group containing 25 rats and 25 mice of each sex was present 1n the 2,3,7,8-TCDD treatment room, and one untreated control group containing 25 rats and 25 mice of each sex was present In the vehicle control room. All surviving animals were killed at 105-107 weeks. In rats, a dose-related depression In mean body weight gain became evident In the males after week 55 of the bloassay and 1n the females after week 45. In mice, the mean body weight gain In the dosed groups was compar- able with that of the vehicle-control groups, but H was lower than that of the untreated controls. In male rats, Increased Incidences of folllcular-cell adenoma or carcinomas In the thyroid were dose related and were significantly higher 1n the low-, mid- and high-dose groups (P=0.001 for high dose) than 1n the vehicles controls (Table 28). A significant Increase In the subcutaneous tissue fibroma was found for the high-dose group (P=0.048). In female rats, the Incidences of folllcular-cell adenomas of the thyroid and hepatocellular carcinomas or neoplastlc nodules of the liver were dose-related, and the Incidence of hepatocellular carcinomas In the high-dose group was significantly higher (P=0.001) than that In the vehicle controls (Table 29), as were the Incidences of subcutaneous tissue flbrosarcoma (P=0.023) and adrenal cortical adenoma (P=0.039). In both male and female mice, Incidences of hepatocellular adenomas or carcinomas were dose related and the Incidences 1n the high-dose groups were higher (P=0.001 male, P=0.002 female) than those vehicle controls (Tables 30 and 31). C-169 1n the corresponding �TABLE 28 Incidence of Primary Tumors In Male Osborne-Mendel Rats (2,3,7.8-TCOD Administered by Gavage)a»b Dose (ug/kq/wk )c Tissue: Types of Neoplastlc Growth Vehicle Control Subcutaneous Tissue; Fibroma or Flbrosarcoma Fibroma Flbrosarcoma 12/75(16) 3/75(4) 9/75(12) 4/50(8) 1/50(2) 3/50(6) 5/50(10) 3/50(6) 3/50(6) 10/50(20) 7/50(14)d 3/50(6) 7/75(9) 4/75(5) 3/50(6) 3/50(6) 1/50(2) 0/50(0) 4/50(8) 4/50(8) Liver: Neoplastlc Nodule or Hepatocellular Carcinoma Neoplastlc nodule 0/74(0) 0/50(0) 0/50(0) 3/50(6) 0/74(0) 0/50(0) 0/50(0) 3/50(6) Pituitary: Adenoma or Chromophobe Adenoma Adenoma 2/61(3) 1/43(2) 3/43(7) 3/40(8) 0/61(0) 1/43(2) 2/43(5) 3/40(8) Adrenal : Cortical Adenoma Pheochromcytoma 6/72(8) 5/72(7) 9/50(18) 0/50(0) 12/49(24) 1/49(2) 9/49(18) 1/49(2) Thyroid; Folllcular Cell Adenoma or Carcinoma Folllcular Cell Adenoma C-Cell Adenoma C-Cell Adenoma or Carcinoma 1/69(1) 1/69(1) 2/69(3) 5/48(10)e 5/48(10) 2/48(4) 8/50(16)f 6/50(12) 4/50(8) 11/50(22)9 10/50(20) 4/50(8) 2/69(3) 2/48(4) 5/50(10) 4/50(8) Parathyroid: Adenoma 0/20(0) 2/41(5) 1/40(3) 1/36(3) Pancreatic Islets: Adenoma 2/70(3) 2/49(4) 3/48(6) 1/50(2) Circulatory System: Hemangloma or Hemanglosarcoma Hemanglosarcoma 0.01 C-170 0.05 0.5 �TABLE 28 (cont.) Tissue: Types of Neoplastlc Growth Mammary Gland: Adenocardnoma Flbroadenoma Dose (uq/kq/wk)c Vehicle Control 0.01 0/75(0) 5/75(7) 0/50(0) 0/50(0) a Source: U.S. DHHS, 1980b ^Values 1n parentheses Indicate percent response. C P = Values calculated using the Fisher Exact test d P = 0.048 e P = 0.042 f P = 0.004 9P < 0.001 C-171 0.05 3/50(6) 1/50(2) 0.5 1/50(2) 0/50(0) �TABLE 29 Incidence of Primary Tumors In Female Osborne-Mendel Rats (2,3,7,8-TCDD Administered by Gavage)a»b Tissue: Types of Neoplastlc Growth Subcutaneous Tissue: Fibroma or Flbrosarcoma Fibroma Flbrosarcoma Dose (uq/kq/wk)c Vehicle Control 0.01 4/75(5} 4/75(5) 0.05 0.5 3/50(6) 0/50(0) 0/75(0) 2/50(4) 0/50(0) 2/50(4) 3/50(6) 5/49(10) 1/49(2) 4/49(8)d Liver: Neoplastlc Nodule or Hepatocellular Carcinoma Neoplastlc nodule 5/75(7) 1/49(2) 3/50(6) 14/49(29)6 5/75(7) 1/49(2) 3/50(6) 12/49(24) Pituitary: Adenoma Chromophobe Adenoma 1/66(2) 5/66(8) 5/47(11) 0/47(0) 2/44(5) 0/44(0) 3/43(7) 1/43(2) Adrenal: Cortical Adenoma or Adenoma Cortical Adenoma or 11/73(15) 8/49(16) 4/49(8) 14/46(30)f Carcinoma or Adenoma 11/73(15) 9/49(18) 5/49(10) 14/46(30) Thyroid: FolUcular Cell Adenoma FolUcular Cell Adenoma 3/73(4) 2/45(4) 1/49(2) 6/47(13) 5/73(7) 7/73(10) 2/45(4) 1/45(2) 1/49(2) 8/49(26) 6/47(13) 6/47(13) 7/73(10) 3/45(7) 8/49(16) 6/47(13) 3/75(4) 27/75(36) 3/50(6) 20/50(40) 2/50(4) 21/50(42) 1/49(2) 17/49(35) 0/75(0) 3/47(6) 0/49(0) 0/48(0) or Carcinoma C-Cell Adenoma C-Cell Adenoma or Carcinoma Mammary Gland: Adenocardnoma Flbroadenoma Brain: Astrocytoma Source: U.S. DHHS, 1980b Values 1n parentheses Indicate percent response. C P = Values calculated using the Fisher Exact test d P = 0.0023; 6P = 0.001; fP = 0.039 C-172 �TABLE 30 Incidence of Primary Tumors In Female B6CF1 Mice (2,3,7,8-TCOO Administered by Gavage) a » b Tissue: Types of Neoplastlc Growth Dose (yg/kq/wk}c Vehicle Control 0.01 0.05 0.5 Subcutaneous Tissue: FIbrosarcoma 1/74(1) 1/50(2) 1/48(2) 5/47(ll)d Lung; Alveolar/Bronchlolar Adenoma Alveolar/Bronchlolar Adenoma or Carcinoma 2/74(3) 3/49(6) 4/48(8) 1/46(2) 2/74(3) 3/49(8) 4/48(8) 2/46(4) Hematopoletlc System; Lymphocytlc Lymphoma H1st1ocyt1c Lymphoma All Lymphoma Lymphoma or Leukemia 5/74(7) 9/74(12) 18/74(24) 18/74(24) 6/50(12) 4/50(8) 11/50(22) 12/50(24) 4/48(8) 8/48(17) 13/48(27) 13/48(27) 6/47(13) 14/47(30)e 20/47(43)f 20/47(43)f 3/73(4) 2/73(3) 1/73(1) 6/50(12) 4/50(8) 2/50(4) 6/48(13) 4/48(8) 2/48(4) 11/47(23)9 5/47(11) 6/47(13) Pituitary: Adenoma 0/62(0) 2/39(5) 0/38(0) 2/33(6) Thyroid: FolUcular-Cell Adenoma 0/69(0) 3/50(6) 1/47(2) 5/46(ll)n Liver: Hepatocellular Adenoma or Carcinoma Hepatocellular Adenoma Hepatocellular Carcinoma a Source: U.S. DHHS, 1980b ^Values In parentheses Indicate percent response. C P = Values calculated using the Fisher Exact test. d P = 0.032; eP = 0.016; fP = 0.029; 9P = 0.002; hP = 0.009 C-173 �TABLE 31 Incidence of Primary Tumors In Male B6CF1 Hlce (2,3,7.8-TCDD Administered by Gavage)a»b Tissue: Types of NeoplasUc Growth Subcutaneous Tissue: Flbrosarcoma or Fibroma Flbrosarcoma Dose (yg/kg/wk )c Vehicle Control 0.01 0.05 0.5 9/73(12) 8/73(11) 6/49(12) 5/49(10) 5/49(10) 4/49(8) 3/50(6) 3/50(6) Lung : Alveolar/Bronchlolar Adenoma or Carcinoma Alveolar/Bronchlolar Adenoma 10/71(14) 2/48(4) 4/48(8) 13/50(26) 7/71(10) 2/48(4) 4/48(8) 11/50(22) Hematopo1et1c System: H1st1ocyt1c Lymphoma Lymphoma or Leukemia 5/73(7) 8/73(11) 0/49(0) 3/49(6) 3/49(6) 4/49(8) 0/50(0) 6/50(12) Circulatory System: Hemanglosarcoma 1/73(1) 2/49(4) 1/49(2) 3/50(6) 15/73(21) 7/73(10) 8/73(11) 12/49(24) 3/49(6) 9/49(18) 13/49(27) 5/49(10) 8/49(16) 27/50(54)d 10/50(20) 17/50(34) FolUcular-Cell Adenoma 0/69(0) 3/48(6) 0/48(0) 0/49(0) F^ye/Lacr1mal Glands: Adenoma 0/73(0) 1/49(2) 1/49(2) 3/50(6) Uver: Hepatocellular Adenoma or Carcinoma Hepatocellular Adenoma Hepatocellular Cardnoma Thyroid: a Source: U.S. DHHS, 1980b ^Values 1n parentheses Indicate percent response. C P = Values calculated using the Fisher Exact test. d P<0.001 C-174 �In female mice, folllcular-cell adenomas 1n the thyroid and hlstocytlc lymphomas In the hematopoletlc system occurred at dose-related Incidences, and the Incidences were significantly higher In the high-dose groups than those 1n vehicle controls. The high-dose group of females also showed a significantly higher Incidence of subcutaneous fIbrosarcomas (P=0.032) and lymphoma or leukemia (P=0.029) (see Table 31). It was concluded that, under the conditions of this bloassay, 2,3,7,8TCDO was carcinogenic for Osborne-Hendel rats, Inducing significant doserelated Increased Incidences of folUcular-cell thyroid tumors In males and liver tumors In females. 2,3,7,8-TCDD was also carcinogenic for B6C3F mice, Inducing significant dose-related Increased Incidences of liver tumors In males and females and of thyroid tumors 1n females (U.S. DHHS, 1980b). C-175 �CRITERION FORMULATION Existing Guidelines and Standards The National Academy of Sciences Committee on Drinking Water and Health (NAS, 1977) suggested an acceptable dally Intake (ADI) of 10"4 yg 2,3,7,8TCDD/kg/day. At that time, 2,3,7,8-TCDD was not considered to be a carcinogen, and the ADI was based on a 13-week feeding study In rats by Kodba et al. (1976). The FDA has Issued a health advisory stating that fish with residues of 2,3,7,8-TCDD >50 ppt should not be consumed, but fish with residues of <25 ppt pose no serious health concern (FDA, 1981, 1983; Cordle, 1981). Federal legal limits for Great Lakes fish distributed In Interstate commerce were deemed unnecessary because most samples analyzed by the FDA contained <25 ppt. Canada has established a 20 ppt concentration limit for 2,3,7,8-TCDO In Lake Ontario commercial fish exported Into the United States to comply with the levels believed by FDA to be safe. lished for 2,3,7,8-TCDD on food crops. No tolerances have been estab- A tolerance of 0.05 ppm hexachloro- phene 1n or on cottonseeds (used as livestock feed), with a proviso that the technical grade of hexachlorophene shall not contain >0.1 ppm 2,3,7,8-TCDD, was published 1n 40 CFR 180.302. The Ministry of Labour of Canada has set a tentative Ambient Air Quality criterion for PCDDs of 30 pg/m3 (Harding, 1982). Current Levels of Exposure The extent of human exposure to 2,3,7,8-TCDD that can be directly attributed to the water route cannot be readily determined. While 2,3,7,8- TCDD does not appear to occur naturally 1n the environment, 1t can be produced with low efficiency from the combustion of 2,4,5-T-conta1n1ng mate- C-176 �rials (Stehl and Lamparskl, 1977); H may also be produced 1n a large variety of normal combustion processes (Anonymous, 1978; Bumb et al., 1980), but It 1s not produced during all combustion processes (Klmble and Gross, 1980). The Impact of these processes on human exposure Is unknown. The high affinity of 2,3,7,8-TCOD for soils with significant organic content would seem to reduce the likelihood of groundwater contamination; however, as the organic content of soil declines the likelihood of groundwater contamination by 2,3,7,8-TCDD Increases. Contaminated beef fat samples have been found to have concentrations as high as 60 ppt of 2,3,7,8-TCDD In one sample (Ross, 1976). 2,3,7,8-TCDD residues also have been detected In the edible portions of fish from the Tlttabawassee, Grand and Saglnaw Rivers, Lake Michigan and the Saglnaw Bay In Michigan at concentrations ranging from 4-695 ppt (Harless and Lewis, 1980). The reports of Incidents of 2,3,7,8-TCDD exposures In Industrial plants and of accidents where 2,3,7,8-TCDD was more widely disseminated are useful In Identifying some of the effects of 2,3,7,8-TCDD exposure 1n man. Unfor- tunately, the existing human data can only roughly estimate the extent and duration of 2,3,7,8-TCDD exposure which produced the toxic symptoms. Special Groups at Risk The most obvious groups at risk are those employed In the manufacture of chemicals In which 2,3,7,8-TCDD may occur as an unwanted by-product. The spraying of herbicides containing traces of 2,3,7,8-TCDD has become less of a problem because of restrictions on the use of such agents. Considering the reproductive toxlclty of 2,3,7,8-TCDD, women of child-bearing age, and especially the fetus, are at high risk from exposures to 2,3,7,8-TCDD. C-177 �Basis and Derivation of Criterion 2,3,7,8-TCDD 1s an unusually toxic compound with demonstrated subacute and chronic effects In animals and man. acute, Acute or subchronlc expo- sures to 2,3,7,8-TCDD can adversely affect the skin, the liver, the nervous system and the Immune system. 2,3,7,8-TCDD displays an unusually high degree of reproductive toxlc1ty. It 1s teratogenVc, fetotoxlc and reduces fertility. In a 3-generat1on reproductive study, Murray et al. (1979) reported a reduction 1n fertility after dally dosing at 0.1 or 0.01 generations of Sprague-Dawley wg rats. Although Murray et al. (1979) con- sidered the lowest dose tested, 0.001 level (NOEL), a re-evaluation 2,3,7,8-TCDD/kg 1n the F, and F_ yg/kg, to be a no-observed-effect of these data by Nlsbet and Paxton (1982), using different statistical methods, Indicated that there was a reduction 1n the gestation Index, decreased fetal weight, Increased liver to body weight ratio, and Increased Incidence of dilated renal pelvis at the 0.001 dose. The re-evaluated vg/kg data would suggest that equivocal adverse effects were seen at the lowest dose (0.001 wg/kg/day) and that this dose should, therefore, represent a lowest-observed-adverse-effect level (LOAEL). Schantz et al. (1979) found reductions 1n fertility and various other toxic effects In rhesus monkeys fed a 50 ppt 2,3,7,8-TCDD diet for 20 months. This corresponds to a calculated dally dose of 0.0015 vg 2,3,7,8-TCDD/kg/ day. These results suggest that monkeys may be somewhat more sensitive than rats, since the effects 1n monkeys were more severe and not equivocal. A tox1c1ty-based criterion has been calculated for comparison with the cancer-based criterion 1n accordance with public comments. Since the data from the limited study by Schantz et al. (1979) are supportive of the findIngs by Murray et al. (1979), 1t seems reasonable to determine an ADI based on the LOAEL. If one selects an uncertainty factor of 100 based on the C-178 �existence of lifetime animal studies and knowledge of effects In man as per NAS (1977) guidelines, and then an additional 10 because a LOAEL 1s used as the basis of this calculation,* then the ADI would be: 10~3 yg/kg/day (LOAEL) ADI = = 1 x 10~« yg/kg/day. 100 x 10 Thus, the acceptable dally Intake for a 70 kg man would be 7.0xlO~5 yg 2,3,7,8-TCOD/day. Using a BCF of 5000 and assuming a dally consumption of 6.5 g of fish, the water concentration corresponding to this ADI would be: 7.0 x 10"5 water concentration = = 2.0 x 10~* yg/l. 2 + (5000 x 0.0065) However, this concentration may not be sufficiently protective of human health since It does not take Into account the demonstrated carcinogenic effects of 2,3,7,8-TCDD In animals and the probability that 2,3,7,8-TCDD 1s a human carcinogen (see cancer-based criterion derivation). The carcinogenic potential of 2,3,7,8-TCDD has been established by feedIng studies In rodents. The results of the study by Van Miller et al. (1977a,b) are summarized In Table 24, and the findings of the more extensive study by Kodba et al. (1978) are summarized 1n Table 25. The Van Miller et al. (1977a,b), the Toth et al. (1979) and recent NCI data (U.S. DHHS, 1980a,b) summarized 1n Tables 28, 29, 30, and 31, reinforce the findings of Kodba et al. (1978) and establish that 2,3,7,8-TCDD 1s an animal carcinogen and 1s probably carcinogenic In humans. *Accord1ng to the methods published by EPA (45 FR 79353), an additional uncertainty factor between 1 and 10 must be used because the calculation 1s based on a LOAEL. An uncertainty factor of 10 was chosen because of the adverse effects seen 1n rhesus monkeys at 0.0015 yg/kg/day, despite the equivocal nature of the effects In rats seen at the 0.001 yg/kg/day dose level. C-179 �Furthermore, the ep1dem1olog1cal findings (Harden, 1977, 1979; U.S. EPA, 1980c; Hardell et al., 1980; Hardell and Sandstrom, 1979; Erlcksson et al., 1979; Thelss and Frentzel-Beyme, 1977; Jlrasek et al., 1973, 1974; Pazderova et al., 1974; Axelson et al., 1980; Zack and Susklnd, 1980; Zack, 1980; Cook et al., 1980a) are consistent with the conclusion from animal studies that 2,3,7,8-TCDO 1s a probable human carcinogen. In addition, 2,3,7,8-TCOD has been shown to be a potent liver cancer promoter (PHot et al., 1980) and a cocardnogen (Kourl et al., 1978). Under the Consent Decree 1n NROC vs. Train, criteria are to state "recommended maximum permissible concentrations (Including where appro- priate, zero) consistent with the protection of aquatic organisms, human health, and recreational activities." 2,3,7,8-TCDD 1s suspected of being a human carcinogen. Because there 1s no recognized safe concentration for a human carcinogen, the recommended concentration of 2,3,7,8-TCOD In water for maximum protection of human health Is zero. Because attaining a zero concentration level may be 1nfeas1ble In some cases, and In order to assist the Agency and states In the possible future development of water quality regulations, the concentrations of 2,3,7,8-TCDD corresponding to several Incremental lifetime cancer risk levels have been estimated. A cancer risk level provides an estimate of the additional Inci- dence of cancer that may be expected 1n an exposed population. A risk of 10~5, for example, Indicates a probability of one additional case of cancer for every 100,000 people exposed, a risk 10~« Indicates one additional case of cancer for every million people exposed, and so forth. In the November 1980 Federal Register notice of availability of ambient water quality criteria (45 FR 79318), the U.S. EPA presented a range of concentrations for carcinogens corresponding to cancer risks of 10~5, 10"' C-180 �or 10"7, based on the upper 95V. confidence level of calculated Incremental risk. The criteria for 2,3,7,8-TCDD are shown In the following table: Exposure Assumptions (per day) 9554 Upper-Limit Risk Levels and Corresponding 95% Lower-Limit Criteria (1) for 2,3,7,8-TCOD 0 1(T' 10"' 10"5 2 8. of drinking water and consumption of 6.5 g fish and shellfish. (2) 0 1.3xl(T9 1.3xlO~8 1.3x10"' Consumption of fish and 0 1.4xlO~9 1.4xlO"8 1.4x10"' 0 2.2xlO"8 2.2x10"' 2.2x10"* shellfish only 2 8. of drinking water only (1) The animal bloassay data used 1n these calculations are presented 1n the Appendix of this document. These levels are calculated by applying a linearized multistage model as discussed In the Human Health Methodology Appendices to the Federal Register notice concerning water quality criteria. Since the extrapolation model Is linear at low doses, the additional lifetime risk Is directly proportional to the water concentration. Therefore, water concentrations corresponding to other risk levels can be derived by multiplying or dividing one of the risk levels and corresponding water concentrations shown In the table by factors such as 10, 100, 1000 and so forth. (2) Approximately 94.2% of the 2,3,7,8-TCDO exposure results from the consumption of aquatic organisms which exhibit an average bloconcentratlon potential of 5000-fold. results from The remaining 5.8% of 2,3,7,8-TCDO exposure drinking water. Correspondingly, 1f no contaminated shellfish or fish are eaten, the water contamination level could be 17 times as high for the same 10"6 upper-limit risk risk level, or 2.2x10"' yg/J. for a level, vs. 1.3xlO~8 yg/8, when contaminated fish and water are consumed. C-181 �Concentration levels were derived assuming a lifetime exposure to various amounts of 2,3,7,8-TCDD, (1) occurring from the consumption of both drinking water and aquatic life taken from waters containing the corresponding 2,3,7,8-TCOD concentrations, (2) occurring solely from consumption of aquatic life grown In waters containing the corresponding 2,3,7,8-TCDD concentrations and (3) occurring from the consumption of drinking water only. Because data Indicating other sources of 2,3,7,8-TCDD exposure and their contributions to total body burden are Inadequate for quantitative use, the figures reflect the Incremental risks associated with the Indicated routes only. The above criteria, which have been calculated on the basis of health effects data, are below the limit of detection for 2,3,7,8-TCDD 1n water by current analytical methods. The detection limit presently Is estimated to be ~3xlO"5 yg/8. (Harless et al., 1980). The detection limit should also be considered when Issuing guidance based on these criteria. Estimates by Others of Carcinogenic Potency and Criteria The U.S. Food and Drug Administration concluded that an advisory level of 25 ppt for Great Lakes fish contaminated with 2,3,7,8-TCDD does not pose an unacceptable risk to public health (FDA, 1981). EPA has reviewed the recent testimony before Congress of Dr. S.A. Miller (FDA, 1983), discussing cancer risk associated with Ingestlon of these fish. The FDA estimate of the 9554 upper-limit carcinogenic potency factor for 2,3,7,8-TCDD 1s q * = 1.75xl04 (mg/kg/dayT1, which q * = 1.56xlOs (mg/kg/day)~a 1s less (see potent than EPA's estimate of Appendix) by a factor of 9. Even though both Agencies used the same data base (Kodba et al., 1978) and risk extrapolation model, some subtle differences In methodology exist which account for this factor of 9. The major part of this difference 1s a factor C-182 �of S.38 which EPA uses for rat-to-man extrapolation on the assumption that dose per unit body surface area, rather than dose per unit body weight, Is an equivalent dose between species (45 FR 79351). Most of the remaining factor of -1.7 Is due to the FDA's use of the Kodba h1stopatholog1cal diagnosis alone, without Including that of Squire, and EPA's adjustment of Us calculations to compensate for the high early mortality observed 1n the Kodba et al. (1978) study (see Appendix). FDA and EPA also differ 1n their assessment of human exposure to 2,3,7,8-TCDD 1n fish, In keeping with their respective regulatory approaches. EPA calculates water quality criteria to protect a body of water as though H were the direct source of 100% of a human population's average dally Intake of water and/or freshwater and estuarlne fish or shellfish. The concentration of a pollutant In the tissues of all such fish or shellfish Is further assumed to be determined by the water concentration and the bloconcentratlon factor (BCF) of the pollutant. FDA, on the other hand, premised Us exposure assessment on the assumption that only limited amounts of fish having 2,3,7,8-TCDD levels at or near the advisory level will actually be consumed. For example, FDA assumed that for this substance, significant contamination problems were limited to bottom feeders such as catfish and carp.* It also assumed that actual average residue levels In the flesh of bottom-feeding species reaching the market would not exceed one-third of the advisory level (I.e., ~8 ppt) and further, that for most Individuals, 90% of the fish consumed would be comprised of other species showing no measurable contamination, or would be taken from uncontamlnated *However, available data Indicate that other species, especially trout and salmon, taken from some areas of the Great Lakes may also have tissue residues of 2,3,7,8-TCDD which exceed 25 ppt (see Table 1). C-183 �areas. Under these assumptions, and using an upper 90 percentlle value for freshwater fish consumption of 15.7 g/day, the FDA potency estimate yields an upper-limit risk estimate of 2.86xlO~6 for consumers of these fish. If the same exposure assumptions were used with EPA's potency estimate a somewhat higher upper limit risk of 2.92xlO~5 would result. The Center for Disease Control (CDC) has also calculated an upper-limit potency value for 2,3,7,8-TCDD (Klmbrough et al., 1983). The CDC estimate Is based on the Squire hlstopathologlcal results, and, like that of FDA, extrapolates from rat to man on a basis of dose equivalence per unit body weight. The CDC difference from both the EPA and FDA approaches Is that the curve fit was done, not on administered dose, but on liver concentration at terminal sacrifice. mortality. Also, like FDA, CDC did not adjust for high early The final result 1s that the CDC 95% upper-limit potency value estimate when converted back to administered dose Is q * = 3.6xl04 (mg/kg/day)"1 which 1s more potent by a factor of 2 than that of FOA and less potent by a factor of 4 than that of the EPA. In January 1984 the three Agencies met to review the differences 1n carcinogenic potency estimation. The three Agencies agreed that they are using virtually the same methodologies for potency estimation although there are differences In some assumptions used. Further, there was agreement that correction for mortality 1s appropriate, making the differences less between the EPA estimate and the other estimates. Lastly, the Agencies agreed that the remaining differences are within the range of uncertainty Inherent 1n the risk assessment process. *The difference between the EPA and FDA risk estimates results from the difference 1n potency estimates, described above, and the use by FDA of an average human body weight of 80 kg, versus 70 kg used by EPA. C-184 �REFERENCES Albro, P.W., J.T. Corbett, M. Harrlss and L.O. Lawson. 1978. Effects of 2,3,7,8-tetrachlorod1benzo~p-d1ox1n on llpld profiles In tissue of the Fischer rat. Chem. B1ol. Interact. 23(3): 315-330. Aldred, J.E. 1978. Report of the Consultative Council on Congenital Abnor- malities 1n the Yarrom District. Minister of Health, Melbourne, Victoria, Aust. (Cited In M1lby et al., 1980). Allen, J.R., D.A. BarsotU, L.K. Lambrecht and J.P. Van Miller. 1979. Reproductive effects of halogenated aromatic hydrocarbons on nonhuman primates. Ann. NY Acad. Scl. 320: 419-425. Allen, J.R., D.A. Barsottl, J.P. Van Miller, L.J. Abrahamson and J.J. Lallch. 1977. Morphological changes In monkeys consuming a diet containing low levels of TCDD. Food Cosmet. Toxlcol. 15: 401. Allen, J.R. and J.J. Lallch. 1962. Response of chickens to prolonged feed- Ing of crude "toxic fat". Proc. Soc. Exp. B1ol. Med. Allen, J.R., J.P. Van Miller and D.H. Norback. 1975. 109: 48-51. Tissue distribution, excretion, and biological effects of [ l4 C]tetrachlorod1benzo-p-d1ox1n In rats. Food Cosmet. Toxlcol. 13(5): 501-505. C-185 �Analzl, N.H. and J.3. Cohen. 1978. The effects of 2,3,7,8-tetrachlorocM- benzo-p-d1ox1n on the renal tubular secretion of phenolsulfonphthaleln. J. Pharmacol. Exp. Ther. 207: 748-755. Anonymous. 1978. 01ox1ns formed by normal combustion. Chem. Eng. News. 56: 7. Anonymous. sarcomas. 1982. Phenoxy herbicides, trlchlorophenols, and soft-tissue May 8. Lancet, p. 1051-1052. Appelgren, L.-E., I. Brandt, E.B. BMttebo, M. Glllner and J.-A. Gustafsson. 1983. Autoradlography of 2,3,7,8-tetrachloro- l4 Cl-d1benzo-p-d1ox1n TCDO: Accumulation 1n the nasal mucosa. Chemosphere. 12(4/5): 545-548. Axelson, 0. 1980. A note on observational bias case-referent studies. Scand. J. Work Environ. Health. 6(1): 80-82. Axelson, 0., L. Sundell, K. Anderson, C. Edllng, C. Hogstedt, and H. K11ng. 1980. Herbicide exposure and tumor mortality: An updated ep1dem1olog1cal Investigation on Swedish railroad workers. Scand. J. Work Environ. Health. 6: 73-79. Barnes, D.G. waste. 1983. "Dloxln" production from combustion of blomass and Presented at the Symp. Energy from Blomass and Wastes VII, Lake Buena Vista, FL, Jan. 24-28. C-186 �Bartsch, H., L. Tomatls and C. Malavellle. 1982. Qualitative and quantitative comparlslons between mutagenlc and carcinogenic activities of chemicals. In: Mutagenlclty: New Horizons In Genetic Toxicology, John A. Heddle, Ed. Academic Press, NY. p. 36-72. Bauer, H., K.H. Schulz and A. Splegelberg. 1961. be1 der herstellung von chlorphenol-verblndunger. BerufUche verglftungen Arch. Gewerbeath. Gewer- behyg. 18: 538-555. (Ger.) Beatty, P. and R.A. Neal. 1976. tetrachlorod1benzo-p-d1ox1n Induction of DT-d1aphorase by 2,3,7,8- (TCOD). Blochem. Blophys. Res. Commun. 12. 68(1): 197-204. Beatty, P.W., W.K. Vaughn and R.A. Neal. 1978. Effect of alteration of rat hepatic mixed-function oxldase (MFO) activity on the toxlclty of 2,3,7,8tetrachlorod1benzo-p-d1ox1n (TCDD). Toxlcol. Appl. Pharmacol. 45(2): 513-519. Benedict, W.F., N. Consldlne and D.W. Nebert. 1973. Genetic differences In aryl hydrocarbon hydroxylase Induction and benzo[a]pyrene-produced genesls 1n the mouse. Mol. Pharmacol. tumorl- 9: 266-277. Benfenatl, E., F. G1zz1, R. Reglnato, R. FanelH, M. Lod1 and R. Taghaferrl. 1983. PolychloMnated d1benzo-p-d1ox1ns (PCDD) and polychlorlnated dlbenzo- furans (PCOF) 1n emissions from an urban Incinerator. II. Correlation between concentration of mlcropollutants and combustion conditions. Chemosphere. (In press) C-187 �Berry, D.L., T.J. Slaga, J. DlG1ovann1 and M.R. Ouchau. 1979. Studies with chlorinated d1benzo-p-d1ox1ns, polybromated blphenyls, and polychlorlnated blphenyls 1n a two-stage system of mouse skin tumoMgenesIs: Potent anticarcinogenic effects. Bert, J., et al. Ann. NY Acad. Sc1. 320: 405-414. 1976. I controlll sanltorl: La sanHa1 Incontrollata. Sapere. 796: 50. (Ita.) Blsantl, L., F. Bonettl, F. Caramaschl, et al. 1980. accident of Seveso. Acta. Morphol. Acad. Sd. Hung. Experiences from the 28(1-2): 139-157. Bishop, C.M. and A.H. Jones. 1981. Non-Hodgkln1 s lymphoma of the scalp In workers exposed to dloxlns. Lancet. 2(8242): 369. Blelberg, J., M. Wallen, R. Brodkln and I.L. Applebaum. 1964. Industrially acquired porphyrla. Arch. Dermatol. 89: 793-797. Bogen, G. 1979. Am. Med. Assoc. Symptoms 1n Vietnam veterans exposed to Agent Orange. J. 242(22): 2391. Bonaccorsl, A., R. FanelH and G. Tognonl. 1978. In the Wake of Seveso. Amblo. 7(5-6): 234-239. Bronzettl, G., C. Bauer, C. Corsl, R. Del Carratare, R. Neerl and M. Paollne. 1983. Mutagenlclty study of TCOD and ashes from urban Incinerator "In vitro" and "In vivo" using yeast 07 strain. Chemosphere. 12: 549-553. C-188 �Brumley, W.C., J.A.G. Roach, J.A. Sphon, et al. 1981. Low-resolution multiple Ion detection gas. Chromatograph1c-mass spectrometrlc comparison of six extraction-cleanup methods for determining 2,3,7,8-tetrachlorod1benzo-pdloxln In fish. J. Agrlc. Food Chem. 29: 1040-1046. Bumb, R.R., W.B. Grummet, S.S. Cutle, et al. fire: A source of chlorinated dloxlns. Science. 1980. Trace chemistries of 210: 385. Burns, L.H., D.M. CUne and R.R. Lasslter. 1981. Exposure analysis modelIng system (EXAMS). Prepared by Environmental Research Laboratory, ORO, U.S. EPA, Athens, GA. Buser, H.R. 1979. Formation of polychlorlnated dlbenzofurans (PCDFs) and dlbenzo-p-dloxlns (PCOOs) from the pyrolysls of chlorobenzenes. Chemos- phere. 8: 415-424. Buser, H.R. and H.P. Bosshardt. dlbenzofurans, and benzenes 1978. Polychlorlnated d1benzo-p-d1ox1ns, In the fly ash of municipal and Industrial Incinerators. Mitt. Geb. Lebens. Hyg. 69: 191-199. Buser, H.R. and C. Rappe. 1980. High resolution gas chromatography of the 22 tetrachlorod1benzo-p-d1ox1n Isomers. Anal. Chem. 52: 2257-2262. Buser, H.R. and C. Rappe. 1983. Isomer-speclf1c separation and analysis of 2,3,7,8-substHuted polychlorlnated d1benzo-p-d1ox1ns (PCDDs) using highresolution gas chromatography and mass spectrometry. review) C-189 Anal. Chem. (In �Canton!, L., M. Salmona and H. R1zzard1n1. 1981. Porphyrogenk effect of chronic treatment with 2,3,7,8-tetrachlorod1benzo-p-d1ox1n In female rats. Dose-effect relationship following urinary excretion of porphyrlns. Tox1col. Appl. Pharmacol. 57: 156-163. Carrerl, V. 1978. Review of the events which occurred In Seveso. in: D1ox1n-tox1colog1cal and Chemical Aspects, F. Cattabenl et al., Ed. Spectrum Publishers, Inc., NY. Carter, C.O., R.D. Klmbrough, J.A. Llddle, R.E. Cllne, M.M. Zack, Jr. and W.F. Barthel. Science. 1975. TCDD: As accidental poisoning episode In horse arenas. 188: 738-740. CEQ (Council on Environmental Quality). 1981. Environmental Quality 1981, 12th Ann. Rep. Council Environ. Qual, Washington, DC. p. 129. Cocuccl, S., F. OIGerolamo, A. Verderlo, et al. 1979. Absorption and translocatlon of tetrachlorod1benzo-p-d1ox1n by plants from polluted soil. Exper1ent1a. 34: 483-484. Cole, P. 1979. The evolving case-control study. 3. Chron. DIs. 32: 15-27. Collins, T.F.S., C.H. Williams and G.C. Gray. with 2,4,5-T and 2,4-D In the hamster. 6(6): 559-567. C-190 1971. Teratogenlc studies Bull. Environ. Contam. Toxlcol. �Cook, R.R. 1981a. Dloxln, chloracne and soft-tissue sarcoma. Lancet. 1: 618-619. Cook, R.R. 1981b. Author's Response. Cook, R.R., et al. 1980. J. Occup. Med. 23: 8. Mortality experience of employees exposed to 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCOD). J. Occup. Med. 22: 530. Cordle, F. 1981. The use of epidemiology 1n the regulation of dloxlns 1n the food supply. Regul. Toxlcol. Pharmacol. Cordle, F. 1983. food supply. 1: 379-387. Use of epidemiology In the regulation of dloxlns In the In: Accidental Exposure to Dloxlns: Human Health Aspects, F. Coulston and F. Pocch1ar1, Ed. Academic Press, NY. p. 245-256. Coulston, F. and E.J. Olajos. Epidemiology of 2,4,5-T. 1980. Panel report: Panel to Discuss the New York City, July 10-11, 1979. Ecotoxlcol. Environ. Safety. 4: 96-102. Courtney, K.D. 1976. Mouse teratology dloxlns. Bull. Environ. Contam. Toxlcol. studies with chlorodlbenzo-p- 16(6): P674-681. Courtney, K.D. 1977. Prenatal effects of herbicides evaluation by the prenatal development Index. Arch. Environ. Contam. Toxlcol. Courtney, K.D., D.W. Gaylor, M.D. Hogan and H.L. Falk. 6(1): 33-46. 1970a. Teratogenlc evaluation of pesticides: Large-scale screening study. Teratology. C-191 3: 199. �Courtney, K.D., D.W. Gaylor, H.D. Hogan, M.L. Falk, R.R. Bates and I. Mitchell. 1970b. Teratogenlc evaluation of 2,4,5-T. Science. 168: 864-866. Courtney, K.D. and O.A. Moore. 1971. Teratology studies with 2,4,5-T and 2,3,7,8-TCDD. Toxlcol. Appl. Pharmacol. 20: 396-403. Courtney, K.D., J.P. Putnam and J.E. Andrews. 1978. Metabolic studies with TCOO (dloxln) treated rats. Arch. Environ. Contain. Toxlcol. 7(4): 385-396. Crosby, D.G. and A.S. Wong. 1977. Environmental degradation of 2,3,7,8- tetrachlorod1benzo-p-d1ox1n (TCDD). Science. 195: 1337-1338. Crosby, O.G., A.S. Wong, J.R. PUmmer and E.A. Woolson. composHlon of chlorinated d1benzo-p-d1ox1ns. Science. 1971. Photode- 173(3998): 748-749. Crow, K.D. 1981. Chloracne and Us potential clinical Implications. CUn. Exp. Oermatol. 6(3): 243-257. Crummett, W.B. and R.H. Stehl. 1973. Determination of chlorinated dlbenzop-d1ox1ns and dlbenzofurans 1n various materials. Environ. Health Perspect. 5: 15-25. Cunningham, H.M. and D.T. Williams. 1972. Effect of 2,3,7,8-tetrachlorodlbenzo-p-d1ox1n on growth rate and the synthesis of Hplds and proteins 1n rats. Bull. Environ. Contam. Toxlcol. 7(1): 45-51. C-192 �Czelzel, E. and J. Klraly. ducing Klorlnol and Bumlnol. plex scientific 1976. Chromosome examinations 1n workers pro- In: The development of a pesticide as a com- task, L. Bankl, Ed. Medldna, Budapest. p. 239-256. (CHed 1n NRCC, 1981). Department of Health, New Zealand. 1980. Report to the Minister of Health of an Investigation Into allegations of an association between human congenital defects and 2,4,5-T spraying In and around Te Ku1t1. New Zealand Med. J. p. 314-315. D10omen1co, A., V. Sllano, G. V1v1ano and G. Zapponl. 1980a. Accidental release of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCOD) 1n Seveso, Italy. Environmental persistence of TCDO 1n soil. V. Ecotoxlcol. Environ. Safety. 4(3): 339-345. DIDomenUo, A., V. Sllano, G. Vlvlano and G. Zapponl. 1980b. Accidental release of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCDO) at Seveso, Italy. IV. Vertical distribution of TCDO In soil. Ecotoxlcol. Environ. Safety. 4(3): 327-338. DIDomenlco, A., V. Sllano, G. Vlvlano and G. Zapponl. 1980c. Accidental release of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCOD) as Seveso, Italy. VI. TCDD levels In atmospheric particles. Ecotoxlcol. Environ. Safety. 4(3): 346-356. Dietrich, R.A., R. Bludeau, M. Roger and J. Schmuck. 1978. aldehyde dehydrogenases. Blochem. Pharmacol. 27: 2343-2347. C-193 Induction of �DIGIovannl, J., A. V1aje, D.L. Berry, T.J. Slaga and M.R. Juchau. 1977. Tumor-Initiating ability of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCDD) and Aroclor 1254 In the two-stage system of mouse skin cardnogenesls. Bull. Environ. Contam. Toxlcol. 18: 552-557. D1G1ovann1, J., D.L. Berry, M.R. Juchau and T.J. Slaga. 1979. 2,3,7,8- tetrachlorod1benzo-p-d1ox1n: Potent antlcarclnogenlc activity 1n CD-I mice. Blochem. Blophys. Res. Commun. 86(3): 577-584. D1Lern1a, R., C. Crlmaudo and G. Pacchettl. 1982. The study of X-rays and TCDD effects on satellite associations may suggest a simple model for application 1n environmental mutagenesls. Hum. Genet. 61(1): 42-47. Dougherty, W.J., M. Herbst and F. Coulston. of 2,4,5-tr1chlorophenoxyacet1c add Bull. Environ. Contam. Toxlcol. Dugols, P. and L. Colomb. Dugols, P. and L. Colomb. The non-teratogen1dty In the Rhesus monkey, Macaca mulatta. 13: 477-482. 1956. Bull. Soc. Franc. Derm. Syph. 1975. Acne1 chlorlque au 2-4-5 trlchlorophenol. 63: 262. (Fre.) 1957. Remarques sur V acne' chorlque (a propos d'une ecloslon de cas provoques par la preparation due 2-4-5 trlchlorophenol). J. Med. Lyon. 38: 899. Dugols, P., et al. Mai. Prof. 19: 626. (Fre.) 1958. Acne chlorlque au 2-4-5 trlchlorophenol. (Fre.) C-194 Arch. �EdUng, C. and S. Granstam. Medical Society. 1979. Summaries transaction of the Swedish XXXVI. Natl. Conf. Med. Soc. sponsored by the Swedish Med. Soc., Stockholmsmasan, Alvlfl, Stockholm, December 5-8, 1979. 88(3): 73-74 (Translation) Elceman, G.A., R.E. Clement and F.W. Karasek. 1979. from municipal Incinerators for trace organic Analysis of fly ash compounds. Anal. Chem. 51(14): 2343-2350. Elceman, G.A., A.C. Vlau and F.W. Karasek. 1980. Ultrasonic extraction of polychlorlnated d1benzo-p-d1ox1ns and other organic compounds from fly ash from municipal Incinerators. Anal. Chem. 52: 1492-1496. Emerson, J.L., D.J. Thompson, C.G. Gerblg and V.B. Robinson. genlc study of 2,4,5-tr1chlorophenoxyacet1c add 1970. Terato- 1n the rat. Toxlcol. Appl. Pharmacol. 17: 317. Emerson, J.L., D.J. Thompson, R.J. Streblng, C.G. Gerblg and B. Robinson. 1971. Teratogenlc studies of 2,4,5-T In the rat and rabbit. Food Cosmet. Toxlcol. 9: 395-404. Eriksson, M., L. Hardell, N.O. Berg, T. Holler and 0. Axelson. 1979. Case- control study on malignant mesenchymal tumors of the soft-tissue and exposure to chemical substances. Lakart1dn1ngen. 76: 3872-3875. C-195 �Eriksson, M., L. Hardell, N.O. Berg, T. Holler and 0. Axelson. 1981. Soft- tissue sarcomas and exposure to chemical substances: A case-referent study. Br. J. Ind. Med. 38: 27-33. Facchettl, S., A. Fornarl and H. Hontagua. 1980. Distribution of 2,3,7,8- tetrachlorodlbenzo-p-dloxln In the tissues of a person exposed to the toxic cloud at Sevesa (Italy). Adv. Mass. Spectrom. 813: 1405-1414. Faith, R.E. and J.A. Moore. 1977. Impairment of thymus-dependent Immune functions by exposure of the developing Immune system to 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCOO). J. Toxlcol. Environ. Health. 3: 451-464. Fanelll, R., M.P. Bertonl, M.G. CastelH, et al. 1980. 2,3,7,8-Tetra- chlorod1benzo-p-d1ox1n toxic effects and tissue levels In animals from the contaminated area of Seveso, Italy. Arch. Environ. Contam. Toxlcol. 9(5): 569-577. FOA (Food and Drug Administration). to monitor d1ox1n-contam1nated fish. 1981. FDA advises Great Lakes States FDA Talk Paper dated August 28. In.: Food Drug Cosmetic Law Reports, Paragraph 41, 321. Commerce Clearing House, Inc. September 8. FDA (Food and Drug Administration). 1983. Statement by Sanford A. Miller, Director, Bureau of Foods, FDA, before the Subcommittee on Natural Resources, Agriculture Research and Environment, U.S. House of Representatives, June 30. C-196 �Field, B. and C. Kerr. defects. Lancet. Flnney, 0.0. 1979. Herbicide use and Incidence of neural-tube 1(8130): 1341-1342. 1971. ProbH Analysis. Cambridge University Press, London. 333 p. Firestone, D. 1977. Determination of polychlorod1benzo-p-d1ox1ns and poly- chlorodlbenzofurans 1n commercial gelatins by gas-Hqu1d chromatography. J. Agrlc. Food Chem. 25: 1274-1280. Firestone, 0., H. Glower, Jr., A.P. Borosettl, R.H. Teske and P.E. Long. 1979. Polychlorod1benzo-p-d1ox1n and pentachlorophenol residues 1n milk and blood of cows fed technical pentachlorophenol. J. Agrlc. Food Chem. 27: 1171. Fowler, B.A., G.W. Luder, H.W. Brown and O.S. HcOanlel. 1973. Ultra- structural changes 1n rat liver cells following a single oral dose of TCDO. Environ. Health Perspect. 5: 141-148. Fries, G.F. and G.S. Marrow. 1975. Retention and excretion of 2,3,7,8- tetrachlorod1benzo-p-d1ox1n by rats. J. Agrlc. Food Chem. 23(2): 265-269. Galston, A.M. 1979. Herbicides: A mixed blessing. Blosdence. 29: 85-90. Gaslewlcz, T.A., H.A. Holscher and R.A. Neal. 1980. The effect of total parenteral nutrition on the toxldty of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n 1n the rat. Toxlcol. Appl. Pharmacol. 54: 469-488. C-197 �Gaslewlcz, T.A. and R.A. Neal. 1979. 2,3,7,8-Tetrachlorod1benzo-p-d1ox1n tissue distribution, excretion, and effects on clinical chemical parameters 1n guinea pigs. Toxlcol. Appl. Pharmacol. Gaslewlcz, T.A. and R.A. Neal. tissue 1982. 51(2): 329-340. The examination and quantHatlon of cytosollc receptors for 2,3,7,8-tetrachlorod1benzo-p-d1ox1n using hydroxyapatlte. Anal. Blochem. 124: 1-11. Gaslewlcz, T.A., O.R. Olson, I.E. Gelger and R.A. Neal. 1983a. Absorption, distribution and metabolism of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCDD) 1n experimental animals. .In: Human and Environmental Risks of Chlorinated D1ox1ns and Related Compounds, R.E. Tucker, A.L. Young and A.P. Gray, Ed. Plenum Press, NY. p. 495-525. Gaslewlcz, T.A., L.E. Gelger, G. Rued and R.A. Neal. excretion, and metabolism of 1983b. Distribution, 2,3,7,8-tetrachlorod1benzo-p-d1oxln 1n C57BL/6J, DBA/23 and B6D2F /J mice. Drug Hetabol. Olsp. (In press) Gelger, L.E. and R.A. Neal. 1981. MutagenlcHy testing of 2,3,7,8-tetra- chlorod1benzo-p-d1ox1n 1n h1st1d1ne auxotrophs of Salmonella typh1mur1um. Toxlcol. Appl. Pharmacol. 59(1): 125-129. G1av1n1, E., M. Pratl and C. Vlsmara. 1982a. Effects of 2,3,7,8-tetra- chlorod1benzo-p-d1ox1n administered to pregnant rats during the prelmplantatlon period. Environ. Res. 29(1): 185-189. G1av1n1, E., M. Pratl and C. Vlsmara. 1982b. Rabbit teratology study with 2,3,7,8-tetrachlorod1benzo-p~d1ox1n. Environ. Res. 27(1): 74-78. C-198 �G1av1n1, E., M. Pratl and C. Vlsmara. 2,3,7,8-tetrachlorod1benzo-p-d1ox1n 1983. administered Embryotoxlc effects of to female rats before mating. Environ. Res. 31: 105-110. Gilbert, P., G. Sa1nt-Ruf, F. Poncelet and M. Herder. 1980. Genetic effects of chlorinated anilines and azobenzenes on Salmonella typhlmurlum. Arch. Environ. Contam. Toxlcol. 9(5): 533-541. G1zz1, F., R. Reglnato, E. Benfenatl and R. FanelH. 1982. Polychlorlnated d1benzo-p-d1ox1ns (PCDD) and polychlorlnated dlbenzofurans (PCDF) 1n emissions from an urban Incinerator. I. Average and peak values. Chemosphere. 11: 577-583. Goldman, P.J. 1972. Schwersta akute chlorakne durch trlchlorphenol- zersetzungsprodukte. Arbeltsmed. Sozlalmed. Arbeltshyg. 7: 12. (Ger.) Goldstein, J.A.m, M. Frlsen, T.M. Scottl, P. Hlckman, J.R. Mass and H. Bergman. 1978. Assessment of the contribution of chlorinated dlbenzo-p- dloxlns and dlbenzofurans to hexachlorobenzene-lnduced toxldty, porphyrla, changes 1n mixed function oxygenases, and hlstopathologlcal changes. col. Appl. Pharmacol. Toxl- 46(3): 633-649. Goldstein, J.A., P. L1nko and H. Bergman. 1982. Induction of porphyrla 1n the rat by chronic versus acute exposure to 2,3,7,8-tetrachlorod1benzo-pdloxln. Blochem. Pharmacol. 31(8): 1607-1613. C-199 �Green, S., F. Moreland and C. Sheu. 1977. Cytogenlc effect of 2,3,7,8- tetrachlorod1benzo-p-d1ox1n on rat bone marrow cells. FDA, Washington, DC. FDA By-Llnes. 6: 292-294. Green, S. and F.S. Moreland. 1975. Cytogenetlc evaluation of several dloxlns In the rat. Toxlcol. Appl. Pharmacol. 33: 161. Greenlee, W.F. and A. Poland. 1979. Nuclear uptake of 2,3,7,8-tetrachloro d1benzo-p-d1ox1n 1n C57BL/6J and DBA/2J mice. Role of the hepatic cytosol receptor protein. J. B1ol. Chem. 254(19): 9814-9821. Grelg, J.B. 1972. Effect of 2,3,7,8-tetrachlorodlbenzo-l,4-d1ox1n on drug metabolism 1n the rat. Blochem. Pharmacol. 21(23): 3196-3198. Grelg, J.B., G. Jones, W.H. Butler and J.M. Barnes. 1973. Toxic effects of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n. Food Cosmet. Toxlcol. 11: 585-595. Gupta, B.N., J.G. Vos, J.A. Moore, O.G. Z1nkl and B.C. Bullock. 1973. Pathologic effects of 2,3,7,8-tetrachlorud1benzo-p-d1ox1n In laboratory animals. Environ. Health Perspect. Hajdu, S.I. lesions. 1983. Presented 5: 125-140. Classification of soft tissue tumors and tumor-Uke at Symp. on Public Health Risks of the Dloxlns, Rockefeller Univ., October 19-20. C-200 �Hanlfy, 3.A., P. Metcalf, C.L. Nobbs and R.J. Worsley. 1981. Aerial sprayIng of 2,4,5-T and human birth malformations: An epldemlologlcal Investigation. Science. Hardell, L. 212: 349-351. 1977. Malignant mesenchymal tumors and exposure to phenoxy- aclds: A clinical observation. Hardell, L. Lakartldnlngen. 74: 2753-2754. 1979. Malignant lymphoma of hlstlocytlc type and exposure to phenoxyacetlc adds or chlorophenols. Lancet. Hardell, L. 1: 55-56. 1983. Letter to D. Bayllss, Carcinogen Assessment Group, U.S. EPA, Washington, DC, August 5. Hardell, L. and A. Sandstrom. 1979. Case-control study: Soft-tissue sar- comas and exposure to phenoxyacetlc acids or chlorophenols. cer. Brit. J. Can- 39: 711-717. Hardell, L. and M. Eriksson. 1981. Soft-tissue sarcomas, phenoxy herbi- cides and chlorinated phenols. Lancet. Hardell, L., M. Eriksson and P. Lenner. 11: 250. 1980. Malignant lymphoma and expo- sure to chemical substances, especially organic solvents, chlorophenols, and phenoxy adds. Lakartldnlngen. 77: 208-210. Hardell, L., M. Eriksson, P. Lenner and E. Lundgren. 1981. Malignant lym- phoma and exposure to chemicals, especially organic solvents, chlorophenols and phenoxy acids: A case-control study. C-201 Br. J. Cancer. 43: 169-176. �Harding, D.H. 1982. Ambient air guideline. Chlorinated dloxlns and chlorinated dlbenzofurans. Issued by: Health Studies Service, Special Studies and Services Branch Ministry of Labour, December 16. Harless, R.L. and R.G. Lewis. 1980. Quantitative determination of 2,3,7.8- tetrachlorod1benzo-p-d1ox1n residues by gas chromatography/mass spectrometry. Presented at Workshop on Impact of Chlorinated Dloxlns and Related Compounds on the Environment. Institute SupeMore dl SanHa, Rome, Italy, October 22-24. Harless, R.L. and R.G. Lewis. 1982. Quantitative determination of 2,3,7,8- tetrachlorod1benzo-p-d1ox1n (TCDO) Isomers. Related Compounds. In: Chlorinated Dloxlns and Impact on the Environment, 0. Hutzlnger, R.W. Fre1, E. Merlan and F. Pocch1ar1, Ed. Pergamon Press, NY. p. 25-35. Harris, M.W., J.A. Moore, J.F. Vos and B.N. Gupta. cal effects of TCDD 1n laboratory animals. 1973. General biologi- Environ. Health Perspect. 5: 101-109. Harvan, D.J., J.R. Hass, 3.L. Schroeder and B.J. Corbett. of tetrachlorod1benzod1ox1ns In air filter samples. 1981. Detection Anal. Chem. 53(12): 1755-1759. Hassoun, E.M. and L. Dencker. 1982. TCDD embryotoxlclty 1n the mouse may be enhanced by «p-naphthflovone, another llgand of the Ah-receptor. Lett. 12: 191-198. C-202 Toxic. �Hay, A. 1976. Toxic cloud over Seveso. Nature. 262: 636-638. Hay, A., Ed. 1982. Toxicology of dloxlns. hi: The Chemical Scythe: Lessons of 2,4,5-T and 01ox1n. Plenum Press, NY and London, p. 41-47. Helling, C.S. 1971. Pesticide mobility 1n soils. II. Applications of soil thin-layer chromatography. Soil. Sc1. Am. Proc. 35: 737-743. Helling, C.S., A.R. Isnesee, E.A. Woolson, et al. 1973. Chlorodloxlns 1n pesticides, soils and plants. J. Environ. Qua!. 2(2): 171-178. Henck, J.W., H.A. New, R.J. Koclba and K.S. Rao. 1981. 2,3,7,8-Tetra- chlorod1benzo-p-d1ox1n: Acute oral toxlclty In hamsters. Toxlcol. Appl. Pharmacol. 59: 405-407. H1les, R.A. and R.O. Bruce. 1976. 2,3,7,8-Tetrachlorod1benzo-p-d1ox1n elimination In the rat: First order or zero order? Food Cosmet. Toxlcol. 14(6): 599-600. Hlnsdlll, R.D., D.L. Couch and R.S. Spelrs. mice Induced by dloxln (TCOO) 1n feed. 1980. Immunosuppresslon 1n J. Environ. Pathol. Toxlcol. 4(2-3): 401-425. Hofmann, H.T. 1957. wasserstoffen. 232: 228. Neuere Erfahrunger mlt hoch-toxlschen chlorkohlen- Naunyn-Schmldeberg's (Ger.) C-203 Arch. Exp. Pathol. Pharmakol. �Holden, C. 1979. Agent Orange furor continues to build. Science. 205: 770-772. Honchar, P.A. and W.E. Halperln. tissue sarcoma. 1981. 2,4,5-TMchlorophenol and soft- Lancet. 1(8214): 268-269. Houk, V.N. 1983. Testimony by Dr. Vernon W. Houk, Director, Center for Environmental Health, CDC, PHS, Dept. Health Human Serv. before the United States House of Representatives Science and Technology Committee, Subcommittee on Natural Resources, Agricultural Research and Environment, March 23. Hueper, W.C. and W.D. Conway. 1964. Chemical cardnogenesls and cancers. Springfield, IL. Thomas. Hutter, R. and M. PhlUppl. 1982. Studies on mlcroblal metabolism of TCDD under laboratory conditions. Pergamon Ser. Environ. Sd. 5: 87-93. Hussaln, S., L. Ehrenberg, G. Lofroth and T. Gejvall. effects of TCDD on bacterial systems. 1972. Mutagenlc Amblo. 1: 32-33. IARC (International Agency for Research on Cancer). 1977. IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Man. Some Fum1gants, the Herbicides 2,4-D and 2,4,5-T, Chlorinated D1benzod1ox1ns and Miscellaneous Industrial Chemicals, Vol. 15. IARC, Lyon, France, p. 41-102. International Classification of Diseases. 1975. 9th revision. WHO, Geneva. C-204 �Isensee, A.R. and G.E. Jones. 1971. Absorption and translocaUon of root and foliage applied 2,4-d1chlorophenol, 2,7-d1chlorod1benzo-p-d1ox1n, and 2,3,7,8-tetrachlorod1benzo-p-d1ox1n. J. Agrlc. Food Chem. Isensee, A.R. and G.E. Jones. 1975. 19: 1210-1214. Distribution of 2,3,7,8-tetrachlorodl- benzo-p-d1ox1n (TCDO) 1n aquatic model ecosystem. Environ. Sd. Technol. 9: 668-672. Ishldate, K., M. Tsuruoka and Y. Nakazawa. 1980. Induction of chollne klnase by polycycllc aromatic hydrocarbon carcinogens In rat liver. B1o- chem. Blophys. Res. Commun. 96: 946-952. Jensen, O.A., M.E. Getzendaner, R.A. Hummel and J. Turley. studies for (2,4,5-trlchlorophenoxy) acetic add Residue and 2,3,7,8-tetrachlorodl- benzo-p-d1ox1n In grass and rice. J. Agrlc. Food Chem. Jlrasek, L., et al. 1973. 1983. 31: 118-122. Acne chlorlna and porphyrla cutanea tarda during the manufacture of herbicides. Cesk. Dermatol. 48: 306-317. Jlrasek, L., J. Kakebsjtm and J. Kubee. 1974. Acne chlorlna, porphyrla cutanea tarda, and other manifestations of general Intoxication during the manufacture of herbicides. Cesk. Dermatol. 49: 145-157. Johnson, F.E., H.A. Kugler and S.M. Brown. 1981. chlorinated phenols. Lancet. 2(8236): 40. C-205 Soft-tissue sarcomas and �Jones, G. and W.H. Butler. 1974. A morphological study of the liver lesion Induced by 2,3,7,8-tetrachlorod1benzo-p-d1ox1n 1n rats. J. Pathol. 112: 93-97. Jones, G. and J.B. Grelg. 1975. Pathological changes In the liver of mice given 2,3,7,8-tetrachlorod1benzo-p-d1ox1n. Exper1ent1a. 31(1)): 1315-1317. Jones, K.G. and Sweeney, G.D. 1980. Dependence of the porphyrogenlc effect of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n upon Inheritance of aryl hydrocarbon hydroxylase responsiveness. Toxlcol. Appl. Pharmacol. 53: 42-49. Josephson, J. 1983. Environ. Sc1. Technol. Chlorinated dloxlns and furans In the environment. 17: 124A-128A. Karasek, F.W., R.E. Clement and A.C. V1an. 1982. Distribution of PCDOs and other toxic compounds generated on fly ash partlculates 1n municipal Incinerators. J. Chromatog. 239: 173-180. Kearney, P.C., E.A. Woolson and C.P. Ellington, Jr. metabolism of chlorodloxlns In soils. 1972. Persistence and Environ. Sd. Technol. 6(12): 1017-1019. Kearney, P.C., et al. 1973. TCDD In the environment: Sourcss, fate and decontamination. Environ. Health Perspect. C-206 5: 273-277. �Kenaga, E.E. 1980. Correlation of bloconcentratlon factors of chemicals 1n aquatic and terrestrial organisms with their physical and chemical properties. Environ. Sd. Technol. 14: 553-556. Khera, K.S., B.L. Huston and W.P. McKlnley. 1971. Pre- and postnatal studies on 2,4,5-T, 2,4-D, and derivatives In Hlstar rats. Toxlcol. Appl. Pharmacol. 19: 369-370. Khera, K.S. and W.P. McKlnley. and 2,4-0 and 1972. Pre- and postnatal studies on 2,4,5-T their derivatives 1n rats. Toxlcol. Appl. Pharmacol. 22: 14-28. Khera, K.S. and J.A. Ruddlck. 1973. Polychlorod1benzo-p-d1ox1ns: Perinatal effects and the dominant lethal test 1n Wlstar rats. Adv. Chem. Ser. 120: 70-84. Klmble, B.J. and M.L. Gross. 1980. Tetrachlorodlbenzo-p-dloxln quantlta- tlon In stack-collected coal fly ash. Science. 207: 59-61. Klmbrough, R.O. 1974. The toxlclty of polychorInated polycycllc compounds and related chemicals. CRC Crlt. Rev. Toxlcol. Klmbrough, R.D., Ed. 1980. 2: 445-489. Occupational exposure. Irr. Halogenated 81- phenyls, Terphenyls, Naphthalenes, Dlbenzodloxlns and Related Elsevler/North-Holland B1omed1cal Press, p. 374-397. C-207 Products. �Klmbrough, R.D., C.D. Carter, 3.A. Llddle and R.E. CUne. 1977. Epidemiology and pathology of a tetrachlorod1benzod1ox1n poisoning episode. Arch. Environ. Health. 32(2): 77-86. Klmbrough, R.D., H. Falk, P. Strehr, C. Portler and G. Fries. 1983. Risk assessment document on 2,3,7,8-TCDD levels 1n soil. Centers for Disease Control, Atlanta, GA. (Draft) K1mm1g, J. and K.H. Schulz. 1957. Berufllche akna (sog. chlorakne) durch chlorlette aromatlsche zykllsche ather. Dermatologla. 115: 540-546. (Ger.) King, M.E. and A.R. Roesler. 1974. Subacute Intubation study on rats with the compound 2,3,7,8-tetrachlorod1ox1n. U.S. EPA. NTIS PB-257 677. p. 27. Klrsch, et al. 1975. Structural and functional studies of Ugandln, a major renal organic anlon-blndlng protein. J. CUn. Invest. 55: 1009. Knutson, J.C. and A. Poland. 1980. 2,3,7,8-Tetrachlorod1benzo-p-d1ox1n: Failure to demonstrate toxldty In twenty-three cell types. Toxlcol. Appl. Pharmacol. 54: 377-383. Kocher, C.W., N.H. Mahle, R.A. Hummel, L.A. Shadoff and M.E. Getzendaner. 1978. A search for the presence of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n 1n beef fat. Bull. Environ. Contam. Toxlcol. 19: 229-236. C-208 �Kodba, R.J., P.A. Keeler, C.N. Park and P.3. GehMng. Tetrachlorod1benzo-p-d1ox1n results of a 13-week 1976. 2,3,7,8- oral tox1c1ty study In rats. Toxlcol. Appl. Pharmacol. 35: 553-573. Kodba, R.J., O.G. Keyes, J.E. Beyer, et al. 1978. Results of a two-year chronic toxlclty and oncogenlclty study of 2,3,7,8-tetrachlorod1benzo-pdloxln In rats. Toxlcol. Appl. Pharmacol. 46(2): 279-303. Kodba, R.J., O.G. Keyes, J.E. Beyer, R.M. Carreon and P.J. Gehrlng. 1979. Long-term toxlcologlc studies of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCDD) In laboratory animals. Ann. NY Acad. Sc1. 320: 397-404. Kondorosl, A., I. Fedorcsak, F. Solymosy, L. Ehrenberg and S. OstermanGolkar. 1973. Inactlvatlon of QBRNA by electrophlles. Mutat. Res. 17: 149-161. Kourl, R.E., H. Ratrle, III, S.A. Atlas, A. Nlwa and O.W. Nerbert. 1974. Aryl hydrocarbon hydroxylase Induction 1n human lymphocyte cultures by 2,3,7,8-tetrachlorod1benzo-p-d1ox1n. Life Sd. Kourl, R. 1976. 15(9): 1585-1595. Relationship between levels of aryl hydrocarbon hydroxy- lase activity and susceptibility to 3-methylcholanthrene and benzo(a)pyreneInduced cancers In Inbred strains of mice. In: Polynuclear Aromatic Hydrocarbons: Chemistry, Metabolism and Cardnogenesls, Vol. 1, R.J. Freudenthal and P.W. Jones, Ed. Raven Press, NY. p. 139. C-209 �Kourl, R.E. and D.U. Nebert. 1977. Genetic regulation of susceptibility to polycycllc hydrocarbon-Induced tumors In the mouse. In: Proc. Symp. Origins of Human Cancer. Cold Springs Harbor, NY. Kourl, R.E., T.H. Rude, R. Joglekar, et al. 1978. 2,3,7,8-Tetrachlorodl- benzo-p-d1ox1n as cocardnogen causing 3-methylcholanthrene-1n1t1ated subcutaneous tumors In mice genetically "nonresponslve" at Ah locus. Cancer Res. 38(9): 2777-2783. Lamb, J.C., J.A. Moore and T.A. Markes. 1980. Evaluation of 2,4-D, 2,4,5-T, and TCDO toxldty In C57BL/6 mice: Reproduction and fertility 1n treated male mice and evaluation of congenital malformation 1n their offspring. NTP, Research Triangle Park, NC. Lamparskl, L.L., N.H. Mahle and L.A. Shadoff. 1978. Determination of pentachlorophenol, hexachlorod1benzo-p-d1ox1n, and octachlorodlbenzo-pdloxln In bovine milk. J. Agrlc. Food Chem. 26(5): 1113-1116. Langhorst, M.L. and L.A. Shadoff. 1980. Determination of part-per-trllllon concentrations of tetra-, hexa-, hepta-, and octa-chlorod1benzo-p-d1ox1ns 1n human milk samples. Anal. Chem. 52: 2037-2044. Lavy, T.L., J.S. Shepard and J.D. Mattlce. 1980. Exposure measurements of applicators spraying (2,4,5-tMchlorophenoxy) acetic add Agrlc. Food Chem. 28: 626-630. C-210 1n the forest. J. �Legraverend, C., B. Mansour, D.W. Nebert and J.M. Holland. 1980. Genetic differences In benzo[a]pyrene-1n1t1ated tumor 1 genes Is 1n mouse skin. Phar- macology. 20: 242-255. Levin, J.O., C.F. Rappe and C.A. Nllssen. 1976. Use of chlorophenols or fungicides In sawmills. Scand. J. Work Environ. Health. Llss, Q.S. and P.G. Slater. 2: 71. 1974. Flux of gases across the air-sea Inter- face. Nature. 247: 181. Luder, G.W., B.R. Sonawane, O.S. McOanlel and G.E.R. Hook. 1975. Post- natal stimulation of hepatic mlcrosomal enzymes following administration of KOD to pregnant rats. Chem. B1ol. Interact. 11: 15-26. Lustenhouwer, J.W.A., K. OHe and 0. Hutzlnger. 1980. Chlorinated dlbenzo- p-d1ox1ns and related compounds In Incinerator effluents: A review of measurements and mechanisms of formation. Chemosphere. 9: 501-522. Luster, M.I., G.A. Boorman, J.H. Dean, et al. 1980. Examination of bone marrow, 1mmunolog1c parameters and host susceptibility following pre- and postnatal exposure to 2,3,7,8-tetrachlorodtbenzo-p-d1oxin (TCDD). Int. J. Immunopharmacol. 2(4): 301-310. Habey, W.R., J.H. Smith, R.T. Podoll, et al. data for organic priority pollutants. 1981. Aquatic fate processes Monitor. Data Support Dlv., Off. Water Regul. Stand., Washington, DC. EPA 440/4-81-014. p. 107-108. C-211 �Mahle, N.H., H.S. Hlgglns and M.E. Getzendaner. 1977. Search for the pres- ence of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n In bovine milk. Bull. Environ. Contain. Toxlcol. 18(2): 123-130. Manara, L., P. Cocda and T. Crocl. 1982. Persistent tissue levels of TCDD In the mouse and their reduction as related to prevention of toxlclty. Drug Metab. Rev. 13(3): 423-446. Mason, M.E. and A.B. Okey. 1982. Cytosollc and nuclear binding of 2,3,7,8- tetrachlorodlbenzo-p-dloxln to the Ah receptor 1n extrahepatlc tissues of rats and mice. Eur. J. Blochem. 123: 209-215. Matsumura, F. and H.3. Benezet. mlcroblal degradation of 1973. Studies on the bloaccumulatlon and 2,3,7,8-tetrachlorod1benzo-p-d1ox1n. Environ. Health Perspect. 5: 253-258. Matsumura, F., J. Quensen and G. Tsushlmoto. of TCDD in a model ecosystem, 1983. Mlcroblal degradation Jji: Human and Environmental Risks of Chlori- nated D1ox1ns and Related Compounds, R.E. Tucker, A.L. Young and A.P. Gray, Ed. Plenum Press, NY. p. 191-219. MatUson, D.R. and S.S. ThorgeUsson. 1979. Ovarian aryl hydrocarbon hydroxylase activity and primordial oocyte toxldty of polycycllc aromatic hydrocarbons 1n mice. Cancer Res. 39: 3471-3475. McCann. 1978. Unpublished study. (Cited 1n Wasson et al., 1978). C-212 �McConnell, E.E. 1980. Acute and chronic toxldty, carclnogenesls, repro- duction, teratogenesls and mutagenesls 1n animals, in: Halogenated B1- phenyls, Terphenyls, Naphthalenes, D1benzod1ox1ns and Related Products, R.D. Klmbrough, Ed. Elsevler/North-Holland Blomedlcal Press, NY. p. 109-150. McConnell, E.E., J.A. Moore, O.K. Haseman and M.W. Harris. 1978a. The com- parative toxldty of chlorinated dlbenzo-p-dloxlns 1n mice and guinea pigs. Toxlcol. Appl. Pharmacol. 44(2): 335-356. McConnell, E.E., J.A. Moore and D.W. Dalgard. 1978b. Toxldty of 2,3,7,8- tetrachlorodlbenzo-p-dloxln In rhesus monkeys (Macacas mulatta) following a single oral dose. Toxlcol. Appl. Pharmacol. 43: 175. McNulty, W.P., K.A. Nielsen-Smith, J.O. Lay, Jr., et al. 1982. Persistence of TCOO 1n monkey adipose tissue. Food Cosmet. Toxlcol. 20: 985-987. McQueen, E.G.. A.M.O. Veale, W.S. Alexander and M.N. Bates. and human birth defects. 1977. 2,4,5-T D1v. Public Health, New Zealand Dept. Health. (Cited 1n MUby et al., 1980). Michigan Department of Public Health. 1983a. Evaluation of congenital malformation rates for Midland and other selected Michigan Counties compared nationally and statewide 1970-1981. Michigan Oept. Public Health, May 4. C-213 �Michigan Department of Public Health. 1983b. nective tissue cancer mortality rates Evaluation of soft and con- for Midland and other selected Michigan counties compared nationally and statewide. Michigan Dept. Public Health, May 4. M1lby, T.H., E.L. Husttng, M.D. Whorton and S. Larson. 1980. Potential health effects associated with the use of phenoxy herbicides: A summary of recent scientific literature. Report for the National Forest Products Assoc. from Environ. Health Associated, Inc., Berkeley, CA. Mllnes, M.H. thermal 1971. Formation of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n by decomposition of sodium 2,4,5-tMchlorophenate. Nature. 232: 395-396. MHchum, R.K., G.F. Moler and W.A. Korfmacher. 1980. Combined capillary gas chromatography/atmospherlc pressure negative chemical 1on1zat1on/mass spectrometry for the determination of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n 1n tissue. Anal. Chem. 52(14): 2278-2282. Moore, O.A., B.N. Gupta, 3.G. 21nkl and J.G. Vos. 1973. Postnatal effects of maternal exposure to 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCDO). Environ. Health Perspect. 5: 81-85. Moore, J.A., M.W. Harris and P.M. Albro. 1976. Tissue distribution of ( l4 C)tetrachlorod1benzo-p-d1ox1n 1n pregnant and neonatal rats. Appl. Pharmacol. 37(1): 146-147. C-214 Toxlcol. �Moore, O.A., E.E. McConnell, D.W. Dalgard, M.W. Harris. 1979. Comparative toxldty of three halogenated dlbenzofurans 1n guinea pigs, mice, and rhesus monkeys. Ann. NY Acad. Sd. 320: 151-163. Hoses, M. and I. Sellkoff. 1981. Soft tissue sarcomas, phenoxy herbicides and chlorinated phenols. Lancet. 1: 1370. Mottura, A., G. 2e1, F. Nuzzo, et al. 1981. Evaluation of results of chromosome analyses on lymphocytes of TCOO exposed subjects after the Seveso accident. Mutat. Res. 85(4): 238-239. Mulcahy, M.T. Aust. 1980. Chromosome aberrations and "Agent Orange". Med. J. 2(10): 573-574. Murray, F.J., F.A. Smith, K.D. NHschke, C.G. Humlston, R.J. Kodba and B.A. Schwetz. 1979. Three-generation reproduction study of rats given 2,3,7,8- tetrachlorod1benzo-p-d1ox1n (TCOD) In the diet. Toxlcol. Appl. Pharmacol. 50: 241-251. NAS (National Academy of Sciences). 1977. Drinking Water and Health: Part II. NAS, Washington, DC. Nash, R.G. and M.L. Beall, Jr. 1980. Distribution of Sllvex, 2,4-D, and TCDD applied to turf In chambers and field plots. 28: 614-623. C-215 J. Agrlc. Food Chem. �Nau, H. and R. Bass. 1981. Transfer of 2,3,7,8-tetrachlorod1benzo-p-diox1n (TCDD) to the mouse embryo and fetus. Toxicology. 20(4): 299-308. Nau, H., R. Bass and D. Neubert. 1982. Transfer of 2,3,7,8-tetrachlorodl- benzo-p-d1ox1n (TCOD) to the mouse embryo, fetus and neonate. nated Dloxlns and Related Compounds. In: Chlori- Impact on the Environment, 0. Hunt- zlnger, R.W. Fre1, E. Merlan and F. PocchlaM, Ed. Pergamon Press, NY. p. 325-337. Neal, R.A., J.R. Olson, T.A. Gas1ew1cz and I.E. Gelger. 1982. The toxlco- klnetlcs of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n 1n mammalian systems. Drug Metab. Rev. 13: 355-385. Nebert, O.W. 1979. Multiple forms of Indudble drug-metabolizing enzymes: A reasonable mechanism by which any organism can cope with adversity. Mol. Cell. Blochem. 27: 27-46. Nebert, D.W. 1982. Pharmacogenetlcs and human cancer. Host factors In human cardnogenesls. IARC Scientific Pub. No. 39. Lyon, France. Nebert, D.W. and N.M. Jensen. 1979. The Ah locus: Genetic regulation of the metabolism of carcinogens, drugs, and other environmental chemicals by cytochrome P-450 mediated monooxygenases. CRC Cr1t. Rev. Blochem. 6: 401-437. Nebert, O.W., et al. 1975. Genetic expression of aryl hydrocarbon hydroxylase activity 1n the mouse. J. Cell. Physio!. 85: 393-414. C-216 �Nebert, D., S. Thorglersson and 0. Felton. 1976. mutagenesls, cardnogenesls, and drug toxldty. Genetic differences In JJK Ln vitro Metabolic Activation 1n Mutagenesls Testing, F. de Serros, J. Folets, J. Bend and R. Phllpot, Ed. Elsevler/North Holland B1omed1cal Press, Amsterdam, p. 105-124. Nebert, D.W., H.J. E1sen, M. Neglshl, M.A. Lang, L.M. HJelmeland and A.B. Okey. 1981. Genetic mechanisms controlling the Induction of polysubstrate monooxygenase (P-450) activities. Ann. Rev. Pharmacol. Toxlcol. 21: 431-462. Nebert, D.W., M. Neglshl and H.J. E1sen. 1983. Genetic differences 1n enzymes which metabolize drugs, chemical carcinogens and other environmental pollutants. In: Human and Environmental Risks of Chlorinated Dloxlns and Related Compounds, R.E. Rucker, A.L. Young and A.P. Gray, Ed. Plenum Press, NY. p. 441-462. Nelson, C.J., J.F. Holson, H.G. Green and D.W. Gaylor. 1979. Retrospective study of the relationship between agricultural use of 2,4,5-T and cleft palate occurrence In Arkansas. Teratology. 19: 377-384. Nestrlck, T.J., L.L. Lamparskl, W.B. Crummet and L.A. Shadoff. 1982. Comments on variations 1n concentrations of organic compounds Including polychlorinated dlbenzo-p-dloxlns and polynuclear aromatic hydrocarbons In fly ash from a municipal Incinerator. Anal. Chem. 54: 823-824. C-217 �Neubert, D. and I. Dlllmann. with 1972. 2,4,5-trlchlorophenoxyacetlc Embryotoxlc effects 1n mice treated add and 2,3,7,8-tetrachlorod1benzo-p- dloxln. Arch. Pharmacol. 272(3): 243-264. Neubert, D., P. Zens, A. Rothenwallner and H.J. Herker. 1973. A survey of the embryotoxlc effects of TCOO 1n mammalian species. spect. Environ. Health Per- 5: 67-79. Nlemann, R.A., W.C. Brumley, D. Firestone and J.A. Spohn. 1983. Analysis of fish for 2,3,7,8-tetrachlorod1benzo-p-d1ox1n by electron capture capillary gas chromatography. Anal. Chem. 55: 1497-1504. Nlsbet, I.C.T. and M.B. Paxton. 1982. Statistical aspects of three- generation studies of the reproductive toxldty of TCOD and 2,4,5-T. Am. Stat. Vol. 36(3): 290-298. Nolan, R.O., F.A. Smith and 3.G. Hefner. 1979. Elimination and tissue distribution of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCOO) 1n female guinea pigs following a single oral dose. Norback, D.H. and J.R. Allen. primate, chicken, and Environ. Health Perspect. rat Toxlcol, Appl. Pharmacol. 48(1): A162. 1973. Biological responses of the nonhuman to chlorinated d1benzo-p-d1ox1n Ingestlon. 5: 233-240. Norman, R.L., E.F. Johnson and U. Huller-Eberhard. 1978. Identification of the major cytochrome P-450 form transplacentally Induced 1n neonatal rabbits by 2,3,7,8-tetrachlorod1benzo-p-d1ox1ns. J. B1ol. Chem. C-218 23: 8640-8647. �NRCC (National Research Council of Canada). 1981. Polychlorlnated dlbenzo-p-dloxlns: Criteria for their effects on man and his environment. Pub. No. NRCC 18574, ISSN 0316-0114. NRCC/CNRC Assoc. Com. Scientific Cri- teria for Environ. Qual. Ottawa, Canada. NTP (National Toxicology Program). 251 p. 1980a. Bloassay of 2,3,7,8-tetrachloro- d1benzo-p-d1ox1n for possible cardnogenlcHy (gavage study). Cardnogene- sls Test. Prog., NCI, NIH, Bethesda, MD and NTP, Research Trlange Park, NC. Pub. No. 82-1765. NTP (National Toxicology Program). 1980b. Bloassay of 2,3,7,8-tetrachloro- d1benzo-p-d1ox1n for possible cardnogenlcHy (dermal study). Cardnogenes1s Test. Prog., NCI, NIH, Bethesda, MD and NTP, Research Triangle Park, NC. Pub. No. 80-1757. O'Keefe, P., C. Meyer, D. Milker, et al. 1983. Analysis of 2,3,7,8-tetrachlorodlbenzo-p-dloxln 1n Great Lakes fish. Chemosphere. Okey, A.G. 12: 325-332. 1983. The Ah receptor: A specific site for action of chlori- nated dloxlns. jn: Human and Environmental Risks of Chlorinated Dloxlns and Related Compounds, R.E. Tucker, A.L. Young and A.P. Gray, Ed. Plenum Press, NY. p. 423-440. Okey, A.B., G.P. Bondy, M.E. Mason, et al. of the Ah locus. J. B1ol. Chem. 1979. Regulatory gene product 254: 11636-11648. C-219 �Okey, A.B., G.P. Bondy, M.E. Mason, et al. 1980. Temperature-dependent cytosol-to-nucleus translocatlon of the Ah receptor for 2,3,7,8-tetrachlorodlbenzo-p-dloxln In continuous cell culture lines. J. Blol. Chem. 255(23): 11415-11422. Okey, A.B. and L.M. Vella. 1982. Binding of 3-methylcholanthrene and 2,3,7,8-tetrachlorod1benzo-p-d1ox1n to a common Ah receptor site 1n mouse and rat hepatic cytosls. Eur. J. Blochem. 127(1): 39-47. Olle, K., M.V.D. Berg and 0. Hutzlnger. and PCDF from combustion processes. 1983. Formation and fate of PCDD Chemosphere. 12: 627-636. Ol1e, K., J.W.A. Lustenhouwer and 0. Hutzlnger. d1benzo-p-d1ox1ns and related compounds 1982. Polychlorlnated 1n Incinerator effluents. In: Chlorinated Dloxlns and Related Compounds: Impact on the Environment, 0. Hutzlnger et al., Ed. Pergamon Press, p. 227-244. Oliver, R.M. 1975. Toxic effects of 2,3,7,8-tetrachlorod1benzo-l,4-d1ox1n 1n laboratory workers. Br. J. Ind. Med. 32(1): 49-53. Olson, J.R. ahd W.E. Blttner. 1983. Comparative metabolism and elimination of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCOD). The Tox1colog1st. 3: 103. Olson, J.R., T.A. Gaslewlcz and R.A. Neal. 1980a. Tissue distribution, excretion, and metabolism of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCDD) 1n the Golden Syrian hamster. Toxlcol. Appl. Pharmacol. 56: 78-85. C-220 �Olson, 3.R., H.A. Holscher and R.A. Neal. 1980b. ToxUHy of 2,3,7,8- Ietrachlorod1benzo-p-d1ox1n In the Golden Syrian hamster. Toxlcol. Appl. Pharmacol. 55: 67-78. Olson, J.R., M. Gudz1now1cz and R.A. Neal. 1981. The in vitro and jjn vivo metabolism of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCDD) 1n the rat. The Tox1colog1st. 1: 69. Olson, 3.R., Gas1ew1cz, T.A., I.E. Gelger and R.A. Neal. 1983. The metabolism of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n 1n mammalian systems, In: Accidental Exposure to D1ox1ns: Human Health Aspects, F. Coulston and F. Pocchlarl, Ed. Academic Press, NY. p. 81-100. Olsson, H. and L. Brandt. 1982. Non-Hodgk1n's lymphoma of the skin and occupational exposure to herbicides. Lancet. 9(81): 579. Orris, P. 1981. Unjustified conclusion? 3. Occup. Hed. 23(1): 7-8. Ott, M.G., B.B. Holder and R.D. Olson. 1980. A mortality analysis of employees engaged 1n the manufacture of 2,4,5-tr1chlorophenoxyacet1c add. J. Occup. Med. 22(1): 47-50. Owens, I.S. 1977. Genetic regulation of UDP-glucuronosyltranferase Induc- tion by polycycllc aromatic compounds 2827-2833. C-221 1n mice. J. Blol. Chem. 252: �Pazderova, J., E. Lukas, M. Nemcova, et al. 1974. Chronic poisoning by chlorinated hydrocarbons formed 1n the production of sodium 2,4,5-tMchlorophenoxyacetate. Prac. Lek. 26: 332-339. Pazderova-Vejlupkova, J., M. Nemcova. J. Plckova, L. Jlrasek and E. Lukas. 1981. The development and prognosis of chronic Intoxication by tetrachloro- d1benzo-p-d1ox1n In men. Arch. Environ. Health. 36(1): 5-10. Pegg, D.G., W.R. Hewitt, K.M. HcCormack and J.B. Hook. 1976. Effect of 2,3,7,8-tetrachlorod1benzo-rho-d1ox1n on renal function 1n the rat. J. Toxlcol. Environ. Health. 2(1): 55-65. Ph1l1pp1, M., V. Krasnobagew, J. Zeyer and R. Huetter. 1981. Fate of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCDO) 1n mlcroblal cultures and soil under laboratory conditions. FEMS Symp. 12: 2210-233. Piper, W.N., R.Q. Rose and P.J. Gehrlng. tribution of 1973. Excretion and tissue dis- 2,3,7,8-tetrachlorod1benzo-p-d1ox1n 1n the rat. Environ. Health Perspect. 5: 241-244. PHot, H.C., T. Goldsworthy and H. Poland. 1980. Promotion by 2,3,7,8- tetrachlorod1benzo-p-d1ox1n of hepatocarclnogenesls from d1ethyln1trosam1ne. Cancer Res. 40: 3616-3620. Pllmmer, J.R., U.I. Kllngeblel, D.G. Crosby and A.S. Wong. Chlorodloxlns -- Origin and Fate. 1973a. In: Advances 1n Chemistry Series No. 120. p. 44. C-222 �Pllmmer, J.R., 3.H. Ruth and E.A. Woolson. 1973b. Mass spectrometlc Ident- ification of the hepta- and octachlorlnated d1benzo-p-d1ox1ns and dlbenzofurans In technical pentachlorophenol. J. AgMc. Food Chem. 21(1): 90-93. Pllmmer, J.R. 1978. Photolysis of TCOD and trlfluraHn on silica and soil. Bull. Environ. Contam. Toxlcol. 20: 87-92. Pocch1ar1, F., V, Sllano and A. Zamplerl. 1979. Human health effects from accidental release of tetrachlorod1benzo-p-d1ox1n (TCOD) at Seveso, Italy. Ann. NY Acad. Sd. 320: 311-320. Polger, H. and C. Schlatter. 1979. Biological degradation of TCOD In rats. Nature. 281(5733): 706-707. Polger, H. and C. Schlatter. 1980. Influence of solvents and adsorbents on dermal and Intestinal absorption of TCDD. Food Cosmet. Toxlcol. 18(5): Polger, H., H.R. Buser, H. Weber, U. Zwelfel and C. Schlatter. 1982a. 477-481. Structure elucidation of mammalian TCDD-metabolltes. Exper1ent1a. 38(4): 484-486. Polger, H., H. Weber and Ch. Schlatter. 1982b. Special aspects of metabolism and kinetics of TCDD 1n dogs and rats. Assessment of toxlclty of TCDD-metabolHe(s) 1n guinea pigs. In: Chlorinated D1ox1ns and Related Compounds. Impact on the Environment, 0. Hutzlnger, R.W. Fre1, E. Merlan and F. Pocch1ar1, Ed. Pergamon Press, NY. p. 317-325. C-223 �Poland, A. and E. Glover. Potent Inducer of 1973. 2,3,7,8-Tetrachlorodlbenzo-p-dloxln. delta-amlnolevuHnlc add synthetase. Science. 179(4072): 476-477. Poland, A. and E. Glover. 1975. Genetic expression of aryl hydrocarbon hydroxylase by 2,3,7,8-tetrachlorod1benzo-p-d1ox1n: Evidence for a receptor mutation 1n genetically non-responsive mice. Hoi. Pharmacol. 11(4): 389-398. Poland, A. and E. Glover. 1979. An estimate of the maximum in vivo cova- lent binding of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n to rat liver protein, Mbosomal RNA and DNA. Cancer Res. 39(9): 3341-3344. Poland, A. and E. Glover. 1980. 2,3,7,8-tetrachlorod1benzo-p-d1ox1n: Segregation of toxldty with the Ah locus. Mol. Pharmacol. 17(1): 86-94. Poland, A. and J.C. Knutson. 1982. 2,3,7,8-Tetrachlorod1benzo-p-d1ox1n and related halogenated aromatic hydrocarbons: Examination of the mechanism of toxlclty. Ann. Rev. Pharmacol. Toxlcol. 22: 517-554. Poland, A.P., D. Smith, G. Metter and P. Posslck. 1971. A health survey of workers In a 2,4-D and 2,4,5-T plant with special attention to chloracne, porphyrla cutanea tarda, and psychologic parameters. 22: 316-327. C-224 Arch. Environ. Health. �Poland, A.P., E. Glover, J.R. Robinson and D.W. Nebert. expression of aryl hydrocarbon hydroxylase activity. 1974. Genetic Induction of monooxy- genase activities and cytochrome Pl-450 formation by 2,3,7,8-tetrachlorodlbenzo-p-d1ox1n In mice genetically "nonresponslve" to other aromatic hydrocarbons. J. B1ol. Chem. 249(17): 5599-5606. Poland, A., E. Glover, A.S. Kende, M. DeCamp and C.M. G1andomen1co. 1976a. 3,4,3,4-Tetrachloroazoybenzene and azobenzene potent Inducers of aryl hydrocarbon hydroxylase. Science. 194(4265): 627-630. Poland, A., E. Glover and A.S. Kende. 1976b. Stereospeclflc, high affinity binding of 2,3,7,8-tetrach1orod1benzo-p-d1ox1n by hepatic cytosol. Evidence that the binding species 1s a receptor for Induction of aryl hydrocarbon hydroxylase. J. Blol. Chem. 251(16): 4936-4946. Poland, A., D. Palen and E. Glover. 1982. Tumour promotion by 1CDD 1n skin of HRS/J mice. Nature. 300(5889): 271-273. Poland, A., J. Knutson and E. Glover. 1983. A consideration of the mechan- ism of action of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n and related halogenated aromatic hydrocarbons. Iji: Human and Environmental Risks of Chlorinated D1ox1ns and Related Compounds, R.E. Rucker, A.L. Young and A.P. Gray, Ed. Plenum Press, NY. p. 539-559. C-225 �Poll, A., G. Grancesclnl, L. Pugllsl and C.R. Slrtorl. 1980. Increased total and high density llpoproteln cholesterol with apoproteln changes resembling streptozotocln diabetes In tetrachlorodlbenzodloxln (TCDD) treated rats. Blochem. Pharmacol. 28(5): 835-838. Poole, C. 1983. Statement of Charles Poole before the United States House of Representatives Science and Technology Committee, Subcommittee on Natural Resources, Agricultural Research and Environment. March 23. Ramsey, J.C., J.G. Hefner, R.J. Karbowskl, W.H. Braun and P.J. GehMng. 1979. The in vivo blotransformatlon of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCDD) 1n the rat. Toxlcol. Appl. Pharmacol. 48: A162. (Abstr.) Ramsey, J.C., J.G. Hefner, R.3. Karbowskl, W.H. Braun and P.O. Gehrlng. 1982. The in vivo blotransformatlon of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCDD) 1n the rat. Toxlcol. Appl. Pharmacol. 65: 180-184. Rappe, C., S. Marklund, P.A. Bergqvlst and P. Hansson. 1983. PolychloM- nated dloxlns, dlbenzofurans and other polychloMnated polynuclear aromatlcs formed during Inclnator and PCB fires. In: Chlorinated Dloxlns and Dlbenzofurans 1n the Total Environment, Vol. I, L.H. Keith, et a!., Ed. Butterworth Publisher. Regg1an1, G. 1980. Acute human exposure to TCDD 1n Seveso, Italy. Toxlcol. Environ. Health. 6(1): 27-43. C-226 J. �R11h1mak1, V., S. Asp, A.M. Seppalalnen and S. Hernberg. 1978. Symptoma- tology, morbidity and mortality of experience of chlorinated phenoxy herbicide (2,A,5, 2,4,5-T) sprayers 1n Finland. logical study. add A clinical and epidemic- Working paper for an IARC working group meeting on coordina- tion of ep1dem1olog1cal studies on the long-term hazards of chlorinated dlbenzodloxlns and chlorinated dlbenzofurans, Lyon, France, Jan. 10-11, 1978. R11h1mak1, V., A. S1sko and S. Hernberg. phenoxyacetlc add 1982. Mortality of 2,4-dlchloro- and 2,4,5-tr1chlorophenoxyacet1c add herbicide appli- cators 1n Finland. Scand. J. Work Environ. Health. 8: 37-42. R11h1mak1, V., S. Asp, E. Pukkala and S. Hernberg. cancer morbidity among chlorinated phenoxy add 1983. Mortality and applicators In Finland. Chemosphere. 12(4/5): 779-784. R1zzard1n1, M., M. Romano, F. Tursl, et al. 1983. lexicological evaluation of urban waste Incinerator emissions. Institute d1 Rlcerche Farmacologlche "Mario NegM". Milan. Italy. Robinson, J.R., 3.S. Felton, R.C. Levitt, S.S. Thorgelrsson and D.W. Nebert. 1975. Relationship between aromatic hydrocarbon responsiveness and the survival times In mice treated with various drugs and environmental compounds. Mol. Pharmacol. 11: 850-865. Rogers, A.M., M.E. Anderson and K.C. Back. 1982. Mutagenlclty of 2,3,7,8- tetrachlorod1benzo-p-d1ox1n and perfluoro-n-decano1c add lymphoma cells. Mutat. Res. 105: 445-449. C-227 In L5178Y mouse �Roll, R. 1971. StutMes of the teratogenlc effect of 2,4,5-T In mice. Food Cosmet. Toxlcol. 9(5): 671-676. Rose, J.Q., J.C. Ramsey, T.H. Wentzler, R.A. Hummel and P.J. Gehrlng. 1976. The fate of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n following single and repeated oral doses to the rat. Toxlcol. Appl. Pharmacol. 36(2): 209-226. Ross, R.T. 1976. Memorandum to R. Oreer and 2,4,5-T dloxlns analytical collaborators. U.S. EPA, July. Ryan, J.J., J.C. PHon, H.B.S. Conacher and D. Firestone. 1983. Inter- laboratory study on determination of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n In fish. J. Assoc. Off. Anal. Chem. 66: 700-707. Sarma, P.R. and J. Jacops. 1981. Thoracic soft-tissue sarcoma In Vietnam veterans exposed to Agent Orange. Lancet. 302(18): 1109. Sawahata, T., J.R. Olson and R.A. Neal. 1982. Identification of metabo- lites 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCOO) formed an Incubation with Isolated rat hepatocytes. Blochem. Blophys., Res. Commun. 105(1): 341-346. Schantz, S.L., D.A. Barsottl and J.R. Allen. 1979. lexicological effects produced In nonhuman primates chronically exposed to fifty parts per trillion 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCDO). 48: A180. C-228 Toxlcol. Appl. Pharmacol. �Schulz, K.H. 1957. Kllnlsche und experlmentelle untersuchungen zur atlologle der chloracne. Arch. Klin. Exp. Dermatol. 206: 589-596. (Ger.) Schwetz, B.A., J.M. NorMs, G.L. Sparschu, et al. chlorinated d1benzo-p-d1ox1ns. Environ. Health Perspect. Seefeld, H.O. and R.E. Peterson. posed mechanism. 1973. 1983. Toxicology of 5: 87-99. TCDD Induced weight loss: A pro- In: Human and Environmental Risks of Chlorinated Dloxlns and Related Compounds, R.E. Tucker, A.L. Young and A.P. Gray, Ed. Plenum Press, NY. p. 405-413. Seller, J.P. 1973. A survey on the mutagenldty of various pesticides. Exper1ent1a. 29: 622-623. Shadoff, L.A., R.A. Hummel, L. Lamparskl and J.H. Davidson. 1977. A search for 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCDD) 1n an environment exposed annually to 2,4,5-tr1chlorophenoxyacet1c add ester (2,4,5-T) herbicides. Bull. Environ. Contam. Toxlcol. 18(4): 478-485. Shum, S., N.M. Jensen and D.W. Nebert. 1979. The Ah Locus. In. utero tox- Iclty and teratogenesls associated with genetic differences In benzofa]pyrene metabolism. Teratology. 20: 365-376. Singer, R., M. Moses, J. Valclukas, R. Lllls and 1.3. SeHkoff. 1982. Nerve conduction velocity studies of workers employed 1n the manufacture of phenoxy herbicides. Environ. Res. 29: 297-311. C-229 �Smith, A.H., J.E. Francis, S.J. Kay and J.B. Grelg. 1981. Hepatic toxldty and uroporphyrlnogen decarboxylase activity following a single dose of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n to mice. Blochem. Pharmacol. 30(2): 2825-2830. Smith, A.H., D.O. Fisher, N.P. Dip and C.J. Chapman. 1982a. Congenital defects and miscarriages among New Zealand 2,4,5-T sprayers. Arch. Environ. Health. 37: 197-200. Smith, A.H., 0.0. Fisher, N. Pearce and C.A. Teague. tural chemicals cause soft-tissue sarcoma? 1982b. Do agricul- Initial findings of a case- control study 1n New Zealand. Community Health Studies. 6(2): 114-119. Smith, A.H., D.O. Fisher, H.J. Giles and N. Pearce. 1983a. The New Zealand soft tissue sarcoma case-control study: Interview findings concerning phenoxyacetlc add exposure. Chemosphere. 12(4/5): 565-571. Smith, J. 1979. EPA halts most uses of herbicide 2,4,5-T. Science. 203(4385): 1090-1091. Smith, F.A., B.A. Schwetz and K.D. NHschke. 1976. Teratogenlclty of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n 1n CF-1 mice. Toxlcol. Appl. Pharmacol. 38(3): 517-523. Smith, R.M., P.M. O'Keefe, K.M. Aldous, D.R. Hllker and J.E. O'Brlan. 1983b. 2,3,7,8-Tetrachlorod1benzo-p-d1ox1n 1n sediment and samples from Love Canal storm sewers and creeks. Environ. Sc1. Techno!. 17: 6-10. C-230 �Sparschu, G.L., Jr., F.L. Dunn, Jr., V.K. Rowe, Jr. 1971a. teratogenldty of 2,3,7,8-tetrach1orod1benzo-p-d1ox1n Study of the 1n the rat. Food Costnet. Toxlcol. 9: 405-412. Sparschu, G.L., F.L. Dunn, R.W. Llsowe and V.K. Rowe. effects 1971b. Study of of high levels of 2,4,5-trlchlorophenoxyacetlc acid on foetal development 1n the rat. Food Cosmet. Toxlcol. 9(4): 527-530. Stalling, D.L., L.H. Smith, J.D. Petty, et al. 1983. Residues of poly- chlorinated d1benzo-p-d1ox1ns and dlbenzofurans 1n Laurentlan Great Lakes fish. In: Human and Environmental Risks of Chlorinated D1ox1ns and Related Compounds, R.E. Tucker et al., Ed. Plenum Publishing Corp. Stehl, R.H. and L.L. Lamparskl. 1977. p. 221-240. Combustion of several 2,4,5-tr1- chlorophenoxy compounds: Formation of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n. Science. 197: 1008-1009. Stephan, C.E. 1980. Memorandum to J. Stara, U.S. EPA, Cincinnati, OH, July. Stevens, K.M. 1981. Agent Orange toxldty: A quantitative perspective. Human Toxlcol. 1(1): 31-39. Sweeney, G.D. and K.G. Jones. chlorinated hydrocarbons. 1978. On the mechanism of prophyrla due to Int. Congr. Ser. Excerpta Hed. 440: 229-231. C-231 �Sweeney, G.O., K.G. 3ones, P.M. Cole, D. Basford and F. Krestynskl. 1979. Iron deficiency prevents liver toxlclty of 2,3,7,8-tetrachlorodlbenzo-pdloxln. Science. 204(4390): 332-335. Swift, L.L., T.A. Gas1ew1cz, G.D. Dunn, P.O. Soul'e and R.A. Neal. 1981. Characterization of the hyperllpldemla 1n guinea pigs Induced by 2,3,7,8tetrachlorod1benzo-p-d1ox1n. Toxlcol. Appl. Pharmacol. 59(3): 489-499. Taylor, J.S. 1979. Environmental chloracne: Update and overview. Ann. NY Acad. Sc1. 320: 295-307. Taylor, M.L., T.O. Tlernan, 3.H. Garrett, G.F. Van Ness and 3.G. Solch. 1983. Assessments of Incineration processes chlorinated hydrocarbons: Concentrations as sources of polychlorlnated dloxlns, dlbenzofurans and possible precursor compounds effluents. of supertoxlc dlbenzo-p- 1n Incinerator In: Chlorinated Dloxlns and Dlbenzofurans In the Total Environ- ment, Vol. I, L.H. Keith et al., Ed. Butterworth Publishers, p. 125-164. Teleglna, K.A. and L.I. Blkbulatova. 1970. Affection of the folllcular apparatus of the skin In workers occupied 1n production of butyl ether of 2,4,5-tr1chlorophenoxyacet1c add. Thelss, A.M. and P. Goldmann. 1977. geschehen 1n der BASFAG vom 13. Medlchem-Kongress, Haifa. Vestn. Dermatol. Veneral. 44: 35-39. Uber das tr1chlorphenol-d1ox1nunfall- November 1953. (Ger.) C-232 in: Vortrag, auf dem IV. �Thelss, A.M. and R. Frentzel-Beyme. exposed November 1977. Mortality study of persons to dloxln following an accident which occurred In the BASF on 13, 1955. Proceed. MEICHEM Congress V., San Francisco, CA, September 5, 1977. Thelss, A.M., R. Frentzel-Beyme and R. Link. 1982. Mortality study of persons exposed to dloxln In a trlchlorophenol-process accident that occurred 1n the BASF AG on November 17, 1953. Am. J. Ind. Med. 3: 179-189. Thlbodeaux, L.J. 1979. Chemodynamlcs-Envlronmental Movement of Chemicals In Air, Water and Soil. John Wiley and Sons, NY. p. 1976-1977. Thlbodeaux, L. 1983. Offslte Transport of 2,3,7,8-Tetrachlorod1benzo-p- dloxln from a Production Disposal Facility. In: Chlorinated 01ox1ns and Olbenzofurans In the Total Environment, L. Keith, Ed. Butterworth Pub- lishers, Woolburn, MA. p. 1-14. Thlgpen, J.E., R.E. Faith, E.E. McConnell and J.A. Moore. 1975. Increased susceptibility to bacterial Infection as a sequela of exposure to 2,3,7,8tetrachlorod1benzo-p-d1ox1n. Infect. Immun. 12(6): 1319-1324. Thomas, P.E., J.J. Mutton and B.A. Taylor. 1973. Genetic relationship between aryl hydrocarbon hydroxylase 1nduc1b1l1ty and chemical carcinogen Induced skin ulceratlon 1n mice. Genetics. 74: 655-659. Thomas, R.F. 1980. Internal memo to P. Cohn, OTS, U.S. EPA, Washington, DC. C-233 �Thompson, D.3., J.L. Emerson and S.L. Sparschu. 1971. Study of the effects of 2,4,5-trlchlorophenoxyacetlc add development. Tlernan, T.O. (2,4,5-T) on rat and rabbit fetal Teratology. 4: 243. 1983. Analytical chemistry of polychlorlnated d1benzo-p- dloxlns and dlbenzofurans: A review of the current status. Jjr. Chlorinated D1ox1ns and Dlbenzofurans In the Total Environment, Vol. I, L.H. Keith et al., Ed. Butterworth Publishers, p. 211-237. Tlernan, T.O., M.L. Taylor, J.G. Solch, G.F. Van Ness and J.H. Garrett. 1982. Characterization of toxic components 1n the effluents from a refuse- fired Incinerator. Res. Conserv. 9: 343-354. Tognonl, G. and A. Bonaccorsl. 1982. Ep1dem1olog1cal problems with 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCDO). Drug Metab. Rev. 13: 447-469. Toth, K., S. Somfal-Relle, J. Sugar and J. Bence. 1979. Carc1nogen1c1ty testing of herbicide 2,4,5-trlchlorophenoxyethanol containing dloxln and of pure dloxln 1n Swiss mice. Nature. 278(5704): 548-549. Toth, K., J. Sugar, S. Somfal-Relle and J. Bence. 1978. Carcinogenic bioassay of the herbicide, 2,4,5-trlchlorophenoxyethanol (TCPE) with different 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (dloxln) content 1n Swiss mice. Blochem. Pharmacol. 14: 82-93. C-234 Prog. �Townsend, J.C., K.M. Bodner, P.P. Van Peenen, R.D. Olsen and R.R. Cook. 1982. Survey of reproductive events of wives of employees exposed to chlorinated dloxlns. Am. J. Ep1dem1ol. 115(5): 695-713. Isushlmoto, G., F. Matsumura and R. Sago. 1982. Fate of 2,3,7,8-tetra- chlorod1benzo-p-d1ox1n (TCDD) 1n an outdoor pond and 1n model aquatic ecosystem. Environ. Toxlcol. Chem. 1: 61-68. Turner, G.N. and D.N. Collins. 1983. Liver morphology 1n guinea pigs administered either pyrolysls products of a polychlorlnated blophenyl transformer fluid or 2,3,7,8-tetrachlorod1benzo-p-d1ox1n. Toxlcol. Appl. Phar- macol. 67: 417-429. United Kingdom Ministry of Agriculture, Fisheries and Food. 1983. Advisory committeee on pesticides: Report on phenoxy add herbicides. Whitehall Place, London, February 7. U.S. OHHS. 1980a. Renamed as: NTP (National Toxicology Program). 1980b. Bloassay of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n for possible carclnogenlclty (dermal study). Cardnogenesls Test. Prog., NCI, NIH, Bethesda, MD and NTP, Research Triangle Park, NC. Publ. No. 80-1757. U.S. DHHS. 1980b. Renamed: NTP (National Toxicology Program). 1980a. Bloassay of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n for possible carclnogenlclty (gavage study). Cardnogenesls Test. Prog., NCI, NIH, Bethesda, MD and NTP, Research Triangle Park, NC. Pub. No. 80-1765. C-235 �U.S. EPA. 1979. Report of assessment of a field Investigation of six-year spontaneous abortion rates 1n three Oregon areas 1n relation to forest 2,4,5-T spray practice. Office of Toxic Substances, U.S. EPA. U.S. EPA. 1980a. Seafood consumption data analysis. SRI Int., Menlo Park, CA. Final rep., Task 11. EPA Contract No. 68-01-3887. U.S. EPA. 1980b. Direct testimony of Dr. Wilbur P. McNulty before the administrator, U.S. EPA. FIFRA Docket Nos. 415 et al., EPA Exhibit No. 106. U.S. EPA. 1980c. Problem oriented report. 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCDD). Carcinogen assessment of Prepared by Carcinogen Assess- ment Group, U.S. EPA, November 24. U.S. EPA. 1980d. D1ox1ns. Industrial Environmental Research Lab., U.S. EPA, Cincinnati, OH. EPA 600/2-80-197. U.S. EPA. 1982. Environmental Monitoring at Love Canal, Vol. I. ORD, U.S. EPA, Washington, DC, May, 1982. EPA 600/4-82-030a. NTIS PB 82-237330. van den Berg, M., K. Olle and 0. Hutzlnger. 1983. Uptake and selective retention 1n rats of orally administered chlorinated dloxlns and dlbenzofurans from fly ash and fly ash extract. Chemosphere. 12(4/5): 537-544. van Logten, M.J., B.N. Gupta, E.E. McConnell and J.A. Moore. 1980. Role of the endocrine system 1n the action of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n (TCDD) on the thymus. Toxicology. 15(2): 135-144. C-236 �van Logten, M.J., B.N. Gupta, E.E. McConnell and J.A. Moore. 1981. The Influence of malnutrition on the toxldty of 2,3,7,8-tetrachlorodlbenzo-pdloxln (TCDD) 1n rats. Toxicology. 21(1): 77-88. Van Miller, J.P., J.J. Lallch and J.R. Allen. 1977a. Increased Incidence of neoplasms 1n rats exposed to low levels of 2,3,7,8-tetrachlorodlbenzorhodloxln. Chemosphere. 6(10): 625-632. Van MUler, J.P., J.J. Lallch and J.R. Allen. 19775. Increased Incidence of neoplasms 1n rats exposed to low levels of 2,3,7,8-tetrachlorod1benzo-pdloxln. Chemosphere. 6(9): 537-544. Van Miller, J.P., R.J. Marlar and J.R. Allen. 1976. Tissue distribution and excretion of trltlated tetrachlorodlbenzo-p-dloxln 1n non-human primates and rats. Food Cosmet. Toxlcol. 14(1): 31-34. Vecchl, A., M. Slronl, M.A. Canegratl, M. Recchla and S. Garattlnl. 1983. Immunosuppresslve effects of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n In strains of mice with different susceptibility to Induction of aryl hydrocarbon hydroxylase. Toxlcol. Appl. Pharmacol. 68: 434-441. Vlnopal, J.H. and J.E. Caslda. 1973. Metabolic stability of 2,3,7,8-tetra- chlorod1benzo-p-d1ox1n 1n mammalian liver mlcrosomal systems and In living mice. Arch. Environ. Contam. Toxlcol. 1(2): 122-132. C-237 �Vos, J.G. and R.B. Beems. 1971. Dermal toxldty studies of technical polychlorinated blphenyls and fractions thereof In rabbits. Toxlcol. Appl. Pharmacol. 19: 617-633. Vos, J.G. and J.H. Koeman. 1970. Comparative toxlcologlc study with polychlorinated blphenyls 1n chickens with special reference to porphyrla, edema formation, liver necrosis, and tissue residues. Toxlcol. Appl. Pharmacol. 17: 656-668. Vos, J.G., J.G. Kreeftenberg, H.W. Engel, A. Mlnderhoud and J.L.M. Van Noorle. 1978. Studies on 2,3,7,8-tetrachlorod1benzo-p-d1ox1n Induced Immune suppression and decreased resistance to Infection: Endotoxln hypersens1t1v1ty, serum zinc concentrations and effect of thymosln treatment. Toxicology. 9(1-2): 75-86. Vos, J.G., J.G. Moore and J.G. 21nkl. 1973. Effect of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n on the Immune system of laboratory animals. Environ. Health Perspect. 5: 149-162. Vos, J.G., J.A. Moore and J.G. Z1nkl. 1974. chlorod1benzo-p-d1ox1n (TCDD) In C57BL/6 mice. Toxlclty of 2,3,7,8-tetraToxlcol. Appl. Pharmacol. 29: 229-241. Ward, C. and F. Matsumura. 1977. Fate of 2,4,5-T contaminant, 2,3,7,8- tetrachlorod1benzo-p-d1ox1n (TCDD) In aquatic environments. 28 p. C-238 NTIS PB-264187. �Ward, C.T. and F. Hatsumura. 1978. Fate of 2,3,7,8-tetrachlorod1benzo-p- dloxln (TCDD) In a model aquatic environment. Arch. Environ. Contarn. Toxlcol. 7: 349-357. Wasson, J.S., J.E. Huff and N. Loprleno. 1978. A review of the genetic toxicology of chlorinated d1benzo-p-d1ox1ns. Hutat. Res. 47(3-4): 141-160. W1pf, H.K. and J. Schmld. 1983. Seveso - An environmental assessment. In: Human and Environmental Risks of Chlorinated D1ox1ns and Related Compounds, R.E. Tucker et al., Ed. Plenum Publishing Corp., NY. p. 255-274. YamaglsM, T., T. M1yazak1, K. Aklyama, et al. 1981. Polychlorlnated d1benzo-p-d1ox1ns and dlbenzofurans 1n commercial dlphenyl ether herbicides, and 1n freshwater fish collected from the application area. Chemosphere. 10: 1137-1144. Young, A.L. 1983. Long-term studies on the persistence and movement of TCDD In a natural ecosystem. In: Human and Environmental Risks of Chlori- nated Dloxlns and Related Compounds, R.E. Tucker et al., Ed. Plenum Pub- lishing Corp., NY. p. 173-190. Young, A.L., H.K. Kang and B.H. Shepard. herbicide contaminants. 1983. Chlorinated dloxlns as Environ. Sc1. Technol. 17: 530A-539A. C-239 �Zack, T.A. 1980. (Referred by Honchar and HalpeMn, 1981, Lancet. 1: 268 and In a letter from E.H. Blair of The Dow Chemical Co., to the Document Control Officer, Chemical Information D1v., OTS, U.S. EPA, Washington, DC, March 6, 1981.) (Unpubl. rep.) Zack, T.A. and R.R. Susklnd. 1980. The mortality experience of workers exposed to tetrachlorodlbenzodloxln 1n a trlchlorophenol process accident. J. Occup. Med. 22: 11. Zleger, E. 1983. Memorandum from Dr. Zleger of Dr. E.E. McConnell on the results of test performed for the Environmental Mutagenesls Development Program. NTP, NIEHS. Z1nkl, J.G., J.G. Vos, J.A. Moore and B.N. Gupta. 1973. Hematologlc and clinical chemistry effects of 2,3,7,8-tetrachlorod1benzo-p-d1ox1n 1n laboratory animals. Environ. Health Perspect. 5: 111-118. C-240 �APPENDIX Summary and Conclusions Regarding the Carclnogenldty of 2.3.7,8-Tetrachlorod1benzo-p-D1ox1n (TCDD1* 2,3,7,8-TCDD Is probably carcinogenic for humans on the basis of animal cardnogenldty studies which were positive In multiple species and organs. Ep1dem1olog1cal studies of workers exposed to chemicals contaminated with 2,3,7,8-TCDD such as 2,4,5-tr1chlorophenoxyacet1c acid and 2,4,5-trlchlorophenol are consistent with the position that 2,3,7,8-TCDD Is probably carcinogenic for humans; the available evidence Indicates an excess Incidence of soft tissue sarcomas. Because 2,3,7,8-TCDD 1s almost always found 1n association with other materials (e.g., chlorophenols, combustion products, etc.), H may never be possible to evaluate the carc1nogen1c1ty of 2,3,7,8-TCDD by Itself 1n humans. SUMMARY OF HUMAN POTENCY ESTIMATES BASED ON PERTINENT DATA A summary of 95% upper-limit human carcinogenic potency estimates for 2,3,7,8-TCDD derived from the Kodba et al. (1978) and NCI (U.S. DHHS, 1980b) studies 1n rats and mice, with two pathologlsts 1 findings for the Kodba study, are given 1n Table 32. These potency estimates have been calculated using the linearized multistage model by a previously described methodology (45 FR 79350-79353). (q *) comes The largest of these potency factors from data In an Independent pathologlst's (Dr. R. Squire) review of the Kodba feeding study of female Sprague-Dawley rats. An adjustment for high early mortality 1n the high dose groups led to a slightly lower estimate. The mean of the two pathologlsts' estimates after mortality adjustment Is: q = [(1.51 x 10s) x (1.61 x 105)]1/2 = 1.56 x 10s (mg/kg/day)"1 *Th1s summary was prepared and approved by an expert peer review panel on dloxlns convened by the U.S. EPA 1n Cincinnati on July 17, 1983. C-241 �TABLE 32 Summary of Human Potency Estimates for 2,3,7,8-TCDD Species Rat Study Sex Kodba et al.a male Pathologist Human Potency Estimate q-|* 1n (mg/kg/day)"1 female 1.73 x 10* Kodba Unadjusted Adjusted for early mortality 2.52 x 10s 1.51 x 10sb Squire Unadjusted Adjusted for early mortality Kodba et al.a 1.47 x 10* Squire Rat Kodba 4.25 x 10s b 1.61 x 10* Rat NCIC female NCI-rev1ewed 3.28 x 10* Mouse NCIC male NCI-rev1ewed 7.52 x 10* Mouse NCIC female NCI-rev1ewed 4.56 x 10* a Source: Kodba et al., 1978 ^Values used to determine the geometric mean of 1.56x10* (mg/kg/day)'1 c Source: U.S. OHHS, 1980b C-242 �These potency estimates were derived from the Kodba feeding study. The responses and parameters of the Kodba feeding study 1n female rats are given below. The number with tumors refers to the number of animals with at least one of liver, lung, hard palate, and/or nasal turblnate tumors. Adjustment for early mortality refers to eliminating those animals which died during the first year of study. The first tumor appeared In the high dose group during the thirteenth month. No. with Tumors/No. Examined Adjusted for Early MortalUy Dose (mg/kg/day) Squire 16/85 8/48 27/48 34/40 0 0.001 x 10~a 0.01 x 10"3 0.1 x 10'» Kodba 9/85 3/48 18/48 34/40 Wh = 70 kg Wo = 0.450 kg R = 5000 I/kg le = 720 days Le = 720 days L = 720 days With these parameters, the mean 95% upper-limit carcinogenic potency factor for humans, q *, 1s 1.56xl05 {mg/kg/day)"1. For a 70 kg human drinking 2 8, water/day and eating 6.5 g of contaminated fish and shellfish, the water concentration should be <1.3xlO~B yg/S, 1n order to keep the upper-limit Individual lifetime cancer risk below 10~6, for example. If fish and shellfish alone are consumed, the corresponding water concentration for this level of risk should be <1.4xlO~e pg/l, and 1f water alone Is consumed, the corresponding water <2.2xlO"7 ng/8.. C-243 concentration should be � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Alvin L. Young Collection on Agent Orange Description An account of the resource <p style="margin-top: -1em; line-height: 1.2em;">The Alvin L. Young Collection on Agent Orange comprises 120 linear feet and spans the late 1800s to 2005; however, the bulk of the coverage is from the 1960s to the 1980s and there are many undated items. The collection was donated to Special Collections of the National Agricultural Library in 1985 by Dr. Alvin L. Young (1942- ). Dr. Young developed the collection as he conducted extensive research on the military defoliant Agent Orange. The collection is in good condition and includes letters, memoranda, books, reports, press releases, journal and newspaper clippings, field logs and notebooks, newsletters, maps, booklets and pamphlets, photographs, memorabilia, and audiotapes of an interview with Dr. Young.</p> <p>For more about this collection, <a href="/exhibits/speccoll/exhibits/show/alvin-l--young-collection-on-a">view the Agent Orange Exhibit.</a></p> Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Box The box containing the original item. 193 Folder The folder containing the original item. 5464 Series The series number of the original item. Series VIII Subseries I Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Description An account of the resource <strong>Corporate Author: </strong>United States Environmental Protection Agency (EPA), Office of Water Regulations and Standards Date A point or period of time associated with an event in the lifecycle of the resource 1984-02-01 Title A name given to the resource Ambient Water Quality Criteria for 2,3,7,8-Tetrachlorodibenzo-p-dioxin ao_seriesVIII https://www.nal.usda.gov/exhibits/speccoll/files/original/c0fdf244b1cd23dcc0c6a2797e2073a6.pdf 28ba802e213b8aef31a024280bbc1c97 PDF Text Text nemDNumber 05480 D Author United States District Court, Eastern District of New Yor RBDOrt/ArtiClB Tltte Amended Complaint, Maureen Ryan, Individually, and as the mother of the infants Kerry Ryan and Michael Ryan, and Michael Ryan, and others similarly situated, Plaintiffs, against the Public Health Service, The Food and Drug Administration, The Veterans Administration, Defendants, In re: "Agent Orange" Product Liability Litigation, Multi District Litigation (MDL) No. 381 Journal/Book Title Year 1984 Month/Day Mav Color n Number of Images ° Dasnrloton NotBS * Also included is a cover letter from Marianne T. Anderson to Alvin L. Young, May 23, 1984. Friday, March 15,2002 Page 5480 of 5571 �CARR, GaansoN & LEE, R C. iei9 PENNSYLVANIA AVENUE,NW MARYLAND DFFICE WASHINGTON, DC 2DDDE 31 WOOD LANE RDCKVILLE, MARYLAND 2DBSD (2D2)4E3-BDE3 LAWRENCE E. CARR, JR. • * WILLIAM J. CARTER » * » VIRGINIA OFFICE _ 4DID UNIVERSITY DRIVE L. PALMER FDRET- * ROBERT W. GaansaN*t FAIRFAX, VIRGINIA 22030 (7D3JESI-B9IB JAME5 F. LEE,JR.'* = MAR3ARETH. WARNER, __ May _ n _ . 23, 1984 KENNETH W. PASSER. DF COUNSEL WILLIAM D. SNEAD, III • • •DC *MD 'VA "NY Dr. Alvin Young Office of Science & Technology Policy White House Dear Al : Just a note to accompany a copy of the complaint filed by Victor Yannacone representing Maureen Ryan, Carrie Ryan, et al. against FDA, VA, and PHS. It appears that Agent Orange has not been resolved as yet. Because of your sustaining interest in the project, I thought you might want to keep apprised of relevant matters. Thus, the complaint is enclosed. If you ever meander down Pennsylvania Avenue, please stop in to see us. Sincerely, Marianne T. Anderson Legal Assistant Encl. �UNITED STATES DISTRICT COURT EASTERN DISTRICT OF NEW YORK In re "AGENT ORANGE" Product Liability Litigation l MAUREEN RYAN, Individually, and as the mother of the infants KERRY RYAN and MICHAEL RYAN, and MICHAEL RYAN, and others similarly situatea ' WEINSTEIN, CH.J. CV-84-2237 Plaintiffs, -againstTHE PUBLIC HEALTH SERVICE, THE POOD AND DRUG ADMINISTRATION, THE VETERANS ADMINISTRATION, Defendants, Multi District Litigation (MDL) No. 381 (All cases) This Master Document submitted by VvW7 YANNACONE &.YANNACONE and PEGALIS & WACHSMAN, P.C., Post Office Drawer #869, Patchogue, New York 11772 (516) 654-2299/487-1990 �UNITED STATES DISTRICT COURT EASTERN DISTRICT OF NEW YORK In re "AGENT ORANGE" igV-84-2237 Product Liability Litigation MAUREEN RYAN, Individually, and as the mother of the infants KERRY RYAN and MICHAEL RYAN, and MICHAEL RYAN, Plaintiffs, -againstTHE PUBLIC HEALTH SERVICE, THE FOOD AND DRUG ADMINISTRATION, THE VETERANS ADMINISTRATION, Defendants, Multi District Litigation (MDL) No. 381 (All cases) This Master Document submitted by YANNACONE & YANNACONE and PEGALIS & WACHSMAN, P.C., Post Office DraWer #869, Patchogue, New York 11772 (516) ^54-2299/487-1990 �VERIFIED COMPLAINT INDEX Page Introduct ion Venue .. 1 . ....... ...... 1 General J r s i t o . . . . . . . . . . . . . . . . . ... 1 - 3 uidcin................ Equitable Jurisdiction.. » 3-4 Class Action Allegations 4-8 The Basis of Plaintiffs' Claims 8-9 Plaintiffs' Claims 10 - 31 Demand for Judgment/Prayer for R l e . . . . . . . . . . 31 - 41 eif.......... Attorney's Verification ~.... ..... -i- 42 �VERIFIED COMPLAINT PLAINTIFFS DEMAND TRIAL BY JURY Plaintiffs allege the following by their attorneys YANNACONE 6 YANNACONE and PE6ALIS & WACHSMAN, P.C. Venue 1. The venue of this action has been determined under the Rules of Procedure of the Judicial Panel on Multidistrict Litigation. General Jurisdiction 2» This action arises under Article VI, section 2, of the Constitution of the United Statesf "This Constitution, and the Laws of the United States shall be made in Pursuance thereof; and all Treaties made, of which shall be made, under the Authority of the United States, shall be the Supreme Law of the Land; and the Judges in every state shall be bound thereby; and Thing in the Constitution of Laws of any State to the Contrary notwithstanding." and involves the declaration and interpretation of the rights retained by the plaintiffs as citizens under the Ninth Amendment of the Constitution of the United states. "The enumeration in the Constitution of certain rights, shall not be contrued to deny or disparage others retained by the people." and under the "due process" clause of the Fifth Amendment to the Constitution of the United States, to the extent that the malfeasance, misfeasance, nonfeasance, and intentional COMPLAINT: Jurisdiction -page 1- �neglect by individual bureaucrats of the defendant federal agencies may deny the individual named plaintiffs and all f others so unfortunate as to be similarly situated, a right of action against the defendants herein for the wrongdoing hereinafter complained of; and under the Eighth -Amendment to the Constitution insofar as plaintiffs' physical and economical afflictions make them d_e facto prisoners of the Veterans Administration System and the wanton and reckless disregard of the Veterans Administration constitutes cruel and unusual punishment. 3. Jurisdiction of this Court is invoke under Title 28 United States Code, Section 1331(a), and where there is an amount in controversy as to any individual member of the class herein, it far exceeds the sum or value of $10,000 exclusive of interest and costs. 4. Jurisdiction of this Court is also invoked under Title 28 United States Code, Section 1346(b), as claims are made hereinunder 28 United States. Code, Sections 2671, et seg., and involve personal injuries, loss of property, or death caused by negligent or wrongful acts or omissions of employees of the defendant federal agencies acting within the scope of their office or employment, under circumstances where a private person would be liable in accordance with the laws of the places where the acts or omissions occurred. COMPLAINT: Jurisdiction -page 2- �5. Upon information and belief, individual bureaucrats as employees of the defendant federal agencies have violated the constitutional rights of the plaintiffs and » ignored the well demonstrated concern of the Congress and People of the United States to provide "general protection" for members of the uniformed services and veterans, as embodied in various federal and state statutes and regulations enacted since the formation of the Continental Army in 1776. 6. Jurisdiction is also predicated under Title 28 United States Code, Section 1361 since mandamus is sought against the appropriate individuals who are responsible for determining policy at the defendant federal agencies. Equitable Jurisdiction 7. This action is brought in equity before this court on certain grounds, stated hereafter, under Title 28 United States Code, Sections 2201 and 2202. 8. The subject matter of this action is essentially equitable in nature since the action is brought for the purpose of restraining the defendants from doing serious, permanent and irreparable injury which cannot be adequately compensated by merely awarding money damages to the class of Plaintiffs, and for mandamus compelling the individuals who are responsible for determining policy at the federal agencies to perform their duties as mandated by Congress. COMPLAINT: Jurisdiction -page 3- �9. The declaratory judgment and injunctive relief demanded on behalf of the class are equitable remedies and the substantive character of the rights sought to be enforced by the Plaintiffs are historically those resolved in a court of equity. •i 10. The law does not afford any adequate remedy for the wrongs of the individual defendant bureacrats. 11. There is no plain, adequate, and complete remedy at law as practical and efficient as the equitable relief sought herein. Class Action Allegations Plaintiffs assert this action is maintainable as a "class action" under the provisions of Rule 23 (b)(l)(a), Rule 23 (b)(l)(b), Rule 23 (b)(2) and Rule 23 (b)(3) of the federal Rules of Civil Procedure. 12. Upon information and belief, during the period from 1962 through 1971, approximately 2*4 million American Servicemen were exposed to contaminated herbicides manufactured, marketed, and distributted for use in Southeast Asia by certain cehmical companies including, THE DOW CHEMICAL COMPANY, MONSANTO CO., HERCULES INCORPORATEDr THOMPSON-HAYWARD CHEMICAL COMPANY, DIAMOND SHAMROCK CORPORATION, UNIROYAL, INC., THOMPSON CHEMICALS CORPORATION, formerly a division of WM» T. COMPLAINT: Jurisdiction -page 4- �THOMPSON CO., HOFFMAN-TAFF, a division of SYNTEX CORP., and HOOKER CHEMICAL COMPANYr a subsidiary Of OCCIDENTAL PETROLEUM CORPORATION, (all hereinafter throughout this complaint referred to collectively as "war contractors"). r 13* Upon information and belief, many servicemen and women now veterans, have manifested symptoms of exposure to toxic synthetic organical chemicals such as 2,3,7,8tetrachloro dibenzo p-dioxin (TCDD or "Dioxin"). 14. Upon information and belief, many servicemen, now veterans suffered traumatic injuries during military service and are therefore entitled to medical care and treatment by the Veterans Administration 15. of the United States. Upon information and belief, unless the equitable relief sought herein is granted by this court, it is reasonable to expect numerous addition claims will be made in the future by individuals so unfortunate as to be similarly afflicted, and it will be necessary for this Court to retain jurisdiction ove the class in order to assure equitable protection of the class. 16. The claims of the plaintiffs are representative of the claims of all the members of the class. COMPLAINT: Class Action -page 5- �17. There are several definable groups of plaintiffs that may be identified within the entire class of plaintiffs t any and each of which may be treated as a class under Rule 23. 18. It is possible to establish criteria for identification of each individual class member by reference to records maintained in the regular course of govenmental operations by various agencies of the United States of America/ in particular, the Department of Defense/ the r Veterans Adminsitration, and the Department of Health/ Education and Welfare. 1. 9. Member of the class are fairly and adequately represented by counsel for the plaintiffs and neither the individual plaintiffs named at this time nor the attorneys for said plaintiffs so named have any interests adverse to those of any individual members of the class of all those who might be entitled to the relief sought herein. 20. There are substantial questions of law and fact common to the class. 21. The questions of law common to the class-of all those who have already been affected/ or who are now or will be so effected/ include/ but are not limited to: (a) the joint and several liability of the defendant federal COMPLAINT: agencies; Class Action -page 6- �(b) the nature and extent of the legal and fiduciary obligations and duties that the defendant federal agencies, by reason of their political power and administrative resources, owe the plaintiffs; • (c) the remedy to be fashioned by the Court to redress the wrongs committed by the defendant federal agencies upon the plaintiffs, and all others similarly afflicted; and (d) 22. the appropriate measure of damages. Some of the questions of fact common to the t class of all those who have already been affected, or who are now, or will be so affected, include, but are not limited to: (a) the extent of the toxicity attributable to the phenoxy herbicides that were contaminated with polychlorinate dibenzo-p-dioxins (PCDDs) and deployed as chemical defoliants in Viet Nam; (b) the characteristics of such toxic effects manifested in human beings; (c) the permanence of such toxic effects; (d) the extension of toxic effects through genetic and somatic damage to succeeding generations; and COMPLAINT: Class Action -page 7- �(e) what knowledge each defendant federal agency had, or with the exercise of reasonable care and concern for the health, safety, and welfare of the plaintiff veterans and their families should have had, concerning such toxic effects. (f) whether the mistreatment, medical mal• practice and hospital neglect rampant throughout the Veterans Administration health care system is so widespread as to constitute a government "policy"* The Basis of Plaintiffs' Claims 23*. Upon information and belief, based on certain statements of fact alleged to be true by the defendant war contractors in their Third Party Complaints against certain departments or agencies of the United States of America (hereinafter referred to as " . . or "United States") and U S * certain officials, employees, and agents of such departments of agencies, (MDL 381 Docket Documents bearing Documents Numbers 108, 109, 110, 111, and 112) subsequent to the time that each of the veterans were discharged from the Armed Forces of the United States, the defendant federal agencies failed to warn the plaintiff of the risks of devastating injury resulting from exposure to toxic synthetic organic f chemicals such as 2, 3, 7, 8-tetrachloro dibenzo p-dioxin COMPLAINT: Claims' Basis -page 8- �(TCDD or "Dioxin") and the defendant federal agencies failed to provide time and accurate information regarding the nature of such hazards to the plaintiff veterans and their families. 24. Upon information and belief, based on certain statements of fact alleged to be true by the war contractors in said Third Party Complaints against the United States, the Veterans Administration and other agencies of the U.S. Government failed to provide adequate medical care and treatment, including genetic counseling to the plaintiff veterans and their familiesr subjecting said plaintiff veterans and their families to unreasonable risk of suffering personal injuries and sustaining serious and permanent . damages. 25. According to said Third Party Claims of the war contractors, the damages sustained by the plaintiff veterans and their families were "caused, in whole or in part, by the negligent .. wanton and reckless conduct" of . the Veterans Administration and other agencies of the United States. COMPLAINT: Claims' Basis -page 9- �Plaintiffs' Claims 26. That upon information and belief, that since 1973, the defendant federal agencies knew, or, with the exercise of reasonable care and concern for the health, safety, and welfare of the plaintiff veterans and their families should have known, the following facts concerning phenoxy herbicides and their use by the United Stated during combat in Vietnam and Southeast Asia. (a) The chlorinated phenoxy herbicidesf such as 2,4,5-trichlorophenoxy acetic acid ( , , - ) and 2 4 5 T 2,.4-dichloro phenoxyacetic acid (2,4-D), are growth regulators with hormone-like activity. (b) They were manufactured, advertised, marketed, and sold since the mid-1940s, shortly after the end of World War II. (c) In the course of synthesizing 2,4,5-T, several chlorinated dioxins, including 2,3,7,8tetrachloro dibenzo p-dioxin ( C D or *Dioxin"), T D are formed as intermediates or by-products of the reaction, and unless destroyed or otherwise removed can, have, and do, contaminate such commercial phenoxy herbicides as 2 4 5 T and 2,4,5-TP ("silvex*) , , (d) In 1970, the Surgeon General of the United States reproted that the use of 2 4 5 T , , - COMPLAINT: Plaintiffs' Claims -page 10- �might be hazardous to human health, and the United States Secretary of Health, Education, and Welfare and the United States Secretary of Agriculture issued a joint order calling for an immediate cessationrof all uses of 2,4,5-T on or around lakes, ponds, and ditch banks and of all uses of liquid formulations recreation areas. around around homes and Another order issued by these two departments cancelled all uses of granular formulation of 2,4,5-T on crops intended for human consumption* (. e) Use of 2,4,5-T was banned in Italy in 1 7 , and its use was also banned in the Netherlands 9 0 and Sweden. (f) The effort to suspend registration of 2,4r5,-T in 1 7 resulted from published studies 9 0 indicating that 2 4 5 T was a teratogen ( . . , , i e , caused birth defects). Subsequent studies, however, indicated that the teratogenic effects charged to t the phenoxy herbicide 2 4 5 T were largely attrib, , utable to the contaminant TCDD. (g) In 1 7 , a report of a Scientific Advisory 9 1 Committee recommended restroation of registration of 2,4,5-T to the status existing prior to April 1970, with the following conditions: . COMPLAINT: Plaintiffs* Claims (1) a 0.1 -page 11- �rag/kg ( . parts per million (ppm)) tolerance of 0 1 2,4,5-T on edible food crops; (2) a limit of 0.5 rag/kg ( . 0 5 ppm) TCDD contamination in 2,4,5-T produced in the future, equivalent to a 0.05 parts per trillion (ppt) TCDD residue on edible food crops, with certified analyses being furnished by the manufacturers to the Environmental Protection Agency (EPA); and (3) provisions for warnings on labels of formulations used around the home. The report recommended additional research on possible soil accumulation and food chain magnification of TCDD and the establishment of monitoring programs for detection of adverse effects that might be associated with continued use of phenoxy herbicides such as 2,4,5-T. (h) On or about 13 April 1972, a new order was issued by the EPA continuing the suspension 2 4 5 T use around homes, recreational , , - of sites, and aquatic areas, and cancelling use of 2,4,5-T on crops used for human food. • (i) In June 1974, the EPA dropped proceedings to ban most uses of 2 4 5 T and cancelled , , - the scheduled hearingsr however, the United States Department of Agriculture decision made in 1970 to cancel use of 2,4,5-T in and around homes, gardens / COMPLAINT: Plaintiffs' Claims -page 12- �and recreational areas was not affected by this action. ( ) j Herbicides were introduced into the armed conflict in Viet Nam in 1962. An estimated 107 million pounds of herbicides, approximately 94% of which were phenoxy herbicides, including 2,4,5-Tr contaminated with toxic synthetic organic chemicals such as TCDD, were aerially disseminated over approximately 6 million acres in South Viet Nam from January, 1962 through February, 1 7 . 9 1 ( ) k American servicemen were exposed to approximately cointained 44 million pounds of 2,4,5-T, which an estimated 368 pounds of the toxic contaminant TCDD. (1) Approximately 96% of all the 2,4,5-T used in Viet Nam was contained in a formulation designated "Herbicide Orange"; the remaining 2 4 5 T was contained in formulations designated , , "Herbicide Green", "Herbicide Pink* and Herbicide Purple". (m) 90% of all the "Herbicide Orange", con- taminated with approximately toxic chemical TCDD, was used 203 pounds of the in defoliation opera- tions on 2 9 million acres of inland forests and . mangrove forests of South Viet Nam. COMPLAINT: Plaintiffs1 Claims -page 13- �(n) The Department of Defense suspended the use of "Herbicide Orange" shortly after the civilian i agencies of government announced the effort to suspect certain uses of the phenoxy herbicide 2 4 5 T , , - . (o) In 1971, the Department of Defense directed that the "Herbicide Orange" in South Viet Mam be returned to the United States and that the entire stock be disposed of in an environmentally safe and efficient manner. 27. That upon information and belief, on or about 9 January 1980, the defendant federal agencies knew, or with, the exercise of reasonable care and concern for the healthr safety, and welfare of the plaintiff veterans and their families, should have known the following 2,3,7,8-tetrachloro dibenzo p-dioxin (a) facts concerning (TCDD or "Dioxin"): The polychlorinated dibenzo-p-dioxins (PCDDs) and the polychlorinated dibenzo furans (PCDFs) are two series of related chemical compounds that exhibit similar chemical and physical properties. • The first chlorinated dioxin was prepared as early as 1872 by two German chemists, Merz and Weith. 1957, In Gilman and Dietrich reported that they had prepared halogenated dioxins in quantities of approximately 20 grams. COMPLAINT: Plaintiffs' Claims -page 14- �(b) TCDD is an unwanted and unnecessary contaminant in the industrial preparation of 'the herbicide 2,4,5-T. (c) Hunan exposure to TCDD has occurred during the production of certain compounds such as the herbicide 2,4,5-T, the fungicide pentachlorophenol, and the germicide hexachlorophene. The dioxins occur as impurites and/or contaminants associated with these commercial products. (d) TCDD is a relatively immobile molecule, not easily decomposed in soil or readily broken down by soil microorganisms. TCDD is capable of being taken up and retained by living organisms and can enter the human body from several contaminated sources, including water. (e) TCDD is one of the most potent low molecular weight toxins known and has been extensively studied. (f) TCDD is a "cellular poison". (g) The pathologic effects produced by the toxic isomers of polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzo furans (PCDFs) vary quantitatively and qualitatively among different species; howeverr within a single species the toxic effects of these compounds are similar, differing COMPLAINT: Plaintiffs' Claims -page 15- �only in the intesity of the toxic effect produced over the time of the study. (h) Chloracne or acneform dermatitis charac- terized by comedones, keratic cysts, pustules, papules, and abscesses, is often associated with and characteristic of human exposure to PCDDs, including TCDD. Squamous metaplasia and kera- tinization of sebaceous glands and hair follicles have also been observed in the skin of a number of animal species following exposre to TCDD. Clinical symptoms of chloracne may appear weeks or months after exposure. (i) In 1 5 , TCDD was identified as the 9 7 agent responsible for causing occupational chloracne in employees factories producing chlorophenol. In 1971 polychlorinated dibenzo-p-dioxins (PCDDs) were implicated as the cause of chloracne in a plant producing 2,4-D and 2 4 5 - . , , , T (j) In 1958, a condition known as "toxic fat syndrome*, characterized by hydropericardiuro, ascites, subcutaneous edema, liver necrosis and death was described ia chickens following accidental administration of toxic fats in their feed. COMPLAINT: The toxic material was called Plaintiffs' Claims -page 16- �"chick edema factor" until it was finally identified by x-ray crystallography as a PCDD. (k) TCDD is metabolically table in mammalian systems, and alters some cellular compenents (particularly endoplasmic reticulum), especially in liver and kidney cells. TCDD and other PCDDs as well as the PCDFs and related compounds such as the polychlorinated biphenyls (PCBs) are known to stimulate a number of enzyme systems, including those responsible for detoxifying foreign compounds metabolizing steroid hormones and other ingested or endogenous lipid-like hormones, and converting several organic compounds to forms that can be eliminated by the kidney. TCDD is known to stimulate the enzymes, delta aminolevulinic acid synthetase (ALAS) and aryl hydrocarbon hydroxylase (AHH). The ability of compounds to stimulate ALAS and AHH correlated closely with their lethal, teratrogenic and acnegenic potency* more potent that the known TCDD is carcinogen 3-methylcholanthrene in stimulating AHH activity in certain animals. •COMPLAINT: Plaintiffs' Claims -page 17- �(1) The synergistic action of TCDD with polycyclic hydrocarbon such as 3-methycholanthrene (MC) has induced cancer in 'different animal strains in direct proportion to the degree of stimulation of the induced (m) enzymes. Exposure to TCDD causes an increased incidence of neoplasms species. in certain animal Upon information and belief, the carcinogenic and tumorigeni? effects of TCDD include: neoplastic nodules of the liver, Cholangiocarcinomasr hepatocellular carcinoma; carcinoma of the ear duct; carcinoma of the kidney; adenoma of the adrenal cortex; metasticizing retroperitoneal histiocytomas; hyperplasia of the epithelium of the lung; sguamous cell carcinoma of the lung; squamous cell carcinoma of the hard palate/nasal turbinates; squamous cell carcinoma of the tongue; altered lymphopoiesis; epithelial chagnes including hypertrophy, hyperplasia and metaplasia. (n) Among the other toxic effects of TCDD on a number of different animal species are: alopecia; hepatic cell necrosis, cirrhosis; hepatic porphyrin accumulation; hypoplasia of lymphoid tissues (with particular involvement of the cortical cells of the thymus resulting in suppression of cell-mediated immunity and reduction of host defenses); COMPLAINT: Plaintiffs' Claims -page 18- �hyperplasia of the lymph tissue and bone marrow; hematological changes including lymphopenia, hypoproteineraia, and increased susceptibility to infection concomitant with the supression of cell-mediated immunity; intestinal hemorrhage; adrenal hemorrhage; hypoplasia of bone marrow and lymph nodes; cellular effects including hypertrophy, hyperplasia and metaplasia in the bronchial tree, bile ducts, pancreatic ducts, salivary-gland ducts and palpebral conjunctivae; gastric hyperplasia and ulceration, hypertrophic gastritis; renal pelvis hyperplasia; ureter and urinary bladder hyperplasia; general debilitation and wasting is associated with exposure to lethal dose of polychlorinated dibenzo-p-dioxins (PCDDS) or polychlorinated dibenzo furans (PCDFS) in a number of animal species which exhibit a chronic and progressive weight loss, parallel mobilization of peripheral fat, increased serum triglyceride levels, and development of fatty liver; Testicular atrophyr necrosis and abnormal spermatocyte development. (o) The erabryotoxic, fetotoxic and teratogenic effects of TCDD in certain animal species include: cleft palate; hemorrhage; edema; fetal kidney abnormalities which may progress into hydronephrosis during the postnatal period; hydrocephalus; lack of eyelid formation (open eye); clubfoot; increased perinatal mortality; interference with lymphatic system development; skeletal abnormalities; increased fetal enzyme activity. (p) .The mutagenic and cytogenic effects of TCDD include: COMPLAINT: Plaintiffs1 Claims -page 19- �increased incidence of reverse mutations in Excherichia coli and Salmonella typhimurium; increased reversion frequency to streptomycin indepdence in Escherichia coli sd-4; induction of frameshift mutations in Salmonella typhimurium, strain TA1532, but not base substitutions in strain TA1530; inhibition of mitosis and chromosomal abnormalities (dicentric bridges and chromatin .fusion with formation of multinuclei or a single large nucleus) were observed in a endosperm cells of the African blood lily (Baemanthus Katherinae Baker). (q) TCDD may form a physical complex with DNA in (r) An outbreak 6f chloracne affected workers cell nuclei. at the 2 4 5 - factory of the Dow Chemical , , , T Company in Midland, Michigan in 1 6 . 9 4 (s) During 1964, certain employees working in a 2,4-D and 2,4,5,-T producing factory of Diamond Alkali Co., a division or subsidiary of Diamond Shamrock Corporation, at Newark, New Jersey, manifested features of cbloracne, increased excretion of uroporphyrins, and elevated serum enzyme levels, hirsutismr hyperpigmentation, increased skin fragility and vesicobullous eruptions on exposed areas of skin* Five years later, in 1969, after the level of TCDD contamination in the 2,4,5,-T had been COMPLAINT: Plaintiffs' Claims -page 20- �reduced, re-examination of the employees of the same factory still revealed evidence ' f chloracne, o hyperpigmentation, facial hypertrichosis, gastrointestinal symptoms (including nausea, vomiting, diarrhea, abdominal pains, blood in the stools), lower extremity weakness, headache and/or decreased auditory acuity, elevated serum cholesterol, elevated serum enzymes, and diminished white blood cell counts. The severity of chloracne was associated with the degree of hypomania as measured on the Minnesota Hultiphasic Personality Inventory hypomanic scale. (t) In 1976, in Seveso, Italy, a chemical reactor involved in the industrial preparation of 2,4,5-T exploded, sending a plume of chemicals,, including TCDD, 30 to 50 meters above the factory. The vapor cooled and came down over an area about 2 kilometers long and 700 meters wide. -According to the calculations of Givaudan, 1CMESA, a Swiss company which owned the plant, between 650 GRAMS and 1700 GRAMS of TCDD were released* For the year prior to the Sevesco accident, only four ( ) 4 cases of congenital malformations were recorded for 3,902 births, for a rate of malformations of 0.12% of live births in the region contaminated with COMPLAINT: Plaintiffs' Claims -page 21- �said toxic synthetic organic chemicals. Upon information and belief, those four (4) cases of congenital malformations included two (2) cases of Down's Syndrome of hypospadias. (mongolism) and two (2) cases For the year after the release of TCDD, 38 malformations of 1.36% of live births occurred in the region. Upon information and belief/ those 38 malformations included two (2) cases of Down's Syndrome and two ( ) cases of hypospadias, together 2 with 34 polymorphic malformations including meningocele (a congenital hernia in which the meninges, the membranes covering the brain and spinal cord, protrude through an opening of the skull or spinal column); pulmonary aplasia (failure of the lungs to properly develop); atresia of the ear (pathologic closing or congenital absence of the ear opening); cardiopathies; ectopic bladder; coeiosomy; abdominal malformations; and anomalies of the skeletal members. 28» Upon information and belief, the defendant federal agencies have conspired to ignore, have continued to ignore, and unless restrained by this Court, will continue to ignore 38 U S C Sections 5001, . . . 38 C.F.R. Sections 610 et.seq. and 3.102, 3.303 et.seq.-which entitle veterans of the Vietnam War to "timely and complete care" COMPLAINT: Plaintiffs1 Claims -page 22- �for any disability which can be considered "within the range of.probability" to be service related and that this indifference to the intent behind, the plain language of, Congressional statutes and the Veteran's Administration's own regulations constitutes a violation of plaintiff's rights assured under the Constitution of the United 29. States. Upon information and belief, the defendant federal agencies have conspired to withhold from public knowledge information about the deleterious effects of toxic synthetic organic chemicals such as TCDD or "Dioxin" to which combat veterans of the armed forces of the United States, r Australia and Mew Zealand were exposed during their military service in Viet Nam and southeast Asia. In particular, that the Public Health Service of the United- Stated with DIAMOND ALKALI COMPANY and/or successor conspired in interest DIAMOND SHAMROCK CORPORATION, to withhold from the Department of Defense and the armed forces of the United States responsible for the decision to deploy phenoxy herbicides as chemical defoliants in Viet Nam, the full extent of the toxic effects upon workers and community residents associated with the Dioxin contamination of phenoxy herbicides produced at the DIAMOND ALKALI Plant in Newark, New Jerseyr and that, the Food and Drug Administration conspired with a number of war contractors to withhold information about the t •COMPLAINT: Plaintiffs' Claims -page 23- �toxic effects of Dioxin contaminated products since sometime in the early 1960s. 30. That the defendant federal agencies have conspired to deprive the plaintiff veterans of the proper medical examinations and treatment and/or the medical evidence necessary to enable them to evaluate the risk of neoplastic disease, genetic damager and other deleterious effects of their exposure to toxic chemicals such as TCDD^ ** 31. That the defendant federal agencies have conspired to pursue a course of conduct of misleading the plaintiff veterans, their families, and all the others so unfortunate as to be similarly situated and afflicted, by misrepresenting the extent of the risk of neoplastic disease, genetic damage, and other deleterious effects of their exposure to toxic synthetic organic chemicals such as ' TCDD* 32. That the defendant federal agencies have conspired to disseminate false and misleading information to the plaintiff veterans and their families and to most of the more than two million Viet Mam combat veterans. misleading information Such is calculated to persuade the veterans and their families not to assert their rights or pursue legal remedies against the war contractors COMPLAINT: their in the first Plaintiffs' Claims -page 24- �instance, and in the case of medical malpractice, hospital negligence, and other wrongs the defendant federal agencies against the United States. Specifically, individual bureau- crats employed by the defendant federal agencies by abuse of their positions of authority within federal bureaucracy have actively discouraged the plaintiff veterans and their families from: • (a) Filing claims for the veterans benefits mandated under the several provision of Title 38, United States Code, where such claims are based on exposure to toxic synthetic organic chemicals such as TCDD during military service in Vietnam and. Southeast Asia; (b) Act Filing claims under Federal Tort Claims (FTCA) of 1946 (28 U S C 2671, et.seq.) for . . . .•» actionable torts committed since the plaintiff veterans completed their military service, by the individual defendant bureaucrats and, upon information and belief, certain of their predecessors and successors in interest, in particular the breach of certain duties owed to the plaintiff veterans and their families. Specifically, these duties include: COMPLAINT: Plaintiffs1 Claims -page 25- �(1) The duty to provide information to, and warn, the plaintiff veterans and 'their families of the risks associated with exposure to toxic synthetic organic chemicals such as TCDD. ( ) 2 The duty to provide medical information, advice, care and treatment, including but not limited to genetic counseling to the plaintiff veterans and their families; (c) Filing suit for compensatory, general, and punitive damages against the individual defendant bureaucrats responsible for the violation of the constitutional rights of the plaintiff veterans and their families; and filing suit against the individual agents, employees and officials of the Veterans Administration responsible for the comission of malpractice on the individual plaintiff veterans and their families. 33. That the defendant federal agencies knew or should have known that their reprehensible neglect of the health, safety and welfare of the plaintiff Vietnam combat veterans and their families violated the spirit and intent of the Congressional mandate of Title 38 of the United States Code of related legislation, and evidenced intent on the part of the individual defendant bureaucrats to deprive the plaintiff veterans, their families, and all those so unfortunate as to be similarly situated of their constitutional rights. COMPLAINT: Plaintiffs' Claims -page 26- �34. That the defendant, VETERANS ADMINISTRATION has conspired to overmedicate Vietnam combat veterans with psychotropic drugs; ignore symptoms of serious, permanent and irreparable damage to the brain, central and peripheral nervous system; ignore clearly discernible symptoms of burgeoning neoplastic disease; and actively neglect.the evidence of significant genetic damage among those Viet Nam combat veterans who have not otherwise been rendered sterile. 35* That by reason of the conduct of the defendant federal agencies, the plaintiffs have suffered serious, permanent and irreparable damage, and, unless this Court grants the equitable relief sought by said plaintiffs will suffer further serious, permanent, and irreparable damage including, but not limited to, brain injury and death of the plaintiff veterans as a result of the misrepresentation of, and overmedication with, psychotropic drugs. 36. Upon information and belief, treatment received by significant numbers of plaintiff veterans incarcerated Veterans Administration in Hospitals or compelled to avail•them- selves of Veterans Administration outpatient services is grossly negligent, less than competent, deliberately indifferent, or nonexistent, as a result of both of policy decisions by Veterans Administration officials and the COMPLAINT: Plaintiffs' Claims -page 27- �actions of subordinates encouraged by the official policy of deliberate indifference. Such systemic and institution- alized malpractice has resulted in a de facto deprivation of the vested right of the plaintiff veterans to "timely and complete care". 37. Under 38 U S C , Section 5001 and 38 U.S.C., . . . Sections 610, e£ seer., the plaintiff veterans of the Viet Nam War are entitled to have "timely and complete care" provided by Veterans Administration hospitals for their servicerelated injuries. 38. Upon information and belief, arbitrary, capricious and inconsistent decisions of Veterans Administration personnel as to what constitutes a service-related injury have resulted in de> facto deprivation of the vested right of the veterans to timely and complete medical care. 3 . 9 That the defendants have denied the plaintiff veterans access to a simple, certain and uniform system of medical care in violation of their rights. 40. Upon information and belief, medical treatment in Veterans Administration Hospitals is neither certain nor uniform, with an arbitrary and capricious standard under which veterans with identical injuries.are afforded treatment in some cases and denied treatment in others on the grounds COMPLAINT: Plaintiffs' Claims -page 28- �that the injuries are or are note service-related/ depending upon the bureaucratic whim of Veterans Administration personnel in various hospitals throughout the country. 41. Upon information and belief, systemwide and c institutionalized medical malpractice by the Veterans Administration has resulted in de facto deprivation of the vested rights of the plaintiff veternas and their families. 42. That the Administrator of the Veterans Admin- istration has abused his discretion. 43. Upon information and belief, certain veterans r are so physically, emotionally and economically helpless as to be de facto prisoners of the Veterans Administration Hospital System in that: . (a) the nature of their mental or physical injuries and/or their treatment renders them incapable of removing themselves from Veterans Administration Hospital facilities. (b) because of the disability resulting from their injuries they are financially unable to avail themselves of other health care and in order to survive must remain "chained" to the Veterans Administration Hospital system. 44. That the Eighth Amendment rights of these de_ facto prisoners are violated by the wanton callousness displayed by the Veterans Administration policies and personnel, AMENDED COMPLAINT: Plaintiffs' Claims -page 29- �45. That upon information and belief, veterans claiming symptoms attributable to service in Viet Nam and Southeast Asia are given cursory physical examinations, minimal blood work and urinalysis, and asked to complete a prefabricated history dealing with "exposure" to herbicides in Viet Nam. 46. Upon information and belief, significant numbers of .Veterans Administration personnel are negligently prescribing psychotropic drugs without justification and without adequate monitoring of the effects of the administration of such drugs. r 47. Upon information and belief, Veterans Adminis- tration personnel are prescribing psychotropic drugs to Vietnam combat veterans in the absence of complete psychiatric and psychological evaluation by licensed psychiatrists and psychologists. 48. Upon information and belief, the veterans medicated with these psychotropic drugs are not fully informed of the effects of such, drugs, or informed of the availability of alternate forms of treatment. 49. Upon information and belief, Veterans Adminis- tration personnel are performing "operant conditioning* therapy on Vietnam combat veterans without first obtaining COMPLAINT: Plaintiffs' Claims -page 30- �informed consent based on a complete disclosure of the nature of the treatment and its effects and the alternative therapeutic modalities available. 50 ,. Upon information and belief, Veterans Adminis- tration personnel are conducting various forms of medical experimentation upon Vietnam combat veterans without obtaining informed consent based upon a full and complete disclosure of the nature of the experiment, its purpose and its effect upon the veteran. DEMAND FOR JUDGMENT AND PRAYER FOR RELIEF; WHEREFORE, the plaintiff veterans and their families, individually- and on behalf of all those so unfortunate as to be similarly situated, seek judgment: APPOINTING and designating, subject to the continuing jurisdiction and direction of this Court, an appropriate legal representative of the plaintiffs for the purpose of representing said plaintiffs as a class in claims and proceedings involving the Veterans Administration and other federal agencies. i Demand for Judgment/Prayer for Relief -page 31- �DECLARING that the very essence of civil liberty consists in the right of every individual Viet Nam combat veteran to claim the protection of the laws, whenever he receives an injury. DECLARING that the right of every individual Vietnam combat veteran to a remedy for the violation of a vested legal right, is a civil liberty and vested right retained by those veterans as citizens of the United States under the United States Constitution. DECLARING that any attempt by the defendant federal agencies to deprive Viet Nam combat veterans of their right to compel the federal government to provide the services to which they *» are entitled violates the rights of the plaintiff veterans assured by the Fifth and Ninth Amendments to the United States Constitution. ~ DECLARING that the tortious conduct of the defendant federal agencies violates the civil and human rights of the plaintiffs. DECLARING that the tortious conduct of the defendant federal agencies violates the rights Demand for Judgment/Prayer for Relief -page 32- �of the plaintiffs under the Fifth, Eighth and Ninth Amendments to the United States Constitution. DECLARING that the continued prescription of psychotropic drugs to the plaintiff veterans without justification and without adequate monitoring of the effects of the administration of such drugs is so wantonly callous as to violate the fundamental human and civil rights of the plaintiff veterans and constitute a form of "cruel and unusual punishment" prohibited by the Eighth Amendment of the United States Constitution. DECLARING that the defendant federal agencies by reason of their political stature and administrative resources, are Trustees of the health, safety, and welfare of the plaintiff veterans and . their families to the extent that such individual defendant bureaucrats, actually had, or now have, or with, the exercise of reasonable concern for the health, safety and welfare of the plaintiff veterans and their families, should have had, or now should have, knowledge of the dangerous properties and toxic characteristics of certain synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin ( C D or "Dioxin") to which the plaintiff T D Demand for Judgment/Prayer for Relief -page 33- �veterans were exposed during their service in the Armed Forces of the United States, Australia and - New Zealand. DECLARING that continued refusal to warn the plaintiff veterans and their families of the risks associated with exposure to toxic synthetic organic chemicals such as TCDD or "Dioxin" is a breach of the fiduciary duty and obligation owed those veterans and their families by the defendant federal agencies. DECLARING that the misrepresentations of fact and law, and the tortious conduct of the defendant federal agencies estop said defendant federal agencies or any other agency of government from asserting any statute or limitations as a bar to the claim of the plaintiff veterans and their families and all others so unfortunate as to the similarly situated for benefits available under Title 38, of the United States Code and associated federal legislation, including but not limited to the Federal Tort Claims Act of 1946, 80 Stat.. 306, 28 U S C 2671, et seq. . . . DECLARING that this complaint constitutes sufficient notice of claim to the defendant federal Demand for Judgment/Prayer for Relief -page 34- �agencies under the Federal Tort Claims Act of 1946, 80 Stat. 306, 28 U S C 2671, et seq. . . . on behalf of each member of the class of plaintiffs sought to be represented until such time as this Court shall direct the particular filing of individual claims. DIRECTING the defendant federal agencies in particulars at the Hospitals maintained by or under the supervision of the Veterans Administration of the United States to take a full and complete history adequate to form the data base for epidemiological analysis from each veteran claiming afflictions invovling exposure to TCDD. DIRECTING the Veterans Administration to provide a simple, certain and uniform system to timely and complete medical care as mandated by Title 38 United States Code, Section 5001. DIRECTING the Veterans Administration of the United States, the Department of Defense, or such other responsible Federal Officials as this Court may direct, to Demand for Judgment/Prayer for Relief -page 35- �(a) NOTIFY all Viet Nam combat veterans and their children, that they are at risk of . genetic and somatic damage, including neoplastic disease, and that there is an increased probability of birth defects among their children; (b) NOTIFY each Viet Nam combat veteran of theirrright to receive medical care, treatment and assistance as a result of exposure to toxic synthetic organic chemicals while serving in Vietnam and the Southeast Asia Theater of Operations; and ( ) NOTIFY each Viet Nam combat veteran, or the c widow, children and parents of each Vietnam combat veteran who has died since returning to the United States from Southeast Asia of the existence of this action and of the chemical company war contractors' allegation that a possible claim exists against the government y (d) NOTIFY each Viet Nam combat veteranr or the widow, childrenr and parents of each Vietnam combat veteran who has died since returning to the United States from Southeast Asia,, of their right to file claims for damage, injury- or death against certain federal officials and federal agencies under the Federal Tort Claims Act of 1946, 80 Stat. 306, 28 O.S.C. 2671, et seq.; 28 O S C 2401(b), and othc . . . statutes by reason of the failure of those federal officials and federal agencies to warn them of the risk associated with their exposure to toxic synthetic organic chemicals while serving in Vietnam and the Southewast Asia Theater of Operations? ( ) PROVIDE each Viet Nam combat veteran, and e the widow, children, and parents of each Vietnam combat veteran who has died since returning to the United States from Southeast Asia, with sufficient detailed instructions for filing of such claims as will enable each such potential claimant to make timely claim against the government* IMPOSING upon the defendant federal agencies, » -• as Trustees of the health, safety and welfare of Demand for Judgment/Prayer for Relief -page 36- �the plaintiff veterans and their families, a non-delegable duty to disclose to the veterans and their families whatever information they now have, or with the exercise of reasonable concern for the health, safety and welfare of the plaintiff veterans and their families should have, concerning the dangerous properties and toxic characteristics of the synthetic organic chemicals to which the plaintiff veterans were exposed during their service in the Armed Forces of the United States. PROHIBITING the defendant federal agencies particularly the staff members at the Hospitals maintained by or under the supervision of the Veterans Administration of the United States from continuing to prescribe psychotropic drugs to Viet Nam combat veterans in the absence of a complete physiological, psychiatric, and psychological evaluation by physicians, psychiatrists, and psychologists all of whom are duly licensed to practice such professions in the State in which • they hospital facility is located. PROHIBITING the defendant federal agencies particularly the staff members at the Hospitals maintained by or under the supervision of the Demand for Judgment/Prayer for Relief -page 37- �Veterans Administration of the United States from continuing to prescribe psychotropic drugs to Viet Nam combat veterans unless and until the veterans have been fully informed of the effects of such drugs, and the opportunity for alternative forms of treatment have been presented to the veteran and his family. PROHIBITING the defendant federal agencies particularly the staff members at the Hospitals maintained by or under the supervision of the Veterans Administration of the United States from continuing to prescribe psychotropic drugs to Viet Nam combat veterans unless and until adequate physiological and psychological monitoring of the level organic"cehmicals such as TCDD during service in Viet Nam and Southeast Asia; COMPELLING the Veterans Administration to conduct complete dermatological examinations; chemical analyses of blood, urine, and other physiological fluids; and such other tests as medical experts would find medically appropriate on those Vietnam combat veterans who present themselves to a Veterans Administration Hospital with symptoms of skin lesions since service in Southeast Asia, and/or Demand for Judgment/Prayer for Relief -page 38- �suggestions of a change in tolerance to alcohol, and/or motor or sensory impairment of central or peripheral nervous system, and/or significant change in bladder or bowel function since returning from Southeast Asia, and/or symptoms of accelerating degenerative processes ("early aging"), and/or complaints of infertility, and/or a history of birth defects among their children. COMPELLING the defendant federal agencies to provide the appropriate representatives of the plaintiffs with, a summary of all the diagnoses of neoplastic disease made and/or confirmed among Viet Nam combat veterans, as well a? veterans under the age off 40 years at the present time and who served in the United States Armed Forces from 1962 through 1972. PROHIBITING the individual defendant bureaucrats, their successors in interest, and the staff members at the Hospitals maintained by.or under the supervisions of the Veterans Administration of the United States from conducting any form of "operant conditioning" therapy on any Vietnam combat veteran without a full and complete disclosure of the nature of the treatment and its effects, and the offer of alternative therapeutic modalities. i Demand for Judgment/Prayer for Relief -page 39- . �PROHIBITING the defendant federal agencies particularly the staff members at the Hospitals "maintained by or under the supervision of the Veterans Administration of the United States and their successors in interest from further disseminating misleading information to the plaintiff veterans and their families about their rights and the legal remedies available to such plaintiff veterans. RESTRAINING the defendant federal agencies and the staff of the Veterans Administration and their successors in interest from any further distribution of a certain document entitled,. "Worried About Agent Orange?", and from the use of the publication Vanguard or films or videotapes, prepared at public expense/ as the means of further dissemination of misleading information about the risk of deleterious effects associated with exposure to toxic syntetic organic chemicals such as TCDD in Viet Nam and Southeast Asia. ALL TOGETHER with such other and further relief as ». to this Court shall seem just and proper under the circumstances, including the costs, disbursements, and reasonable attorney's 4 Demand for Judgment/Prayer for Relief -page 40- �ATTORNEY'S VERIFICATION STATE OF NEW YORK COUNTY OF SUFFOLK ) )ss.i ) VICTOR JOHN YANNACONE, jr., duly affirming under the penalties of perjury on this 17th day of May, 1984, states that he has prepared the foregoing Complaint on behalf of the plaintiff veterans and their families, and knows the contents thereof, that the same are true to his own knowledge except as to those portions therein stated to be alleged upon information and belief and as tp/'those portions he believes them to be true, YANNACONE, jr Demand for Judgment/Prayer for Relief -page 42- � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Alvin L. Young Collection on Agent Orange Description An account of the resource <p style="margin-top: -1em; line-height: 1.2em;">The Alvin L. Young Collection on Agent Orange comprises 120 linear feet and spans the late 1800s to 2005; however, the bulk of the coverage is from the 1960s to the 1980s and there are many undated items. The collection was donated to Special Collections of the National Agricultural Library in 1985 by Dr. Alvin L. Young (1942- ). Dr. Young developed the collection as he conducted extensive research on the military defoliant Agent Orange. The collection is in good condition and includes letters, memoranda, books, reports, press releases, journal and newspaper clippings, field logs and notebooks, newsletters, maps, booklets and pamphlets, photographs, memorabilia, and audiotapes of an interview with Dr. Young.</p> <p>For more about this collection, <a href="/exhibits/speccoll/exhibits/show/alvin-l--young-collection-on-a">view the Agent Orange Exhibit.</a></p> Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Box The box containing the original item. 194 Folder The folder containing the original item. 5480 Series The series number of the original item. Series VIII Subseries I Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Description An account of the resource <strong>Corporate Author: </strong>United States District Court, Eastern District of New York Date A point or period of time associated with an event in the lifecycle of the resource 1984-05-01 Title A name given to the resource Amended Complaint, Maureen Ryan, Individually, and as the mother of the infants Kerry Ryan and Michael Ryan, and Michael Ryan, and others similarly situated, Plaintiffs, against the Public Health Service, The Food and Drug Administration, The Veterans Adm ao_seriesVIII https://www.nal.usda.gov/exhibits/speccoll/files/original/41f10218f85c4830371ffa5d4991aa62.pdf c0d9b37a5c14c81544b188825648ff65 PDF Text Text Item D Number 05254 Author D Yannacone, Victor John, Jr. MotScannBfl Corporate Author Report/Article TitlB Amended Verified Complaint, January 5, 1979 Journal/Book Title Year 1979 Month/Day January 5 Color D Number of Images 104 DeSCriptOU Notes Complaint Against Dow Chemical Company, Hercules Inc., and Northwest industries regarding toxic effects of phenoxy herbicides Friday, March 01, 2002 Page 5254 of 5263 �AMENDED VERIFIED COMPLAINT Victor John Yannacone, jr, YANNACONE & YANNACONE Professional Corporation of counsel to Plaintiffs' Attorneys Office & P.O. Address Post Office Drawer §109 Patchogue, New York 11772 (area code 516) 654-2299 O'HAGAN REILLY & GORMAN Plaintiffs' Attorneys Office & P.O. Address 444 Main Street Islip, New York 11751 (area code 516) 581-8687 �INDEX Venue • Jurisdiction The Plaintiffs Other Interested Parties Class Action Phenoxy Herbicides \\1. 1(2. - 1(6. 1(7. - 1(15. if 16*. - 1(20. • 1(21. - 1(29. 1(30. "- 1(59. page 4 5 7 12 14 17 p-dioxin (TCDD or "dioxin") 1(60. - 1(86. Defendant THE DOW CHEMICAL COMPANY Corporate History 1(87. - 11114. Subsidiaries 1(115. Joint Ventures 11116. - K128. Business & Products 11129. - 11132. Principal Properties 11133. - 1(137. 24 2,3,7,8-tetrachloro dibenzo Defendant HERCULES INCORPORATED. Corporate History 11138. Subsidiaries . 1(172. Joint Ventures 1(174. Business & Products 1(184. Principal Properties 1(211. - 30 34 36 39 45 . . " 1(171. 1(173. 1(183. 1(210. 1(218. 48 53 54 57 63 • Defendant NORTHWEST INDUSTRIES, INC. Corporate History Business & Products Operations Plants & Property 1(219. 1(225. 1(266. 1(283. - 1(224. 1(265. 1(282. 1(286. Plaintiffs Complain 1(287. - 1(299. Damages 1(300. - 1(313. Declaratory Judgment Claims 1(314. - 1(321. Equitable Jurisdiction 1(322. - 1(327. Demand for Judgment Declaring Rights • -' Imposing a Trust _ Directing Disclosure • . Restraining Defendants Establishing Reserves Appointing a Trustee Reimbursing United States ' Compensating Plaintiffs Reimbursing Utilities bthor and Further Appropriate Relief 66 .67 75 78 . ' ^ . -page 3-. • Amended Verified-Complaint - . 5 Jnnnarv . 9 83 87 9'2 95 6 99 99 99 100 102 102 - 103 103 104 �AMENDED VERIFIED COMPLAINT The plaintiff, now deceased, by his Legal Representative, individually and on behalf of all those so unfortunate as to have been and now to be similarly situated at risk, not only during this generation but during those generations yet to come, from the toxic effects of exposure to phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") and manu- . factured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC., jointly and severally, individually and collectively, alone or in concert with others, in wanton and reckless disregard of the public health, safety and welfare, sets forth and alleges the following by his attorneys: Venue 1. The venue of this action is determined under Title 28 United States Code S§1391(a), 1391(b), 1391(c). -page 4Amonded Verified Complaint 5 January 1S79 �Jurisdiction 2. Jurisdiction of this Court is invoked under Title 28 United States Code §1331(a). "The district courts shall have'original jurisdiction of all civil actions wherein the matter in controversy exceeds the sum or value of $10,000 exclusive of interest and costs, and arises under the Constitution, laws or treaties of the United States." because the amount in controversy herein far exceeds the sum or value of $10,000 exclusive of interest and costs, and this cause of action arises under the Constitution, laws-or treaties of the United States, as more fully and at length hereinafter set forth and alleged. 3. This action arises under Article VI, section 2, of the Constitution of the United States and involves the declaration and interpretation of the rights of the now deceased plaintiff individually, and all those others so unfortunate as to have been or now to be'similarly situated, to be free of risk from the toxic effects of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES,. INC. although contaminated with the synthetic organic chemical 2,3,7,8-tetrachloro dibenzo pdioxin (TCDD or "dioxin"), which rights are retained by said plaintiffs under the Ninth Amendment of the Constitution of the United States "The enumeration in the Constitution of certain rights, shall not be construed to deny or disparage others retained by the people." and protected under the "due process" clause of the Fifth Amendment to the Constitution of the United States, and under the "due process" and "equal protection" clauses of the Fourteenth Amendment of the Constitution of the United States. . 4. Jurisdiction of this Court is also invoked under Title 28 -page 5Amended Verified Complaint 5 January 19.79 " . • ' �United States Code §1332. -5. Jurisdiction of this Court is also invoked under Title 42 United States Code §1983: "Every person who, under color of any statute, ordinance, regulation, custom, or usage, of any State or Territory, subjects or causes to be subjected, any citizen of the United States or other person within the jurisdiction thereof to the deprivation of any rights, privileges or immunities secured by the Constitution and law, shall be liable to the party injured in an action at law, suit in equity, • or other proper proceeding for redress." 6. Jurisdiction of this Court is also invoked under Title 28 United States Code §1337, "The district courts shall have original jurisdiction of any civil action or proceeding arising under any Act of Congress regulating commerce or protecting trade and commerce against restraints and monopolies." since this proceeding may involve consideration of Title 15 United States Code §§15, 26, 22: "Any person who shall be injured in his •• business or property by reason of anything forbidden in the antitrust laws may sue therefor in any district court of the United States in the district in which the defendant resides or is found or has an agent, without respect to the amount in controversy, and shall recover threefold the damages by him sustained, and the cost of suit, including reasonable attorney's fees." "Any -person . . . shall be entitled to sue for and have injunctive relief, in any court of the United States having jurisdiction over the parties, against threatened loss or damage by a violation of the antitrust laws. .. . . when and under the same conditions and principles as injunctive.relief against threatened 'conduct that will cause loss or damage is granted by courts of equity under the rules governing such proceedings. ..." "Any suit, action, or proceeding under the • " • -page 6Amcneled Verified Complaint 5 January 1979* �antitrust laws against a corporation may be brought not only in the judicial district whereof it is an inhabitant, but also in any district wherein it may be found or transacts business; and all process in such cases may be served in the district of which it is an inhabitant." The statutes relevant to this proceeding are Title 15 United States Code §§1, 22: "Every . . . combination in the form of trust or otherwise, or conspiracy in restraint o.f trade or commerce among the several States or with foreign nations, is declared to be illegal. . . ." and Title 15 United States Code §§45(a)(l), 52, 55(a)(l), 55(c): "* * *(a) It shall be unlawful for any person, partnership, or corporation to disseminate, or cause to.be disseminated, any false advertisement. "(b) The dissemination or the causing to be disseminated of any false advertisement within the provisions of subsection (a) of • this section shall be an unfair or deceptive trade act or practice in commerce within the meaning of §45 of this Title." "The term false advertisement means any advertisement, other than labeling, which is misleading in a material respect; and in determining whether any advertisement is misleading, there shall be taken into account (among other things) not only representations made or suggested by statement, word, design, device, sound, or any combination thereof, but also the extent to which the advertisement fails to reveal facts material in the light of such representations or material with respect to consequences which may result from, the use of the commodity to which the advertisement relates under the conditions presentee in said advertisement or under such conditions as are customary or usual. . ." and Title 15 'United States Code §§1125(a), 1121: "Any person who shall affix, apply, or annex, or use in connection with, any goods or services, or any container or containers for goods . . . any false description or rcpresen-pagc 7Amended Verified Complaint 5 January 1979 . t - �tation, including words or other symbols tending 'falsely to describe or represent the same, and shall cause such goods or services to enter into commerce, and any person who .shall with knowledge of the falsity of such description or representation cause or procure the same to be transported or used in commerce or deliver the same to any carrier to be transported or used, shall be liable to a civil action ... by any person who believes that . he is likely to be damaged by the use of any such false description or representation." "The district courts ... of the United States shall have original jurisdiction . . . of all actions arising under this chapter, without regard to the amount in controversy or to diversity or lack of .diversity of the citizenship of the parties." -page 8Amonded Verified Complaint 5 January 1979 �The Plaintiffs 7. That the deceased plaintiff' PAUL REUTERSHAN was .a member of the armed forces of the United States and was exposed to phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin {TCDD or "dioxin"). 8. Upon information and belief, as many as 4.2 million (4,200,000) members of the armed forces of the United States may have been exposed to phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"). 9. That the deceased plaintiff PAUL REUTERSHAN has made a claim for veterans service benefits as a result of disability attributable' to phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin ,(TCDD'or "dioxin"). 10. Upon information and belief, other members 'of the armed forces of the United States have made similar claims for veterans service benefits as a result of disability attributable to phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the -page 9Amended Verified Complaint 5 January 1979 �corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"). 11. Upon information and belief, the phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) which were contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") and to which the deceased plaintiff and other members of the armed forces of the United States were,exposed were manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. with knowledge of such contamination. 12. That the plaintiff PAUL REUTER3HAN suffered from a malignant condition, and believed that his malignant condition was atributable to his exposure to phenoxy herbicides such as 2,4,5~trichloro phenoxyacetic acid (2,4,5-T) which were contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") as a result of the negligence and carelessness of the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. 13. That the plaintiff PAUL REUTERSHAN eventually died as a result of a malignant condition, attributable to his exposure to phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid : (2,4,5-T) which were contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or ."dioxin") as a result of the negligence and carelessness of the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. -i- . 14. That the contamination of phenoxy herbicides such as 2,4,5trichloro phenoxyacetic acid (2,4,5-T) manuEactured, formulated, -page 10Amended Verified Complaint 5 January 1979 �advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. with the toxic synthetic organic chemical 2,3,7,8tetrachloro dibenzo p-dioxin (TCDD or "dioxin"), may contribute to the illness and eventual death of all those so unfortunate as to have been exposed. • . 15. Upon information and belief, phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") were purchased by public utility companies for use and application •• throughout the United States of America. -page 11Amended Verified Complaint 5 January 1979 �Other Interested Parties 16. Upon information and belief, the United States Department of Defense, in particular the Veterans Administration and the United States Department of Health, Education and Welfare in particular the Social Security Administration, have been or will be responsible for at least part of the cost of medical care and treatment required by and rendered unto the deceased plaintiff as a result of the toxic effects of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES, INC., and NORTHWEST INDUSTRIES, INC. and con-, taminated with toxic synthetic organic chemicals such as 2,3,7,8tetrachloro dibenzo p-dioxin (TCDD or "dioxin")17. Upon information and belief, the United States Department of Defense, in particular the Veterans Administration and the United States Department of Health, Education and Welfare in particular the Social Security Administration, have been, now are, or will eventually be responsible for at least part of the cost of medical care and treatment required by and rendered unto the other plaintiffs so unfortunate as to have been exposed to phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"). 18. Upon information- and belief, the United States Department of Defense, in particular the Veterans Administration and the United States Department of Health., Education and Welfare in particular the Social Security Administration, hav-e made, and-may make payments to or on behalf of the deceased plaintiff PAUL REUTERSHAN and/or his family and all those others so unfortunate as to be similarly entitled to -page 12Amendc-d Verified Complaint r v Januarv 1979 �receive such benefits as compensation or benefits for disability or death attributable to the toxic effects of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"). 19. Upon information and belief, the public utility companies which purchased, used and applied phenoxy herbicides such as 2,4,5trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or."dioxin") may be liable for damages attributable to the use of such contaminated products. 20. Upon information and belief, the stockholders, rate payers, and subscribers of the services provided by the public utility companies which purchased, used and applied phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") may eventually be required to bear the economic burden resulting from the damages attributable to the use of such contaminated products. . -page 13Amended Verified Complaint 5 January 1979 �Class Action 21. This action is brought on behalf of the deceased Plaintiff individually and on behalf of all those so unfortunate as to have been, now to be, or who may be similarly situated at risk, from the toxic effects of exposure to phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised-, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with the synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"). 22. This action is brought on behalf of the deceased Plaintiff individually and on behalf of all those so unfortunate as to have been and now to be similarly situated at risk, from the toxic effects of exposure to phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5~T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with the synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"), all of whom are entitled to the protection of their personal health, safety and welfare from injury and damage attributable to the contamination of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"). 23. This action is brought on behalf of the deceased plaintiff,and on behalf of all those stockholders, rate payers and subscribers of public utility companies which used phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted/ marketed and sold by the corporate defendants THE: DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with toxic synthetic organic. -page 14Atnended V e r i f i e d C o m p l a i n t 5 J a n u a r y 1979 �chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"), all of which stockholders, rate payers and subscribers may eventually be called upon to pay for the economic, loss and damages attributable to the use of such phenoxy herbicides as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"). 24. The claims of the deceased plaintiff are representative of the claims of all the members of the class of all those so unfortunate as to be similarly situated, and the actions of all the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES, INC., and NORTHWEST INDUSTRIES, INC. have substantial effect upon all members of the class exemplified by the deceased plaintiff, thereby making declaratory judgment appropriate and the equitable relief and injunction sought in this action the proper remedy for the class as a whole. 25. -That at the present time, the deceased plaintiff, PAUL REUTERSHAN, is one of an indeterminate number of the approximately 4.2 MILLION American servicemen who are at risk from exposure to phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5~-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") but who are only now beginning to manifest the long-term toxic effects attributable to such exposure. 26. Upon information and belief, it can be reasonably expected that a number of claims similar to those made by the deceased Plaintiff, PAUL REUTERSHAN, will be made for veterans service benefits as a result of disability attributable to and resulting from exposure to phenoxy herbicides such as 2,4,5-trichlcro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES -page 15Araendeci Verified Complain! 5 January 1979 �INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") and that many of those claims will involve disability attributable to or resulting from exposure to such contaminated phenoxy herbicides. 27. Upon information and belief, it can be reasonably expected that a number of claims similar to those made by the deceased Plaintiff, PAUL,REUTERSHAN, will be made for social security disability benefits as a result of disability attributable to and resulting from exposure to phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with toxic synthetic organic chemicals such as 2,3,7,8tetrachloro dibenzo p-dioxin (TCDD or "dioxin") and that many of those claims will involve disability attributable to and resulting from exposure to such contaminated phenoxy herbicides. 28. The members of the class exemplified by the deceased plaintiff PAUL REUTERSHAN are so numerous as to make it impracticable to bring them all before this court, however, there are substantial questions of law and fact common to the class and common equitable relief on behalf of all members of the class is sought. 29. The members of the class exemplified by the deceased plain- tiff PAUL REUTERSHAN are fairly and adequately represented by the legal representatives of said deceased plaintiff and the counsel for said plaintiffs and neither the deceased plaintiff nor the attorneys for the deceased plaintiff ha^e any interests adverse to that of any individual, member of the class of all those who might be entitled to the relief sought herein. -page 16Ainended V e r i f i e d Complaint 5 J a n u a r y 1979 �Phenoxy Herbicides 30. Upon information and belief, the effectiveness of the phenoxy plant growth regulators 2,4-dichloro phenoxyacetic acid (2,4-D) and 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) as "herbicides was determined in mid-1944 field trials at Beltsville and Camp (now Fort) Detrick, Maryland. 31. Although herbicides were not used in tactical military • operations in World War II, upon information and belief, a small program for screening potential herbicides for military use continued after that War. 32. Upon information and belief, the phenoxy herbicides such as 2,4-dichloro phenoxyacetic acid (2,4-D) and 2,4,5-trichloro phenoxyacetic acid (2,4,5-T), in particular their salts and esters were manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DON CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC., individually and collectively, jointly and severally, alone or in concert with others since the mid-1940s. shortly after the end of World War II. 33. Upon information and belief, by 1951, personnel at Fort Detrick had determined that the vegetation control chemicals of choice were mixtures of the butyl esters of the phenoxyacetic acids 2,4-D and 2,4,5-T. 34. Upon information and belief, in 1959, the Crops Division, Fort Detrick, conducted the first large-scale military defoliation effort over four square miles at Fort Drum, New York. - . 35. Upon information arid belief, phenoxy herbicides such as0 2,4,5-trichloro phonoxyacetic acid (2,4,5-T) were .introduced into the armed conflict in Vietnam in 1962. -page 17Amended Verified Complaint 5 January 1979 �36. Upon information and belief, an estimated 107 million pounds of herbicides were aerially disseminated on 6 million acres in South Vietnam from January 1962 through February 1971. Upon information and belief, approximately 94 percent of all herbicides sprayed in Vietnam were 2,4-D or 2,4,5-T. 37. Upon information and belief, approximately 96% of all the 2,4,5-T so used was contained in a formulation designated Herbicide' Orange; the remaining 2,4,5-T was contained, upon information and belief, in formulations designated Herbicide Green, Herbicide Pink, and Herbicide Purple. 38. Upon information and belief, Herbicide Green, Herbicide Pink, and Herbicide Purple contained approximately 40% of the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") which was present as a contaminant of the phenoxy herbicide such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T). Upon information and belief, Herbicide Green, Herbicide Pink, and Herbicide Purple were sprayed as defoliants on less than 90,000 acres from 1962 through 1964. 39. Upon information and belief, the 44 million pounds of 2,4,5-T to which American servicemen such as the deceased plaintiff herein and all those others so unfortunate as to be similarly exposed, contained an estimated 368 pounds of the toxic contaminant, 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or dioxin). 40. Upon information and belief, 90% of all the Herbicide Orange which, upon information and belief, contained 38.3 million pounds of2,4,5-T and 203 pounds of the-toxic 'synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin JTCDD or "dioxin") were used in defoliation operations on 2.9 millipn acres of inland forests and mangrove forests of South Vietnam. 41. Upon information and belief, flight mechanics and crew chiefs -page 18Araended V e r i f i e d C o m p l a i n t 5 January 1979" �were the most likely military personnel exposed.to the phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"). 42. Upon information and belief, in April.1970, the United States Secretary of Agriculture, the United States Secretary of Health, Education^and Welfare, and the United States Secretary of the Interior jointly announced suspension of the registration of liquid formulations of 2,4,5-T for use around the home and recreational areas and for use on lakes, ponds and ditch banks. 43. Upon information and belief, through the efforts of the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC., jointly or severally, individually or collectively, alone or in concert with others, registration was not suspended for use of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) on range and pasture lands, nonagricultural lands, or in weed and brush control programs on communications and highway rights-of~way. 44. Upon information and belief, the effort to suspend registration of 2,4,5-T in 1970 resulted from published studies indicating that 2,4,5-T was a teratogen, however, upon information and belief, subsequent studies revealed that the teratogenic effects attributable to the phenoxy herbicide 2,4,5-T resulted from a toxic contaminant of the 2,4,5-T, which was identified as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"). . 45. Upon information-and belief, the Department of Defense suspended the use of Herbicide Orange shortly after the civilian agencies of government announced suspension of certain uses for the phenoxy herbicide 2,4,5-T. 46. Upon information and belief, in September 1971, the Depa.rt- -page 19Amended Verified Complaint 5 January 1979 • �merit of Defense directed that the Herbicide Orange in South Vietnam be returned to the United States and that the entire stock be disposed of in an environmentally safe and efficient manner. 47. Upon information and belief, more than 1 million gallons of Herbicide Orange containing phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or. "dioxin") were moved from South Vietnam to Johnston Island, Pacific Ocean, for storage (Project PACER IVY) in April 1972. 48. Upon information and belief, during the nine year period from 1961 through 1969, approximately 78 million pounds of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") were used domestically in the United States. 49. Upon information and belief, various techniques for destruction and recovery of the phenoxy herbicides were investigated by the Department of Defense from 1971 to 1974. 'Upon information and belief, destructive techniques studied included soil biodegradation, high temperature incineration, deep well injection, burial in underground nuclear test cavities, sludge burial and microbial reduction; while techniques to recover a. useful product included use, return to manufacturers, fractionation and chlorinolysis. Of the.techniques studied, only high temperature incineration was, upon information and belief, sufficiently developed at that time to warrant further investigation. 50. . - Upon information and belief, in December 1974, the United -page 20' . Amended Verified Complaint 5 January 1979 �States Air Force filed a final Environmental Impact Statement with the President's Council on Environmental Quality, for the disposition of Herbicide Orange by destruction aboard a specially designed inciner- ' ation vessel in a remote area of the Pacific Ocean west of Johnston Island. 51. Upon information and.belief, the United States Environmental Protection Agency held a public meeting in February 1975 to consider an ocean incineration permit application submitted by the United States Air Force in accordance with the Marine Protection, Research and Sanctuaries Act of 1972 as amended, (33 U.S.C. §§1401, 'et seq.) 52. Upon information and belief,.during this meeting, testimony was presented which indicated that techniques for chemically reprocessing the herbicide to remove unacceptable quantities of the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") might have been developed and the Environmental Protection Agency indicated that the option for reprocessing should be further explored as a means of disposition prior to making a decision to destroy the contaminated herbicides by means of incineration. 53. Upon information and belief, the United States Air Force subsequently undertook an investigation into the feasibility of reprocessing Herbicide .Orange and pilot plant studies including selective activated carbon adsorption, were conducted from the Fall of 1975 to'July 1976, however, a feasible and environmentally acceptable method of safely disposing of the TCDD-laden activated carbon was not demonstrated. By February 1971, the United States Air Force, upon information and belief, concluded that the option of reprocessing was not feasible, timely or cost-effective since a technique for tho ultimate disposal of the activated carbon was not currently available or anticipated in the foreseeable future. 54. Upon information and belief, on 9 March 1977, the United States Air Force requested the Environmental Protection Agency to -page 21Amemled Verified Complaint 5 January 1979 �convene the public hearings. As a result of the public hearing held on 7 April 1977, the EPA issued a research permit to the USAF to move stocks of Herbicide Orange from the Naval Construction Battalion Center, Gulfport, Mississippi to a designated site in the North Pacific Ocean for the purpose of at-sea incineration in accordance with the provisions of the Marine Protection, Research and Sanctuaries Act of 1972, as amended. 55. Upon information and belief, the vessel contracted for the at-sea incineration of the contaminated herbicides'was the Dutch-owned M/T Vulcanus, a ship registered in Singapore and previously used in the North Atlantic Ocean and the Gulf of Mexico to destroy chlorinated hydrocarbon wastes. Upon information and belief/ a total of three herbicide loadings were required to incinerate the total stocks of Herbicide Orange: one loading from Gulfport, Mississippi and two loadings from Johnston Island. 56. Upon information and belief, during the summer of 1977 the United States Air Force disposed of 2.22 million gallons of Herbicide Orange .by high temperature incineration at sea. This operation, was, upon information and belief, designated Project PACER HO. 57. Upon information and belief, phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") have been and still are manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. as herbicides for application to a-nd upon rights-ofway, forests, hay, pasture and rangelands, farms, lawns and turf. 58. Upon information and belief, the principal civilian use of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) is for brush control, while the principal civilian use of phenoxy herbicides such as 2,4-dichloro phenoxyacetic acid (2,4-D) is -poqo 22Amendcd V e r i f i e d Comolaint 5 January 1979 �for broadleaf weed control in corn and other grains. 59. Upon information and belief, phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES/ INC. and contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") were used by the Department of Defense and o'ther government departments and agencies, quasi-governmental entities, and public utilities. -page 23Amended Verified Complaint 5 January 1979 �2,3,7,8-tetrachloro dibenzo p_-dioxin (TCDD or "dioxin") 60. Upon information and belief, statements on the fate of the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") in the environment are predicated upon the detection of the compound in environmental substrates. 61. Upon information and belief, prior to 1973 the detection limit for the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") in soils was 100 parts per trillion (0.1 parts per billion). 62. Upon information and belief, prior to 1973 the detection limit for the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") in biological tissue was 50 parts per billion (0.05 parts per million). 63. Upon information and belief, the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") is a nonpolar, chlorinated aromatic molecule, without biologically labile' functional groups and soluble in water only to the extent of approximately 200 mg/liter. _ • • • 64. Upon information and belief, the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-<5ioxin (TCDD or "dioxin") has a lower partition coefficient in lipids than certain environmental toxicants such as 1,1,1-trichloro 2,2-bis p-chloroph-enyl ethane (DDT). 65. Upon information and belief, the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") is capable of bioaccumulation (being taken up by a living organism and retained). 66. • Upon information and belief, the toxic synthetic organic chemical 2 , 3 , 7 , 8-tetcachloro dibenzo p-dioxin (.TCDD or "dioxin") has a -page 24Amcn-Jr-d Verified Complaint '. - 5 January 1979 �half-life in certain soil of at least one year. 67. Upon information and belief, the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") is not biosynthesized by microbial condensation reactions, and, upon information and belief, the toxic synthetic organic chemical 2,3,7,8tetrachloro dibenzo p-dioxin (TCDD or "dioxin") is not readily metabolized by soil microorganisms. 68. Upon information and belief, volatilization of the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") is a major pathway of dissipation for the toxicant from soil and other sites of-application. 69. Upon information and belief, the synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") is one of the most potent low molecular weight toxins and teratogens known. 70. Upon information and belief, 2,3,7,8-tetrachloro dibenzo pdioxin (TCDD or_"dioxin").is the most toxic of the chlorinated dibenzo p-dioxins. • • 71. Upon information and belief, the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") is a "cellular poison." 72. Upon information and belief, among the toxic effects of the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD 'or "dioxin") are to be found: hepatic cell necrosis, cirrhosis; biphasic decline in body weight; increased tissue amyloid; thymic atrophy, decrease in thymus weight, depletion of rlymphocytes in the thymic cortex; alopecia; subcutaneous edema; acne; hemorrhage and atrophy of adrenal zona glomerulosa; -page 25Amendod Verified Complaint . "5 January 1979 �increased levels of 8- and 7-carboxyporphyrins in the liver, liver enlargement, progressive necrotic centilobular liver lesions; gastric hyperplasia and ulceration; impairment of cellular immunity, reduced host defenses; prolongation of aliographic rejection times; depletion of lymphoid organs, suppression of cell mediated immunity in the lymphoid system; hematological" changes including lymphopenia and thrombocytopenia, hemorrhage, anemia; decreased spleen weights; hyperplasia of the lymph tissue and bone marrow; hypocellularity of the bone marrow and lymph nodes; hydropericardium; progressive leukopenia and hypoproteinemia; pancytopenia; and •increased plasma albumin, total protein, iron, urea nitrogen, cholesterol and triglycerides; 73. Upon information and belief, the embryotoxic, fetotoxic and teratogenic effects of the toxic synthetic organic chemical 2,3,7,8tetrachloro dibenzo p-dioxin (TCDD or ."dioxin") include: cleft palate; kidney anomalies, renal hydronephrosis; hydrocephalus; _ lack of eyelid formation (open eye); clubfoot; fetal edema; and fetal hemorrhage, intestinal hemorrhage; 74. Upon information and belief, the carcinogenic and tumorigenic effects of the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") include: " . ' neoplastic nodules and cholangiocarcinomas of the liver, hepatocellular carcinoma of"the liver, carcinoma of the liver; carcinoma of the ear duct; carcinoma of the kidney; rnetasticizing retroperitoneal- histiocytomas; hyperplasia of the epitheliuia of the lung, squamous cell carcinoma of the lung; squamous cell carcinoma of the hard palate/nasal turbinates; squamous cell carcinoma of the tongue; altered lymphopoiesis; epithelial changes including hypertrophy, -page 26nd'v] Verified Complaint :> January 1979 �hyperplasia, and metaplasia; 75. Upon information and belief, many chemically nonreactive carcinogens are enzymatically converted to biologically active carcinogens, and the enzyme aryl hydrocarbon hydroxylase (AHH) has been strongly implicated in this process. 76. Upon information and belief, the synergistic action of 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") with 3-methylcholanthrene (MC) has induced cancer in different animal strains in direct proportion to the degree of elevation of the induced hydroxylase activity and the associated cytochrome content. 77. Upon information and belief, the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"} is more potent than 3-methylcholanthrene (MC) as an inducer of hydroxylase activity in cultured human lymphocytes. 78. Upon information and belief, the toxic synthetic organic chemical 2,3,778-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"} may be orders of magnitude more potent on a molar basis than 3-methylcholanthrene (MC) as aan inducer of hepatic aryl hydrocarbon hydroxylase (AHH) activity. 79. Upon information and belief, the mutagenic and cytogenic effects of the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") include increased incidence of reverse mutuations in Eschorchia coli and Salmonella typhimurium and chromosomal abberations. 80. Upon information and belief, the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p.-dioxin (TCDD or "dioxin") may form a physical complex by "intercalation" in DMA, leading to ' fcameshiLt mutation. 81. npon information and belief, there have been at least 23 -page 27Amendetl Verified Complaint: �industrial incidents involving over 1,100 human beings, generally adult males, in exposure to the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"), however, there are, upon information and belief, no significant scientific long term studies of the individuals so exposed. 82. Upon information and belief, in August 1972 waste oil contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") was sprayed on a horse arena in eastern Missouri and 54 of 57 horses exposed to the arena died of an illness characterized by skin lesions, severe weight loss and hepatotoxicity.. Birds, dogs, cats, insects, and rodents were also found dead in and around the arena. 83. Upon information and belief, a six year old girl exposed to the waste oil. contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") and sprayed on the Eastern Missouri Horse Arena developed epistaxis, gastrointestinal complaints, and severe hemorrhagic cystitis. 84. Upon information and belief, on or about 10 July 1976, in Seveso, Italy, a small' town north of-Milan, a safety pressure disk ruptured sending a plume of chemicals, including the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") 30 to 50 meters above the factor. Upon information and belief, the vapor cooled and came down over a cone shaped-area about 2 kilometers long and 700 meters wide. 85". Upon information and belief, according to the calculations of Givaudan ICMESA, a Swiss company and owner of the plant in Seveso, Italy, between 650 GRAMS and 1700 GRAMS of the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") was released. 86. : Upon information and belief, according to reports from -page 28cl Verified Complaint 5 January 1979 �eveso, Italy, animals, including birds, rabbits, and chickens were beginning to die two to three days after the accident, and although the temporary and acute effects of the incident have passed, there is no estimate yet of the long term effects. -page 29Amended Verified Complaint .5 January 1979" �Defendant THE DOW CHEMICAL COMPANY: Corporate History 87. Upon information and belief, defendant THE DOW CHEMICAL COMPANY was incorporated in Delaware on or about June 11, 1947, and by merger agreement dated June 27, .1947, succeeded to all assets and business of a company bearing the same name which had been incorporated in Michigan on or about May 18, 1897 to build a bleach plant at Midland, Michigan utilizing the waste liquids of Midland Chemical Co. 88. Upon information and belief, on or about March 12, 1900 defendant THE DOW CHEMICAL COMPANY purchased the Dow Process Co. which had been formed in 1895 and upon information and belief, on or about August If 1900 said defendant purchased Midland Chemical Co. which had been formed in 1892. Upon information and belief, both Dow Process Co. and Midland Chemical Co. had been founded by Herbert H. Dow. 89. Upon information and belief, on or about August 2, 1928, Jones Chemical Co., Inc. was incorporated in Louisiana to manufacture iodine from the waste brines of Louisiana oil fields. Upon .information and belief, its operations were moved to California oil fields in June, 1932, and in 1933 Dow acquired the minority interests in this company. Upon information and belief, in 1934 the name of Jones Chemical Co., Inc. was changed to I0-Dow Chemical Co., Inc., which, upon information and belief, was absorbed by the parent on or ab'out March 31, 1939. 90.. Upon information and belief-, on or about' December 31, 1938 defendant THE DOW CHEMICAL COMPANY by paying 86,988 shares of its common stock at the rate of 3/16ths share of its common stock for-each' preferred share of Great Western and one share of its common stock for each share, of Great Western common acquired Great Western ElectroChemical Co. which had been formed in 1916. -page 30~ .--. Amended Verified Complaint 5 January -1979 �j)l. Upon information and belief, on or about June 5, 1942, defendant THE DOW CHEMICAL COMPANY formed Dow Chemical of Canada, Ltd. under Canadian laws. 92. Upon information and belief, on or about February 11, 1943 defendant THE DOW CHEMICAL COMPANY formed the Dow Corning Corp. undar Michigan laws jointly with Corning Glass Works. w' 93. Upon information and belief, on or about December VA,-1951, defendant THE DOW CHEMICAL COMPANY formed Dow Chemical Inter-American, Ltd. and Dow Chemical International Ltd. as wholly-owned subsidiaries under Delaware laws. 94. Upon information and belief, in May, 1952, Dow Chemical International Ltd., a subsidiary of defendant THE DOW CHEMICAL COMPANY formed Asahi-Dow Ltd. jointly with Asahi Chemical Industry Co., Ltd. of Japan. • •. 95. Upon information and belief, on or about August 20, 1954 defendant THE DOW CHEMICAL COMPANY organized Distrene Ltd. jointly with Distillers Co. Ltd. under the laws of Great Britain. 96. Upon information and belief; on or about January 21, 1955, defendant THE DOW CHEMICAL COMPANY formed Nederlandsche Dow Maatschappij N.V., under Netherlands law. 97. Upon information and belief, as of September 1, 1957, for 386,921 shares of its common stock defendant THE DOW"CHEMICAL COMPANY acquired the-assets of Dobeckmun Co. 98. Upon information and belief, in May, :L958 defendant THE DOW CPIEMICAL COMPANY formed Dow Badische Co. for the production of chemicals and synthetic fibers jointly with BASF Overzee N.V. 99. Upon information and belief, in its fiscal year 1959, -page 31Arnended V e r i f i e d C o m p l a i n t -5 J a n u a r y 1979 �defendant THE DOW CHEMICAL COMPANY acquired: Dow Quimica Argentina S.A.; a 93.75% interest in Cosraoclor S.A.I.C.; and a 74% interest in Dow Agrochemicals Ltd., of England (now 100% owned). 100. Upon information and belief, in fiscal year 1959, defendant THE DOW CHEMICAL COMPANY organized Dow Chemical International, S.A.., in Venezuela and Dow Chemical A.G., in Switzerland. 101. Upon information and belief, in September 1959, Dow Chemical A.G. a subsidiary of defendant THE DOW CHEMICAL COMPANY formed Plasttchimie S.A. jointly with Pechiney S.A., a French chemical and electro-metallurgical firm to construct and operate a multimillion dollar plant at Ribecourt %near Paris, France) for manufacture of polystyrene and saran. 102. Upon information and belief, on or about December 30, 1960, for 533,116- shares of its common stock, defendant THE DOW CHEMICAL COMPANY acquired the assets of Allied Laboratories, Inc., which is, upon information and belief, now operated as a division of said defendant. 103. Upon information and belief, in February, 1965, defendant THE DOW CHEMICAL COMPANY acquired a majority interest in Life, S.A. of Quito, Ecuador, a producer of Pharmaceuticals for human and veterinary use. . • • ' • . . 104. Upon information and belief, in 1965 defendant THE DOW CHEMICAL COMPANY formed Dow Banking Corp.'in Zurich, Switzerland as a wholly-owned-subsidiary. 105. Upon information and belief, in 1966 defendant THE DOW CHEMICAL COMPANY sold a 50% interest in Dow-Smith, Inc. to A.O. Smith Corp. 'for cash. .106. Upon information and belief, in 1966, defendant THE DOW -page 32Amencled Verified Complaint -5 January 1979 �CHEMICAL COMPANY formed Dolco Packaging Corp. jointly with Olson Brothers, Inc., a producer of foamed polystyrene food packages and, upon information and belief, sold that company in 1974 to Olson Brothers. 107. Upon information and be-lief, in February, 1967, defendant'THE DOW CHEMICAL COMPANY acquired a controlling interest in Bio-Science Laboratories, and upon information and belief defendant THE DOW CHEMICAL COMPANY acquired total interest in said company in 1969. . Upon information and belief, defendant THE DOW CHEMICAL COMPANY now operates that company so aquired as Bio-Science Enterprises. 108. Upon information and belief, in 1967', defendant THE DOW CHEMICAL COMPANY acquired controlling interest (a substantial interest having been acquired in 1964) in Gruppo Lepetit S.P.A., of Milan, Italy, a pharmaceutical company operating in about 35 countries. 109.. Upon information and belief, in 1972, defendant THE DOW CHEMICAL COMPANY acquired Wanda Petroleum Co. from Ashland Oil, Inc. 110. Upon information and belief, in March, 1974, defendant THE DOW CHEMICAL COMPANY acquired 650,000 shares of Magma Power Co. and a warrant to purchase additional shares. 111. Upon information and belief, in July, 1974 defendant THE DOW CHEMICAL COMPANY sold its 45% interest in Banco Cidade De .Sao Paulo, / Brazil and its 45% interest in Multi-Bankir.*; Corp. a commercial bank in Zurich, Switzerland to Swiss Bank Corp. 112. Upon information and belief, in 1974 defendant THE DOW CHEMICAL COMPANY sold 10% of Dov;-Bank to Fuji Bank, together with-an option to acquire an additional 40% of Dow Bank. 113. Upon information and belief, in January, 1975, defendant THE DOW CHEMICAL COMPANY formed three environmental divisions which -page 33Amended Verified Complaint "5 January 1979 �together became Hydroscience Associates, Inc. 114. Upon information and belief, in January, 1977 for $15,250,000 defendant THE DOW CHEMICAL COMPANY sold an interest it held in 151 gas and oil wells and approximately 63,000 undeveloped leasehold acres to Texas Oil & Gas Corp. Defendant THE DOW CHEMICAL COMPANY: Subsidiaries. 115. Upon information and belief, defendant THE DOW CHEMICAL COMPANY functions chiefly as an operating company but, upon information and belief, as of December 31, 1976 defendant THE DOW CHEMICAL COMPANY owned 100% of the voting control of the following " firms except where otherwise noted: . . • Bio-Science Enterprises, California; Bio-Science Hospital & Clinic Laboratories, California; Bio-Science Laboratories, California; " Bio-Science Laboratories, Inc., Illinois; Bio-Science Laboratories, Inc., Maryland; Bio-Science Laboratories, Inc., Mississippi; Bio-Science Laboratories, Inc., Missouri; Bio-Science Laboratories, Inc., New York; Bio-Science Laboratories, Inc., Pennsylvania; — Bioscientia Biochemictry Dienstleistungs- . Gesellschaft mit beschrankter Haftung, Germany (50%) Boston Medical Laboratory, Inc., Massachussets; BLMG, Inc., California; Jennings Laboratory, Inc., California; Kopp Laboratories Ltd., Canada Pathology Associates, Inc., Hawaii Photovolt Corp., New York; Bronco Pipeline Co., Texas; Chemical Transportation Co., Texas; Compagnie des Services Dcwell Schlumberger, France (50%); Cordis Dow Corp., Delaware (50%); Dow Badische Co., Delaware (50%); Dow Banking Corp., Switzerland (90%); Dow Chemical A.G., Switzerland; Asahi-Dow Ltd., Japan (50%); Dow Chemical A.B., Sweden; Dow Chemical Africa (Pty.) Ltd., South Africa; Dow Chemical A/S., Denmark; Dow Chemical (Australia) Ltd,, Australia; Dow Chomical Belgium, Belgium; Dow Chemical Company Ltd., United Kingdom; -page 34.Amended Verified Complaint 5 January 1979 • �Dow Chemical (Europe) S.A., Switzerland; Dow Chemical Export A.G., Switzerland; Dow Chemical France, France; Dow Chemical Geselschaft m.b.H., Austria; Dov/ Chemical GmbH., Germany; Dow Chemical Rheinwerk GmvH., Germany; Dow Chemical Schwinge GmbH., Germany; Dow Chemical Handels-und Verwaltungs-Gesellschaft mbH, Germany; • . • Dow Chemical (Hellas), A.E., Greece; Dow Chemical Iberica, S.A./Spain (98%); Ferro Enamel Espanola S.A., Spain (50%); Dow Chemical (Nederland) B.V., Netherlands; Dow Chemical (Norway) A/S, Norway; Dow Chemical Pacific Ltd., Hong Kong; Dow Chemical S.P.A., Italy; Dow Quimica Argentina S.A., Argentina; _ .Ivon Watkins-Dow Ltd., New Zealand (51%); Petroquimica-Dow S.A., Chile (60%); Dow Chemical Arizona Investment Corp., Arizona; • Dow Chemical (Hong Kong) Ltd., Hong Kong; Dow Chemical Inter-American Ltd., Delaware; Dow Chemical International Inc., Panama; Dow Chemical International Inc. of Delaware; Dow Chemical. International Ltd., Delaware; Dow Chemical Japan Ltd., Japan; Dow Chemical N.V., Netherlands Antilles; Coral Navigation Co., Inc., Panama; Dow Chemical Korea Ltd., South Korea; Dow Colombiana, S.A., Columbia; Dow Quimica Centro Americana S.A., Costa Rica; Dow Quimica de Colombia, S.A., Colombia; Dow Quimica de Guatemala Limitada, Guatemala; Dow Quimica S.A., Brazil; Indoquim, S.A., Argentina; Korea Pacific Chemical Corp., South Korea (50%); Laboratories Industriales Farmaceuticos Ecuatorianos L'IFE, Ecuador (59%); Laboratories LIFE de El Salvador S.A.,-E1 Salvador; Negocios e Industrial S.A. Neisa,"Ecuador; . Pacific Chemicals Berhad, Malaysia (51%);. Pacific Chemical -(Taiwan) Ltd., Republic of China; Pacific Plastics (Thailand) Ltd., Thailand; Prcductos Quimicos Peruanos S.A., Peru; Propenasa-Produtos Petroquimicos Nacionais S.A., Brazil (801); F.T. Pacific Chemicals Indonesia, Indonesia; Tecnica Petroquimica Venezuela C.A., Venezuela; Dov/ Chemical of Canada, Ltd., Canada; G.S.. Blakeslee Equipment Ltd., Canada; Dov/ Chemical Engineering Ltd-., Canada; -page 35Amended Verified Complaint - 5 January 1979 �Maligne Resources Ltd., Canada; Dow Pipeline Ltd., Canada; Waviskaw Explorations Ltd., Canada (50%); Dow Chemical Overseas Capital N.V., Netherlands Antilles; Dow Chemical Overseas Management Co., Delaware; Dow Chemical (Quebec) Ltd., Delaware; Dow Chemical (Singapore) Private Ltd., Singapore; Dow Corning Corporation, Michigan (50%); Dowell Schlumberger Corp., Panama (50%); Dow Engineering Co., Delaware; Dow Engineering Inc., Michigan; Dow Insurance Ltd., Bermuda; Dow Quimica Chilena S.A., Chile; Petroquimica-Dow S.A., Chile (10%); Dow Quimica Mexicana, S.A. de C.V., Mexico; Terminales Maritimas, S.A. de C.V., Mexico; Generic Pharmaceuticals Inc., Virgin Islands; •Gruppo Lepetit S.p.A., Italy (79%); Administration de Participations Etrangeres S.A., Luxembourg; • Laboratories Mexico; Laboratories Laboratories Laboratories Lepetit de Mexico, S.A. de C.V., Lepetit S.A. (Brazil), Brazil; Lepetit S.A. (Sapin), Spain (75%); Lepetit S.A.C. (Chile), Chile; Lepetit Belgica S.A., Belgium; — • Lepetit de Colombia S.A., Colombia; Lepetit de Panama S.A., Panama; Lepetit de Venezuela C.A., Venezuela; Lepetit (Iran) Private Co. Ltd., Iran (41%); Lepetit B.V., Netherlands; Lepetit Japan Ltd., Japan; Lepetit Pharmaceuticals Ltd., United Kingdom; Lepetit S.A., France (99%); Lepetit S.A.Q.I.y.C., Argentina; Neofarma S.A. C.I.F., Argentina (76%); Lepetit-(Thailand) Co. Ltd., Thailand (83%); Lepiran Private Co. Ltd., Iran; Neofarma S.A. C.I.F., Argentina (24%); Sociedade Quimica Lepetit S.A.R.L., Portugal; Lepetit (Iran) Private Co.. Ltd., Iran (52%); Hydroscience Associates, Inc., New Jersey; Midland Pipeline'Corp., Delaware; Mineracao e Quimica do Nordeste S.A., Brazil; Pacific Chemical (Thailand) Ltd., ThailandPinto Pipeline Co., Texas; The Kartridge Pak Co., Iowa (50%); Wanda Petroleum Co., Texas. Defendant THE DOW CHEMICAL COMPANY: Joint Ventures -page 36Amonded Verified Complaint 5 January 1979 J �^6. Upon information and belief, defendant THE DOW CHEMICAL COMPANY engages in a number of joint ventures with other firms. 117. Upon information and belief, Dow Corning Corp. is owned jointly by defendant THE DOW CHEMICAL COMPANY and Corning Glass Works and manufactures silicone products and hyperpure silicone. 118. Upon information and belief/ defendant THE DOW CHEMICAL COMPANY and BASF Overzee N.V. jointly own Dow Badische Co. which produces certain organic chemicals and synthetic textile fibers. 119. Upon information and belief, Kartridge Pak Co. is owned jointly by defendant THE DOW CHEMICAL .COMPANY with Oscar Mayer &'Co. Inc. and produces packaging equipment for aerosol, food, dairy and meat products. 120. Upon information and belief, Cordis Dow Corp., is owned jointly by defendant THE DOW CHEMICAL COMPANY with Cordis Corp. and produces hollow fiber artificial kidneys. 121. Upon information and belief, Defendant THE DOW CHEMICAL COMPANY owns a 30% interest in Oasis Pipe Line Co. which, upon information and belief, was formed in 1971 to construct and operate a natural gas gathering and transmission system. 122. Upon information and belief, defendant THE DOW CHEMICAL COMPANY has significant interests in foreign companies engaged in production and manufacture of various related products. 123. Upon information and belief, in March, 1976 defendant THE DOW CHEMICAL COMPANY was building a complex of 10 Petrochemical plants on an island off the Adriatic Coast in partnership with INA a petrochemical manufacturer which is, upon information and belief, the largest company in Yugoslavia. Upon information and belief, Yugoslavia will have a 51% controlling interest, and the investment of -page 37Amended Verified Complaint 5 January 1979"" �defendant THE DOW CHEMICAL COMPANY in the project will be approximately $750,000,000. Upon information and belief, construction of the complex was to begin in 1977 and is scheduled for completion in 1982. 124. Upon information and belief, in March, 1977 defendant THE DOW CHEMICAL COMPANY announced that its subsidiary Dow Chemical Europe and the Saudi Basic Industries Corp." reached an interim agreement to continue investigating an $800,000,000 joint venture petrochemical complex in Jubail, Saudi Arabia. Upon information and belief, the proposed petrochemical complex is.scheduled for completion in the mid-1980's. . . • 125. Upon information and belief, .in September, 1973, Dow Chemical (Europe) S.A. established a joint enterprise in Zagreb, Yugoslavia with Organsko Kemijska Industries/Oki/ for manufacture of polystyrene and expandable polystyrene granules. Upon information and belief, the total initial investment by defendant THE DOW CHEMICAL COMPANY will be about .$17,000,000, including technology, land, fixed and working capital and the enterprise will be jointly managed. Upon information and belief, the plant will supply raw material to the Yugoslavian plastics molding industry. 126. Upon information and beliefj in September, 1975 a subsidiary of defendant THE DOW CHEMICAL COMPANY, Dow Chemical of Canada, Ltd., together with the Alberta Gas Trunk.Line Co. Ltd., and Dome Petroleum, Ltd. reached basic agreement with the Government of Alberta, Canada, for the first phase of a major petrochemical manufacturing project in that province. Upon information and belief, said first phase is expected to involve capital outlays by the joint venturers,"including defendant THE DOW CHEMICAL COMPANY on the ordar of $1,500,000,000 and would produce products that are forecast to improve Canada's balance of payments position by approximately $750,000,000 a year in 1979. Upon information and belief, said project meets the provincial government's policies on petrochemical development and no government subsidy will bo required. Upon information and belief, the project is based -page 38Amendod Verified Complaint 5 January 1979 �on the manufacture of ethylene and its derivatives in Alberta from the ethane component of natural gas streams in the province. 127. Upon information and belief, in January, 1977, the government of South Korea authorized two ventures involving defendant THE DOW CHEMICAL COMPANY to obtain $61,000,000 in loans from a group of foreign banks led by European Banking Corp. to help finance construction of two plants in Korea. Upon information and belief, $37,000,000 of the total will go to Dow Chemical Korea a subsidiary of defendant THE DOW CHEMICAL COMPANY which is building a $10,000,000 plant at Yochion, Korea capable of producing approximately 210,000 tons of chlorine and .231,000 tons of caustic soda a year. Upon information and belief, the remaining .$24,000,000 will be used to help finance construction of a 100,000-ton low density polyethylene plant at Yochon, Korea .by Korea Pacific Chemical Corp. in which defendant THE DOW CHEMICAL COMPANY has a 50% interest. 128. Upon information and belief, defendant THE DOW CHEMICAL COMPANY owns a 90% interest in Dow Banking Corp., a Swiss bank, and a minority interest in a bank in the Netherlands. Defendant THE DOW CHEMICAL COMPANY: Business and Products 129. Upon information and belief, Dow was organized in 1897 to extract chemicals from the native 'brine deposits of central Michigan. From an initial concentration on products derived from these chemicals, the company expanded its activities into the manufacture and sale of chemicals and metals, plastics and packaging, bioproducts and consumer products. 130. Upon information and belief, the principal chemical and metal products manufactured and services supplied by defendant THE DOW CHEMICAL COMPANY, and the major uses thereof, include: Acetone, which, upon information and belief, is used in the manufacture of methyl methacrylate and solvents. -page 39Amended Verified Complaint 5 January 1979 �Antimicrobial agents which, upon information and belief, are used by industry and agriculture for the control of bacteria and fungi. Aspirin, other pharmaceutical salicylates and fine organics. Automotive chemicals which, upon information and belief, ace used as antifreeze/coolants, brake fluids, and de-icing products. Caustic soda which, upon information and belief, is used in production of soaps, chemical intermediates, fibers, alumina, paper, and petroleum products. Chlorine and hydrochloric acid which, upon information and belief, are used primarily in the products manufactured, formulated, advertised, promoted, marketed and sold and services furnished by defendant THE DOW CHEMICAL COMPANY. Chlorobenzenes which, upon information and belief, are used in agricultural intermediates, synthetic rubber, chemical intermediates, and textile dye carriers. Upon information and belief, Dow Industrial Service provides services, technology, and products for cleaning industrial facilities. Upon information and belief, Dowell Division of defendant THE DOW CHEMICAL-COMPANY provides specialized servicing of oil and gas wells, including fracturing, acidizing, and cementing. Epichlor.ohydrin which, upon information and belief, is used in the manufacture of epoxy resins and specialty chemicals. Ethanolamines and ispropanolarnines which, upon information and belief, are used in natural gas purification and as an intermediate in the manufacture of detergents, surfactants, and cosmetics. Ethylene amines which, upon information and belief, are used in chelating agents, oil aclditives, paper resins, epoxy curing, and agricultural intermediates. Ethylene dibromide which, upon information and belief, is used in leaded gasolines and as a soil fumigant. -page 40A:n ended Verified Complaint -5 January 1979 �Ethylene dichloride which, upon information and belief, is used primarily in products manufactured, formulated, advertised, promoted, marketed and sold by defendant THE DOW CHEMICAL COMPANY. Ethylene and propylene glycols and polyglycols which, upon information and belief, are used in antifreeze, polyester fibers & resins, plasticizers, lubricants, humectants, mold release agents, and personal care products. Ethylene and propylene oxides which, upon information and belief, are used in the production of glycols, urethane polyols and industrial polyglycols, alkanolamines, glycol ethers, and surfactants. Flame retardent chemicals which, upon information and belief, are used in flame-retarding plastics, fabrics, and other combustibles. Flocculents and polyacrylamides which, upon information and belief, are used in mining, paper and food industries, and for water and wastewater clarification. Glycerine which, upon information and belief, is used in the manufacture of alkyd resins, cellophane, dynamite/ food/ and tobacco products. Glycol ethers which, upon information and belief, are used as solvents, hydraulic fluids, penetrants., and chemical intermediates. . Heat transfer products which are, upon information and belief, organic fluids thermally stable at high temperatures and low pressure. Highway products particularly calcium chloride which, upon information and belief, is used for ice and dust control, concrete roadway insulation, paving systems, pavement expansion material/ and bridge median barrier. Upon information and belief, Hydroscience Inc. provides complete package engineering services including base surveys for industrial and municipal pollution control, v/aste treatment, conceptual design for wastewater treatment plants, and . environmental assessment of streams, estuaries, rivers, and other receiving waters. -page 41Amondsd Verified Complaint 5 January 1979 �Inorganic bromides which, upon information and belief, are used in production and processing of film and fibers. Ion exchange resins which, upon information and belief, are used in water conditioning, demineralization, industrial processing, and treatment of radioactive and other liquid wastes. Magnesium extrusions, sheet and plate which, upon information and belief, are used in commercial, industrial, and military products. Magnesium ingot which, upon information and belief, is used in alloying aluminum, desulfurizing and modularizing ferrous products, cathodic protection/ die casting, chemical and structural applications, metal reductant and production of magnesium sheet for the printing industry. Mining products which, upon information and belief, include frothers, flotation agents, and metal collectors. • . Methylcellulose which, upon information and belief, is used in a variety of industries as a thickener, suspending agent, protective colloid, binder, stabilizer, film former, and sizing. Upon information and belief, defendant THE DOW CHEMICAL COMPANY, through its divisions, subsidiaries and joint ventures provides services, technology and products for cleaning of nuclear facilities. Olefins and aromatics, including ethylene, propylene, butadiene, and benzene which, upon information.and belief, are used primarily in products manufactured, formulated, advertised, promoted, marketed and sold by defendant THE DOW CHEMICAL COMPANY. • Organic bromides which, upon information "and belief, are used in production of chemical and fiber intermediates. Pharmaceutical bromides and methoxyflourane anesthetic. Phenol which, upon information and belief, is used in making phenolic resins and as an intermediate for nylon, weed killers, fungicides, plasfcicizers, oil additives and detergents, and -page 42. „ Amended Verified Complaint 5 January 1979 �dyeing intermediate. Process products which, upon information and belief, include catalysts and membrane systems. Pusher chemicals which, upon information and belief, are used to enhance oil recovery. Solvents including perchloroethylene, trichloroethylene, 1,1,1-trichloroethane, methylene chloride, and propylene dichloride which, upon information and belief, are advertised, promoted, marketed and sold by defendant THE DOW CHEMICAL COMPANY for drycleaning, metal degreasing, cold cleaning, aerosol and paint removers, among other promoted uses. Surfactants and chelating agents which, upon information and belief/ "are used in detergents, disinfectants, photography, textiles, polymerization, water treatment' formulations, and agricul.tural applications. Urethane polyols and isocyanates which, upon information and belief, are used in the manufacture of structural and cushioning urethane foams. Upon information and belief, Wanda Petroleum Co., a subsidiary of defendant THE DOW CHEMICAL COMPANY engages in trading and marketing petroleum products. 131. Upon information and belief, the principal plastic and packaging materials and products manufactured, formulated, advertised, promoted, marketed and sold by defendant THE DOW CHEMICAL COMPANY include: Coating materials such as epoxy, vinyl ester and other resins which, upon information and belief, are used in coatings, reinforced plastics, laminates and electronic devices; latexes which, upon information.and belief, ace used by the textile, paper and packaging industries; and electroconductive resins which, upon information and belief, are used for paper noating. Construction materials such as polystyrene foam insulation for construction and structural insulating panels. Molding and extrusion materials such as -page 43Amended Verified Complaint 5 January 1979 �polystyrene, low and high density polyethylene, chlorinated polyethylene, saran, acrylonitrilebutadiene-styrene (ABS), styrene-acrylonitrile (SAN) and other plastic resins which, upon information and belief, are used in injection molding, blow molding and extrusion processes for fabrication of articles used in the automotive, appliance, packaging, wire and cable, housewares, toy, and construction businesses. Monomers and coating materials such as styrene, vinyl toluene, divinylbenzene, vinyl chloride, vinylidene chloride and other monomers, which, upon information and belief, are used in producing polymer and copolymer coatings and resins. .. Packaging materials such as polystyrene film for envelope windows, packaging and disposable dinnerware; saran-containing films for commercial packaging; polyethylenee films for packaging, industry, and agriculture; and expandable polystyrene foam strands for loose-fill packing. Plastic foams such as polystyrene foam for insulation, flotation, and floral craft uses; and polyethylene foam for cushion packaging, flotation, and recreational applications. Plastic lined piping products including, upon information and belief, pipe, fittings, valves and accessories for corrosive piping systems. 132. Upon information and belief-, defendant THE DOW CHEMICAL COMPANY has manufactured', formulated, advertised, promoted, marketed and sold "Bioproducts" and "consumer" products, in the following areas, and is continuing so to do: Agricultural products which, upon, information and belief, include weed, brush and grass killers;, nitrogen stabilizers; soil, space and grain fumigants; insecticides; nematocides; fungicides; nutritional, and feed additive products for animals and poultry. ) Consumer products which, upon information and belief, include SARAN WRAP and HANOI-WRAP plastic household film, DOW Bathroom Cleaner, DOW Oven Cleaner, ZIPLOC bags, AZTEC sun care products. Health care products and services including -page 44Amcndcd Verified Complaint - 5 January 1979 �pharmaceutical product such as RIFADIN and RIFOCIN which/ upon information and belief, are proprietary antibiotics, and other antibiotic formulations; PLASIL, a digestant, EQUILID an anti-depressant; cough and cold preparations; LORELCO, a cholesterol reducing agent; diagnostic reagents and equipment. Upon information and belief, Bio-Science Enterprises, a wholly-owned and subsidiary of defendant THE DOW CHEMICAL COMPANY provides diagnostic services to the medical profession. Defendant THE DOW CHEMICAL COMPANY: Principal Properties 133. Upon information and belief, defendant THE DOW CHEMICAL COMPANY and its wholly and principal partly owned companies have plants and holdings which include 47 manufacturing locations in 22 of the United States, located, upon information and belief, in Arkansas at Magnolia and Russellville; in California at Concord, Costa Mesa, Fresno, Pittsburg, Torrance, and Van Nuys; in Colorado at Denver; in Connecticut at Gales Ferry and Trumbull; in Florida at Miami; in Georgia at Dalton, Gainsville and Sylvania; in Illinois at Joliet; — in Indiana at Indianapolis and Zionsville; in Iowa at Davenport; in Kentucky at Carrollton and Elizabethtown; in Louisiana at Breaux Bridge, Napoleanville, and Plaguemine; in Michigan at Bay City, Hemlock, Ludington and Midland; in Missouri at Cape Girardeau, and Peveley; in New Jersey at Carteret; in New York at New York City; in North Carolina at Boonville, Greensboro, and Mebane; in Ohio at Findlay, Hebron and Ironton; in Oklahoma at Tulsa; in Pennsylvania at Royersford; in South Carolina at Allendale and Anderson; in Texas at Freeport and Oyster Creek; and in Virginia at Willisamsburg. 134. Upon information and belief, defendant THE DOW CHEMICAL COMPANY and its wholly and principal partly owned companies have plants and holdings which include 31 manufacturing Ic.j-...;.::: i'. 13 -page 45-Amended Verified Complaint • 5 January 1979 . �European countries located, upon information and belief, in Belgium at Seneffe and Tessenderio; in France at Drusenheim and Seclin; in Germany at Rheinrnuenster, Munich, and Stade; in Greece at Lavrion; in Iran at Tehran; in Italy at Anagni, Brindisi, Cinisello, Garessio, Livorno, Milan, and Martellago; in Morocco at El Jdida; . in The Netherlands at Amsterdam, Rotterdam, Terneuzen, and Roden; in Portugal at Lisbon; in Spain at Bilbao, Madrid, and Tarragona; • in Sweden at NorrRoping? in United Kingdom at King's Lynn, England, and Barry", Wales; .in Yugoslavia at Zagreb. 135. Upon information and belief, defendant THE DOW CHEMICAL COMPANY and its wholly and principal partly owned companies have plants, and holdings which include 19 manufacturing locations in 6 Latin. American countries, located, upon information and belief, __ in Argentina at Buenos Aries and San Lorenzo; in Brazil at Aratu, Franco da Rocha, Guaruja, and Sao Paulo; in in in in Martin. Chile at Concepcion and Santiago; Colombia at Cartagena and Bogata; Ecuador at Quito; and Mexico at Cuernavaca, Mexico City, and San 136. Upon information and belief, defendant THE DOW CHEMICAL COMPANY and its wholly and principal partly owned companies have plants and holdings which include 14 manufacturing locations in four Provinces of Canada, located, upon information and belief, in Alberta at Fort Saskatchewan and Two Hills; in British Columbia at Delta; in Ontario at Arnprior, Cambridge, Cornwall, Kitchener, Sarnia, Toronto, Mississauga, and - Richmond Hill; and in Quebec at Varennes. -page 46- Amcncied Verified Complaint 5 January 1979 �137. Upon information and belief, defendant THE'DOW CHEMICAL COMPANY and its wholly and principal partly owned companies have plants and holdings which include 29 manufacturing locations in 9 Pacific countries, located, upon information and belief, _ in Cairns, in in in in Kanuma, Suzuka, in . in in 'in Australia at Altona, Brisbane, Blacktown, Mackay, Newcastle, and Smithfield; Hong Kong; - Hong Kong; India at Bombay and Nira; Indonesia at Medan; Japan at Ashigara, Hyuga, Totsuka, Chiba, Kawasaki, Mizushima, Nobeoka, Sapporo, and Ageo; Korea at Ulsan; Malaysia at Kuala Lumpur; New Zealand at New Plymouth; and Thailand at Bangcock. -page 47Amended Verified Complaint -5 January 1979 �Defendant HERCULES INCORPORATED: Corporate History 138. Upon information and belief, defendant HERCULES INCORPORATED was incorporated in Delaware on or about October 18, 1912, as Hercules Powder Co. pursuant to judicial decree in a suit brought by the United States of America against E.I. d-u Pont de Nemours and others in which suit Atlas Powder was also segregated from du Pont. 139. Upon information and belief, defendant HERCULES INCORPORATED adopted its present name on or about May 1, 1966. Pursuant to the provisions of the decree in the suit brought by the United States of America, all of the-outstanding $6,500,000 of bonds and 65,000 shares of.stock at par value $100 were issued to du Pont which immediately distributed all of the stock and one-half of the bonds to its common stockholders. The remaining bonds were, upon information and belief, disposed of shortly thereafter by du Pont. 140. Upon information and belief, defendant HERCULES INCORPORATED started in'1913 as a single-department, single-technology, singleconcept company^. Its administration consisted of a president, vice president and general manager, a vice president in chage-of sales, an assistant general manager, and a secretary-treasurer. Little change occurred until 1928, when some changes in organization were made to improve the handling of the expanding business. 141. Upon information and belief, during this period from 1913 to 1928, defendant HERCULES INCORPORATED broadened its -technological base by branching off from explosives into cellulosics and naval stores. As these new businesses were acquired and absorbed, said defendant operate'd them as separate business entities, upon information and belief, forming three operating departments in 1928 and establishing an Executive Committee and a Finance Committee to monitor the businesses and tie them together under the office of the president. 142. Upon information and belief, over the next 50 years, this -page 48* Amended Verified Complaint -5 January 1979 . - • �basic structure was followed by defendant HERCULES INCORPORATED with the addition of new departments as new businesses were acquired or developed and the merger or dissolution of other departments and operating groups was required. 143. Upoa information and belief, the defendant HERCULES INCORPORATED made some changes in operations as it became a , multinational corporation. 144. Upon information and belief, in December,'1956, defendant HERCULES INCORPORATED diversified is operations by acquiring Huron Milling Co. of Harbor Beach, Michigan with annual sales of approximately $12,000,000 through the production and sale of hydrolyzed vegetable proteins, proteins and wheat starches among other products. . 145. Upon information and belief, in January, 1959, defendant HERCULES INCORPORATED acquired Young Development Laboratories of Rocky Hill, New Jersey, producers of filament-wound glass fiber and reinforced plastic which company was subsequently sold, upon information and belief, in 1966. 146. Upon information and belief, in April, 1960 for 509,802 of its class A shares defendant HERCULES INCORPORATED acquired the assets of Imperial .Color Chemical's Paper Corp. of Glens Falls, New York a producer of pigments and quality wallpaper. 147. Upon information and belief, defendant HERCULES INCORPORATED also acquired during April, 1960, Nitroform Agricultural Chemical Co. of Woonsocket, Rhode Island. 148. Upon information and belief, in December, 1961 defendant HERCULES INCORPORATED acquired Reasor-Hill Corp. of Jacksonville, Arkansas, producers of herbicides. 149. Upon information and belief, in March, 1962 defendant -page 49Amended Verified Complaint 5 January 1979 �HERCULES INCORPORATED sold the wallpaper business of Imperial Color & Chemical Department. 150. Upon information and belief, in exchange for 110,309 shares of its $1.65 cumulative convertible class A shares and 32,820 shares of its common stock, respectively, in 1964 defendant HERCULES INCORPORATED acquired Haveg Industries, Inc. and MHD Research, Inc. 151. Upon information and belief, for 280,495 shares of its reacquired common stock in 1966, defendant HERCULES INCORPORATED acquired the net assets of Crystal Preforming & Packaging, Inc., B.F.. Drakenfeld & Co., Inc. and Lincomatic Industries, Inc. ' 152. Upon information and belief, in 1967 defendant HERCULES INCORPORATED acquired, Aquatrol, Inc. of Texas, a company which specializes in products and engineering services for the treatment of industrial processed water. 153. Upon information and belief, in July, 1969 defendant HERCULES INCORPORATED acquired Radiant Color Company a producer of fluorescent pigments and related products. 154. Upon information and belief, in 1969 defendant HERCULES INCORPORATED acquired Philip Bock Co., Inc. 155. Upon information and belief, in March, 1970"defendant HERCULES INCORPORATED acquired an environmental engineering firm, Black Crow and Eidsness, Inc. of Gainesville, Florida. 156. Upon information and belief, in 1970 defendant HERCULES INCORPORATED acquired Modular Structures, Inc. 157. Upon information and belief, on or about April 1, 1970 defendant HERCULES INCORPORATED sold its 50°o interest in Hystron Fibers, Inc. to Farbwerke Hoescht A.G. -page 50Amended Verified Complaint '5 January 1979 �158. Upon information and belief, in early 1972, for about $6,500,000 in cash, defendant HERCULES INCORPORATED acquired from AVC Corp. a group of their overseas subsidiaries. Upon information and belief, 1005 of the stock of Oceanchem International A/S of Denmark, Kedea A.B., of Sweden and Genu Products Canada Ltd. were also involved in the sale. 159. Upon information and belief, in July, 1972 defendant HERCULES INCORPORATED -acquired AB Purac of Lund, Sweden and its German subsidiary AdKaMatic G.m.b.H. of Giessen and the water treatment assets of Mibis AB, Sweden. 160. Upon information and belief, in December, 1972 defendant HERCULES INCORPORATED acquired Artizans, Inc. of Calhoun, Georgia, a carpet dyeing and finishing firm. 161. Upon information and belief, in March, 1973 for approximately $1,250,000, Haskon Inc., a subsidiary of defendant HERCULES INCORPORATED acquired Seaboard Manufacturing Laboratories, Inc., a subsidiary of Universal Container Corp. 162. Upon information and belief, on or about May 31, 1973 in exchange for 480,000 shares of its common stock, defendant HERCULES INCORPORATED acquired Pennsylvania Industrial Chemicals Corp. 163. Upon information and belief, on or about October 1, 1973 for 732,587 shares of its common stock, defendant HERCULES INCORPORATED acquired Polak's Frutal Works, of Middletown, New York. 164. Upon information and belief, on or about November 15, 1973, for 70,977 shares of its common stock, defendant HERCULES INCORPORATED acquired Scott-Wise Industries of Delaware, Inc. 165. Upon information and belief, at the start of the year 1977, defendant HERCULES INCORPORATED consisted of five domestic operating -page 51Amended Verified-Complaint 5 January 1979 �.departments and three geographic operating units. Upon information.--and belief, the international entities, representing an investment 'af $278 million, constituted a significant portion of the overall busi^.; ' ness of defendant, HERCULES INCORPORATED. ' 166. Upon information and belief, a companywide reorganization^was announced by defendant HERCULES-INCORPORATED in fourth quarter 197-7 > which, according to the 1977 annual report of said defendant HERCULES; INCORPORATED was the first fundamental reorganization in nearly 5.0 : -;i years of operation, although considerable restructuring had taken v. place during that period. 167. Upon information and belief, the multiplicity of products manufactured, formulated, advertised, promoted, marketed and sold by defendant HERCULES INCORPORATED led to a diverse organization in which departmental sales organizations frequently overlapped. '. 168. Upon information and belief, several departments of the defendant HERCULES INCORPORATED used the same plants for the manufacture, formulation and distribution of products. ' "• 169. Upon information and belief, an "office of the president" has been formed by the Board of Directors of defendant HERCULES INCORPORATED made up of the president, the chairman of the Board, and two senior vice presidents. 170. Upon information and belief, other senior vice presidents of the defendant HERCULES INCORPORATED have assumed the duties of" managing the operations of said defendant with one in charge of • : domestic operations and the other in charge of international operations, including the three international operating divisions -i Hercules Europe, Hercules Americas, and Hercules Far East. • •,- . 171. Upon information and belief, seven new vice presidents were named as a part of the reorganization of defendant HERCULES -page 52Amended Verified Complaint 5 January 1979 �INCORPORATED and the positions of general manager and assistant general manager of five operating departments were eliminated. Defendant HERCULES INCORPORATED: Subsidiaries 172. Upon information and belief, defendant HERCULES INCORPORATED functions chiefly as an operating company but as of December 31, 1976/ upon information and belief, owned 100% (except as otherwise noted ' hereinafter) of the voting control of the following subsidiary enterprises: Black, Crow & Eidsness, Inc. of Florida Blackhawk Resin & Chemical Co. of Georgia Ceratonia S.A. of Spain Cesalpinia, S.p.A. of Italy Genu Products Canada Ltd. of Canada Haveg Industries, Inc. of Delaware Hercofina (75%) Hereofood Europe A/A Denmark Hercoform, inc. of Delaware Hercoform Marketing, Inc. of Delaware Hercules Andino, S.A. of Venezuela Hercules Bahamas Corp. of Delaware Hercules B.V. of Netherlands Hercules Hercules Hercules Hercules Hercules Hercules Canada Ltd. of Canada Chemical Corp. of Delaware Chemicals N.V. of Belgium do Brasil Produtos Quiraicos Ltda. of.Brazil Europe, S.A. of Belgium Far East Ltd. of Japan Hercules France, S.A. of France Hercules G.m.b.H..of Germany Hercules Industries, Inc. of Puerto Rico Hercules International Finance Corp. of Delaware Hercules International Trade Corp. Ltd. of Bahamas Hercules Kemiska Aktiebolag of Sweden Hercules Overseas Corp. of Delaware Hercules Powder Co. of Delaware • -- Hercules Powder Co. Ltd. of England Hercules Taiwan Co. Ltd. of Taiwan Hercules Trading Corp. of Delaware Herdata, Inc. of California Imperial Paper & Color Corp. (Canada) Ltd. of'Canada Kedea, A.B. of Sweden Maxchem Holding, A.G. of Switzerland Organa Trust .of Liechtenstein OY Hercofina AB of Finland Patex Holding G.m.b.H. of Austria . Pennsylvania Industrial Chemical Corp. of Delaware PFW Inc. of Delaware -page 53Amended Verified Complaint 5 January 1979 �Picco, S.A. of Switzerland Polak's Frutal Works Inc. of New York Quimica Hercules S.A. de C.V. of Mexico Rohe S.A. of Spain Technical Packaging B.V. of Netherlands Ten Horn Pigment Chemie B.V. of Netherlands 173. Upon 'information and belief, defendant HERCULES INCORPORATED also owns partial interests in the following companies: Italy 80% interest in Bewoid Italiano & Callegaro S.p.A. of 59%. in Coon Canyon Irrigation Co. of Utah 60% in Australian Chemical Holdigns, Ltd. 51% in Pakistan Gum Industries, Ltd. of Pakistan 60% in Hercules De Centroamerican, S.A. of Nicaragua 50% in Abieta Chemie GmbH of Germany Adria Laboratories, Inc. of Delaware DIC-Hercules Chemicals of Japan Holden Vale Manufacturing Co., Ltd. of England Nelson Acetate of England; Neofil S.p.A. of Italy Hercor Chemical Corp. of Puerto Rico St. Croix Petrochemical Corp. of Virgin Islands Texas Alkyls, Inc. of Delaware Organa Commercial Trust, of Liechtenstein Hind Gums Ltd. of India. , Defendant HERCULES INCORPORATED: Joint Ventures 174. Upon information and belief, in March, 1963, defendant HERCULES INCORPORATED and Teijin Ltd. of Osaka, Japan formed Teijin Hercules Chemical Co. Ltd. of Kubushiki Kaisha, to manufacture dimethyl terepthalate a chemical which is used in the manufacture of a polyester fiber for textiles, fabrics, tire cord and other items. Upon information and belief, a 30,000,000 Ib. a year plant was completed in 1964, at Matsuyama, an island of Shikoku, Japan. Upon 'information and belief, a second plant, for the manufacture of dimethyl terephthalate was completed in 1968. Upon information and belief, a third dimethyl terephthalate plant started operations in September of 1970 increasing the DMT capacity o'f the venture to approximately 180,000 metric tons. . Upon information and belief, the share of defendant HERCULES INCORPORATED in the Joint venture is 49%. 175. Upon information and belief, in February, 1965, defendant -page 54Aniondcd V e r i f i e d Complaint 5 J a n u a r y 1979 �HERCULES INCORPORATED and Commonwealth Oil Refining Co., Inc., jointly formed a Puerto Rican company, Hercor Chemical Corp., to build paraxylene production facilities at the Guayanilla Bay facility of Commonwealth Petrochemicals, Inc. which supplies feedstock and other process ingredients. Upon information and belief, in October, 1972, defendant HERCULES INCORPORATED announced plans to more than double plant .capacity to 525,000,000 pounds per year. Upon information and belief, in 1976 defendant HERCULES INCORPORATED sold a 25% interest in Hercor Chemical Corp. 176. Upon information and belief, in 1965 Hercules de Centroamerica, S.A., jointly owned by defendant HERCULES INCORPORATED and Institute de Fomento Nacional, the Nicaraguan Government development agency, was organized to produce toxaphene insecticide for sale in.the' five nation area.of the Central American Common Market. 177. Upon information and belief, in February, 1968 Herdillia Chemicals Ltd., jointly-owned by defendant HERCULES INCORPORATED BP Chemicals (UK) Ltd., EID-Parry of India and other Indian Investors, completed a new plant near Bombay, India, to produce phenol, phthalic anhydride, acetone, diacetone alcohol, and other petrochemical products to be used in plastics, paints, fungicides, and other commercial products. 178. Upon information and belief, in 1968 defendant HERCULES INCORPORATED acquired a 40% interest in Dawood Hercu.les Chemicals Ltd., Pakistan which was formed for production of urea fertilizer at a plant near Lahore with annual production capacity of"345,000 tons. 179. Upon information and belief, in 1969 defendant HERCULES INCORPORATED and Montecatini, S.A. of Italy formed Neofil S.p.A. which manufactures polyolefin fibers at Terni, Italy. 180. Upon information and belief, in March, 1973, defendant HERCULES INCORPORATED announced that a Taiwanese group planned to sign -page 55Amendod Verified Complaint .5 January 1979 �a contract with said defendant HERCULES INCORPORATED for a $15,000,000 polypropylene plant on Taiwan, in which facility defendant HERCULES INCORPORATED will have a 40% interest. Upon information and belief, three Taiwanese investing groups will hold the remaining 60% in the joint venture. 181. Upon 'information and belief, in April, 1973 defendant HERCULES INCORPORATED and Mexican Investment interests formed Petrocel, S.A., Upon information and belief, defendant HERCULES INCORPORATED is a 40% shareholder and the Mexican interests own 60%. Upon information and belief, Petrocel has built a multimillion dollar plant at Altamira, Tamaulipas, Mexico for the production of DMT (dimethyl terephthalate) and TPA (terephthalic acid) with a combined production capacity of 137,000 metric tons. Upon information and belief, both products used in .the manufacture of polyester film and polyester fiber. 182. Upon information and belief, in February, 1974 defendant HERCULES INCORPORATED and a U.S. affiliate of Montedison S.p.A. of Milan, Italy, formed Adfia Laboratories Inc. Upon information and belief, Adria will perform clinical testing for the purpose .of obtaining U.S. Food & Drug Administration approval for drugs already developed and being sold in Europe by Montedison's pharmaceutical affiliates. 183. Upon information and belief, on or about August 31, 1976 defendant HERCULES INCORPORATED and American Petrofina, Inc. formed two joint ventures, Hercofina and Hercofina Europe,' for production and marketing of terephthalates. Upon information and belief, Defendant HERCULES, INC. sold to American Petrofina a 25% interest in its terephthalate assets for a 75% interest in the joint ventures. Upon information and belief, defendant HERCULES INCORPORATED interest will be reduced as American Petrofina elects to invest additional money for capital expansion. -page 56Amended Verified Complaint -5 January 1979 �defendant HERCULES INCORPORATED: Business and Products 184. Upon information and belief, defendant HERCULES INCORPORATED produces many diversified products through its Coatings & Specialty Products Department: • Coatings: " Nitrocellulose . • "Parlon" Chlorinated Rubber Ethyl Cellulose Ethyl Hydroxyethyl Cellulose Penaerythritols Formaldehyde "Imperial" Organic and Inorganic Pigment Colors •Ceramic Colors "Radiant" Fluorescent Products Specialty Products: Chemical Cotton • Cellulose Gum and CMC . Hydroxyethyl. Cellulose Hydroxypropyl Cellulose Carrageenan, Guar and Guar Derivatives Magnetic Iron Oxides "Aqualon" Super Absorbant Polymer 185. Upon information and belief, defendant HERCULES INCORPORATED produces many diversified products through its Food and Fragrance Development Department: Hydrocolloids: Cellulos.e Gum And CMC "HVP" Hydrolyzed Vegetable Protein , Carrageenen, Pectin, and Locust Bean and Guar Derivatives Exudate Gums "Klucel" Hydroxypropyl Cellulose ". Polak's Frutal Works, Inc.: Flavors and Fragrances » 186. Upon information and belief, defendant HERCULES INCORPORATED produces many diversified products through its Industrial Systems Department: -page 57Amended Verified Complaint 5 January 1979 �Industrial Group: Explosives and Blasting Agents Smokeless Powder Nitrogen Products Water Treatment Chemicals and Equipment "Hercofloc" Polymers Lightweight Aggregates Plastic Products G'roup: "Vantage" Polypropylene Structural Foam Oriented Polypropylene Bottles "Delnet" Plastic Netting • "Trimpak" Polypropylene Cups Custom and Proprietary Plastic Moldings Dairy and Food Processing Machinery Haveg Industries, Inc.: ' • 'Silica Fabrics Corrosion Resistant Plastics and Chemicals Equipment . . Asbestos-Reinforced Epoxy and Phenolic Pipe Filament-Wound Fiber Glass Products Plastic Missile and Rocket Structures High-Temperature Insulated Materials and Fabrics High-Temperature Wire and Cable Electrical Harnesses and Assemblies jjystems Group: Solid Propellant Rocket Motors Graphite Fiber Structural Shapes and Components and Fabricated Structures from Graphite and Glass Fiber Composites 187. Upon information and belief, defendant HERCULES INCORPORATED produces many diversified products through its Organics Department: Resins: Rosin a n d Related Resins ' . . . Terpenes Pine Oil Resins o "Stwlube" Synthetic Esters Plasticizers Wash Emulsions Defoamers 1979 �Emulsifiers Hydroperoxide Catalysts Organic Peroxides Synthetic Rubbers "Kerigraph" Systems for Letter Press and Flexographic Printing Agricultural Chemicals: Insectides, which include Toxephene, Delnav and Torak "Nitroform" Urea-Fomaldehyde Plant Food 188. Upon information and belief, defendant HERCULES INCORPORATED produces many diversified products.through its Polymers Department: Plastics: '"Pro-Fax" Amorphous Polypropylene "1900" Ultrahigh Molecular Weight Polyethylene Film: Hercules Film: pylene Films Balanced Oriented Polypro- Home Furnishings: "Herculon" Olefin Fibers Finished Carpet Commission Dyeing, Printing and Finishing 189. Upon information and belief, polypropylene is used in many applications to reduce weight and give corrosion resistance to a wide variety of vehicle parts including battery casings, fan shrouds, and fender lines, as well as upholstery, carpeting and decorative applications in automobile interiors. 190. Upon information and belief, according to statements published- by defendant HERCULES INCORPORATED, graphite fibers manufactured, formulated, advertised, promoted, marketed and sold by said defendant are finding new applications in both the automotive and the aircraft industries as a substitute for steel and other metals where reduction of weight, couple-d with high strength and stiffness is required. Upon information and belief, Magnamite graphite fiber, silver-knit fabric of Herculon fiber, Pro-fax polypropyl'ene structural foam, metal laminates, and Herclor elastomer are finding new uses in • . -page 59Amended Verified Complaint 5 January 1979 �the transportation field. 191. Upon information and belief, defendant HERCULES INCORPORATED produces many diversified products through its Synthetics Department including the chemical intermediates Methanol, Terephthalates, and Dimethyl Terephthalate. 192. Upon information and belief, defendant HERCULES INCORPORATED corporate operations are based upon "value added" chemistry which takes readily available and relatively low cost raw materials and adds value to them by upgrading their chemical characteristics. • 193. Upon information and belief, other companies utilize the products manufactured by defendant HERCULES INCORPORATED to provide consumer products. 194. Upon information and belief, in the area of shelter, defendant HERCULES INCORPORATED not only makes the pigments, coatings, and preservatives that are basic ingredients in paints and stains, but also produces chemicals that are part of "a myriad of other products that make homes-more attractive and more comfortable," according to the annual report of defendant HERCULES INCORPORATED Upon information and belief, Herculon olefin fiber is_used by many manufacturers to produce popular upholstery fabric and carpeting for residential and commercial uses. 195. Upon information and belief, defendant HERCULES INCORPORATED makes adhesives and tackifiers to bond plywood, paneling, formica, and tile, as well as specialty ceramic pigments to decorate glassware and make possible the touch-sensitive controls on electric applicances. Such basic construction materials as cement, wallboard, and linoleum also, upon information and belief, include products of defendant HERCULES INCORPORATED technology. 196. Upon information and belief, production and operation of -page 60Amondod Verified Complaint - 5 January 1979 �cars, buses, trucks, trains, and airplanes depend upon the technology developed and marketed by defendant HERCULES INCORPORATED. 197. Upon information and belief, published statements by defendant HERCULES INCORPORATED claim road signs are easier to read because of Radiant fluorescent pigments. According to the published statements of said defendant, bridges" last, longer and ships can be in service for more voyages without dry-docking because of Parlon chlorinated rubber products that defendant HERCULES INCORPORATED supplies to paint and coatings manufacturers. 198. Upon information and belief, published statements by defendant HERCULES INCORPORATED claim tires., fan belts, ignition and other electrical wires, and oils and lubricants in all phases of the transportation industry exist, in part, because of the resins, rosins, fibers, plasticizers, stablizers, modifiers, and other "value added" ingredients that Hercules manufactures. 199. Upon information and belief, few shoppers can leave a grocery store .without taking with them products manufactured, formulated, advertised, promoted, marketed and sold by the defendant HERCULES INCORPORATED. Upon information and belief, such products include the coatings on frozen food packages; stablizers in ice cream; thickeners in canned gravies; strength additives in paper products; or palatability and flavoring ingredients in soft drinks and desserts. Upon information and belief, products manufactured, formulated, advertised, promoted, marketed, and sold by the defendant HERCULES INCORPORATED are found on virtually every aisle of every.food store in the country. 200. Upon information and belief, products manufactured, formulated, advertised, promoted, marketed and sold by the defendant HERCULES INCORPORATED play an important part in the manufacture of detergents, disinfectants, soaps, cleaners and paper towels. 201. Upon information and belief, tobacco products, snack foods, -page 61Amended Verified Complaint -5 January 1979 �and variety items not only contain various products manufactured, formulated, advertised, promoted, marketed and sold by the defendant HERCULES INCORPORATED, but often come to market protected and packaged in Hercules polypropylene film. 202. Upon information and belief, defendant HERCULES INCORPORATED participates in the clothing industry through Hereofina's DMT (dimethyl terephthalate) and PTA (purified terephthalic acid) used to manufacture polyester fibers. 203. Upon information and belief, fertilizers manufactured, formulated, advertised, promoted, marketed and sold by the defendant HERCULES INCORPORATED are applied to the cotton crop. 204. Upon information and belief, pesticides manufactured, formulated, advertised, promoted, marketed and sold by the defendant HERCULES INCORPORATED are applied to many crops and in many areas for weed control, among other uses. 205. Upon information and belief, according to statements published by the defendant HERCULES INCORPORATED, wherever you turn in a clothing store, you will find materials manufactured, formulated, advertised, promoted, marketed and sold by said defendant. 206. Upon information and belief, carboxymethylcellulose (CMC) manufactured, formulated, advertised, promoted, marketed and sold by the defendant HERCULES INCORPORATED is used in weaving cotton into cloth. Upon information and belief, defendant HERCULES INCORPORATED brought a new warp size plant on stream in 1977 to meet industry demand. 207. Upon information and belief, Indalca gums manufactured, formulated, advertised, promoted, marketed and sold by the defendant HERCULES INCORPORATED are involved in the designs in printed fabrics. 208. Upon .information and belief, pigments and paper •'•page 62Amended V e r i f i e d C o m p l a i n t 5 J a n u a r y 1979 additives �manufactured, formulated, advertised, promoted, marketed and sold by the defendant HERCULES INCORPORATED are found in the packaging, and even as part of the price tag attached by the retailer. 209. Upon information and belief, Delnet propropylene netting manufactured, formulated, advertised, promoted, marketed and sold by the defendant HERCULES INCORPORATED is used as a bonding and stiffening agent in collars, shoes, and gloves, while cross-linking agents are used to make raincoats, boots, and belts. 210. Upon information and belief, defendant HERCULES INCORPORATED manufactures and markets products which smooth and thicken toothpaste and can be found in -mouthwashes, deodorants and other personal health care items. Defendant HERCULES INCORPORATED: Principal Plants & Properties 211. Upon information and belief, defendant HERCULES INCORPORATED owns, operates and maintains plants and owns property at a number of locations in the United States and throughout the world. 212. Upon information and belief, defendant HERCULES INCORPORATED owns, operates and maintains plants and owns property for the manufacture, formulation, and sale of products by its Coatings & Specialty Products Department in the following locations: Glen Falls, New York; Harbor Beach, Michigan; Hopewell, Virginia; Louisiana, Missouri; Parlin, New Jersey; Pulaski, Virginia"; Richmond, California; and Washington, Pennsylvania; 213. Upon information and belief, defendant HERCULES INCORPORATED owns, operates and maintains plants and owns property for the manufacture, formulation, and sale of products by its Industrial Systems -page 63Amencied Verified Complaint " 5 January 1979 �Department in the following locations: Bacchus, Utah; Bessemer, Alabama; Buena Park and Santa Fe Springs, California; Burlington, Vermont; Carthage and Louisiana, Missouri; Donora, Pennsylvania; Hopewell and Snowden, Virginia; Houston, Texas; Ishpeming and and Williamston, Michigan; Kenvil, New Jersey; Marshallton and Middletown, Delaware; Oxford, Georgia; Port Ewen, New York; Rocket Center, West Virginia; St. Paul, Minnesota; . ' 'and San Leandro, California; Taunton, Massachussetts; and Terre Haute, Indiana. 214. Upon information and belief, defendant HERCULES INCORPORATED owns, operates and maintains plants and owns property for the manufacture, formulation, and sale of products.by its Organics Department at the following locations: __ Baton Rouge, Louisiana; Brunswick and Savannah, Georgia; . . Burlington, New Jersey; Chicopee, Massachussetts; Clairton and West Elizabeth, Pennsylvania; Franklin, Virginia; Gibbstown, New Jersey; Hattiesburg, Mississippi; Kalamazoo, Michigan; Milwaukee, Wisconsin; and Portland, Oregon. 215. Upon information and belief, defendant HERCULES INCORPORATED owns, operates and maintains plants and owns property for the manufacture, formulation, and sale of products by its Polymers Department at the following locations: Bayport, Texas;' Calhoun, Georgia; Covington, Virginia; Crowley and Lake Charles, Louisiana; Oxford, Georgia; and Tcrre Haute, Indiana; *-- -page 64Amenued Verified Complaint 5 Jonuarv 1979 �216. Upon information and belief, defendant HERCULES INCORPORATED owns, operates and maintains plants and owns property for the manufacture, formulation, and sale of products by Hercofina at Plaqueraine, Louisiana and Wilmington, North Carolina. 217. Upon information and belief, defendant HERCULES INCORPORATED owns, operates and maintains plants and owns property for the manufacture, formulation, and sale of products by its Food & Fragrance Development Department at the following locations: Middletown, New York; Amersfoort, The Netherlands; London, England; and 'Sydney, Australia 218. Upon information and belief, defendant HERCULES INCORPORATED owns, operates and maintains plants and owns property for the manufacture, formulation, and sale of products by its Europe Department and International Department in a number of countries, including: Belgium Denmark _ England Finland France Germany Italy ' _" Spain" . Sweden The Netherlands; Brazil, and Canada. -page 65d Verified Complaint 5 January 1979 . • ' . �DEFENDANT NORTHWEST INDUSTRIES, INC.: Corporate History 219. Upon information and belief, defendant NORTHWEST INDUSTRIES, IMC. was incorporated in Delaware on or about August 3, 1967 and in April 1968 became the parent corporation of Northwest Chemco, Inc. ("Chernco") holding 99.9% of the -outstanding shares presently owned,220. Upon information and belief, defendant NORTHWEST INDUSTRIES, INC. became the parent corporation of Philadelphia and Reading Corporation ("Philadelphia-Reading") holding 96.5% of the outstanding shares presently owned). 221. Upon information and belief, in April 1971 defendant NORTHWEST INDUSTRIES, INC. purchased for cash all of the outstanding capital stock of The Buckingham Corporation ("Buckingham"), the exclusive United States importer and selling agent of Cutty Sark brand Scotch whisky. 222. Upon information and belief, in April 1973 defendant NORTHWEST INDUSTRIES, INC. purchased for cash approximately 98% of the common stock of General Battery Corporation ("General Battery") and, upon information and belief, defendant NORTHWEST INDUSTRIES, INC. now 'owns 100% of the outstanding shares of that company. 223. Upon information and belief, in February 1976 defendant -NORTHWEST INDUSTRIES, INC. purchased for cash approximately 83% of the common stock of Microdot Inc. ("Microdot") and, upon information and belief, defendant NORTHWEST INDUSTRIES, INC. now owns 100% of the. outstanding s'hares of that company. 224. Upon information and belief, in November 1977 defendant NORTHWEST INDUSTRIES, INC. purchased for cash 97.4% of the common and common equivalent shares of Coca-Cola Bottling Company of Los Angeles ("Coca-Cola of Los Angeles"). -page 66Amenueu Verilitci Complaint 5 January 1979 �Defendant NORTHWEST INDUSTRIES, INC.: Business and Products 225. Upon information and belief, defendant NORTHWEST INDUSTRIES, IMC. is a holding and management company owning, directly or indirectly, eleven principal operating subsidiaries. 226.-Upon information and belief,. defendant NORTHWEST INDUSTRIES, INC. is headquartered in Chicago. 227. Upon information and belief, control of day-to-day operations of the defendant NORTHWEST INDUSTRIES, INC. is decentralized, with responsibility in the hands of operating company managements. .228. Upon information and belief, corporate management of defendant NORTHWEST INDUSTRIES, INC. retains broad policy authority, provides advisory and staff functions, reviews plans and budgets, monitors performance, and maintains overall financial controls, including the allocation of capital and the conduct of all financing for its several operating subsidiaries. 229. Upon information and belief, the business activities of defendant NORTHWEST INDUSTRIES, INC. were carried on in 1977 by the following principal operating subsidiaries: Acme Boot Company, Inc.; Buckingham; Coca-Cola of Los Angeles; • Fruit of the Loom, Inc.; General Battery; Imperial Reading Corporation; Union Underwear Company, Inc. ("Union Underwear"); Lone Star Steel Company .("Lone Star"); Microdot; Universal Manufacturing Corporation ("Universal"); and Velsicol Chemical Corporation ("Velsicol Chemical"); . in the following three product groups: Consumer: apparel, batteries and beverages; Industrial: principally tubular steel products and ingot moulds, connecting devices and electrical -page 67Aiuenaed V e r i f i e d CornplainL 5 J a n u a r y 1979 �lamp ballasts; and . Chemical: principally pesticides and industrial chemicals and resins. 230. Upon information and belief, the principal consumer products manufactured, fabricated, advertised, promoted, marketed and sold by defendant NORTHWEST INDUSTRIES, INC. are men's and boys' underwear; western, leisure and casual boots; automotive replacement batteries; soft drinks and bottled water; and Scotch whisky. Upon information and belief, other products and operations in this group manufactured, fabricated, advertised, promoted, marketed and sold by the defendant NORTHWEST INDUSTRIES, INC. include jeans and shirts; industrial batteries and battery chargers; industrial water treatment; and French wines a n d vodka. . . . 231. Upon information and belief, defendant NORTHWEST INDUSTRIES,. INC. through a subsidiary, owns the Fruit of the Loom and B.V.D. trademarks, which are registered in the United States and in many foreign countries. Upon information and belief, these trademarks are used primarily on men's and boys' underwear and knit shirts marketed by Union Underwear and, upon information and belief, said trademarks, are also licensed to nonaffiliated manufacturers of wearing apparel, home furnishings and other household products under agreements which typically call for royalty payments based on net sales with fixed annual minimum fees. 232. Upon information and belief, the underwear boot, replacement battery, soft drink, bottled water and Scotch whisky operations of the defendant NORTHWEST INDUSTRIES, INC. rank among the leaders in their repsective f-ields. Underwear is sold principally under 'the Fruit of the Loom and B.V.D. trademarks and boots are sold principally under the Acme, Dingo and Dan Post labels. 233. Upon information and belief, the principal markets for apparel products manufactured, fabricated, advertised, promoted, marketed and sold by the defendant NORTHWEST INDUSTRIES, INC. are department, discount, mail order, variety, chain and independent. r' -page 68- Amend-ed Vcritied Complaint 5 January 1.979 �specialty stores, and distribution is direct. 234. Upon information and belief, automotive replacement batteries manufactured, fabricated, advertised, promoted, marketed and sold by the defendant NORTHWEST INDUSTRIES, INC. are marketed largely through auto parts retail stores, large -retailers, and tire and oil companies, principally under private labels. 235. Upon information and belief/ most soft drinks manufactured, formulated, advertised, promoted, marketed and sold by the defendant NORTHWEST INDUSTRIES, INC. are sold directly to retailers, principally under the Coca-cola and Canada Dry trademarks. Upon information and belief, the franchise areas for the soft drink products manufactured, formulated, advertised, promoted, marketed and sold by the defendant NORTHWEST INDUSTRIES, INC. include Southern and Central California, four of the five main Hawaiian Islands, and portions of Nevada (including Las Vegas), Missouri, Kansas, Wisconsin, Illinois, Iowa and' Nebraska. Upon information and belief, bottled water manufactured, formulated, advertised, promoted, marketed and sold by the defendant NORTHWEST INDUSTRIES, INC.is sold directly to consumers in Southern California and Houston, Texas. 236. Upon information and belief, Scotch whisky manufactured, formulated, advertised, promoted, marketed and sold by the defendant NORTHWEST INDUSTRIES, INC. is sold under the Cutty. Sark and Cutty 12 trademarks. Upon information and belief, Scotch whisky, French wines, and vodka manufactured, formulated, advertised, promoted, marketed and Sold by the defendant NORTHWEST INDUSTRIES, INC. are sold'to distributors throughout the United States 'and directly to state governments for resale in state-controlled liquor stores. 237. Upon information and belief, the principal raw materials for apparel manufactured by defendant NORTHWEST INDUSTRIES, INC. are cotton staple, polyester staple, finished textile fabrics and trim, leather, leather substitute and.rubber products, all of which are -page 69Aniendeci V e r i r i e d Complaint 5 January J979 �regularly available. 238. Upon information and belief, Union Underwear is a fully integrated manufacturer, spinning most of its yarn requirements and manufacturing most of its cloth and elastic needs. 238. Upon information and belief, lead, the most important component in the manufacture of batteries by defendant NORTHWEST INDUSTRIES, INC. .is obtained as a raw material from internal secondary smelting operations primarily, and also from various external sources. 239. Upon information and belief, soft drink syrup and concentrate are purchased by defendant NORTHWEST INDUSTRIES, INC. primarily from franchisers under contracts which generally cover an unlimited period of time. 240. Upon information and belief, Cutty Sark Scotch whisky is purchased by defendant NORTHWEST INDUSTRIES, INC. pursuant to a long-term, exclusive United States distributorship agreement. 241. Upon.information and belief, the Consumer Products.Group operations of defendant NORTHWEST INDUSTRIES, INC. employed approximately 25,000 persons at December 31, 1977. 242. Upon information and belief, the principal industrial products manufactured, fabricated, advertised, promoted, marketed and sold by the defendant-NORTHWEST INDUSTRIES, INC. are tubular steel . products, connecting devices, ingot moulds and ballasts (small transformers) for flourescent and high-intensity discharge lamps. Upon information and belief, in all of these operations defendant NORTHWEST INDUSTRIES, INC. ranks as a leader. 243. Upon information and belief, tubular steel products manufactured, fabricated, advertised, promoted, marketed and sold by the defendant NORTHWEST INDUSTRIES, INC. include casing and tubing for -page 70Amencicd Veriried Complaint • 5 January 1979 �oil and gas wells, line pipe for gathering and transporting petroleum, standard pipe for construction, spiral weld pipe for water mains, and mechanical tubing. 244. Upon information and belief, oil country tubular goods manufactured by defendant NORTHWEST INDUSTRIES, INC. are marketed principally through distributors in the Southwestern, South Central, and Western states, while other tubular steel products are marketed primarily through distributors to end-users throughout the United States except for New England and the Mid-Atlantic states. . 245. Upon information and belief, tubular steel products manufactured, fabricated, advertised, promoted, marketed and sold by the defendant NORTHWEST INDUSTRIES, INC. are sold principally under the Lone Star and A.O. Smith labels, and ballasts are sold under the Universal label. 246. Upon information and belief, essential raw materials for the manufacture of tubular steel products by defendant NORTHWEST INDUSTRIES, INC. are iron ore, coal, coke, limestone and steel scrap. Upon information and belief, iron ore reserves in close proximity to the principal plant are owned either in fee or under long-term lease and are estimated to be sufficient for ninety years of operation. To supplement this supply it is believed that there are long-term purchase contracts for concentrated ore pellets. Upon information and belief, defendant NORTHWEST INDUSTRIES, INC. maintains coal 'reserves in Oklahoma and arkansas, both owned and leased, which represent more than eighty million tons, and, upon information and belief, these deposits are not being mined currently. Upon information and belief, defendant NORTHWEST INDUSTRIES, INC. purchases high, medium and low ' volatile coals in the required mix"under long-term contracts. While some coke to -supplement production is presently purchased by defendant NORTHWEST INDUSTRIES, INC., upon information and belief, approximately $100 million will be spent by defendant NORTHWEST INDUSTRIES, INC. for a new battery of 70 coke ovens and new iron ore sintering facilities -page 71Ainendeci Verified Complaint �in order to add productive capacity for the necessary intermediate products. Upon information and belief, there are numerous limestone quarries within two hundred miles of the plant. Upon information and belief, steel scrap is usually purchased by defendant NORTHWEST INDUSTRIES, INC. from several dealers in the Texas, Oklahoma, Arkansas, Louisiana area. . 247. Upon information and belief, new tubular steel facilities have been added by defendant NORTHWEST INDUSTRIES, INC. including two electric .furnaces, a continuous caster, and two extrusion presses - one of which is the largest of its kind in the world. <~" 248. Upon information and belief, 'in 1977 a $13.5 million expenditure was authorized for additional specialty tubing capacity which will provide for production of this tubing in sizes up to 15 inches in diameter. 249. Upon information and belief, connecting devices manufactured, fabricated, advertised, promoted, marketed and sold by the defendant NORTHWEST INDUSTRIES, INC. consist of specialty fasteners, electrical and electronic~~connectors, and fluid seals and couplings, which, upon information and belief, are sold primarily to the automotive and aerospace markets, as well as to other major' industries. . 250. Upon information and belief, steel is the principal raw material for most connecting devices manufactured by defendant 'NORTHWEST INDUSTRIES, INC., while silver, brass, aluminum, gold, rubber, and plastics are also used. 251. Upon information and belief, the ingot molds manufactured, fabricated, advertised, promoted, marketed and sold by the defendant NORTHWEST INDUSTRIES, INC. are large cast iron forms into which molten steel' is poured and allowed to solidify to produce ingots. 252. Upon information and belief, Hicroclot is the largest of three -page 72Mmenueu Verifieu Complaint "5'January 1979 �major independent ingot mold manufacturers, which supply approximately 50% of domestic steel industry requirements. Both connecting devices and ingot molds are marketed through Microdot's own sales force. 253. Upon information and belief, Microdot owns a number of domestic and foreign patents which are important in the aggregate, but not individually. Upon information and belief, ballasts and some of the processes involved in their production are covered by numerous patents, of which the principal ones are owned by two major competitors. Universal believes it is adequately licensed to conduct its business. 254. Upon information and belief, ballasts manufactured, advertised, promoted, marketed and sold by the defendant NORTHWEST INDUSTRIES, INC. are sold primarily directly to original equipment lamp manufacturers throughout the country. 255. Upon information and belief, essential raw materials for ballasts manufactured, fabricated, advertised, promoted, marketed and sold by the defendant NORTHWEST INDUSTRIES, INC. are copper, steel and aluminum. Upon information and belief, all of these requirements are . met from major producers and dealers in the United States and abroad. 256. Upon information and belief, the Industrial Products operations of defendant NORTHWEST INDUSTRIES, INC. employed approximately 15,000 persons at December 31, 1977. 257. Upon information and belief, the principal chemical products manufactured, formulated, advertised, promoted, marketed and sold by the defendant NORTHWEST INDUSTRIES, INC. are specialty pesticides, industrial chemicals, and repins. 258. Upon information and belief, chemical products are sold by defendant NORTHWEST INDUSTRIES, INC. under various brand names both directly to manufacturers and through distributors, dealers, and -page 73Amended Verified Complaint r 7-,nn-.rv 1979 �foreign governments to a wide variety of customers. 259. Upon information and belief, all chemical products manufactured, formulated, advertised, promoted, marketed and sold by the defendant NORTHWEST INDUSTRIES, INC. are higly competitive, however/ according to reports published by defendant NORTHWEST uINDUSTRIES, INC., based upon available market research, the chemical products group of defendant NORTHWEST INDUSTRIES, INC. ranks as a major producer of specialty pesticides, brine-derived chemicals, including flame retardents, benzoic acid derivatives and, petroleum hydrocarbon resins. . 260. Upon information and belief, Velsicol Chemical has patent protection expiring from 1978 to 1986 on several of its pesticide products. " • 261. Upon information and belief, raw materials are generally available to defendant NORTHWEST INDUSTRIES, INC. from other chemical manufacturers, petroleum processors, and owned brine wells. 262. Upon information and belief, the chemical product operations of defendant NORTHWEST INDUSTRIES, INC. employed approximately 2,000 persons at December 31, 1977. 263. Upon information and belief, approximately $13,900,000 was spent by defendant NORTHWEST INDUSTRIES, INC. on research and development in 1977, compared with approximately $12,300,000 in 1976. 264. According to forms filed with the Securities & Exchange Commission of the United States by defendant NORTHWEST INDUSTRIES, INC. for 1977, "Compliance with present environmental regulations, has not had, nor is it expected to have, any material effect on said defendant NORTHWEST INDUSTRIES, INC." 265. Upon information and belief, according to the report of the -page 74Aiaendcd Verified Co:nplainc 5 January 1979 �May 9, 1978 Annual Meeting of defendant NORTHWEST INDUSTRIES, INC., 88 banks own about 3.3 million shares and 37 funds own another 1.6 million shares of the stock of defendant NORTHWEST INDUSTRIES, INC., in all about one-third of the outstanding stock of the corporation. Defendant NORTHWEST INDUSTRIES, INC.: Operations 266. Upon information and belief, sales of defendant NORTHWEST INDUSTRIES, INC. for 1977 increased $307,600,000, up 19.6% over 1976, while net earnings increased $11,800,000, up 10%. Upon information and belief, the operations of Coca-Cola of Los Angeles are Included in consolidated results only from November 1, 1977, the effective date of acquisition, however,- even excluding the Coca-Cola of Los Angeles contribution, sales and earnings, were at record levels and improved $267,600,000, or +17.1%, and $10,800,000, or 4-9.2%, respectively, over 1976. 267. Upon information and belief, sales of all three product groups of defendant NORTHWEST INDUSTRIES, INC. exclusive of Coca-Cola of Los Angeles improved sharply over 1976. 268. Upon information and belief, the Industrial Group (up 20%) was led by record tubular steel sales,•especially in oil country products, as a result of increased drilling activity and because field inventories, which were high in 1976, came into better balance. Upon information and belief, connecting device and electrical lamp ballast sales were also records. 269. Upon information and belief, the Chemical Group's improvement (up 20% from a weak 1976) stemmed from higher pesticide sales in both the domestic and international markets. 270. Upon information and belief, the Consumer. Group's improvement (up 12.5%) over 1976 resulted principally from record battery and underwear sales, while other apparel sales were down. -page 75Amended Verified Complaint �271. Upon information and belief, gross earnings of defendant NORTHWEST INDUSTRIES, INC. increased $63,431,000, or 16.3%,-during 1977 slightly less than the 17.1% sales increase, as lower consumer margin outweighed the effect of chemical margin improvement, which was attributable according to statements by management of defendant NORTHWEST INDUSTRIES, INC., "to-increased sales of relatively high• margin pesticides." 272. Upon information and belief, during 1977 Industrial Group margin was approximately the same as 1976. 273. Upon information and belief, the major factors causing the consumer margin decline during 1977 were the higher cost of Scotch whisky and costs related to an 18-week stike at Acme Boot Company, Inc. . 274. Upon information and belief, earnings of defendant NORTHWEST INDUSTRIES, INC. before income taxes increased $29,700,000, or +14.6%, during 1977, a slower rate of increase than gross earnings principally as a result of_higher interest expense attributable to the cost of treasury stock purchases,, approximately $4,800,000, and higher interest rates. ' . 275. Upon information and belief, selling, administrative and general expenses were 10.2% of sales in both 1976 and 1977 in spite of significantly increased Cutty Sark advertising and higher chemical insurance and legal expenses mainly because higher tubular steel sales required minimal additional expense. 276. Upon in.t'ormation and belief, the 9.2% increase in net earnings during 1977 was sharply lower than the pretax increase as a result of lower investment tax credit. 277. Upon information and belief, there were a substantial number of the significant changes in the balance sheet of defendant NORTHWEST -page 76Amenueu Verifies Complaint 5 January 1979 �INDUSTRIES, INC. during 1977 compared with 1976, attributable to tho acquisition of Coca-Cola of Los Angeles and purchases of treasury stock. Excluding these effects, the balance sheet reflected growth in line with the strong increases in sales and earnings. 278..Upon information and belief, sales of defendant NORTHWEST INDUSTRIES, INC.for 1976 increased 32% over 1975 while net earnings increased 16%. Even excluding the results of Microdot, a manufacturer of connecting devices and ingot molds, in consolidated results from February 11, 1976, the date this company was acquired by defendant NORTHWEST INDUSTRIES, INC., sales and earnings registered improvement during 1976 over the record results of 1975, notwithstanding an extremely weak pesticide business and lower oil country tubular goods sales. 279. Upon information and belief, during 1976 the major favorable factor for the defendant NORTHWEST INDUSTRIES, INC. was significantly higher Consumer Group sales and earnings. 280. Upon information and belief, excluding the contribution of Microdot to 1976 results, record Consumer Group sales increased 16% over 1975 as demand for underwear, batteries, boots and jeans was strong. Industrial sales were about even with 1975. According to the reports published by defendant NORTHWEST INDUSTRIES, INC., lower oil country tubular goods sales caused by a temporary oversupply of field inventories were virtually offset by strong sales of flourescent ballasts. Defendant NORTHWEST INDUSTRIES, INC. stated in its reports that the oversupply of field inventories of oil country goods resulted in part from a generally unfavorable and uncertain legislative climate which affected oil and gas well drilling in the United States.- According to reports published by defendant NORTHWEST INDUSTRIES, INC., Chemical Group sales were also about the same in 1976 as 1975 with the lower sales of pesticides being offset by strong sales of industrial chemicals. According to the reports published by defendant NORTHWEST INDUSTRIES, INC., tho pesticide business was affected by high field -page 77reci V e r i f i e d Complaint .Tanuary 1979 �inventories as well as a severe drought in Europe and adverse regulatory action. 281. According to reports published by defendant NORTHWEST INDUSTRIES, INC., gross earnings increased at a slower rate during 1976 than sales principally from- the lower sales of relatively high . margin pesticides and oil country pipe; while consumer group margin improved as the underwear and replacement battery markets recovered from the 1975 recession-affected levels. ^Increases in selling, administrative and general expenses were primarily volume related and interest expense was slightly lower than 1975 because of lower rates. 282. According to reports published by defendant NORTHWEST INDUSTRIES, INC., net earnings for 1976 also benefited from a higher investment tax credit, a large part of which was generated by the expansion of Lone Star Steel's production capacity; while the significant increase in the accrued tax liability over 1975 resulted from the accrual for currently payable federal income taxes which were not required in recent years because of the tax loss carryforward resulting from_the 1972 sale of transportation assets. Defendant NORTHWEST INDUSTRIES, INC.: Plants and Property 283. Upon information and belief, defendant NORTHWEST INDUSTRIES, INC. owns or leases property at the following locations as described: In Alabama, six facilities: four cotton mills which are owned; a leased shirt plant, which lease, upon information and belief, expires .in 1985; and an automotive battery plant which is owned. In Arizona, an agricultural chemical plant which is owned. In Arkansas, a bromine manufacturing facility and brine wells which are owned. .In California, fifteen facilities: two connecting device plants which are leased and which leases, upon information and belief, expire in 1982; a leased automotive battery plant which lease, upon information and belief, expires in -page 78Verified Cumy.laint 5 January 1979" ~ �1980; eleven bottling plants, nine of which are owned, and two of which are -leased and which leases, upon information and belief, expire in 1980; and a leased canning plant which lease, upon • information and belief, expires in 1987. In Colorado, a small diameter pipe mill which is owned. In Connecticut, a capacitor plant which is leased and which lease, upon information and belief/ expires in 1983. In Costa Rica, a connecting device plant which is owned. In England, a connecting device plant which is leased and which lease, upon information and belief/ expires in 1991. In France, two facilities: a connecting device plant which is owned, and a connecting device plant which is leased and which lease, upon information and belief, expires in 1981. In Hawaii, a bottling and canning plant which is owned. In Illinois, ten facilities: four connecting device plants which are owned and three connecting device plants which are leased and which leases, upon information and belief, expire from 1979-1985; an ingot mold plant which is leased and which • lease/ upon information and belief, expires in 1987; an agricultural and industrial chemical plant, and a research center which are owned. In Indiana, an automotive battery plant which is owned. In Iowa, a bottling plant which is owned. In Kansas, three facilities: an.automotive battery plant which is leased and which lease, upon information-and belief, expires in 2001; and two bottling and canning plants v/hich are owned. In Kentucky, five facilities: two underwear plants which are owned; an underwear plant which is leased and which lease, upon information and belief, expires in 1986; and two connecting device pl.ants which are owned. -page 7.9AmendecS V e r i f i e d Complaint 5 J a n u a r y 1979 �In Louisiana, three facilites: two underwear plants which are leased and which leases, upon information and belief, expire in 1986 and 1990; and a lead smelter plant which is owned. In Massachusetts, a footwear plant which is owned. In Michigan, -seven facilities: five connecting device plants, four of which are owned, and one of which is leased and which lease, upon information and belief, expires in 1983; a general manufacturing facility and brine wells, and a research center all of which are owned. In Mississippi, five facilities: a ballast assembly plant which is leased and which lease, upon information and belief, expires in 1989; a wire mill which is leased and which lease, upon information and belief, expires in 1997; a steel slitting mill which is owned; a leased underwear plant which lease, upon information and belief, expires in 1990; and a leased underwear distribution center which lease, upon information and belief, expires in 1988. In Missouri, a leased bottling plant which lease, upon information and belief, expires in 1982. In Nebraska, a leased bottling plant which • lease, upon information and belief, expires in 1996. In Nevada, a bottling plant which is owned. In New Hampshire, a connecting device plant which is owned. In New Jersey, tv/o ballast assembly plants which are leased and which leases, upon information and belief, expire in 1983 and 1989. In North Carolina, four facilities: two cotton mills which are owned; an underwear plant^ which is owned; and a connecting device plant which is ov/ned. In Ohio, four facilities: a connecting device plant which is owned; two ingot mold plants which are owned; and a battery charger plant which is owned. -page 80Verified Complaint 5 January 1979 �In Oklahoma, a leased underwear plant which lease, upon information and belief, expires in 1985. In Oregon, an automotive battery plant which is owned. In Pennsylvania, six facilities: an automotive battery plant which are owned; a leased automotive battery plant, which lease, upon information and belief, expires in 1986; an industrial battery plant which is owned-; a lead smelter plant which is owned; an injection molding plant which is owned; and a leased connecting device plant, which lease, upon information and belief, expires in 1985.. In Puerto Rico, a leased automotive battery plant, which lease, upon information and belief, expires in 1982. In South Carolina, two facilities: a woven elastic plant which is owned and an automotive battery plant which is owned. In Tennessee, twelve facilities: a leased jeans plant, which lease, upon information and belief, expires in 1988; a leased shirt plant, which lease, upon information and belief, expires in 1986; a leased jeans and shirt plant, which lease, upon information and belief, expires in 1984; five leased boot plants which leases, upon information and belief, expire from 1978-1984; two connecting device plants which are owned; and two agricultural and industrial chemical plants -which are owned. In Texas, eight facilities: two agricultural and industrial chemical plants which are owned.; basic- steel manufacturing facilities consisting of a blast furnace, 78 coke ovens, a five-furnace open hearth shop, standard ingot casting facilities, a slabbing mill, two electric furnaces, continuous caster, two extrusion presses, and mills and machines for the production of skelp, pipe and other products; an ore beneficiation plant and a tailing basin all of"which facilities ere owned; a jeans plant which is leased and which lease, upon information and belief, expires in 1982; a leased boot plant, which lease, upon information and belief, expires in 1979; a leased automotive battery plant, which lease, upon information and belief, expires in 1984; a lead smelter plant which is owned; and a bottling plant which is owned. -page 8'1,Amenued Veritied Complaint 5 January 1979 �In Utah, a connecting device plant which is leased and which lease, upon information and belief, expires in 1985. In Virginia, three jeans and ladies' fashion plants which are owned. In Wisconsin, a bottling plant which is ov/ned. 284. Upon information and belief, the properties in its Industrial Group are operated by defendant NORTHWEST INDUSTRIES, INC. at approximately 70-75% of capacity. 285. Upon information and belief, those properties in its Chemical Group are operated by defendant NORTHWEST INDUSTRIES, INC. at approximately 60% of capacity. 286. Upon information and belief, those properties in its Consumer Group are operated by defendant NORTHWEST INDUSTRIES, INC. at approximately 80-100% of capacity. • -page 82VeriLied Complaint '5 January 1979 • �Plaintiffs Complain 287. Upon information and belief, the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. warranted and represented that phenoxy herbicides such as 2,4,5trichloro phenoxyacetic acid (2",4,5-T) even though contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo pdioxin (TCDD or "dioxin") were safe and fit for the uses advertised and promoted for such contaminated phenoxy herbicides by said corporated defendants. 288. Upon information and belief, the phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") was not safe and fit for the uses advertised and promoted and was therefore mislabled, misbranded and misrepresented by said corporate defendants. 289. Upon information and belief, the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. knew, or with the exercise of reasonable concern for the public health, safety and welfare should have known, that phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo pdioxin (TCDD or "dioxin") were inherently dangerous and toxic. 290. Upon information and belief, the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, anr! NORTHWEST INDUSTRIES, INC. deliberately, intentionally and with knowledge of the potential toxic effects, manufactured, formulated, advertised, promoted, marketed and sold phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with the toxic synthetic organic -page 83Amenaeu V e r i f i e d Complaint 5 January.1979 �chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") in wanton and reckless disregard of the public health, safety and welfare. 291. Upon information and belief, the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC., individually and collectively, -jointly and severally, their several agents, servants and/or employees failed to warn the deceased plaintiff PAUL REUTERSHAN and all those other 4.2 million American service persons so unfortunate as to be similarly at risk, of toxic effects attributable to phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"). 292. Upon information and belief, the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. failed to warn the United States Department of Defense, the United States Department of Health, Education and Welfare, the United States Department of Interior, and/or the United States Department of Agriculture, much less the American people of the dangers associated with the handling and use of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"). 293. The corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. failed to exercise reasonable care and caution or demonstate resposible corporate concern for the public health, safety and welfare commensurate with the economic stature of said corporate, defendants when they failed to warn the deceased plaintiff PAUL REUTERSHAN and all the other 4.2 million American servicemen so unfortunate .as to be similarly at risk of the dangers associated with exposure to phenoxy herbicides such as 2,4,5trichloro phenoxyacetic acid (2,4,5-T) contaminated with the toxic -page 84Ariieri(.J«ti V e r i f i e d Cowjjiaant 5 J a n u a r y 1979 �synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") . 294. Upon information and belief, the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. failed to adequately test the phenoxy herbicides such as 2,4,5trichloro phenoxyacetic acid (2,4,5-T) which said corporate defendants knew, or with the exercise or reasonable concern for the public health, safety, and welfare should have known, were contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo pdioxin (TCDD or "dioxin") in order to ascertain the dangerous nature and toxic characteristics of such contaminants, before advertising, promoting and marketing such contaminated phenoxy herbicides. 295. Upon information and belief, in the event the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. did conduct tests to ascertain the dangerous nature and toxic characteristics of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") said corporate defendants deliberately, intentionally, wilfully, and wantonly failed to publish the results of such tests in a manner which would have alerted the American people to the extent of the risk associated with exposure to such contaminated phenoxy herbicides. 296. Upon information and belief, the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. have refused or failed to remove phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic' acid (2,4,5-T) contaminated with the toxic synthetic organic chemical 2-, 3 , 7 , C--tetrachloro dibenzo p-dioxin (TCDD or "dioxin") from international and interstate commerce and local markets when said corporate defendants knew, or with reasonable concern for the public health, safety and welfare should have known, of the risk associated with exposure to such contaminated phenoxy -page 85Vfi." i L itrij Coiujjlciin t 5 J a n u a r y 1979 �i cides. 297. Upon information and belief, the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. have violated the common law and/or the statutory laws of the United States of America and/or -the laws of the several states in • intentionally advertising, promoting, marketing, and selling phenoxy herbicides such as 2,4,5-trichloro phenoxy acetic acid (2,4,5-T) con- taminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") which such contaminated herbicides were, and still are, inherently dangerous and unsafe for use as advertised and promoted. 298. Upon information and belief, the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES, INC., and NORTHWEST INDUSTRIES, INC., their subsidiaries, associated venturers, agents and employees, will continue to manufacture, distribute, advertise, market, sell and promote the use of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with toxic synthetic organic chemicals such as 2,3, 7, 8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") unless otherwise restrained by this Court. 299. That the continued manufacture, distribution, advertisement, marketing, sale, and promotion of phenoxy herbicides such as 2,4,5trichloro phenoxyacetic acid (2,4,5-T) contaminated with toxic synthetic organic chemicals such as 2, 3, 7, 8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES, INC., and NORTHWEST INDUSTRIES, INC. presents an imminent danger of serious, permanent, and irreparable damage to the public health, safety and welfare. -page 86rtnienceu V e r i f i e d 'Complaint 5 J a n u a r y 1979 �300. That the the contamination of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T). manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") represents a significant potential for illness, disability and death to all those so unfortunate as to be situated at risk from such exposure to phenoxy herbicides so contaminated. 301. The many human beings such as the deceased plaintiff PAUL REUTERSHAN, and all those others so unfortunate as to have been similarly situated were placed in jeopardy as a result of the manufacture, distribution, advertising, marketing," promotion and sale of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES, INC., and NORTHWEST INDUSTRIES, INC. 302. That the serious, permanent, and irreparable damage, to the public health, safety, and welfare resulting from the contuinued efforts of. the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES, INC., and NORTHWEST INDUSTRIES, INC. to manufacture, distribute, advertise, market, sell and promote the use of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) con-. taminated with toxic synthetic organic chemicals such as 2,3,7,8tetrachloro dibenzo p-dioxin (TCDD or "dioxin") in wanton and reckless disregard for the toxic effects of such contaminants upon the public health, safety and welfare violates the constitutional rights of those so unfortunate as to be at risk from the toxic effects of such contaminants not only during this generation but in succeeding generations, which rights have been retained by the People under the Ninth Amendment of the Constitution and as such are protected under the "due -page 87G V«2r i i. i«=J Coii 5 January 19.79 �process" and "equal protection" clauses of the Fifth Amendment of the Constitution and the "privileges or immunities," "due process" and "equal protection" clauses of the Fourteenth Amendment of the Constitution. 303. That the deceased plaintiff PAUL REUTERSHAN, and all those others so unfortunate as to have been similarly situtated, have already been, and to the extent that such individuals are still alive will continue to be, seriously, permanently and irreparably damaged as a result of the actions of the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES, INC., and NORTHWEST INDUSTRIES, INC. which individually and collectively, alone and in concert- with others, did manufacture, distribute, advertise, market, sell and promote the use of phenoxy. herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) which said corporate defendants knew, or with the exercise of reasonable care and caution should have known, were contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxiri (TCDD). 304. That all those so unfortunate as to be at risk as a result of the toxic effects of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with toxic synthetic organic chemicals such as 2,3,7,8tetrachloro dibenzo p-dioxin (TCDD or "dioxin") will continue to be seriously, permanently and irreparably damaged as a result of the actions of the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES, INC., and NORTHWEST INDUSTRIES, INC. as those corporate defendants, individually "and collectively, alone and in concert with others, continue to manufacture, distribute, advertise, market, sell and promote the use of phenoxy herbicides such as 2,4., 5-tr ichloro phenoxyacetic acid (2,4,5-T) contaminated with'toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD) without adequate regard'for its toxic effects and in wanton and reckless • -page 38cJ Vet i i" ic.-u .CcjYipi i> in t 5 J a n u a r y 1979 �disregard of the public health, safety and welfare. 305. Upon information and belief, the estimated cost of treating ' any of the malignant conditions resulting from exposure to phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) con- • taminated with toxic synthetic organic chemicals such as 2,3,7,8tetrachloro dibenzo p-dioxin (TCDD or "dioxin") will exceed $10,000 for each year that such condition continues during the life of the victim. 306. Upon information and belief, the estimated economic loss to the families of those victims of malignant conditions or cancers resulting from exposure to phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") will exceed $10,000 for each year that such victims could have been reasonably expected to have been employed and contributed to the maintenance and care of their families. 310. Upon information and belief, the deceased plaintiff PAUL REUTERSHAN was exposed to phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, fabricated, advertised, promoted, marketed and sold by the corporate defendants corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"), which exposure contributed to his illness, disability and and eventual death, while as many as 4.2 million other American servicemen may have been similarly exposed and as a'result of such exposure may similarly become ill, disabled, and eventually die as a result of exposure to phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marKeted and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with the toxic synthetic organic chemical -page 89VeriLieo Complaint 5 January J979 �2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"). 311. That as a result of exposure to phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") the deceased plaintiff PAUL REUTERSHAN became ill and disabled, suffered pain and economic loss, and incurred medical expenses before being released eventually from such anguish and torment through death. 312. That as a result of exposure to phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") as many as 4.2 million American servicemen and an indeterminate number of other persons are at risk of becoming ill and disabled, suffering pain and economic loss, and incurring medical expenses and eventually dying as a result of such illness as did the deceased plaintiff, PAUL REUTHERSHAN. 313. That the illness, disability and eventual death of the plaintiff PAUL REUTERSHAN was attributable, directly or indirectly to the acts of the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC., which individually and collectively, jointly and severally, independently and in concert with each other manufactured, "formulated, advertised, promoted, marketed and sold phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") in wanton and reckless disregard of the public health, safety and welfare; as a result of which, the deceased plaintiff: PAUL REUTERSHAN has sustained, -page 90Amendod Verified Complaint �and all the other 4.2 million American servicemen together with the indeterminate number of other persons, all of whom are so unfortunate as to be similarly at risk, may sustain, general and special damages. • -. • l"\ " �Declaratory Judgment Claims 314. This is a proceeding seeking a Judgment declaring the rights and legal relations of the parties to the matter in controversy specifically: 315. DECLARING the rights of the deceased Plaintiff and all the other people of the United States, to the protection of their personal health, safety and welfare from the toxic effects of phenoxy herbicides such as 2,4,5-trichlorophenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"). 316. DECLARING the rights of the deceased Plaintiff and all the people of the United States, not only of this generation but of those generations yet unborn, to protection from the wanton and reckless disregard of the public health, safety and welfare by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES, INC., and NORTHWEST INDUSTRIES, INC., their several subsidiaries, associated business venturers, agents, and employees, jointly or severally, individually or in concert with each other in the manufacture, formulation, advertising, promotion, marketing and sale of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"). 317. DECLARING the rights'of. the deceased Plaintiff and all the people of the United States not only during this generation, but those generations yet unborn, to the full benefit, use and enjoyment of the Environment (considered as the comp-lex system resulting from the union, in the mathematical sense of that term, of natural, social, and societal resources together with all the interactions among those • • -page 92Amended Verified Complaint : 5 January 1979 �resources in accordance with the natural, social, and societal processes operative upon those resources) without damage or degradation attributable to the toxic effects of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") and manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. . 318. DECLARING the rights of all the people of the United States, not only of this generation, but of those generations yet unborn, to the protection of their personal health, safety and welfare, and the protection of the Environment (considered as the complex system resulting from the union, in the mathematical sense of that term, of natural, social, and societal resources together with all the interactions among those resources in accordance with the natural, social, and societal processes operative upon those resources) on which they each and all depend, from injury, damage, and degradation as a result of the toxic effects of phenoxy herbicides such as 2,4,5trichloro phenoxyacetic acid (2,4,5-T) contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") and manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. 319. DECLARING that the actions of the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES, INC., and NORTHWEST INDUSTRIES, INC. in marketing, distributing, advertising, selling and promoting the use of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acic] (2,4,5-T) contaminated with toxic synthetic o-rganic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") represented and still now represents a threat to the-public health, safety and welfare and so violates the rights retained by the deceased .plaintiff and all the people of the United States not only of this generation, -page 93Amended Vor-iCied Complaint b January 1979 �but of those generations yet unborn, under the Ninth Amendment of the Constitution of the United States, and as such protected under the "due process" and "equal protection" clauses of the Fifth Amendment and the "due process," "equal protection," and "privileges or immunities" clauses of the Fourteenth Amendment of the Constitution of the United States. • . 320. DECLARING that the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES, INC., and NORTHWEST INDUSTRIES, INC. are Trustees of the public health, safety and welfare to the extent that, such corporate defendants actually had, or now have, or with the exercise of reasonable concern for-the public health, safety and welfare should have had or should now have, knowledge of the dangerous properties and toxic characteristics of the products they manufacture, advertise, promote, market, and sell, in particular phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"). 321. DECLARING that the continued promotion of the use of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate .defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. while known to be contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") is an intentional tort. -page 94Amonded Verified Complaint 5 January 1979 �Equitable Jurisdiction 322. This action is brought in equity before this court on the following grounds: 323. The subject matter is equitable in nature. • 324. This action is brought for the purpose of restraining the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES, INC., and NORTHWEST INDUSTRIES', INC. from doing serious, permanent and irreparable injury which cannot be adequately compensated in merely money damages to the class represented by now deceased Plaintiff.. 325. The declaratory judgment and injunctive relief demanded on behalf of the class are equitable remedies and the substantive character of the rights sought to be enforced by the Plaintiff are historically those resolved in a court of equity. 326. The law does not afford any adequate remedy for the wrong resulting from the actions of the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES, INC., and NORTHWEST INDUSTRIES, INC. 327. There is no plain, adequate and complete remedy at law as practical and efficient as the equitable relief sought herein, nor are the damages sustained by the class on whose behalf this action is brought as a result of the actions of the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES, INC., and NORTHWEST INDUSTRIES, INC. capable of adequate recovery in any action at law. . -page 95Amended Verified Complaint . 5 January 1979 - �WHEREFORE judgment is demanded on behalf of the plaintiff class herein against the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES, INC., and NORTHWEST INDUSTRIES, INC., jointly and severally, individually and collectively: DECLARING the rights of the deceased Plaintiff and all the other people of the United States, to the protection of their personal health, safety and welfare from the toxic effects of phenoxy herbicides such as 2,4,5-trichlorophenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, . promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"). DECLARING the rights of the deceased Plaintiff and all the people of the United States, not only of this generation but of those generations yet unborn, to protection from the wanton and reckless disregard of the public health, safety and welfare by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES, INC., and NORTHWEST INDUSTRIES, INC., their several subsidiaries, associated business venturers, agents, and employees, jointly or severally, individually or in concert with each other in the manufacture, formulation, advertising, promotion, marketing and sale of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"). . DECLARING the rights of the deceased Plaintiff and all the people of the United States not only during this generation, but those generations yet unborn, to the full benefit, use and enjoyment of the Environment (considered as v- • ' -page 96Amended Verified Complaint -1/7/79- �the complex system resulting from the union, in the mathematical sense of that term, of natural, social, and societal resources together with all the interactions .among those resources in accordance with the natural, social, and societal processes operative upon those resources) without damage or degradation attributable to the toxic effects of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") and manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. DECLARING the rights of all the people of the United States, not only of this generation, but of those generations yet unborn, to the protection of their personal health, safety and welfare, and the protection of the Environment (considered as the complex system resulting from the union, in the mathematical sense of that terra, of natural, social, and societal resources together with all the interactions among those resources in accordance with the natural, social, and societal processes operative upon those resources) on which they each and all depend, from injury, damage, and degradation as a result of the toxic effects of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with the toxic synthetic organic chemical 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") and manufactured, formulated, advertised, promoted, marketed and sold by.the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. DECLARING that the actions of the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES, INC., and NORTHWEST -page 97A.TIPndod Ve r i £ i ocl Coir.p 1 a'ln t -1/7/79- * . �INDUSTRIES, INC. in marketing, distributing, advertising, selling and promoting the use of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") represented and still now represents a -threat to the public health, safety, and welfare and so violates the rights retained by the deceased plaintiff and all the people of the United States not only of this generation, but of those generations yet unborn, under the Ninth Amendment of the Constitution of the United States, and as such protected under the "due process" and "equal protection" clauses of the Fifth Amendment and the "due process," "equal protection," and "privileges or immunities" clauses of the Fourteenth Amendment of the Constitution of the United States. DECLARING that the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES, INC., and NORTHWEST INDUSTRIES, INC. are Trustees of the public health, safety and welfare to the extent that such corporate defendants actually had, or now have, or with the exercise of reasonable concern for the public health, safety and welfare should have had or should now have, knowledge of the dangerous properties and toxic characteristics of the products they manufacture, advertise, promote, market, and sell, in particular.phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"). DECLARING that the continued'promotion of the use of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the 'corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED,"and NORTHWEST INDUSTRIES, INC. while known to be contaminated with toxic G 98-. Ar.ionded V e r i f i e d Complaint -1/7/79- �synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin '(TCDD or "dioxin") is an intentional tort. IMPOSING upon the corporate defendants TH'E DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES,INC., as Trustees of the public health,.safety and welfare, a non-delegable duty to disclose to the public as part of their efforts to advertise, promote, market and sell phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) whatever information they now have, or with the exercise of reasonable concern for the public health, safety and welfare should have, concerning the dangerous properties and toxic characteristics of such phenoxy herbicides which are contaminated with the toxic synthetic organic chemical 2,3,7,8tetrachloro dibenzo p-dioxin (TCDD or "dioxin"). DIRECTING the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. as Trustees of 'the public health, safety, and welfare TO DISCLOSE to the general public as part of their efforts to advertise, promote, market and sell phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") whatever information they have or with the exercise of reasonable concern for the public health, safety, and welfare 'should have, concerning the dangerous properties and toxic character of such products in order to permit the public to-accurately essess the risk associated with the continued use of such products as advertised and promoted. • RESTRAINING the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. from selling, offering for salo, marketing, distributing, advertising and promoting the use of phenoxy -pag c 99Amendod Verified Complaint -.1/7/79- �herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin") unless and until such time as said corporate defendants can establish that such advertised and promoted use will not present a "risk of injury and damage to the public health, safety and welfare. DIRECTING the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. to establish out of current corporate earnings a fund in the nature of a reserve against losses sufficient: 1. to REIMBURSE the United States Department of Defense, in particular the Veterans Administration, and the United States Department of Health, Education and Welfare, in particular the Social Security Administration, for the full and complete cost of any and all medical care and treatment rendered by and through such agencies unto the deceased plaintiff and any others so unfortunate to be similarly in need of treatment to the extent that such costs .may be attributable to the toxic effects of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with toxic synthetic organic chemicals such as 2,3,7,8tetrachloro dibenzo p-dioxin (TCDD or "dioxin"); 2. to REIMBURSE the United States Department of Defense, in particular the Veterans Administration, and the United States Department of Health, Education and Vi'olfaco, in particular the Social Security Administ r a t i o n , for any and all payments made to the deceased _ -page 100A:nonded V o r i E i o d Corr.pl.a in L -1/7/79- �plaintiff and/or his family and all those others so unfortunate as to be similar recipients of such benefits as compensation or 'benefits for disability attributable to the toxic effects of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated," advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"); 3. to COMPENSATE the deceased plaintiff and the legal representative of said deceased plaintiff together with all those others so unfortunate to be similarly situated, and their families, for the full and complete economic loss and damages resulting from any disability attributable to the effects of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"); 4. to REIMBURSE all the stockholders, rate payers and subscribers of public utilities for any costs arid expenses incurred by such public utilities as a result of the use "of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid ("2,4,5-T} manuf actured, formulated, advertised, promoted, marketed and.sold by the corporate defendants THE DOK CHEMICAL COMPANY, HERCULES INCORPORATED, and' NORTHWfc'.ST INDUSTRIES, INC. and contaminated with toxic synthetic • organic chemicals such as 2,3,7,8-pnO'1 .101Amendod V e r i f i e d Complaint -L/7/79- �tetrachloro dibenzo p-dioxin (TCDD or "dioxin"); TO APPOINT AND DESIGNATE, subject to the continuing jurisdiction and direction of this Court/ an appropriate representative of the deceased Plaintiff as Trustee of such a fund for the purpose of administering such a fund for the benefit of said deceased plaintiff and all those others so unfortunate as to be at risk not only during this generation but during those generations yet to come from the toxic effects of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with toxic synthetic organic chemicals such as 2,3/7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"). DIRECTING the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES, INC., and NORTHWEST INDUSTRIES, INC. to make payments from such a fund in accordance with the further directions of this Court: 1. to REIMBURSE the United States Department of Defense, in particular the Veterans Administration, and the United States Department of Health, Education and Welfare, in particular the Social Security Administration, for the full and complete cost of any and all medical care and treatment rendered by and through such agencies unto the deceased plaintiff and any others so unfortunate to be similarly in need of treatment to the extent that such costs may be attributable to the toxic effects.of phenoxy herbicides such as 2,4,5-trichloro phenoxya-cct.ic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed ana sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPO- -page 102Verified Cono.laint -1/7/79- - �RATED, and NORTHWEST INDUSTRIES, INC. and contaminated with toxic synthetic organic chemicals such as 2,3,7,8tetrachloro dibenzo p-dioxin (TCDD or "dioxin"); 2. to REIMBURSE the United States Department of Defense, in particular the Veterans Administration, and the United States Department of Health, Education and Welfare, in particular the Social Security Administration, for any and-all payments made to the deceased plaintiff and/or his family and all those others so unfortunate as to be similar recipients of such benefits as compensation or benefits for disability attributable to the toxic effects of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with toxic synthetic organic chemicals such as 2,3,7,8-tetrachloro dibenzo p-dioxin (TCDD or "dioxin"); 3. to COMPENSATE the deceased plaintiff and the legal representative of said deceased plaintiff together with all those others so unfortunate to be similarly situated, and their families, for the full and complete economic loss and damages resulting from any disability attributable to the effects of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured, formulated, advertised, promoted, marketed and sold by the-corporate defendants 'THE DOW CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. an_d contaminated with toxic synthetic organic chemicals such as 2,3>7,8-tetrachloro dibenzo p-dioxin (TCDD or" "dioxin" ).; 4. to_ REIMBURSE all -the stockholders, rate payers" -page 101An;cnded Verified Complaint -1/7/79- �and subscribecs of public utilities for any costs and expenses incurred by such public utilities as a result of the use of phenoxy herbicides such as 2,4,5-trichloro phenoxyacetic acid (2,4,5-T) manufactured,, formulated, advertised, promoted, marketed and sold by the corporate defendants THE DOW.CHEMICAL COMPANY, HERCULES INCORPORATED, and NORTHWEST INDUSTRIES, INC. and contaminated with toxic synthetic organic chemicals such as 2,3,7,8tetrachloro dibenzo p-dioxin (TCDD or "dioxin"); ALL TOGETHER with such other and further relief as to this Court shall seem just and proper under the circumstances. DATED at Patchogue, New York,' 5 January 1/979; VictorNjohn\Yannaeone, jr YANNACONE \JYANNACONE Professional Corporation of counsel to Plaintiffs' Attorneys Office & P.O. Address Post Office Drawer £109 Patchogue, New York 11772 (area code 516) 654-2299 REILLY & GORMAN Plaintiffs' Attorneys Office & P.O. Address 444 Main Street Islip, New York 11751 ' (area code 516) . 581-8687 -page 104-Ver if led -Corrplaint -1/7/79- �ATTORNEYS' VERIFICATION STATE OF NEW YORK ) COUNTY OF SUFFOLK ) VICTOR JOHN YANNACONE, jr., duly affirming under the penalties of perjury states that he is an attorney duly licensed to practice law in the State of New York, and admitted to practice before the United States District Court for the Southern District of New York, and the United States District Court for the Eastern District of New York, and the United States Court of Appeals for the Second Circuit and the Supreme Court of the United States; that he has read the foregoing amended complaint and knows the contents thereof; that the same are true to his own knowledge based upon documents contained in affirmant's files, except as to those portions therein stated to be alleged upon information and belief, and as to those portions, affirmant believes them to be true. Affirraant submits this affirmation based upon information contained in the files of his firm and because his office is located in Suffolk County, New York and the home of the deceased plaintiff and his family is in Westchester County. Duly affirmed under the penalties of perjury this 5th day of January, 1979 Attorney Xor Plaintiffs xj I i � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Alvin L. Young Collection on Agent Orange Description An account of the resource <p style="margin-top: -1em; line-height: 1.2em;">The Alvin L. Young Collection on Agent Orange comprises 120 linear feet and spans the late 1800s to 2005; however, the bulk of the coverage is from the 1960s to the 1980s and there are many undated items. The collection was donated to Special Collections of the National Agricultural Library in 1985 by Dr. Alvin L. Young (1942- ). Dr. Young developed the collection as he conducted extensive research on the military defoliant Agent Orange. The collection is in good condition and includes letters, memoranda, books, reports, press releases, journal and newspaper clippings, field logs and notebooks, newsletters, maps, booklets and pamphlets, photographs, memorabilia, and audiotapes of an interview with Dr. Young.</p> <p>For more about this collection, <a href="/exhibits/speccoll/exhibits/show/alvin-l--young-collection-on-a">view the Agent Orange Exhibit.</a></p> Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Box The box containing the original item. 179 Folder The folder containing the original item. 5254 Series The series number of the original item. Series VIII Subseries I Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Creator An entity primarily responsible for making the resource Yannacone, Victor John, Jr. Date A point or period of time associated with an event in the lifecycle of the resource January 5 1979 Title A name given to the resource Amended Verified Complaint, January 5, 1979 ao_seriesVIII Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Alvin L. Young Collection on Agent Orange Description An account of the resource <p style="margin-top: -1em; line-height: 1.2em;">The Alvin L. Young Collection on Agent Orange comprises 120 linear feet and spans the late 1800s to 2005; however, the bulk of the coverage is from the 1960s to the 1980s and there are many undated items. The collection was donated to Special Collections of the National Agricultural Library in 1985 by Dr. Alvin L. Young (1942- ). Dr. Young developed the collection as he conducted extensive research on the military defoliant Agent Orange. The collection is in good condition and includes letters, memoranda, books, reports, press releases, journal and newspaper clippings, field logs and notebooks, newsletters, maps, booklets and pamphlets, photographs, memorabilia, and audiotapes of an interview with Dr. Young.</p> <p>For more about this collection, <a href="/exhibits/speccoll/exhibits/show/alvin-l--young-collection-on-a">view the Agent Orange Exhibit.</a></p> Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Box The box containing the original item. 197 Folder The folder containing the original item. 5544 Series The series number of the original item. Series VIII Subseries I Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Creator An entity primarily responsible for making the resource Olexa, Michael T. Paul W. Bergman Title A name given to the resource An Ounce of Prevention is Worth a Pound of Cure: Chemigation and Legal Liability ao_seriesVIII https://www.nal.usda.gov/exhibits/speccoll/files/original/48d7d20bf51eceaf8b6cf60681f46ee2.pdf f8bfb3f0433306a9be117423add26327 PDF Text Text Item D Number °5447 Author Barnes, Donald Not Scanned Corporate Author Report/Article Title An Overview on Dioxin Journal/Book Title EPA journal Year 1983 Month/Day November Color D Number of Images ° Descrfyton Notes Friday, March 15, 2002 Page 5447 of 5571 �An Overview on Dioxin By Donald Barnes (Excerpts from testimony before the Subcommittee on Natural Resources, Agricultural Research and Environment of the House Committee on Science and Technology, June 30, 1983) ly testimony this morning focuses on 2,3,7,8-tetrachloro-dibenzo-p-dioxin (2,3,7,8TCDD), what is generally considered to be the most toxic of the 75 chlorinated dibenzop-dioxins (CDDs). Although other CDDs are present in the environment and are being addressed, most public attention is being focused on this particular dioxin. My remarks are divided into three sections. In the first, I will briefly describe some of the situations in which EPA has been involved with 2,3,7,8TCDD in the environment. Next, I will discuss the data and the methods we have used to assess the potential for human health effects in these cases. Finally, I will describe some of the gaps in our knowledge about 2,3,7,8-TCDD and the research that would help fill those gaps. EPA first became aware of the hazards associated with 2,3,7,8-TCDD through laboratory animal studies conducted in the 1960s and early 1970s. At that time, the scope of the Agency's "dioxin problem" was defined by the presence of 2,3,7,8-TCDD as an unavoidable contaminant in certain pesticide products. During the 1970s, the Agency took action to restrict the use of certain of those pesticide products and to obtain more information about the toxicity of 2,3,7,8-TCDD and analytical methods for detecting its presence in the environment. Some of these efforts involved extensive cooperation between EPA, various academic institutions and environmental groups, other Federal agencies, and industry. By the end of the decade, this cooperative venture had succeeded in developing a reliable method to detect TCDD in some media in the low parts per trillion range. (One part per trillion is roughly equivalent to the thickness of a human hair compared to the distance across the United States.) Donald Barnes is Science Advisor to the Assistant Administrator for Pesticides arid Toxic Substances. He has been cha/iman since 1980 of EPA's Chlorinated Dioxins Work Group, which has been assisting in the coordination of FPA's involvement in dioxin-related matters Barnes has also been EPA's representative on the Cabinet Council's Agent Orange Work Group. 16 In 1979, based on extensive animal data and epidemiologic information, the Agency took emergency action to suspend certain uses of 2,4,5-T and Silvex, two pesticide products which contain 2,3,7,8-TCDD. That ban remains in effect at this time. Also in the late 1970s, the Agency took action in connection with a series of dumpsites along the Niagara River in New York, some of which were found to contain 2,3,7,8-TCDD wastes. These,wastes were found along with a range of other hazardous substances which had resulted from previous manufacturing operations in the area. During the same time period, the Agency provided technical assistance in the successful cleanup of a smaller dioxin-contaminated dumpsite in Missouri. By 1979, the possibility of a range of dioxin emissions from combustion processes had become an issue. During this period, the Agency carefully investigated the question of the emission of 2,3,7,8-TCDD, other TCDD isomers, and tetrachlorinated dibenzofurans (TCDFs) during the combustion of polychlori nated biphenyls (PCBs) at two hazardous waste incinerators in the midwest. In 1980 and 1981, the Agency participated on a United States team, headed by the Food and Drug Administration (FDA), which met with Canadian officials to determine the presence of 2,3,7,8-TCDD in fish in the Great Lakes, assess the significance of these findings, and discuss ways to reduce or remove any sources of contamination. In 1980, the Agency issued a rule that requires 60-day notification to EPA prior to the disposal of most 2,3,7,8-TCDD contami nated manufacturing wastes. This 60-day period gives the Agency the opportunity to assess the risks associated with the proposed disposal and to take action if those risks are judged too unreasonable. In 1981, furthering its assessment of the emissions from combustion processes, EPA completed a series of studies of TCDD emissions during the combustion of municipal wastes. TCDDs, including small amounts of 2,3,7,8-TCDD, were detected at four of five facilities sampled. An interim evaluation of the significance of these TCDD emissions for human health was issued in November 1981, and it was concluded thai the emissions "do not present a public heal hazard for residents living in the immediate vicinity" of the facilities tested. More recently, the Agency has been acti* in identifying sites in several states, predominantly in Missouri, which have beor contaminated with 2,3,7,8-TCDD as a consequence of manufacturing activities or the injudicious disposition of wastes. Finally, I would like to mention that an EPA Task Force on Dioxins, with represent, lives from several program offices, is currently developing an overall strategy which will recommend specific actions and coordination mechanisms to address the wide range of dioxin questions. The top * management at EPA now has this strategy under review. I have included this chronology neither ti seek commendation nor to evoke sympathy but rather to illustrate that the Agency is n< stranger to 2,3,7,8-TCDD and the challenge, it presents to those required to make decisions regarding unreasonable risks to human health and the environment. The data base on 2,3,7,8-TCDD toxicity i extensive, but certainly not exhaustive. Mm of what we know has been obtained from animal studies. For example, we know that the material is lethal to a variety of animal species when administered in single, small doses (less than a millionth of a gram in some species). We know that there is a 1000-fold range of toxic response among various species in these lethality studies. W know that 2,3,7,8-TCDD is carcinogenic in rats and mice at very low doses (via both ingestion and dermal absorption), resulting a variety of tumors in these animals. We know that, as a carcinogen, it can at least behave as a promoter, a compound capabU of eliciting frank carcinogenesis in animals which have been previously exposed to oth carcinogens, and as a cocarcinogen. We know that the compound can interfere with reproductive success in females, especially pregnant ones, of several species (including rats, mice, rabbits, and monkeys), often at very low doses. We know that the material can affect elements of the immune system in test animals. In addition, there are a EPAJOURN/s �number of other effects which have been observed, including organ damage (for example, to the liver arid thymus), metabolic disruptions, and significant enzymatic changes. In the area of human health effects, our folder of known information is somewhat slimmer. This is partially due to the fact that most human data are obtained from occupational exposure and industrial accidents. In these cases, it has been difficult to estimate the level of 2,3,7,8-TCDD exposure in individual cases or to distinguish the effect of concommitant exposure to other chemicals. In any event, there is general agreement in the scientific community that chloracne, a persistent, acne-like condition which can be disfiguring but which is not life-threatening, is associated with persons acutely exposed to 2,3,7,8-TCDD. Chloracne can also be evoked by a number of chlorinated hydro- NOVEMBER 1983 carbon chemicals in addition to 2,3,7,8TCDD. Other effects which have been associated with these exposure incidents, and which are generally considered to be short-terrn, include liver dysfunction, effects on the immune system, and various neurological complaints. A series of reports has associated human exposure to 2,3,7,8-TCDD-containing chemicals and a rare form of cancer, soft tissue sarcoma. This possible link was first reported in a pjiir of studies of Swedish workers, and additional, but not definitive, support for the association was found by re-assessing studies completed here in the United States, in which isolated cases of soft tissue sarcomas have been found in 2,3,7,8TCDD exposed populations. However, other studies both here and abroad have failed to confirm this association. This possible association is being explored in a number of current or planned studies by various government agencies. In considering risk, one must remember that it is a function of two variables: hazard and exposure. A reduction in the size of either variable will result in a comparable reduction in risk. For example, even the most hazardous substance will be of no risk, if its exposure to people and the environment can be reduced to zero. In evaluating risk, the Agency combines hazard information (data on inherent toxicity) and exposure data to arrive at quantitative estimates of risk. To illustrate, I will briefly discuss how the Agency assesses carcinogenic and reproductive effects. The Agency assesses the excess risk of cancer using the methods'of the Cancer Assessment Group (CAG), whose guidelines were published in 1976. Briefly, the Agency first examines the data base to make a determination as to whether the chemical substance is a carcinogen. In addition to the qualitative question, a quantitative extrapolation to low environmental doses is performed in order to estimate a rough upper bound for the risk, using a linear, non-threshold procedure. This presumes that the initiation of cancer is a non-threshold phenomenon; that is, there is some risk, perhaps very small, at any exposure above zero. In the case of 2,3,7,8-TCDD the Agency has based its quantitative analysis primarily on the linerarized, multi-stage extrapolation model, although several others have also been used on occasion. It should be pointed out that these procedures result not in an absolute prediction of the risk, but rather a "ballpark" estimate of the upper limit of risk which we do not believe will be exceeded. The actual risk is likely to be some value less than this upper limit, possibly zero. These extrapolation procedures indicated that 2,3,7,8-TCDD was quite potent compared to many other carcinogens evaluated using the same techniques. 17 �The magnitude of the risk depends heavily on the level of 2,3,7,8-TCDD to which people are exposed and the likelihood that this exposure results in an absorbed, lexicologically active dose. Usually, we do not have definitive information on human exposure. In lieu of such data, the Agency makes certain assumptions, usually of the "reasonable worst case" variety, so as to err on the side of public safety. For example, in the case of TCDDs emitted from combustion sources, the Agency assumed that a person might spend his entire life at the spot of highest estimated ground level concentration, that all TCDDs inhaled would be retained, and that TCDDs attached to particles would be completely biologically active. "Reasonable worst case" estimates of exposure, when combined with the extrapolation results, teed to an estimate of the upper limit of rtsk. In contrast to caflcer, the Agency has generally regarded reproductive hazard as one for which there exists a level of exposure below which it i* not expected that an adverse effect wtB occur, the so-called "threshold assumption." In assessing this type of risk, the scientist uses an adequate study in which an administered dose level resulted in no observed adverse effects (NOAEL) in test animals and compares it to the generally snrmHer level of estimated human exposure. The ratio of the NOAEL to this estimated human exposure is referred to as the margin of safety. In the case of 2,3,7,8-TCDD, the Agency used a study In which rats were followed over three generations to determine the effect of 2,3,7,8-TCDD on the reproductive success of the animals. Although the authors of the study reported that no consistent adverse effects were observed at the lowest dose tested, EPA scientists concluded that statistically significant effects were observed at that dose and that the study lacked sufficient statistical power to conclusively demonstrate a NOAEL. This issue has been the source of considerable debate. Therefore, in comparing the lowest dose tested to the estimated exposure dose in humans, the Agency speaks of a "confidence ratio," 18 instead of a "margin of safety." In recent decisions associated with the Comprehensive Environmental Response, Compensation and Liability Act (CERCLA, or "Superfund"), the Agency has also made use of risk evaluations generated by the Centers for Disease Control (CDC). In general, CDC uses methods which are comparable to those of the Agency. In a typical 2,3,7,8-TCDD-related situation, Agency scientists provide decisionmakers with the results of a risk assessment; i.e., estimated upper limits of cancer risk and confidence ratios for reproductive effects associated with various exposure scenarios. The assumptions and limitations of.the approach should be explicitly stated. At this point, risk assessment ends and risk management begins. The distinction between risk assessment and risk management has been highlighted in the recent report of the National Academy of Sciences (NAS) entitled Risk Assessment in the Federal Government. Generally, risk assessment is an objective, scientific evaluation of the magnitude of the risk, independent of considerations of what should be done about that risk. Risk management is the decisionmaking process, involving more subjective, societal judgments which consider certain non-risk factors when selecting an appropriate response to the risk. In a speech delivered at the NAS, EPA Administrator Ruckelshaus emphasized this difference, stating that the two processes should remain separated within a regulatory agency. In its letter of invitation to these hearings the Subcommittee has asked EPA specifically, "What evidence on the effects of dioxin on human health justifies establishing a dioxin concentration standard of one part per billion in soil and how should such a standard be interpreted and used?" I believe your question may have been promoted by EPA's recent relocation action under "Superfund," and EPA welcomes the opportunity this hearing affords to clarify what has been erroneously characterized by some press reports as an EPA "safe" level of dioxin. First of all, there is no simple level which will give rise to equivalent risks in all cases. Even if one were to decide on an acceptable level of risk, the key question of exposure must be addressed on a site-specific basis before making any estimate of an acceptabl level of contamination in the soil. For example, a decisionmaker could conclude that greater than 1 ppb of 2,3,7,8-TCDD in the soil of one person's front yard might represent an unreasonable risk, since such i person might not be expected to easily or reasonably limit his exposure to this soil. That same decisionmaker could conclude, however, that many times that level is acceptable in an isolated spot at a manufac turing site or at the bottom of a reclaimed dumpsite where people are unlikely to be exposed. In sum, the determination of an acceptable level is dependent upon many factors, and it is an oversimplication to soola universally applicable level. Second, the act of establishing a level is no longer in the realm of risk assessment; instead, this is the province of risk management. In assessing risks, scientists can, for example, present the decisionmaker with a graphical summary illustrating the possible range of risks associated with various exposure scenarios and contamination level: in the soil. In reaching the risk managemem decision, the decisionmaker weights all the elements of the risk assessment; i.e., the qualitative case, the quantitative case, the exposure assessment, and the limitations at uncertainties involved. In addition, the decisionmaker factors in non-risk considera tions, which might include feasibility and cost of clean-up, possible alternative action: consistency with regulation of other risks, and concerns of the affected community. Ir sum, while the scientists may agree that a certain spectrum of risk is associated with different levels of contamination and exposure, precisely where on that spectrum a decisionmaker determines the appropriate level to be will vary as factors specific to a given situation are considered. Thus, EPA has not adopted a generally applicable action level for "Superfund" purposes; rather the Agency continues to make decisions on a site-by-site basis, takin into consideration both the CDC health advisories and any special on-site circum- EPAJOURNA �stances in determining action at individual sites. Finally, I would like to address some of the gaps in our scientific knowledge about toxicity of 2,3,7,8-TCDD in the environment and what type of research would improve the scientific data base for decisionmaking. 1. Toxicity of 2,3,7,8-TCDD in complex mixtures Most of the data generated to date has been with 2,3,7,8-TCDD alone. In the environment, however, we usually encounter the compound in combination with other materials and associated with particulates; e.g., soil or fly ash. The effects of dioxins in the presence of these other materials need to be investigated to answer questions of synergism and bioavailability. Promising techniques for assessing "TCDD equivalents" of NOVEMBER 1983 such complex mixtures should be developed further. 2. Exposure issues We need to know more about the ways 2,3,7,8-TCDD moves in the environment; e.g., possibility of volatilization, bioaccumulation from soils into fish, dermal penetration, and the amount of soil children might ingest. 3. Disposal/destruction methods Currently, adequately tested and practical methods for disposal and/or destruction of dioxin contaminated materials are limited. Much work remains to be done to determine how best to deal with this material once it has been discovered in the environment. 4. Epidemiological studies Various Federal agencies are now conducting epidemiological studies to investigate the possible effects of 2,3,7,8-TCDD in humans. There are additional studies which could be conducted, involving populations near more recently discovered contamination sites. 5. Background levels It would be helpful to know the background level of 2,3,7,8-TCDD in various parts of the environment, such as land, various foods, and human tissue. This information could serve as valuable benchmarks. 6. Related compounds There are 74 other chlorinated dibezno-pdioxins and 135 chlorinated dibenzofurans, some of which are also of concern and appear in the environment. Activities need to be encouraged to deal with these compounds on a rational, deliberate basis. 7. Mechanisrn-of action studies and pharmacokinetics Important information is currently being deduced about the first stages of toxicity induced by 2,3,7,8-TCDD and related compounds. As we obtain more fundamental knowledge about what is happening at the molecular and cellular level, the possibility of our being able to understand exactly how and why 2,3,7,8-TCDD exerts its toxicity increases. This information may help us explain the basis of the »|>ecies variability and where humans fall in this range of reactions. Moreover, we may then be able to assess the toxic potential of literally hundreds of related toxic chemicals without devoting to each individual compound the mass of resources we have had to dedicate to 2,3,7,8-TCDD. As a final word, I would like to observe that it is important that we keep the dioxin problem in a proper perspective. I believe we need to address the dioxin issue in a rational, deliberate manner. At the same time, we should not permit this legitimate concern to cause us to neglect other legitimate concerns, such as those embodied in the pools, pits, and lagoons of abandoned dumpsites, the emission of toxic pollutants into our air and water, and the potential for unreasonable risks associated with chemicals to which we are exposed daily. As scientists and regulators, we have an obligation to maintain a balance among all of these concerns. D 19 �Office of Public Affairs (A-1071 Washington, DC 20460 unitea states Environmental Protection Agency Volume 9 Number 3 November 1383 v/EPA JOURNAL William D. Ruckelshaus, Administrator Josephine S. Cooper, Assistant Administrator for External Affaifientral Office Library (14201) Jean Statler, Director, Office of Public Affairs Charles D. Pierce, Editor John M. Heritage, Managing Editor l-'tC 13 1983 Veterans Administration Washington, D.C. 20420 Articles EPA is charged by Congress to protect the Nation's land, air and water systems. Under a mandate of national environmental laws, the Agency strives to formulate and implement actions which lead to a compatible balance between human activities and the ability of natural systems to support and nurture life. Controlling Hazardous Waste 2 Enforcement Drive Mounts 7 An Interview with Assistant Administrator Lee Thomas By Tom Kelley Recent major EPA activities By Carl Gagliardi The State of Hazardous and Municipal Waste Control by U.S. Senator Jennings Randolph, D.-W.Va. Making the Superfund List 14 The Effective Regulation of Hazardous Waste by U.S. Representative James T. Broyhill, R.-N.C. The EPA Journal is published quarterly by the U.S. Environmental Protection Agency. The Administrator of EPA has determined that the publication of this periodical is necessary in the transaction of the public business required by law of this Agency. Use of funds for printing this periodical has been approved by the Director of the Office of Management and Budget through 4/1/84. Views expressed by authors do not necessarily reflect EPA policy. Contributions and inquiries should be addressed to the Editor (A-107) Waterside Mall, 401 M St., S.W., Washington, D.C. 20460. No permission necessary to reproduce contents except copyrighted photos and other materials. EPA's Opportunity to Communicate 27 An Overview on Dioxin 16 By David Cohen By Assistant Administrator Josephine S. Cooper River At Risk 28 By Donald Barnes The Towers of Tacoma 20 A Photo Essay Putiet oou;id nt sunset with ducks Photo Credits Photn, Steve Dolaney, Eric Muyerson and Christ ophtu Moffett of I-PA llcgion 10; Design Credits: Robert Flanagan, Ron Farrah. EPA JOURNAL Subscriptions I hi; annual rate for subscribers in the U.S. for the quarterly EPA Journal is $7.50. The charge to subscribers in foreign countries is $9.40 a year. The price of. a single copy of the Journal is $2.50 in this country and $3.15 if sent to a foreign country. Prices include mailing costs. Subscriptions to EPA Journal as well as to other Kederal'Governrnent magazines, are handled only by the U.S. Government Printing Office. Anyone wishing to subscribe to the Journal should fill in the form at right and enclose a check or money order payable to the Superintendent of Documents. The request should be mailed to: Superintendent of Documents, GPO, Washington, D.C. 20402. 22 Update 24 The First Five Cleanups 9 The View from Capitol Hill 4 New EPA Appointments Please Print Ntime \'\r'.;\. Last I .. 1 i . , I... . 1 . 1 j i j 1 _!._ i | I I . I I I 1.1 1 1 Company Name or Additional Address Lir 1 L -jr." — 'Ai jJ.-UiiJ S t r e e t Address State City L.... - 1 ..L .1 I I Payment enclosed (Make check payable to Superintendent of Documents) i 'Charge to my Deposit Account No. Zip Code � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Alvin L. Young Collection on Agent Orange Description An account of the resource <p style="margin-top: -1em; line-height: 1.2em;">The Alvin L. Young Collection on Agent Orange comprises 120 linear feet and spans the late 1800s to 2005; however, the bulk of the coverage is from the 1960s to the 1980s and there are many undated items. The collection was donated to Special Collections of the National Agricultural Library in 1985 by Dr. Alvin L. Young (1942- ). Dr. Young developed the collection as he conducted extensive research on the military defoliant Agent Orange. The collection is in good condition and includes letters, memoranda, books, reports, press releases, journal and newspaper clippings, field logs and notebooks, newsletters, maps, booklets and pamphlets, photographs, memorabilia, and audiotapes of an interview with Dr. Young.</p> <p>For more about this collection, <a href="/exhibits/speccoll/exhibits/show/alvin-l--young-collection-on-a">view the Agent Orange Exhibit.</a></p> Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Box The box containing the original item. 193 Folder The folder containing the original item. 5447 Series The series number of the original item. Series VIII Subseries I Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Creator An entity primarily responsible for making the resource Barnes, Donald Source A related resource from which the described resource is derived EPA Journal Date A point or period of time associated with an event in the lifecycle of the resource 1983-11-01 Title A name given to the resource An Overview on Dioxin ao_seriesVIII https://www.nal.usda.gov/exhibits/speccoll/files/original/d205ca78ad089482ff3793305792b3eb.pdf 3807b98f708558d9a6bf31a1091fa8bb PDF Text Text Item D Number 05526 Author Stanley, John S. Corporate Author D Midwest Research Institute RBDOTt/ArtiClO Tltb Analysis for Polychlorlnated Dlbenzo-p-DioxIns (PCDD) and Dibenzofurans (PCDF) in Human Adipose Tissue: Method Evaluation Study: Final Report with attached letter transmitting the report to Alvin L Young, from Janet C. Remmers, Field Studies Branch, Exposure Evaluation Division, United States Environmental Protection Agency Journal/Book Title Year 1986 Month/Day October Color D Number of Images 149 DUSCrbtOn NOtBI EPA Prime Contract No. 68-02-3938, Work Assignment No. 46, MRI Project No. 8501-A(46). See Item 5522 for draft version Tuesday, March 19, 2002 Page 5526 of 5611 �UNITED STATES ENVIRONMENTAL PROTECTION AGENCY WASHINGTON, D.C. 20460 JAN I 3 1987 OFFICE OF PESTICIDES AND TOXIC SUBSTANCES A l v i n Young, P h . D . , Lt. Col. , USAF Senior Policy Analyst for Life Sciences Executive Office of the President Office of Science and Technology Policy Room 5005 New Executive Office Building Washington, DC 20506 Dear Dr. Young: Enclosed for your information is the final report entitled "Analysis for Polychlorinated Dibenzo-p_-dioxins (PCDD) and Dibenzofurans (PCDF) in Human Adipose Tissue: Method Evaluation Study." Sincerely, Janet C. Remmers Field Studies Branch Exposure Evaluation Division (TS-798) Enclosure �United Stale-. Environment'*! Aqency Toxic Sub.itan &EPA ANALYSIS FOR POLY-CHLORINATED DIBENZO-p-DIOXINS (PCDD) AND DIBENZOFURANS (PCDF) IN HUMAN ADIPOSE TISSUE: METHOD EVALUATION STUDY 80t- C o co w *C 0) o C o o TJ C 3 O 95% Confidence Limits for Individual Analyses 2, 3, 7, 8-TCDD I I 50 Spiked Concentration (pg/g) �ANALYSIS FOR POLYCHLORINATED DIBENZO-£-DIOXINS (PCDD) AND DIBENZOFURANS (PCDF) IN HUMAN ADIPOSE TISSUE: METHOD EVALUATION STUDY by John S. Stanley, Randy E. Ayling, Karin M. Bauer, Michael J. McGrath, Thomas M. Sack, and Kelly R. Thornburg FINAL REPORT EPA Prime Contract No. 68-02-3938 Work Assignment No. 46 MRI Project No. 8501-A(46) and EPA Prime Contract No. 68-02-4252 Work Assignment No. 24 MRI Project No. 8824-A(01) Field Studies Branch, TS-798 Office of Toxic Substances U.S. Environmental Protection Agency 401 M Street, S.W. Washington, DC 20460 Attn: Ms. Janet Remmers, Work Assignment Manager Dr. Joseph J. Breen, Project Officer �DISCLAIMER This document has been reviewed and approved for publication by the Office of Toxic Substances, Office of Pesticides and Toxic Substances, U.S. Environmental Protection Agency. The use of trade names or commercial products does not constitute Agency endorsement or recommendation for use. �PREFACE This report provides a summary of the results from a method evaluation study for the determination of 2,3,7,8-substituted polychlorinated dibenzo-p_-dioxins (PCDD) and dibenzofurans (PCDF) in human adipose tissues at the parts-per- trill ion (ppt) level. This method evaluation is an integral part of a collaborative program between the U.S. Environmental Protection Agency's Office of Toxic Substances and the Veterans Administration to determine if significant differences exist in the 2,3,7,8-substituted PCDD and/or PCDF levels in human adipose tissues for Vietnam veterans compared to the general adult male population. The study design will focus on specimens within EPA's National Human Adipose Tissue Survey (NHATS) repository. The method evaluation described in this report was necessary to establish method performance (accuracy and precision) before proceeding with actual sample analysis. This method evaluation study was completed under EPA Contract Nos. 68-02-4252, Work Assignment 24 and 68-02-3938, Work Assignment 46, "Analysis for Dioxins and Furans in Human Adipose Tissue," Ms. Janet Remmers, Work Assignment Manager, and Dr. Joseph Breen, Project Officer. MIDWEST RESEARCH INSTITUTE aul C. Constant Program Manager Approved: Jack Balsinger^ Quality Assurance Coordinator JoMi E. Going, Director Chemical Sciences Department October 30, 1986 m �TABLE OF CONTENTS I. Introduction 1 II. Summary 3 III. Recommendations 4 IV. Experimental 5 A. Preparation of Homogenized Tissue B. Analytical Standards 1. Calibration Standards 2. Spiking Solutions 5 5 6 6 C. Analytical Procedure 10 D. 12 HRGC/MS Analysis E. Data Interpretation 1. Qualitative 2. Quantitation F. Quality Assurance/Quality Control (QA/QC) G. Preliminary Method Studies 1. Gravimetric Studies 2. Carbon-14 Recovery Studies V. Results 23 A. Analytical Results B. Statistical Analysis 1. Recovery of Internal Quantitation Standards. . 2. Estimation of Background Levels of PCDDs and PCDFs 3. Day-to-Day HRGC/MS Analysis Precision VI. Quality Assurance/Quality Control (QA/QC) A. Initial Calibration B. Daily Verification of Response Factors C. Blanks D. 17 17 17 20 20 21 21 23 23 57 57 59 63 63 65 65 Absolute Recoveries of the Internal Quantitation Standards 77 VII. Glossary of Terms 82 VIII. References 84 Appendix A - Analytical Protocol for Determination of PCDDs and PCDFs in Human Adipose Tissue A-l �LIST OF FIGURES Figure 1 2 3 4 5 6 7 8 9 10 11 12 Page Comparison of the HRGC/MS-SIM reconstructed ion chromatogram (RIC) from the analysis of unspiked homogenized human adipose tissue matrix and a calibration standard for PCDDs and PCDFs 30 Example of the TCDF (m/z 304) and TCDD (m/z 320) HRGC/MS-SIM elution profiles in unspiked and spiked human adipose 31 Example of the PeCDF (m/z 338) and PeCDD (m/z 354) HRGC/MS-SIM elution profiles in unspiked and spiked human adipose 32 Example of the HxCDF (m/z 374) and HxCDD (m/z 390) HRGC/MS-SIM elution profiles in unspiked and spiked human adipose 33 Example of the HpCDF (m/z 408) and HpCDD (m/z 424) HRGC/MS-SIM elution profiles in unspiked and spiked human adipose 34 Examples of the OCDF (m/z 442) and OCDD (m/z 458) HRGC/MS-SIM elution profiles in unspiked and spiked human adipose 35 Measured concentrations versus concentrations of 2,3,7,8TCDD spiked into the homogenized human adipose lipid matrix 36 Measured concentrations versus concentrations of 1,2,3,7,8-PeCDD spiked into the homogenized human adipose lipid matrix 37 Measured concentrations versus concentrations of 1,2,3,4,7,8-HxCDD spiked into the homogenized human adipose lipid matrix 38 Measured concentrations versus concentrations of 1,2,3,6,7,8-HxCDD spiked into the homogenized human adipose lipid matrix 39 Measured concentrations versus concentrations of 1,2,3,7,8,9-HxCDD spiked into the homogenized human adipose lipid matrix 40 Measured concentrations versus concentrations of 1,2,3,4,6,7,8-HpCDD spiked into the homogenized human adipose lipid matrix 41 �LIST OF FIGURES (continued) Figure 13 14 15 16 17 18 19 20 21 22 23 24 25 Page Measured concentrations versus concentrations of OCDD spiked into the homogenized human adipose lipid matrix. . 42 Measured concentrations versus concentrations of 2,3,7,8-TCDF spiked into the homogenized human adipose lipid matrix 43 Measured concentrations versus concentrations of 1,2,3,7,8-PeCDF spiked into the homogenized human adipose lipid matrix 44 Measured concentrations versus concentrations of 2,3,4,7,8-PeCDF spiked into the homogenized human adipose lipid matrix 45 Measured concentrations versus concentrations of 1,2,3,4,7,8-HxCDF spiked into the homogenized human adipose lipid matrix 46 Measured concentrations versus concentrations of 1,2,3,6,7,8-HxCDF spiked into the homogenized human adipose lipid matrix 47 Measured concentrations versus concentrations of 2,3,4,6,7,8-HxCDF spiked into the homogenized human adipose lipid matrix 48 Measured concentrations versus concentrations of 1,2,3,7,8,9-HxCDF spiked into the homogenized human adipose lipid matrix 49 Measured concentrations versus concentrations of 1,2,3,4,7,8,9-HpCDF spiked into the homogenized human adipose lipid matrix 50 Measured concentrations versus concentrations of l,2,3,4,6,7,8~HpCDF spiked into the homogenized human adipose lipid matrix 51 Measured concentrations versus concentrations of OCDF spiked into the homogenized human adipose lipid matrix. . 52 Method accuracy estimates as determined from the slopes of the least squares regression lines for the 17 target PCDD and PCDF analytes 54 Control charts showing response factors by date for 2,3,7,8-TCDF and 2,3,7,8-TCDD 66 VII �LIST OF FIGURES (continued) Figure 26 27 28 29 30 31 32 33 34 Page Control charts showing response factors by date for 1,2,3,7,8-PeCDF and 2,3,4,7,8-PeCDF 67 Control chart showing response factors by date for 1,2,3,7,8-PeCDD 68 Control charts showing response factors by date for 1,2,3,4,7,8-HxCDF and 1,2,3,6,7,8-HxCDF 69 Control charts showing response factors by date for 2,3,4,6,7,8-HxCDF and 1,2,3,7,8,9-HxCDF 70 Control charts showing response factors by date for 1,2,3,4,7,8-HxCDD and 1,2,3,6,7,8-HxCDD 71 Control chart showing response factors by date for 1,2,3,7,8,9-HxCDD 72 Control charts showing response factors by date for 1,2,3,4,6,7,8-HpCDF and 1,2,3,4,7,8,9-HpCDF 73 Control chart showing response factors by date for 1,2,3,4,6,7,8-HpCDD 74 Control charts showing response factors by date for OCDF and OCDD 75 v i ii �LIST OF TABLES Table 1 Page Analytical Standards Available for the Method Evaluation Studies 7 2 Concentration Calibration Solutions 8 3 Native PCDD and PCDF Spiking Solution 9 4 Internal Standard Spiking Solutions 11 5 HRGC/LRMS Operating Conditions for PCDD/PCDF Analysis . . . 13 6 Ions Monitored for HRGC/MS Analysis of PCDD/PCDF 14 7 Typical Daily Sequence for PCDD/PCDF Analysis 16 8 Ion Ratios for HRGC/MS Analysis of PCDD/PCDF 18 9 Summary of the Results of the Sample Preparation Method Evaluation Using Carbon-14 PCDDs 22 Spiked Versus Measured Concentrations of 2,3,7,8-TCDF and 2,3,7,8-TCDD in Homogenized Human Adipose Lipid Samples 24 Spiked Versus Measured Concentrations of 1,2,3,7,8-PeCDF, 2,3,4,7,8-PeCDF, and 1,2,3,7,8-PeCDD in Homogenized Human Adipose Tissue Samples 25 Spiked Versus Measured Concentrations of 1,2,3,4,7,8-; 1,2,3,6,7,8-; 2,3,4,6,7,8-; and 1,2,3,7,8,9-HxCDF in Homogenized Human Adipose Lipid Matrix 26 Spiked Versus Measured Concentration of 1,2,3,4,7,8-; 1,2,3,6,7,8-; and 1,2,3,7,8,9-HxCDD in Homogenized Human Adipose Lipid Samples 27 Spiked Versus Measured Concentrations of 1,2,3,4,6,7,8HpCDF, 1,2,3,4,7,8,9-HpCDF, and 1,2,3,4,6,7,8-HpCDD in Homogenized Human Adipose Lipid Samples 28 Spiked Versus Measured Concentrations of OCDF and OCDD In Homogenized Human Adipose Lipid Samples 29 16 Regression Line Slopes with 95% Confidence Limits 55 17 Results of the Analysis of the Low and High Level Native Spike Solutions 56 10 11 12 13 14 15 IX �LIST OF TABLES (continued) Table Page 18 Background Level Estimates with 95% Confidence Limits . . . 58 19 Day-to-Day Precision of Analysis of Specific Sample Extracts for Tetra- and Pentachloro PCDF and PCDD . . . . 60 Day-to-Day Precision of Analysis of Specific Sample Extracts for Hexa- and Heptachloro PCDF and PCDD 61 Day-to-Day Precision of Analysis of Specific Sample Extracts for OCDF and OCDD 62 Relative Response Factors (Grand Means) Determined from Multipoint Concentration Calibration Standards 64 Summary of Results from the Analysis of a Laboratory Method Blank 76 Recovery of Radiolabeled PCDDs from Precleaned Activated Alumina 78 Absolute Recoveries of the Internal Quantitation Standards from the Human Adipose Lipid Matrix 79 Recovery of Carbon-14 Labeled 2,3,7,8-TCDD, 1,2,3,4,7,8HxCDD, and OCDD as a Function of Final Concentration Conditions 81 20 21 22 23 24 25 26 �I. INTRODUCTION The Environmental Protection Agency Office of Toxic Substances (EPA/OTS) and the Veterans Administration (VA) have established an interagency agreement to study the level of polychlorinated dibenzo-£-dioxins (PCDDs) and dibenzofurans (PCDFs) in human adipose tissues. The occurrence and levels of PCDDs and PCDFs with chlorine substitution in the 2,3,7,8 positions (especially 2,3,7,8-TCDD) of the parent molecules are of primary interest. As part of this interagency effort, it has been proposed to use selected adipose tissue samples that were collected for the Field Studies Branch (FSB) of EPA's Office of Toxic Substances (OTS) through the National Human Adipose Tissue Survey (NHATS) to determine exposure to PCDDs and PCDFs. The available adipose tissues include specimens obtained from young men whose age indicates that they could have served in Vietnam and could have been exposed to Agent Orange. The tissues were originally collected as part of a broadly based and statistical random sampling of the continental United States. The analysis of these tissues may provide information on the differences of exposure of the general adult male population and Vietnam veterans to the 2,3,7,8-substituted PCDDs and PCDFs. The overall objectives of the proposed EPA/VA collaborative studies are: 1. Evaluate the reliability, accuracy, precision, and sensitivity of a proposed method for the determination of 2,3,7,8substituted PCDDs and PCDFs (tetra- through octachloro homologs) in human adipose tissue at the parts-per-tri1 lion (ppt) level. 2. Determine if these compounds can be detected in adipose tissues of the American male adult population; and 3. Determine if individuals with military service in Vietnam have significantly different levels of 2,3,7,8-substituted PCDDs and PCDFs (particularly 2,3,7,8-TCDD) than other American men. As a prelude to this work assignment, MRI conducted an extensive literature review of applicable analytical methods and conducted a meeting with recognized experts in this field to identify critical aspects of analytical methodology.1'2 Based on the information gathered through the literature review and the meeting with the recognized experts, a special report was prepared for OTS proposing a framework for an analytical method for analysis of human adipose tissues.3 Several studies have been completed since the issuance of that report which reflect the advances in analytical techniques for adipose tissue analysis.4 16 The salient features of these methods have been combined into a single protocol for the routine analysis of tetra- through octachloro PCDDs and PCDFs at the low-parts-per-tril1 ion level for the EPA/VA tissue study. This report focuses on a method evaluation study that was conducted to achieve the first objective of the interagency agreement. Clarification �of method performance is necessary before proceeding with the analysis of actual samples retrieved from the NHATS repository. This report includes a summary of the method evaluation study results (Section II). Recommendations to be implemented before proceeding with the actual tissue samples from the NHATS repository are presented in Section III. A description of the actual experimental procedures is provided in Section IV. Results of sample analyses are summarized in Section V, and quality assurance/quality control (QA/QC) aspects of the study are detailed in Section VI. Pertinent references are listed in Section VII. Appendix A contains the detailed analytical protocol that will be followed for the analysis of the NHATS specimens designated in the study design to be provided by EPA/VA. �II. SUMMARY The results of the replicate analysis of spiked and unspiked homogenized human adipose tissue matrix demonstrate that the analytical method produces accurate and precise data for 17 specific 2,3,7,8-substituted PCDD and PCDF (tetra- through octachloro homologs) compounds. Accuracy of the analytical method was demonstrated to range from 90 to 120% for the 17 2,3,7,8-substituted PCDD and PCDF compounds. Data are reported for three or four replicate analyses of samples spiked at three different concentration levels. The endogenous or background levels of the PCDD and PCDF congeners in the homogenized adipose lipid matrix were estimated through regression analyses of measured (found) versus spiked concentrations for each compound. The analytical method is capable of providing quantitative data for tetra- through octachloro PCDD and PCDF congeners to concentration levels as low as 1 pg of the tetrachloro congeners per gram of adipose tissue. However, an interference was noted at m/z 304 which coeluted with 2,3,7,8-TCDF, resulting in a detection level of approximately 4 pg/g. Average absolute recoveries of the internal quantisation standards ranged from 52% for 13C12-TCDD up to 89% for 13C12-OCDD. The agreement of the measured concentrations versus the spiked concentrations for each PCDD and PCDF congener demonstrates that the internal standard quantitation procedure provides an accurate measure of concentration which is independent of the absolute recovery. Final concentration conditions were noted to have pronounced effect on the absolute recoveries of the lower chlorinated compounds, particularly 2,3,7,8-TCDD. Experiments with carbon-14 labeled 2,3,7,8-TCDD demonstrated that final concentration at temperatures of 55 to 60°C resulted in recoveries as low as 54% while the same procedure conducted at ambient conditions resulted in greater than 90% recovery. Analysis of method and reagent blanks provided information on potential artifacts in the sample preparation scheme. Additional experiments were conducted with carbon-14 labeled PCDDs to evaluate the cleanup efficiency and recovery of PCDDs from chromatographic materials, particularly acidic alumina. �III. RECOMMENDATIONS Some minor modifications have been made in the written protocol (Appendix A) that were not included in this phase of the method validation. These include: a cleanup procedure for activated acidic alumina prior to fractionation of sample extracts to remove artifacts; and final concentration of the sample extracts using nitrogen blowdown at room temperature rather than heating to 55-60°C. The spiking solutions used to prepare the spiked quality control samples should be submitted for replicate (minimum of three/per spike level) HRGC/MS analysis to assist the interpretation of positive or negative bias in the accuracy of QC sample data. The accuracy bounds should be extended to 50-130% from 50-115% as specified in the draft quality assurance program plan. The method should include additional internal quantitation standards to pair with the HpCDF 13 OCDF congeners. Also, an additional internal recovand ery standard, possibly C12-l,2,3,4,7,8-HxCDD, is required to provide better estimates of absolute method recovery. These additional compounds, if available, will be incorporated into the method before initiating sample analyses. Analysis for 2,3,7,8-TCDF may require high resolution mass spectrometry to avoid interferences occurring at m/z 304. This will require modification of the HRGC/HRMS portion of the protocol to include the specific acquisition parameters for the characteristic ions of 2,3,7,8-TCDF. �IV. EXPERIMENTAL A. Preparation of Homogenized Tissue A bulk lipid sample was prepared from the extracts of human adipose tissue samples collected through the NHATS program. The adipose tissue samples have been stored in a deep freezer at approximately -10°C since collection. The homogenized tissue extract or bulk lipid was used in this method evaluation study for preparation of replicate samples spiked with varying levels of specific PCDD and PCDF isomers. This homogenized matrix will also be used for preparing control and spiked quality control samples for the actual NHATS sample analysis phase of the program. A total of 2,465 g of adipose tissue was extracted, dried, and concentrated to yield 1,652 g (62% of original weight) of homogenized lipid. Specific procedures for preparing this matrix are described below. The adipose tissue samples were thawed at room temperature for 1 to 2 h. Portions of the samples were added to a blender cup of a Waring® blender and covered with methylene chloride. The volume of methylene chloride was approximately equal to the sample volume (100 to 200 ml). This mixture was blended at high speed for approximately 10 min, and the contents were transferred to a 500-mL Erlenmeyer flask and further blended with a Tekmar® Tissumizer, also at high speed for 10 min. A powder funnel was plugged with a wad of glass wool (silanized, methylene chloride extracted) and filled with •v 50 g of sodium sulfate (heated overnight to 600°C in a muffle furnace). The sodium sulfate was wetted with methylene chloride prior to elution of the sample extract. The dried effluents were refiltered in the same way using a fresh bed of sodium sulfate to remove particulate and residual water. The samples were transferred to 1-L round bottom flasks, and the solvent was removed by rotary evaporation. The water bath on the rotary evaporator was kept at 60°C using a thermostatted heating element. Once the solvent appeared to have been removed (constant volume in flask, no visible condensation in condenser), the heating and evaporation process was continued for at least 2 h. The flask and contents were removed and stored in a refrigerator. The extracted lipid solidified upon refrigeration and was visually checked for homogeneity. No precipitates or phase separation was observed. The lipid residue was allowed to liquify at room temperature and was transferred to a 4-L glass bottle with a Teflon®-!ined lid. The lipid residue was brought to room temperature and heated just enough to allow the lipid to achieve an oily state prior to aliquotting portions for the method evaluation studies. B. Analytical Standards Analytical standards including native PCDD and PCDF congeners, stable isotope (carbon-13) labeled standards and radiolabeled (carbon-14) standards were purchased from Cambridge Isotope Laboratories, Woburn, Massachusetts, and Pathfinder Laboratories, St. Louis, Missouri. The 2,3,7,8TCDD was received from the EPA Reference Materials Branch as a solution in �isooctane. The other native PCDD and PCDF congeners were received as 1-mg neat standards. The stable and radiolabeled isotopes were received as solutions in n-nonane or isooctane and toluene, respectively. Table 1 provides a summary of the standards used for this study. Stock solutions of the individual PCDD and PCDF congeners were prepared from the neat standards. The neat materials were weighed using a Cahn 27 electrobalance calibrated versus a 1-mg (Class M) standard. The neat compounds were transferred to glass vials and were dissolved in 2.0 to 3.0 mL of toluene (Burdick and Jackson, distilled in glass). Toluene was added to each standard using volumetric pipettes (Class A). The OCDD required dilution to 10.0 ml using a 50:50 mixture of toluene and anisole. A working solution consisting of the 17 native PCDD and PCDF congeners was prepared in toluene at a concentration of 2 (J9/mL for the TCDD, TCDF, PeCDD, and PeCDF congeners, 5 ug/mL for the HxCDD, HxCDF, HpCDD, and HpCDF congeners, and 10 ug/mL for the OCDD and OCDF. The working solution was used to prepare both the lipid matrix spiking solution and the calibration standards. The stable isotope labeled internal standards were obtained as solutions13in n-nonane or isooctane at 50 ug/mL concentration with the exception of the C12-OCDD, which was provided at 10 ug/mL. Separate working solutions containing mixtures of the carbon-13 labeled PCDDs and PCDFs were prepared in isooctane for use in the calibration standards and the sample spiking solutions. The carbon-14 radiolabeled PCDDs were used for preliminary method evaluation studies. The specific activity of the 14C-2,3,7,8-TCDD (117.56 mCi/mmole) was high enough to allow recovery studies at spike levels equivalent to 10 pg/g for a 10-g sample. 1. Calibration Standards Eight concentration calibration standards containing the 17 native and the 9 carbon-13 labeled internal standards were prepared for determining the consistency of response factors for the native PCDDs and PCDFs versus the corresponding carbon-13 congeners. Table 2 presents a summary of the calibration standards prepared for the method calibration study. The solution concentrations (pg/pL) can also be considered as equivalent to residue levels in picograms per gram of adipose. For example, a 1 pg/uL concentration standard corresponds to a tissue concentration of 1 pg of PCDD or PCDF congener per gram of adipose assuming a 10-g sample is available for analysis. 2. Spiking Solutions a. Native PCDD and PCDF A solution containing the 2,3,7,8-substituted PCDD and PCDF congeners was prepared in isooctane for spiking the homogenized lipid materials for the method evaluation study. Table 3 specifies the levels of each of the native PCDD and PCDF congeners present in this solution. �Table 1. Analytical Standards Available for the Method Evaluation Studies Source Compound Native 2,3,7,8-TCDD 2,3,7,8-TCDF 1,2,3,7,8-PeCDD 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2, 3,4,7, 8-HxCDF 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,4,6,7,8-HpCDD 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF OCDD OCDF EPA QA Reference Materials Branch Cambridge Isotope Laboratories Cambridge Isotope Laboratories Cambridge Isotope Laboratories Cambridge Isotope Laboratories Cambridge Isotope Laboratories Cambridge Isotope Laboratories Cambridge Isotope Laboratories Cambridge Isotope Laboratories Cambridge Isotope Laboratories Cambridge Isotope Laboratories Cambridge Isotope Laboratories Cambridge Isotope Laboratories Cambridge Isotope Laboratories Cambridge Isotope Laboratories Cambridge Isotope Laboratories Cambridge Isotope Laboratories Lot/Code 20603 AWN 1203-74/EF-903C MLB-706-53/ED-950C AWN-729-21/EF-953C AWN-729-45/EF-956C 830244/ED-961C MLB-706-47/ED-960C MLB-706-73/ED-969C AWN-729-20/EF-964C MB 13106-7/EF-962-C MB 13106-47/EF-967-C MB 13106-3/EF-968-C MLB-706-21/ED-971C AWN-729-22/EF-973C MB-13-106-77/EF-975C 8465-F-982-C/EF-982C F2832/ED-980C 13 C12-Internal stand;irds Cambridge 2,3,7,8-TCDD Cambridge 2,3,7,8-TCDF 1,2,3,7,8-PeCDD Cambridge 1,2,3,7,8-PeCDF Cambridge Cambridge 1,2,3,6,7,8-HxCDD Cambridge 1,2,3,4,7,8-HxCDF 1,2,3,4,6,7,8-HpCDD Cambridge OCDD Cambridge 37 Isotope Isotope Isotope Isotope Isotope Isotope Isotope Isotope Laboratories Laboratories Laboratories Laboratories Laboratories Laboratories Laboratories Laboratories R00208/ED-900 R00236/EF-904 R00241/ED-955 R00221/EF-952 R00249/ED-966C R00234/EF-963C R00248/ED-972 R00263/ED-981 Cl-Internal standaird 37 C14-1,2,3,4,6,7,8HpCDD KOR Isotopes 580012/SSY-4-32 l4 C 12 -Radiolabeled standards 2,3,7,8-TCDD Pathfinder Laboratories Pathfinder Laboratories 1,2,3,4,7,8-HxCDD OCDD Pathfinder Laboratories S.A. = specific activity. S.A. = 117.56 mCi/mmole S.A. = 24.16 mCi/mmole S.A. = 20.50 mCi/mmole �Table 2. Concentration Calibration Solutions' Concentration in calibration solutions in p flA iL CS1 CS2 CS3 CS4 CS5 CS6 CS7 CSS Compound Native 200 200 200 200 200 500 500 500 500 500 500 500 500 500 500 1 ,000 1 ,000 25 5 5 5 5 5 12. 5 12. 5 12. 5 12. 5 12. 5 12. 5 12. 5 12. 5 12. 5 12. 5 25 25 100 100 100 100 100 250 250 250 250 250 250 250 250 250 250 500 500 50 50 50 50 50 125 125 125 125 125 125 125 125 125 125 250 250 25 25 25 25 62. 5 62. 5 62. 5 62. 5 62. 5 62. 5 62. 5 62. 5 62. 5 62. 5 125 125 Internal quantitation standards 13 50 50 C12-2,3,7,8-TCDD 13 C12-2,3,7,8-TCDF 50 50 13 50 50 C12-l,2,3,7,8-PeCDD 13 C12-l,2,3,7,8-PeCDF 50 50 13 C12-l,2,3,6,7,8-HxCDD 125 125 13 C12-l,2,3,4,7,8-HxCDF 125 125 13 C12-1,2,3,4,6,7,8125 125 HpCDD 13 C12-OCDD 250 250 50 50 50 50 125 125 125 50 50 50 50 125 125 125 50 50 50 50 125 125 125 50 50 50 50 125 125 125 50 50 50 50 125 125 125 50 50 50 50 125 125 125 250 250 250 250 250 250 50 50 50 50 50 50 2,3,7,8-TCDD 2,3,7,8-TCDF 1,2,3,7,8-PeCDD 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,4,6,7,8-HpCDD 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF OCDD OCDF 10 10 10 10 10 25 25 25 25 25 25 25 25 25 25 50 50 2. 5 2. 5 2. 5 2. 5 2. 5 6. 25 6. 25 6. 25 6. 25 6. 25 6. 25 6. 25 6. 25 6. 25 6. 25 12. 5 12. 5 1 1 1 1 1 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 5 5 Internal recovery standard 13 C12-1,2,3,4-TCDD Prepared in tridecane. 50 50 �Table 3. Native PCDD and PCDF Spiking Solution3 Compound 2,3,7,8-TCDD 2,3,7,8-TCDF 1,2,3,7,8-PeCDD 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,4,6,7,8-HpCDD 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF OCDD OCDF a Prepared in isooctane. Concentration (pg/pL) 5 5 5 5 5 12.5 12.5 12.5 12.5 12.5 12.5 12.5 12.5 12.5 12.5 25 25 �b. Internal Standards Two different internal standard spiking solutions were prepared for quantitation of native PCDD and PCDF congeners. The compositions of each of the spiking solutions are presented in Table 4. The internal quantitation standards were spiked into the lipid aliquots prior to any cleanup procedures and hence were carried throughout the method exactly as the corresponding native congeners. The internal recovery standard was added in 10 uL of a keeper solution (tridecane) during final extract concentration prior to analysis. The recovery standard was used to measure the absolute method recoveries of the internal quantitation standards. C. Analytical Procedure The homogenized human adipose lipid matrix was allowed to come to room temperature and then warmed in a water bath until the matrix changed to an oily state. Approximately 10.0 g of the oily material was transferred by pipette to preweighed glass vials, and the actual weight of the lipid was determined to the nearest 0.01 g by difference using an analytical balance. Four 10.00-g aliquots were spiked with 20 uL of the native spiking solution presented in Table 3, another four aliquots were spiked with 50 (jL of the same solution, and three additional aliquots were spiked with 100 p.L of native PCDD and PCDF solution. These spikes were equivalent to concentrations ranging from 10, 25, and 50 pg/g in the lipid matrix for the tetra- and pentachloro PCDD and PCDF congeners up to 50, 125, and 250 pg/g for the OCDD and OCDF for the low, medium, and high level spikes. In addition to the spiked samples, three aliquots of the lipid material were transferred for determining the endogenous levels of each of the PCDD and PCDF congeners in the control matrix. Each of the sample aliquots was fortified with 100 |jL of the internal quantitation standard spiking solution (Table 4). The spiked samples were each quantitatively transferred to 500-mL Erlenmeyer flasks using hexane. The residues were diluted with a total of 200 mL of hexane, and 100 g of sulfuric acid (H2S04) modified silica gel (40% w/w) was added to each solution with stirring. The mixtures were stirred for approximately 2 h, and the supernatants were decanted and filtered through filter funnels packed with anhydrous sodium sulfate (Na2S04). The H2S04 modified silica adsorbents were washed with at least two additional aliquots of hexane and dried by elution through Na2S04. The combined hexane extracts for each sample were eluted through a column consisting of the 40% H2S04 modified silica gel (4.0 g) and silica gel (1.0 g). The eluates were concentrated to approximately 15 mL and added to columns of acidic alumina (Bio-Rad, AG-4, 6.0 g). The acidic alumina columns were eluted first with 20 mL of hexane, which was collected but not analyzed, followed by elution with 30 mL of 20% methylene chloride in hexane. The PCDDs and PCDFs were eluted from the acidic alumina using the 20% methylene chloride in hexane. The PCDDs and PCDFs in the eluates were isolated from other chlorinated planar aromatics using columns (5-mL disposable pipettes containing 500 mg of 18% Carbopak C and Celite-545). The Carbopak C/Celite 10 �Table 4. Internal Standard Spiking Solutions Concentration (pg/uL) Compound Internal quantisation standard 13 5 13 5 13 5 13 5 C12-2,3,7,8-TCDD C12-2,3,7,8-TCDF C12-l,2,3,7,8-PeCDD C12-l,2,3,7,8-PeCDF 13 12.5 13 12.5 13 12.5 C12-l,2,3,6,7,8-HxCDD C12-l,2,3,4,7,8-HxCDF C12-l,2,3,4,6,7,8-HpCDD 13 C12-OCDD 25 Internal recovery standard 13 C12-1,2,3,4-TCDD 50 .Solution prepared in isooctane. Solution prepared in tridecane. 11 �columns were pre-eluted with 2 ml of toluene, 1 mL of 75:20:5 methylene chloride/methanol/benzene, 1 mL of 1:1 methylene chloride/cyclohexane, and 2 ml of hexane. The sample extracts (30 ml) were added to the columns, which were eluted with 2 mL of hexane, 1 mL of 1:1 methylene chloride/cyclohexane, and 1 mL of the 75:20:5 methylene chloride/methanol/benzene. These eluents were collected and combined but were not analyzed. The Carbopak C/Celite columns were turned upside down, and the PCDDs and PCDFs were eluted with 20 mL of toluene. The toluene was concentrated to less than 1 mL using flowing nitrogen and a heated water bath (55-60°C) and transferred to 1.0-mL conical vials using a solution of 1% toluene in methylene chloride. Tridecane (10 uL) containing 500 pg of the internal recovery standard 13C12-1,2,3,4-TCDD was added as a keeper when the solution had concentrated to approximately 200 uL. The extracts were concentrated to final volume using nitrogen and the heated water bath. D. HRGC/MS Analysis The analyses of the spiked and unspiked lipid samples were completed using a Kratos MS50TC double-focusing magnetic sector mass spectrometer. The determination for the tetra- through octachloro homologs was achieved in a single analysis using the conditions described in Table 5. Table 6 provides the characteristic ions monitored for each PCDD and PCDF homolog. As noted from Table 6, the analysis requires five different parameter descriptions that were switched automatically during the course of the analysis. Parameters monitored included two characteristic molecular ions for each PCDD and PCDF homolog and the corresponding carbon-13 labeled internal standard. In addition, a fragment ion of perfluorokerosene (PFK), m/z 380.976, was monitored throughout each analysis to ensure that proper mass calibration was maintained. The parameter descriptors also included an ion characteristic of specific homologs of chlorinated diphenyl ethers to demonstrate that responses meeting the qualitative criteria for specific PCDF congeners were not due to these potential interferences. Triplicate analyses of six of the eight calibration solutions (Table 2) were completed, and the variability in relative response factors across this range was calculated. The analyst was required to demonstrate on a daily basis that the relative response factors (RRF) were in agreement within ± 20% of the established averages for 2,3,7,8-TCDD and 2,3,7,8-TCDF and within ± 30% of the average RRF values for the other congeners. The equation used to calculate the relative response factors for each PCDD and PCDF congener are discussed later in this report (Section E - Data Interpretation, 2 - Quantitation, p. 17). The analyst was also required to determine column performance by analyzing a mixture of TCDD isomers before proceeding with sample analysis. Table 7 gives an example of the typical daily sequence for PCDD/PCDF analysis. 12 �Table 5. HRGC/LRMS Operating Conditions for PCDD/PCDF Analysis Mass spectrometer 8,000 V 500 MA 70 eV -1,800 V 280°C Accelerating voltage: Trap current: Electron energy: Electron multiplier voltage: Source temperature: Resolution: Overall SIM cycle time: ^ 3,000 (10% valley definition) 1s Gas chromatograph Column coating: Film thickness: Column dimensions: He linear velocity: He head pressure: DB-5 0.25 60 m •v 25 1.75 Injection type: Split flow: Purge flow: Injector temperature: Interface temperature: Injection size: Initial temperature: Initial time: Temperature program: Splitless, 45 s 30 mL/min 6 mL/min 270°C 300°C 1-2 pL 200°C 2 min 200°C to 330°C at 5°C/min 13 (jm x 0.25 mm ID cm/sec kg/cm2 (25 psi) �Table 6. Ions Monitored for HRGC/MS Analysis of PCDD/PCDF Descriptor Al ID TCDF 13 C12-TCDF TCDD 13 C12-TCDD HxCDPE PFK (lock mass) A2 TCDF TCDD PeCDF 13 C12-PeCDF PeCDD 13 C12-PeCDD PFK (lock mass) HpCDPE A3 HxCDF PFK (lock mass) 13 C12-HxCDF HxCDD 13 C12-HxCDD OCDPE Mass Nominal dwell time (sec) 303.902 305.899 315.942 317.939 319.897 321.894 331.937 333.934 373.840 380.976 0.090 0.090 0.090 0.090 0.090 0.090 0.090 0.090 0.090 0.090 303.902 305.899 319.897 321.894 337.863 339.860 349.903 351.900 353.858 355.855 365.898 367.895 380.976 407.801 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.035 0.035 373.821 375.818 380.976 385.861 387.859 389.816 391.813 401.856 403.853 443.759 0.080 0.080 0.080 0.080 0.080 0.080 0.080 0.080 0.080 0.080 �Table 6 (continued) Descriptor A4 Mass ID PFK (lock mass) HxCDD 380.976 389.816 391.813 407.782 409.779 419.822 421.819 423.777 425.774 435.817 437.814 429.768 431.765 477.720 C12-HpCDF HpCDD 13 37 C12-HpCDD Cl4-HpCDD NCDPE A5 PFK (lock mass) OCDF 13 C12-OCDF OCDD 13 C12-OCDD DCDPE 15 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 380.976 441.743 443.740 453.784 455.781 457.738 459.735 469.779 471.776 511.681 HpCDF 13 Nominal dwell time (sec) 0.060 0.070 0.070 0.070 0.070 0.070 0.070 0.070 0.070 0.060 �Table 7. Typical Daily Sequence for PCDD/PCDF Analysis 1. Tune and calibrate mass scale versus perf1uorokerosene (PFK). 2. Determine column performance by injecting the TCDD isomer mixture. 3. Inject concentration calibration solution 2.5 to 12.5 pg/|A (CS-7) solution. 4. Inject blank (tridecane). 5. Inject samples 1 through "n". 6 Inject concentration calibration solution 2.5 to 12.5 pg/pL (CS-7) solution. 16 �E. Data Interpretation 1. Qualitative The HRGC/MS elution profiles of the tetra- through octachloro PCDD and PCDF homologs were established through the analysis of environmental sample extract (fly ash from a municipal waste incinerator). The characteristic ions for each homolog were plotted within the retention window established using this mixture. The criteria for identification of a response as a PCDD or PCDF were the coincidental response of the characteristic ions monitored within the established retention window, and within ± 20% of the theoretical ion ratio. Table 8 presents the range of ion ratios used for the qualitative criteria for the specific PCDD and PCDF homologs and internal standards. 2. Quantisation Quantisation of the specific PCDD and PCDF congeners was achieved using the respective internal quantisation standards. For example, TCDD was quantitated versus the 13C12-2,3,7,8-TCDD; PeCDD versus the 13C12-1,2,3,7,8PeCDD, etc. The HpCDF and OCDF responses were quantitated versus the carbon-13 labeled hepta- and octachlorodibenzo-p_-dioxin internal standards since the corresponding dibenzofuran internal standards were not available for this study. The absolute recovery of the internal quantisation standards was achieved using 13C12-1,2,3,4-TCDD. A second internal recovery standard, 37 Cl4-l,2,3,4,6,7,8-HpCDD, was evaluated but was not used for recovery measurements due to interference arising from the corresponding native HpCDD. Relative response factors (RRF) were calculated for each of the native PCDD and PCDF compounds listed in Table 2. The RRF values were calculated as shown in Equation 1. "crn x STD RRF = A R T<^ 1S X c IS X LSTD Eq. 1 where Arjn = the sum of the area responses for the two characteristic ions of the standard compound; AT<- = the sum of the area responses for the two characteristic ions of the corresponding internal quantisation standard; CIS = concentration (pg/|jL) of the internal quantisation standard; and = concentration (pg/uL) of the standard compound. The relative response factors for the internal quantisation standards (RRFjS) were calculated as shown in Eq. 2. 17 �Table 8. Compound TCDF 13 C12-TCDF TCDD 13 C12-TCDD PCDF 13 C12-PeCDF PeCDD 13 C12-PeCDD HxCDF 13 C12-HxCDF HxCDD 13 C12-HxCDD HpCDF 13 C12-HpCDF HpCDD 13 C12-HpCDD OCDF 13 C12-OCDF OCDD 13 C12-OCDD Ion Ratios for HRGC/MS Analysis of PCDD/PCDF Ions monitored Theoretical ratio 304/306 316/318 320/322 332/334 338/340 350/352 354/356 366/368 374/376 386/388 390/392 402/404 408/410 420/422 424/426 436/438 442/444 454/456 458/460 470/472 0.76 0.76 0.76 0.76 0.61 0.61 0.61 0.61 1.22 1.22 1.22 1.22 1.02 1.02 1.02 1.02 0.87 0.87 0.87 0.87 Acceptable range is ± 20% of the theoretical value. 18 Acceptabl e range' 0.61 - 0.91 0.61 - 0.91 0.61 - 0.91 0.61 0.49 0.49 0.49 0.49 0.98 0.98 0.98 0.98 0.82 0.82 0.82 0.82 0.70 0.70 0.70 0.70 - 0.91 0.73 0.73 0.73 0.73 1.46 1.46 1.46 1.46 1.22 1.22 1.22 1.22 1.04 1.04 1.04 1.04 �TC * RRF IS IS = QG. Eq AH RS x CLjg X "2 where A JS and GIS are defined as in Equation 1 and C10 = concentration (pg/uL) of the internal recovery standard, DC 13 C12-1,2,3,4-TCDD, and A RS = the sum of the area responses for the two characteristic ions (m/z 332 and 334) corresponding to the internal recovery standard. A calibration curve was established using six concentration levels of standards; for example, the calibration curve for 2,3,7,8-TCDD was initially established with standards at concentrations of 1, 2.5, 5, 10, 50, and 100 pg/(jL. The 2.5 pg/uL standard was analyzed daily to verify response factors and instrument sensitivity. The RRF values for each of the internal quantisation standards were calculated versus the internal recovery standard, 13 C12-1,2,3,4-TCDD, using Equation 2. The concentration of a PCDD or PCDF congener in a composite sample was calculated as shown in Equation 3. _ Asamp1e x QIS Vr AIS x RRF x Wt r where P tq> ~ * CUT = wet tissue concentration of the PCDD or PCDF congener in each tissue (pg/g); A sample = sum of the area responses for the two characteristic ions of , thfi pC[JD Qr pCDF congener; A-,<- = sum of the area responses for the two characteristic ions of the respective internal quantitation standard; Q T r = amount of the internal quantitation standard added to the 1:> 13 sample (500 pg of 13C12-TCDD,13 C12-TCDF, 13C12-PeCDD, 13 and 13C12-PeCDF; 1,250 pg of C12-HxCDD, 13C12-HxCDF, and C12-HpCDD; or 2,500 pg of 1§C12-OCDD); RRF = the relative response factor for the PCDD or PCDF congener from Equation 1; and Wt = mass of the sample (grams). The absolute recovery of the internal quantitation standard was calculated using Equation 4. 19 �L Recovery (%) = •? * ^ ARS x RRF IS x QIS where x 100 Eq 4 AR<- = sum of the area responses for the two characteristic ions of the internal recovery standard, 13C12-1,2,3,4-TCDD; QR<- = amount of the internal recovery standard (13C12-1,2,3,4-TCDD) added to the final extract (500 pg); and RRF IS = response factor of the internal quantisation standard relative to the internal recovery standard. These values are calculated as defined in Equation 2. The RRF T< - values were all calculated versus 13C12-1,2,3,4-TCDD. All data were qualified to reflect that the response for a particular compound was a positive quantifiable parameter, present as a trace value only, or was not detected. Positive quantifiable values were identified for responses greater than 10 times background signal to noise. Trace (TR) values were assigned to responses which were in the range of 2.5 to 10 times background signal to noise. A value of not detected (ND) was used to reflect that a response was not detected at greater than 2.5 times signal to noise. A limit of detection (LOD) was calculated for all trace and not detected values using the peak height response of the respective internal standard and the average measured signal to noise for the characteristic ions of the PCDDs and PCDFs. F. Quality Assurance/Quality Control (QA/QC) The QA/QC procedures included analysis of multipoint calibration concentration standards to establish relative response factor (RRF) curves for each of the 17 native PCDD and PCDF congeners. Triplicate analyses of 6 concentration calculation standards (Table 2; CS2, CSS, CSS, CS6, CS7, and CSS) were determined to vary by less than ± 20% for TCDD and TCDF and ± 30% for all other PCDD and PCDF congeners. The mean RRF values were also determined to vary by less than this criteria over the entire calibration range. The mean RRF (RRF) values and instrument sensitivity were verified daily by bracketing the sample analyses with an injection of a standard that ranged from 2.5 pg/uL for 2,3,7,8-TCDD and 2,3,7,8-TCDF up to 12.5 pg/(jL for OCDD and OCDF. The criterion for continuing with the sample analysis was agreement of the measured RRF value with the mean RRF within ± 20% for 2,3,7,8TCDD and TCDF and ± 30% for all other PCDD and PCDF congeners. Other activities included the analysis of laboratory method blanks and reagent blanks and measurement of the absolute recoveries of the internal quantisation standards. Laboratory method blanks were samples that were handled exactly as an adipose sample except no lipid matrix was used. G. Preliminary Method Studies Prior to analysis of the homogenized human adipose lipid matrix by HRGC/MS, several experiments were conducted to confirm that the sample preparation scheme was feasible. 20 �1. Gravimetric Studies The first concern was the efficient removal of up to 10 g of lipid matrix from extracted adipose tissue. A series of experiments was conducted with 10-g lipid aliquots to demonstrate removal of lipid using the H2S04-Si02 slurry technique. Initially, 50 g of the acid modified silica was added to the lipid extract in 100 ml of hexane. The acid modified silica was noted to turn dark brown immediately on contact with the lipid solution. The hexane was recovered and the adsorbent was extracted with additional hexane. The extracts were combined and concentrated to 5 mL with Kuderna-Danish evaporators. The extract was eluted through a column of 4.0 g of acid modified silica and 1.0 g of silica with 45 ml of hexane. The acid modified silica was noted to be highly discolored throughout, and the extract required a second slurry treatment of the eluent with an additional 50 g of acid modified silica gel. The adsorbent from the second slurry procedure was noted to discolor significantly, indicating that lipid materials had not been efficiently removed from the first step of the procedure. The hexane supernatant from the second slurry cleanup was reduced in volume and taken to dryness in a preweighed glass vial. The final residue was measured at approximately 10 mg for duplicate samples, which translates into a removal efficiency of 99.9% based on the initial 10-g aliquot. This lipid cleanup procedure was modified such that 100 g of acid modified silica gel was used in the initial slurry cleanup, followed by elution of the resulting extract through a column containing 4.0 g of acid modified silica and 1.0 g of silica gel. The lipid removal efficiency of duplicate samples through the cleanup procedure was determined to average 99.8% (20 to 30 mg of the initial 10-g lipid remained after cleanup). The column cleanup step in this procedure did not exhibit any significant color change. Thus this step of the procedure was incorporated into the method as a check of the efficiency of lipid removal to prevent overloading of the acidic alumina fractionation column. 2. Carbon-14 Recovery Studies The carbon-14 radio!abeled PCDD standards listed in Table 1 were used to estimate overall method recoveries for the tetra- through octachloro homologs prior to proceeding with the HRGC/MS evaluation. Triplicate analyses (10-g aliquots of lipid materials) were conducted with each of the three radiolabeled standards. The first experiment addressed the recovery of the compounds from bulk lipid cleanup. Triplicate analyses using 10-g aliquots were completed for the three compounds at the following concentrations: 14 C-2,3,7,8-TCDD, 10 pg/g; 14C-l,2,3,4,7,8-HxCDD, 100 pg/g; and 14C-OCDD, 250 pg/g. The results of these analyses indicated that all compounds were recovered in the range of approximately 70 to 80%. Following this experiment, the total sample preparation procedure described earlier in this report was evaluated using triplicate analysis of 10-g lipid samples. Table 9 provides a summary of the results from this study. These data indicate that overall method recovery is limited by the initial bulk lipid removal procedure. This assumption is based on the similar recoveries of the carbon-14 labeled compounds noted for evaluation of the bulk lipid removal step as compared to the total sample preparation scheme. 21 �Table 9. Summary of the Results of the Sample Preparation Method Evaluation Using Carbon-14 PCDDs Spike levels Bulk lipid removal, recovery Analytes (pg/g) Total method a recovery ( ) % 14 C-2,3,7,8-TCDD 10 68 75 14 100 79 66 250 82 76 C-l,2,3,4,7,8-HxCDD 14 C-OCDD Average value for triplicate analyses taken through the total sample preparation scheme. Precision of measurements varied by less than ± 10% (relative standard deviation). Average value for triplicate analyses taken through bulk lipid cleanup only. Precision of measurement varied by less than ± 6% (relative standard deviation). 22 �V. RESULTS A. Analytical Results The analytical results for the quantisation of the 17 target PCDD and PCDF 2,3,7,8-substituted congeners in the spiked and unspiked homogenized human adipose lipid samples are presented in Tables 10 to 15. These data demonstrate that 13 of the 17 congeners were definitely detected in the unspiked lipid matrix. Although 2,3,7,8-TCDF is reported as not detected, responses for the characteristic ions (m/z 304 and 306) greater than 10 times signal to noise were noted to be coincident with the internal standard, 13C122,3,7,8-TCDF. The ratio of the responses (m/z 304/306) in each of the triplicate analyses of the unspiked matrix were well outside the acceptable ratio of 0.90 to 0.61 established in Table 8. Figure 1 provides a comparison of the HRGC/MS-SIM responses noted for the unpsiked human adipose lipid matrix as compared to a concentration calibration standard. Figures 2 through 6 provide examples of the individual PCDD and PCDF responses observed for the unspiked lipid samples as compared to fortified matrices. In general, the precision of the replicate measurements at each spike level is good (relative standard deviations typically less than 10%) for PCDD and PCDF congeners that were detected with responses greater than 10 times signal to noise. The precision of the measurements for the unspiked matrices for 1,2,3,7,8-PeCDF (Table 11), 1,2,3,7,8,9-HxCDF (Table 12), and OCDF (Table 15) ranges from 21.6% to 43.1% as a result of little or no response at the specified retention windows. B. Statistical Analysis The regression results for each of the 17 specific PCDD and PCDF congeners are plotted separately in Figures 7 to 23. These plots provide the results for the individual sample analyses, a line defining the results of a least squares regression analysis, and boundaries that depict the confidence limits for the range of spiked concentrations. The regression lines were obtained by the method of least squares using the sample measurements at the three spiking levels and the unspiked level. Two types of upper and lower 95% confidence limits or bounds were calculated for the least square regressions of measured (found) concentrations versus spiked levels. The first set of confidence limits (defined by the inner pair of curves closest to the regression line) is the 95% confidence bounds for the regression line. These bounds are interpreted as follows: The true regression line (as would be determined if the experiment were repeated a countless number of times at the same spiked levels) lies within these confidence limits unless the analytical results are sufficiently unusual to be among those expected to occur less than 5% of the time. The second set of confidence bounds, depicted by the outer pair of lines, constitutes the 95% confidence limits for the result of a single analysis at a particular spiking level. The interpretation is as follows: the result (reported value) of a single analysis of a sample spiked at a given level can be predicted to fall between these 95% confidence bounds unless the analytical result is among those sufficiently unusual to be expected less than 5% of the time. 23 �Table 10. Spiked Versus Measured Concentrations of 2,3,7,8-TCDF and 2,3,7,8-TCDD in Homogenized Human Adipose Lipid Samples 13 13 C12-TCDF absolute recovery ( ) % 2,3,7,8-TCDD spike level (pg/g) (pg/g) C12-TCDD absolute recovery (%) (.) 41* ( . )a 41? (.) 40 (.) 41 0.1 1.8 59 63 59 60.3 2.3 3.8 0 0 0 10.7 11.4 13.1 11.7 1.2 10.5 53 52 46 50.3 3.8 7.5 10 10 10 10 14.3 14.8 13.6 13.8 14.1 0.5 3.9 61 67 71 78 69.3 7.1 10.3 10 10 10 10 23.4 22.8 24.7 22.5 23.3 1.0 4.1 53 53 53 62 55.3 4.5 8.1 25 25 25 25 30.8 30.8 30.7 28.7 30.2 1.1 3.5 59 75 62 72 67.0 7.7 11.5 25 25 25 25 40.8 40.6 40.3 38.3 40.0 1.2 2.9 48 53 51 60 53.0 5.1 9.6 50 50 50 57.7 59.4 55.8 57.6 1.8 3.1 64 48 58 56.7 8.1 14.3 50 50 50 65.8 72.1 67.6 68.5 3.3 4.8 55 43 48 48.7 6.0 12.4 2,3,7,8-TCDF spike level 2,3,7,8-TCDF concentration (pg/g) 0 0 0 Mean STD RSD (%) ro Mean STD RSD ( ) % Mean STD RSD (%) Mean STD RSD (%) (pg/g) ND ND ND ND 2,3,7,8-TCDD concentration ND = not detected. Value in parentheses is the estimated limit of detection. A response of greater than 10 times signal-to-noise was noted for both characteristic ions (m/z 304 and 306) at the appropriate retention time for 2,3,7,8-TCDF. However, the ion ratio was considerably greater than the acceptable range of 0.61 to 0.90. �i au ic J.A. Table 11. 1,2,3.7,8-PeCDF spike level (pg/g) 0 0 0 STO RSD ( ) % 10 10 10 10 Mean STO RSD ( ) % 25 25 25 Mean STD RSD ( ) % a ND (l.l) ND ( . ) 08 ND ( . ) 08 12.2 11.5 11.9 11.4 29.2 30.1 28.5 25.5 50 51.3 55.9 55.5 54.2 2.5 4.7 C12-PeCDF absolute 2,3,4,7,8-PeCDF concentration (pg/g) (pg/g) 0 0 0 20.8 21.6 19.0 75 76 80 20.5 1.3 6.5 10 10 10 10 27.6 recovery (%) 1,2,3,7,8-PeCDD concentration (pg/g) (pg/g) 0 0 0 20.2 19.9 18.1 51 54 57 19.4 1.1 5.7 54.0 3.0 5.6 32.1 37.6 30.8 30.4 60 57 55 62 30.2 1.9 6.3 58.5 3.1 5.3 48.0 43.5 55 60 57 64 46.7 59.0 2.5 5.4 3.9 6.6 3.4 . 32.4 4.9 15.2 50 50 50 54.9 48.5 41.7 48.2 62 81 87 84 78.5 11.3 14.4 74.6 62.9 71.9 64 69.8 69.3 5.0 7.3 10 10 10 10 78.0 11.7 15.1 48.4 5.4 11.1 25 25 25 25 36.3 90 63 84 75 37.0 28.7 6.1 8.8 NO = not detected. The value in parentheses is the estimated limit of detection. 74 70 13 1,2,3,7,8-PeCDO spike level 77.0 2.6 28.3 2.0 7.0 50 50 Mean STD RSD (%) (pg/g) 13 2,3,4,7,8-PeCDF spike level 11.8 0.4 3.1 25 CJl 1,2,3,7,8-PeCDF concentration ,t versus neasureij uunueniT at IUMS ui .L ,£ , j , / ,o rex-ur f &(j , / ,o rev,Lrr „ anu x ,£. , i ,o rc^uu t j, iii Homogenized Human Adipose Tissue Sampl es ND (0.9) 0.2 21.6 Mean ro j\j 1 1«;u 25 25 25 25 50 50 50 46.1 49.3 C12-PeCDO absolute recovery (%) 72.2 72.8 58 54 56 71.6 56.0 69.7 1.6 2.3 2.0 3.6 �Table 12. in Homogenize(1 Human Adipose L ipid Matrix 1,2,3,4,7,8HxCDF spike level (P9/Q) 0 0 0 Mean STD RSD ( ) % Mean STD RSD (X) Mean STD RSD ( ) % 125 125 125 1,2,3,6,7,8HxCOF concentration 2,3,4,6,7,8HxCDF spike level 2,3.4,6.7.8HxCDF concentration 1,2,3,7,8,9HxCDF spike level 1,2,3,7,8,9HxCDF concentration 13 C12-HxCDF (pg/g) (pg/g) (pg/g) (pg/g) (pg/g) (pg/g) (pg/g) absolute recovery () % 22.0 22.1 22.4 0 12.3 12.4 12.7 0 0 0 4.9 4.2 4.2 0 NO (0.5)a NO ( . ) 07 NO ( . ) 09 55 57 52 ND ( . ) 07 0.2 25.3 55.7 2.5 4.6 concentration 0 0 46.7 52.0 48.8 49.0 25 25 25 25 90.2 89.1 90.3 90.7 156.3 0.9 0.6 25 25 25 25 62.5 62.5 62.5 62.5 83.7 80.5 79.6 76.0 62.5 62.5 62.5 62.5 3.2 3.9 151.0 157.7 149.1 32.3 33.6 32.0 32.5 25 25 25 25 74.7 75.5 74.2 71.1 125 125 125 152.6 4.5 2.9 ND = not detected. The value in parentheses reflects the estimated limit of detection. 141.9 141.6 151.0 144.9 5.3 3.7 34.8 28.5 30.9 29.2 30.9 2.8 • 9.1 59 57 60 67 60.8 4.3 7.2 125 125 125 82.7 89.4 82.2 76.3 60 63 63 70 82.7 5.4 6.5 62.5 62.5 62.5 62.5 73.9 2.0 2.6 79.9 125 125 125 0 32.6 0.7 2.1 39.0 1.8 4.7 90.1 0.7 0.7 157.2 155.4 156.4 36.6 38.6 40.8 39.8 0 4.4 0.4 8.9 12.4 0.2 1.7 49.1 2.2 4.4 62.5 62.5 62.5 62.5 Mean STD RSD (X) 1,2,3,6,7,8HxCDF spike level 22.2 0.2 1.1 25 25 25 25 no en 1,2,3,4,7,8HxCDF 64.0 4.2 6.6 143.0 144.1 163.4 57 54 57 150.1 11.4 7.6 56.0 1.7 3.1 �Table 13. in Homogenized Human Adipose Lipid Samples 1 1,2,3,4,7,8-HxCDD concentration 1,2,3,7,8,9-HxCDD 1,2,3,7,8,9-HxCDD concentration spike level concentration (pg/g) 1,2,3,6,7,8-HxCDD spike level 1,2,3,6,7,8-HxCDO spike level (pg/g) (pg/g) (pg/g) (pg/g) (pg/g) 0 21.6 22.7 20.3 0 0 0 157.0 162.0 154.0 0 0 0 19.1 26.0 2,3,4,7,8-HxCDD 0 0 Mean STD RSD ( ) % 25 25 25 25 Mean STD RSD ( ) % 62.5 Mean STD RSD ( ) % 82.9 96.7 90.3 81.9 141.1 150.8 146.0 145.9 4.9 3.3 184.0 165.0 198.0 193.0 62.5 62.5 62.5 62.5 220.0 239.0 220.0 220.0 288.0 299.0 266.0 284.3 16.8 5.9 63.5 46.1 40.8 57.3 65 61 65 70 51.9 10.4 20.0 65.3 3.7 5.6 3.7 25 25 25 25 125 125 125 114.0 99.9 101.2 107.4 64 66 66 77 105.6 6.5 6.1 62.5 62.5 62.5 62.5 224.8 9.5 4.2 125 125 125 58 60 58 58.7 1.2 2.0 23.2 185.0 14.5 7.9 87.9 7.0 7.9 125 125 125 Mean STD RSD ( ) * 25 25 25 25 51.6 2.6 5.1 62.5 62.5 62.5 ro --J 47.8 52.0 53.7 52.7 C12-HxCDD absolute recovery (%) 15.8 157.7 4.0 2.6 21.5 1.2 5.5 24.7 13 68.3 5.9 8.7 141.3 152.7 189.4 62 57 61 161.1 .25.2 15.6 60.0 2.6 4.4 �Table 14. Spiked Versus Measured Concentrations of 1,2,3,4,6,7,8-HpCDF, 1,2,3,4,7,8,9-HpCDF, and 1,2,3,4,6,7,8-HpCDD in Homogenized Human Adipose Lipid Samples 1 ?,3,4,6,7,8-HpCDF 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF spike level concentration (pg/g) 0 0 0 Mean RSO ( ) % 25 25 25 25 Mean STD RSD ( ) % ro 00 (pg/g) (pg/g) (pg/g) (pg/g) (pg/g) 30.6 27.3 28.6 0 0 0 NO (1.3)* NO ( . )a 11* NO ( . ) 10 0 0 0 214.7 210.8 215.5 71 74 69 213.7 2.5 1.2 71.3 2.5 3.5 214.7 239.0 243.9 248.5 83 75 76 78 243.3 4.0 1.7 78.0 3.6 4.6 281.5 288.1 269.9 274.9 77 84 90 99 278.6 8.0 2.9 87.5 9.3 10.7 353.2 355.4 343.7 62 61 69 350.8 6.2 1.8 64 4.4 6.8 48.6 48.6 51.0 56.3 ND ( . ) 11" 0.2 13 25 25 25 25 Mean STD RSD ( ) * 86.3 85.7 80.9 83.4 62.5 62.5 62.5 62.5 84.1 2.5 2.9 125 125 125 154.5 157.3 154.4 155.4 1.6 1.0 ND - not detected. 26 23 23 25 25 25 25 25 24.1 1.4 5.7 51.1 3.6 7.0 62.5 62.5 62.5 62.5 C12-HpCDD absolute recovery ( ) % spike level 1,2,3,4,6,7,8-HpCDD spike level 28.9 1.6 5.7 STO l3 1,2.3,4,7,8,9-HpCDF concentration 60 60 59 58 62.5 62.5 62.5 62.5 59.0 0.8 1.3 125 125 125 119 126 121 122.4 3.7 3.0 The value in parentheses is the estimated limit of detection. 125 125 125 1,2,3.4,6,7,8-HpCDD concentration �Table 15. Spiked Versus Measured Concentrations of OCDF and OCDD in Homogenized Human Adipose Lipid Samples 13 OCDF spike level OCDF concentration spike level OCDD concentration (pg/g) (pg/g) (pg/g) (pg/g) 0 0 0 4.9 0 0 0 804 833 781 88 94 806.1 26.1 3.2 91.0 3.0 3.3 Mean STD RSD (%) 50 50 50 50 Mean STD RSD (%) 125 125 125 125 Mean STD RSD ( ) % 250 250 250 Mean STD RSD ( ) % 2.3 2.6 OCDD 3.2 1.4 43.1 44.2 45.0 45.9 49.6 50 50 50 50 849 856 876 860 860.4 11.4 46.2 2.4 5.2 111.1 107.8 110.7 113.1 1.3 125 125 125 125 () % 91 100 87 91 91 92.3 5.5 6.0 932 934 944 90 96 102 907 110.7 2.2 2.0 227.8 231.0 228.6 C12-OCDD abolute recovery 104 929.1 15.7 1.7 250 250 98.0 6.3 6.5 1,080 1,140 1,070 1,096 34.7 3.2 250 229.1 1.7 0.7 29 69 67 74 70.0 3.6 5.2 �Umpiked Human Adipose Tiisue 13 C , i,2.: TCDF ,3.7.B-Kf>F ,?.3.4-?CDO icon .3.7.IMCDU B-fcCDr 13C| ?,3.< 8-PfCDf 8-N-CDD 'Vl' . 1,2.: tt. t.2,3 ,3.6,7,B-M.COD 9-Hi-COlJ 9'lliiCDf 7.B-HpCDF 7,i-HpCt>n ,3,«.4.7,ft-hpCDD .7.3.4,«,7,B-HpCDD ,7,B,9-HpCDF 75 OCOO 36 '3C17-OCDO 37 OCDf R1C 21,22.23 16.17 1B19 W^W**< JlJVWsMM. izee Calibrolion Slaiidorcl 1,12 RIC 15,16,17 6,7 21,22.23 9.10 roe 12M 14m 16M 2«M SOW Figure 1. Comparison of the HRGC/MS-SIM reconstructed ion chromatogram (RIC) from the analysis of unspiked homogenized human adipose tissue matrix and a calibration standard for PCDDs and PCDFs. 30 �Umpik.d Human Adipau 384 . Spiked Human Adipaic 2.3.7,8-TCDF 304 . 9M nee 958 use 1286 1256 SCAN Umpiktd Humon Adipou 328 . r !3 C|2-lCDF ^2.3,7,8-ICDO A. A. iee.e-i Spiked Humon Adipoie 32« . >3C|2-2.3.7,8-TCDFi 9flO tew -2.3.7,8-TCDD urn 1280 1230 SCAN Figure 2. Example of the TCDF (m/z 304) and TCDD (m/z 320) HRGC/MS-SIM elution profiles in unspiked and spiked human adipose. The spiked concentrations for TCDF and TCDD in these chromatograms were 25 pg/g each. 31 �Urapik.d Human Adlpow 2.3.4.7.B-P.CDF 338 . 1.2.3,7.8-P.CDF \ Spiked Human Adipnia 2.3.4,7,8-P.CDF 1.2,3,7,8-P.CDF-, 338 . nee use 12*9 izse I3*e t3se 1400 SCMN Unspiktd Human Adipose 3C|2-I.2.3.7,8-P.CDF 354 . (1.2,3,7.88-P.CDD .. A . Spik«d Human AdipoM 354 . .2.3.7.8-P.COD 1308 13M I4W SCAN Figure 3. Example of the PeCDF (m/z 338) and PeCDD (m/z 354) HRGC/MS-SIM elution profiles in unspiked and spiked human adipose. The spiked concentrations for each isomer of PeCDF and PeCDD in these chromatograms were 25 pg/g. 32 �Umpiked Human Adipose |l.2,3,<.7.8-HxCDf h,2.3.6.7.8-HxCDF 374 . 2.3.4.6,7,8-HxCDF Spiked Human Adipote l.2,3.<.7,8-HxCLIF | | l.2.3.6.7.8-HxCDF r2,3,4,6.7.6-. \H»CDF 374 . 1.2,3.7,8.9-HxCDF 1343 1238 1366 1363 1358 1558 U_B9 1468 MSB 1588 1558 1668 SMI Umpiked Human Adipole 1.2,3.6.7.B-HxCDD 1,2.3,4.7.8-llxCDD- 339 . '3C)2-l,2.3,4,7,8-hxCDF A 1,2.3.7.8,9-Hx Spiked Human Adi ,2.3,6.7.fl-HxCDD 3M . 1.2.3,4,7.8-HxCDD \ 1.2,3,7.6.9-HxCDD l3C|2-l,2.3,<.7,8-HxCDFl 1398 1356 1488 1450 1588 1558 16!« 1E58 SOW Figure 4. Example of the HxCDF (m/z 374) and HxCDD (m/z 390) HRGC/MS-SIM elution profiles in unspiked and spiked human adipose. The spiked concentrations for each isomer of HxCDF and HxCDD were 62.5 pg/g. 33 �Umpik.d Human AdlpoM ,2,3.4.6,7,8-HpCDF 468 . Splk.d Human Adipotl l,2.3.4.6.7.8-HpCDF 488 . Jl.2.3.4.7.8,9-HpCDF MM ISM 1556 1CM I65B 1768 [750 18M SCAN Umpikcd Human Adipau 1,2.3.4.6,7,8-HpCDD 424 . Spik.d Human Adipoit 1.2,3,4.6.7,8-MpCDO 424 . 1588 1558 1688 1638 17W 1758 1868 1858 SCtH Figure 5. Example of the HpCDF (m/z 408) and HpCDD (m/z 424) HRGC/MS-SIM elution profiles in unspiked and spiked human adipose. The spiked concentrations for each isomer of HpCDF and HpCDD were 62.5 pg/g. 34 �ll3c,2-OCDD Unipik.d Human Adipotc 442 . Spik.d Hunan Adipau 13 C|2-OCDD| AJ 17W 1758 ieee ISM 1658 1358 2988 SCAM Untpikad Human Adipose OCDD 458 . Spik.d Human Adipow OCOD 458 . 1968 2MB SCAN Figure 6. Examples of the OCDF (m/z 442) and OCDD (m/z 458) HRGC/MS-SIM elution profiles in unspiked and spiked human adipose. The spiked concentrations for OCDD and OCDF were 125 pg/g each. 35 �80 70 - Regression Line- t» tt a vx d o •H +> 60 J 95% Confidence Limifs for Regression Line 50 -J td 0 CO a ti o u 30 - 95% Confidence Limifs for Individual Analyses 10 0 I 20 10 D 30 i 40 Spiked Concentration (pg/g) lipid sample meas. Figure 7. Measured concentrations versus concentrations of 2,3,7,8-TCDD spiked into the homogenized human adipose lipid matrix. 50 �80 70 - Regression Line- ue (4 95% Confidence Limits for Regression Line 60 - o •H 50 - d 4) 0 GO -vl d 0 u •d d 4:0 - 30 - •95% Confidence Limits for Individual Analyses 10 i 10 r 20 D Figure 8. r 30 i 40 Spiked Concentration (pg/g) lipid sample meas. Measured concentrations versus concentrations of 1,2,3,7,8-PeCDD spiked into the homogenized human adipose lipid matrix. 50 �160 150 Regression Line- 140 130 C4 a 95% Confidence Limits for Regression Line 130 110 - o •H 100 90 - d 80 - fl 0 70 - 0 0 CO CO u 60 - •a d 50 95% Confidence Limits for Individual Analyses 40 30 20 10 0 I 40 I 20 D Figure 9. T 60 T 80 100 Spiked Concentration (pg/g) lipid sample meas. Measured concentrations versus concentrations of 1,2,3,4,7,8-HxCDD spiked into the homogenized human adipose lipid matrix. 120 �a v_/ o *H 55 u ti CO o u •d fl 330 310 300 390 380 370 360 350 340 330 330 310 300 190 180 170 160 150 140 H 130 Regression Line. 95% Confidence Limitsfor Regression Line 95% Confidence Limits for Individual Analyses I 40 30 D i 60 80 100 Spiked Concentration (pg/g) lipid sample meas. Figure 10. Measured concentrations versus concentrations of 1,2,3,6,7,8-HxCDD spiked into the homogenized human adipose lipid matrix. 130 �(4 M a 0 •H 4> ti u 0 0 d 0 u •0 C 200 190 180 170 160 150 140 130 120 110 100 90 80 H 70 60 50 40 30 20 10 0 -10 Regression Line 95% Confidence Limitsfor Regression Line -i - •95% Confidence Limits for Individual Analyses i 40 l 20 D Figure 11. i 60 i 80 100 Spiked Concentration (pg/g) lipid sample meas. Measured concentrations versus concentrations of 1,2,3,7,8,9-HxCDD spiked into the homogenized human adipose lipid matrix. 120 �Lfr Fovmd. (Q c n> T3 (t) _.. 0) 7T- (/> ro c a. -? n> - . Q. 1 3 r^ n o o r*- o 3- n> fD 3 c* 3" -5 o a> 3 r+ 0 -"• CQ o fD 3 3 tn N < fD fD O. -i (ft 3" C c in 01 n 3 o 3 o> n a. n> O -5 v> Q> fD r+ _j. —' O ->• 3 •O V> Q. O -h Q) I-1 -5 ro 00 •o o a a a Ul •a I—i M« •-" te* us SP- Sg 3 S •Eg a> 5*i BR o jr- ft O M 9 •o 9Q m Concentration �1.15 Regression Line- 1.1 M ft 95% Confidence Limitsfor Regression Line 1.05 1 0 n •H _l 41 ti H C 0.95 g 0 0.9 U~ 0.85 -| (4 ti ~ •a 3 •95% Confidence Limits for Individual Analyses 0.8 0 0.75 0.7 I 40 80 D 120 160 200 Spiked Concentration (pg/g) lipid sample meas. Figure 13. Measured concentrations versus concentrations of OCDD spiked into the homogenized human adipose lipid matrix. T 340 �60 Regression Line- 50 bl 95% Confidence Limitsfor Regression Line a d o 40 - 30 ti 0 GO a d o u •0 d •95% Confidence Limits for Individual Analyses 10 - 0 I 10 I 20 n Figure 14. 30 40 Spiked Concentration (pg/g) lipid sample meas. Measured concentrations versus concentrations of 2,3,7,8-TCDF spiked into the homogenized human adipose lipid matrix. 50 �60 Regression Line' 50 H 95% Confidence Limitsfor Regression Line M a 40 -^ c 0 30 fl 0 0 C 20 - 0 u •d a 95% Confidence Limits for Individual Analyses 10 - -10 D Spiked Concentration (pg/g) lipid sample meas. Figure 15. Measured concentrations versus concentrations of 1,2,3,7,8-PeCDF spiked into the homogenized human adipose lipid matrix. �90 Regression Line- 80 - tfl d o d 95% Confidence Limitsfor Regression Line 70 - 60 50 - 0 0 cn d u 0 30 •95% Confidence Limits for Individual Analyses 10 I I i 10 20 30 D Figure 16. i 40 Spiked Concentration (pg/g) lipid sample meas. Measured concentrations versus concentrations of 2,3,4,7,8-PeCDF spiked into the homogenized human adipose lipid matrix. 50 �160 150 140 M Regression Line- 130 95% Confidence Limitsfor Regression Line ti o 110 100 90 - ti 0) o ti 0 u •d fl 80 70 60 - •95% Confidence Limits for Individual Analyses 50 40 30 30 10 i 40 T 20 D Figure 17. i 60 T 80 100 Spiked Concentration (pg/g) lipid sample meas. Measured concentrations versus concentrations of 1,2,3,4,7,8-HxCDF spiked into the homogenized human adipose lipid matrix. 120 �160 150 -^ 140 M 130 - (4 130 - a 0 •H +> (fl u Regression Line 95% Confidence Limitsfor Regression Line 110 100 90 80 - 0 70 - 0 60 - ti u •d ti 50 •95% Confidence Limits for Individual Analyses 40 30 30 10 J 0 0 I 30 I 40 D Figure 18. i 60 r 80 100 Spiked Concentration (pg/g) lipid sample meas. Measured concentrations versus concentrations of 1,2,3,6,7,8-HxCDF spiked into the homogenized human adipose lipid matrix. 130 �/-s tt d o 0 0 CO ti 0 u •d a 160 150 140 130 130 110 100 90 80 70 60 50 40 30 20 10 0 -10 - 0 Regression Line- 95% Confidence Limitsfor Regression Line 95% Confidence Limits for Individual Analyses 20 40 D 60 80 100 Spiked Concentration (pg/g) lipid sample me as. Figure 19. Measured concentrations versus concentrations of 2,3,4,6,7,8-HxCDF spiked into the homogenized human adipose lipid matrix. 120 �170 Regression Line ttt 95% Confidence Limits for Regression Line * ft o •H 0 0 a 0 u •d fl 95% Confidence Limits for Individual Analyses 100 D Spiked Concentration (pg/g) lipid sample meas. Figure 20. Measured concentrations versus concentrations of 1,2,3,7,8,9-HxCDF spiked into the homogenized human adipose lipid matrix. 120 �130 120 - 110 01 100 - Regression Line- 90 - o •H 4* 95% Confidence Limitsfor Regression Line 80 70 60 - 4) 0 en O 50 - o 0 40 - TJ ti 30 - a 95% Confidence Limits for Indjvidual Analyses 20 10 H 0 -10 0 i 40 I 20 D i 60 i 80 100 Spiked Concentration (pg/g) lipid sample meas. Figure 21. Measured concentrations versus concentrations of 1,2,3,4,7,8,9-HpCDF spiked into the homogenized human adipose lipid matrix. 120 �Found Concentration. �240 220 200 U Regression Line > 180 - 95% Confidence Limitsfor Regression Line 160 - ti o 140 120 - d 100 -i fi 0 80 - 0 0 en ro u •d ti •95% Confidence Limits for Individual Analyses 60 40 20 - -20 r 80 I 40 0 D Figure 23. 120 160 200 Spiked Concentration (pg/g) lipid sample meas. Measured concentrations versus concentrations of OCDF spiked into the homogenized human adipose lipid matrix. 240 �The slopes of the calculated regression lines from the data points in each of the 14 analyses can be used as an indication of the accuracy of the analytical method for the 17 target analytes. Figure 24 is a plot of the slope of regression lines versus the 17 individual compounds. Table 16 provides a key to specific compounds associated with a number on the x-axis of this plot. The plot presents the estimated slope from each least squares regression line as well as the upper and lower 95% confidence limits for the slope. The slope of the regression line can be interpreted as a measure of accuracy with a value of 1.00 equivalent to 100% agreement of the measured concentration with the theoretical values (background plus spike level). The plot of the 95% confidence limits presents some confirmation on the precision of measurements across the four spike levels. These confidence bounds can also be used to determine whether the accuracy of the measurements (slope of regression line) is significantly different from 100% (or 1.00). If the vertical line connecting the lower and upper 95% confidence limits intersects with the horizontal line at 1, then the accuracy of the method (as determined from the regression line) is not significantly different from 100% (slope = 1.00). The results plotted in Figure 24 demonstrate that the method accuracies for 7 of the 17 analytes are not significantly different from 100%. On the other hand, if the upper and lower confidence limits are both greater than or both less than 1.00, then the accuracy of the method is significantly different from 100%. The data presented graphically in Figure 24 indicate that some positive bias (greater than 100%) is associated with the method accuracies for 9 of the 17 analytes while the measurements for a single analyte (OCDF) result in a slightly negative (less than 100%) bias. Table 16 provides a key to the compound identification in Figure 24 and tabulates the slope of the regression lines and the upper and lower 95% confidence limits for each of the 2,3,7,8-substituted PCDD and PCDF analytes. As noted from Table 16, method accuracy (as defined by the regression line slope) ranges from 90% for OCDF up to 121% for 1,2,3,7,8,9-HxCDF. The accuracies for all other measurements fall within a range of 97 to 115%. The overall method accuracies meet the initial accuracy objective of 50-115% identified in the project quality assurance program plan. However, the predicted accuracy results for individual analysis as defined by the 95% upper confidence limits indicate that this range should be adjusted to 50-130%. The bias in the accuracy of the measurements may be a result of slight differences in the concentration calibration standards and the internal quantisation standard and native PCDD and PCDF spiking solutions. As a preliminary check on these differences, solutions of the low level and of the high level native spike combined with the internal quantitation standards were analyzed. The results of these analyses are provided in Table 17. Accuracy was calculated as measured/spiked x 100. The results of these analyses suggest that bias observed in overall method accuracy is attributed to the differences in the spiking solutions versus the calibration standards. For instance, the four HxCDF isomers demonstrated a consistent positive bias to method accuracy based on the least squares regression analysis. The analysis of the spiking solutions, submitted as samples, also indicates a definite positive bias for the same four HxCDF isomers. 53 �ACCURACY ESTIMATES 1.3 - 1.2 n M £ o> « II -I- 1.1 - §• II :TT + 0.9 - 0.8 4(014(024(034(044(054(064(0741084(094110#11 4(124(134>144(154(1 64(17 D slope estimate Compound Number 4 lower 95* CL o upper 953C CL Figure 24. Method accuracy estimates as determined from the slopes of the least squares regression lines for the 17 target PCDD and PCDF analytes. Refer to Table 16 for the key to compound number. 54 �Table 16. Regression Line Slopes with 95% Confidence Limits Significantly Lower 95% different confidence Compound no. Compound Upper 95% confidence Slope from 1.00? limit limit 1.08 1.13 1.07 0.98 yes yes yes no 1.04 1.08 1.02 0.82 1.11 1.19 1.12 1.13 01 02 03 04 2,3,7,8-TCDF 2,3,7,8-TCDD 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 05 1,2,3,7,8-PeCDD 1.04 no 0.98 1.11 06 07 08 09 10 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 1,2,3,4,7,8-HxCDD 1.07 1.12 1.12 1.21 0.98 yes yes yes yes no 1.06 1.09 1.09 1.12 0.92 1.09 1.16 1.15 1.29 1.05 11 1,2,3,6,7,8-HxCDD 1.01 no 0.86 1.16 12 13 14 15 16 17 1,2,3,7,8,8-HxCDD 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF 1,2,3,4,6,7,8-HpCDD OCDF OCDD 1.12 1.01 0.97 1.08 0.90 1.15 no no no yes yes yes 0.94 0.95 0.95 1.01 0.88 1.00 1.30 1.07 1.00 1.16 0.92 1.30 55 �Table 17. Compound Results of the Analysis of the Low and High Level Native Spike Solutions Low level spike Measured Spike concentration concentration (pg/uL) (pg/uL) High level spike Accuracy Spike concentration (pg/uL) Measured concentration (pg/uL) Accuracy 108 114 2,3,7,8-TCDF 2,3,7,8-TCDD 13 12 130 120 50 50 54 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 10 10 10 11 10 12 110 1,2,3,7,8-PeCDD en 10 10 100 120 50 50 50 52 49 53 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 25 25 25 25 25 25 25 27 32 35 40 27 24 125 125 125 125 125 125 134 137 152 39 108 128 140 160 108 96 156 125 123 132 148 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF 1,2,3,4,6,7,8-HpCDD 25 25 25 22 25 26 88 100 104 125 125 125 116 121 130 97 104 OCDF OCDD 50 50 44 48 88 96 250 250 247 250 99 100 57 185 104 98 106 107 110 122 148 98 106 118 93 �Similar trends are noted for other compounds in Table 17 compared to the data presented in Figure 24 and Table 16. The limited number of analyses of the spiking solutions does not provide an adequate comparison with the sample data to confirm the bias. However, it is recommended that at least triplicate measurements of the spiking solutions at each fortification level should be analyzed at the outset of the actual NHATS sample analysis program. This will be necessary to account for any biases that will be observed from the determination of PCDD and PCDF residue levels in spiked QC samples. It should be noted that additional homogenized spiked samples will be prepared prior to initiation of the NHATS sample analyses. 1. Recovery of Internal Quantitation Standards The absolute recoveries for the carbon-13 labeled internal quantitation standards were determined for each sample by comparing the responses to the internal recovery standard, 13C12"1,2,3,4-TCDD. The average recoveries of the compounds13in Tables 10 to 15 range from 52.1% for 13C12-2,3,7,8-TCDD up to 88.9% for C12-OCDD. The results for the absolute recoveries compared to the overall method accuracy for each compound indicate the importance of the internal standard quantisation technique for analysis of the PCDDs and PCDFs in human adipose. 2. Estimation of Background Levels of PCDDs and PCDFs The estimated background levels of the various PCDD and PCDF congeners were determined as the intercept obtained from the least squares linear regression analyses. Table 18 provides a comparison of the average measured values for the unspiked matrix and the background concentration estimates from linear regression analysis of the data. In general, the measured and estimated background levels are in good agreement. However, several analytes with concentrations of less than 5 pg/g, particularly OCDF, demonstrate some disagreement in the measured versus estimated concentrations. This apparently arises from the fact that the first spike level is significantly greater than the actual background concentration. In the case of OCDF, the first spike level estimated by linear regression was 50 pg/g compared to an average measured value of 3.2 pg/g. In order to provide a better estimate of the background level based on the linear regression analysis, additional spike levels between 5 and 50 pg/g would be required. Table 18 also provides the upper and lower 95% confidence limits for the background levels estimated by the linear regression analysis of the data. These estimated background levels and confidence limits can be viewed as the intersections of the regression line and its upper and lower 95% confidence bounds, respectively, with the y-axis (measured or found concentration). These values will be used as the initial data points for developing control charts of the unspiked lipid matrix which will be analyzed with each batch of samples throughout the EPA/VA study. 57 �Table 18. Background Level Estimates with 95% Confidence Limits Compound no. 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 Measured background level (pg/g)a Compound 2,3,7,8-TCDF 2,3,7,8-TCDD 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,7,8-PeCDD 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 1,2, 3,4,7, 8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,6,7,8HpCDF 1,2,3,4,7,8,9HpCDF 1,2,3,4,6,7,8HpCDD OCDF OCDD ND ( . ) 41C 11.5 ND (0.9) 20.5 19.4 22.2 12.4 4.4 ND (0.7) 21.5 157.75 23.2 28.9 Estimated background level (pg/g) Level significantly different from zero? Lower 95% confidence limit Upper 95% confidence limit (pg/g) (pg/g) 4.5 13.3 2.5 26.4 21.5 23.5 13.7 6.5 8.2 29.3 169.5 38.7 29.6 (3.6) 11.8 ND (1.1) 22.2 19.8 22.4 11.2 4.3 ND (2.3) 24.9 159.4 26.5 25.7 yes yes no yes yes yes •j yes •J yes 2.7 10.4 -0.2 17.9 18.2 21.3 8.6 2.1 -3.6 20.5 149.3 14.3 21.9 ND (0.3) no -2.1 1.5 213.7 214.0 yes 208.9 219.1 3.2 806.1 ND (1.0) 799.7 no yes -1.7 779.4 3.8 820.0 ND (1.1) ND yes *j no yes •j yes yes .The measured background levels are the averages of the triplicate analyses of the unspiked matrix. The estimated background levels were derived from the linear regression analysis of data. ND = not detected. The value in parentheses reflects the estimated method detection limit. �3. Day-to-Day HRGC/MS Analysis Precision In addition to the analysis of the replicate spiked samples, four extracts were analyzed by HRGC/MS on two different dates. The results of the duplicate HRGC/MS analyses of these four samples for the 17 target compounds are presented in Tables 19 to 21. Concentration values from the second analysis date were included in the statistical analysis of data presented earlier in this section. 59 �Table 19. Day-to-Day Precision of Analysis of Specific Sample Extracts for Tetra- and Pentachloro PCDF and PCDD Analysis date 4/22/86 4/28/86 Spike level (pg/g) (pg/g) 0 0 RPD ( ) %b 4/22/86 4/28/86 0 0 RPD ( ) % ND (3.1) ND (4.1) 28 0 0 ND (3.3) ND (4.0) 1,2,3,7,8PeCDF 2,3,7,8TCDD 2,3,4,7,8PeCDF 1,2,3,7,8PeCDD (pg/g) (pg/g) (pg/g) (pg/g) 10 ND ( . 4 08) ND (1.12) 24 21 18 20 13 11 23 22 17 20 4 16 16 19 17 18 11 10 29 ND (0.78) ND (0.75) 10 11 4 10 ND (0.81) ND (0.75) 10 13 19 RPD (%) 4/22/86 4/28/86 ND ( . ) 30a ND (4.1) 29 RPD ( ) % 4/22/86 4/28/86 2,3,7,8TCDF 10 10 26 8 17 6 12 14 21 23 11 12 26 28 27 19 18 9 8 7 35 .ND = not detected. Value in parentheses is the estimated limit of detection. Relative percent difference. Calculated as the difference of the two values divided by the mean of the two values times 100%. 60 �Table 20. Day-to-Day Precision of Analysis of Specific Sample Extracts for Hexa- and Heptachloro PCDF and PCDD Analysi s date 4/22/86 4/28/86 Spike level 1,2,3, 4,7,8HxCDF 1,2,3, 6,7,8HxCDF 2,3,4, 6,7,8HxCDF 1,2,3, 7,8,9HxCDF 1,2,3, 6,7,8HxCDD 1,2,3, 7,8,9HxCDD 1,2,3,4, 6,7,8HpCDF 1,2,3,4, 7,8,9HpCDF 1,2,3,4, 6,7,8HpCDD (pg/g) 1,2,3, 4,7,8HxCDD (pg/g) (pg/g) (pg/g) (pg/g) (pg/g) (pg/g) (pg/g) (pg/g) (pg/g) (pg/g) 0 0 21 22 12 12 5 0 22 22 12 12 0 0 20 23 12 13 14 8 10 44 47 36 37 6 3 RPD ( ) %b 4/22/86 4/28/86 0 0 RPD ( ) % 4/22/86 4/28/86 0 0 RPD ( ) % 4.2 4.9 ND (0.33)a ND (0.51) ND (0.83) ND (0.36) 210 216 20 22 149 170 16 19 28 31 10 13 17 10 21 23 150 178 18 26 25 27 17 36 8 21 20 176 171 18 25 27 29 71 5 3 33 7 116 3 28 32 30 35 43 49 171 181 46 63 46 49 23 25 235 241 13 15 13 6 31 6 8 3 15 4.0 4.2 5 3.9 4.3 43 ND (0.32) ND (0.74) 79 ND (0.41) ND (0.86) 9 ' 79 ND ( . 9 10) ND (1.14) 4 ND (1.06) ND (0.28) 3 207 211 2 223 216 CTl 4/22/86 4/28/86 RPD ( ) % a 25 25 . Relative percent difference. Calculated as the difference of the two values divided by the mean of the two values times 100%. �Table 21. Day-to-Day Precision of Analysis of Specific Sample Extracts for OCDF and OCDD Analysis date OCDF concentration OCDD concentration (pg/g) (pg/g) (pg/g) 0 0 4.9 3.5 811 810 Spike level 4/22/86 4/28/86 RPD ( ) %a 33 2.2 2.3 0 0 4/22/86 4/28/86 4 RPD ( ) % 1.9 2.6 0 0 4/22/86 4/28/86 RPD ( ) % 2 788 784 1 43 44 834 848 2 50 50 RPD ( ) % . 819 836 31 4/22/86 4/28/86 a 0.1 2 . values divided by the mean of the two values times 100%. 62 �VI. QUALITY ASSURANCE/QUALITY CONTROL (QA/QC) As discussed in the experimental section of this report, the QA/QC activities included the analysis of a multipoint calibration curve, daily verification of relative response factors for each analyte, analysis of a method blank and reagent blanks along with the samples, and determining the absolute recoveries of each of the internal quantitation standards for every sample. Each of these QA/QC activities is discussed below. A. Initial Calibration At the outset of sample analysis activity, six calibration concentration standards containing each of the target PCDDs and PCDFs at varying levels and constant concentrations of the internal quantitation and recovery standards were analyzed in triplicate. The relative response factors (RRF) for each native compound and internal quantitation standard were determined for each standard analysis. An average RRF and relative percent standard deviation (RSD) were determined for each concentration level. The average RRF values from each of the six concentration calibration standards were then used to calculate a grand mean RRF value for each compound in the calibration solution. Table 22 presents a summary of the grand mean RRF values for each component in the standards. As noted from Table 22, the average RRF values for native PCDDs and PCDFs generally varied by less than ± 10% (RSD) with the exception of the pentachloro congeners. These results fall well within the criteria established in the draft quality assurance program plan which required the variability of RRF values for the tetrachloro homologs to be within ± 20% (RSD) while the RRF criterion on all other compounds was set at ± 30% (RDS). The variability of the RRF values for the internal quantitation standards, on the other hand, was noted to increase with the degree of chlorination. This is a result of the measurement of all internal quantitation standards versus the single internal recovery standard, 13C12-1,2,3,4-TCDD. A second internal recovery standard, 37Cl4-l,2,3,4,6,7,8-HpCDD, was evaluated. However, problems resulting from contribution of native HpCDD to the characteristic ions of this internal standard resulted in variabilities in the RRF value up to 50%. Hence, this internal standard was not used for any calculations. It is anticipated that an additional internal recovery standard, such as 13C12-l,2,3,4,7,8-HxCDD, will reduce the variability in the RRF values of the higher chlorinated internal quantitation standards. This compound will be incorporated into the method if available. The sensitivity of the Kratos MS-50TC to the tetra- through octachloro PCDDs and PCDFs was demonstrated through the triplicate analysis of the low level standard (CS-8, Table 2) that ranged in concentration from 1 pg/pL for the tetra- and pentachloro congeners up to 5 pg/pL for the octachloro congeners. These solution concentration values of 1 pg/pL and 5 pg/uL correspond to residue levels in tissue of 1 pg/g and 5 pg/g, respectively. Table 22 provides an indication of the observed signal-to-noise ratio for each of the native PCDD and PCDF congeners. These data demonstrate that the low level standard is well above the instrument detection limit, which is defined as the amount of a particular compound necessary to give a signal 2.5 times the background signal to noise for each of the characteristic ions while meeting the qualitative criteria for ion ratios. 63 �Table 22. Relative Response Factors (Grand Means) Determined from Multipoint Concentration Calibration Standards Compound Wa 1.00 0.80 0.98 1. .06 1. ,33 0.94 0.93 0.86 0.86 1. ,31 1. ,44 1. ,61 2. 33 1. .89 ,19 1. 1. ,38 1.04 RRF controj RSD (%) limits 2,3,7,8'-TCDD 1,2,3,7 ,8-PeCDF 2,3,4,7 ,8-PeCDF 1,2,3,7 ,8-PeCDD 1,2,3,4 ,7,8-HxCDF 1,2,3,6 ,7,8-HxCDF 2,3,4,6 ,7,8-HxCDF 1,2,3,7 ,8,9-HxCDF 1,2,3,4 ,7,8-HxCDD 1,2,3,6 ,7,8-HxCDD 1,2,3,7 ,8,9-HxCDD 1,2,3,4 ,6,7,8-HpCDF 1,2,3,4 ,7,8,9-HpCDF 1,2,3,4 ,6,7,8-HpCDD OCDF OCDD 13 C12-l,2,3,4-TCDD ia C12-2,3,7,8-TCDD ia Ci2-l,2,3,7,8-PeCDD ia Ci2-l,2,3,6,7,8-HxCDD ia C12-l,2,3,4,6,7,8- HpCDD *C14-1,2,3,4,6,7,8- 3 5.7 6.2 5.2 10.1 11.3 3.1 2.4 2.7 6.9 4.9 3.0 1.0 4.0 3.6 4.7 3.3 2.5 0.80-1.20 0.64-0.96 0.68-1.28 0.74-1.38 0.93-1.73 0.66-1.22 0.65-1.21 0.60-1.12 0.60-1.12 0.92-1.70 1.01-1.87 1.13-2.09 1.63-3.03 1.32-2.46 0.83-1.55 0.97-1.79 0.73-1.35 0.70 1.28 0.41 0.33 7.6 4.7 4.5 7.7 15.8 19.6 25.8 1.58-2.38 1.38-2.08 0.95-1.77 0.49-0.91 0.90-1.66 0.29-0.53 0.23-0.43 0.12 51.8 0.24 2,3,7,8'-TCDF 28.0 00 98 73 36 Signal-to-noise ratio for low level standard 12 6.5 11 8.9 5.7 32 30 29 17 13 14 14 35 26 13 31 21 Calibration range (pg/uL) 1-100 1-100 1-100 1-100 1-100 2.5-250 2.5-250 2.5-250 2.5-250 2.5-250 2.5-250 2.5-250 2.5-250 2.5-250 2.5-250 5-500 5-500 50 50 50 50 50 125 125 125 125 0.17-0.31 250 = grand mean RRF. RRF control limits designate the acceptable range of values based on the criteria for ± 20% of the KRF for 2,3,7,8-TCDD and 2,3,7,8-TCDF and ± 30% of the REP~for all other PCDD and PCDF compounds. Value for signal-to-noise ratio based on observed response for the major characteristic ion for each native PCDD or PCDF congener (Data File ,8501017X02). Internal recovery standard. 64 �B. Daily Verification of Response Factors Before proceeding with analysis of samples, the analyst was required to verify the existing response factor calibration through the analysis of a calibration standard (CS-7, Table 2). Criteria for proceeding with sample analysis required that the measured RRF value for 2,3,7,8-TCDD and 2,3,7,8-TCDF were within ± 20% (and all other congeners within ± 30%) of the mean RRF established from the calibration curve. This standard was also analyzed at the end of each working day to demonstrate that the calibration had been maintained. All RRF values were tabulated to generate RRF control charts for each specific PCDD and PCDF congener. Figures 25 through 34 are plots (control charts) of the RRF values established for the 17 individual target analytes. The RRF data are plotted versus time of analysis. These plots contain 28 individual data points, 18 of which were generated for triplicate analysis of 6 concentration calibration solutions from initial calibration and 10 analyses of solution CS-7 (Table 2) injected over the 5 days for which actual samples were analyzed. The upper and lower boundaries (dashed lines) represent a relative standard deviation of approximately ± 10% with the exception of the plot for 1,2,3,7,8-PeCDD, for which the boundaries are plotted as ± 20%. It should be noted that the actual control limits as specified in the project QAPP were set at ± 20% for 2,3,7,8-TCDD and 2,3,7,8-TCDF and ± 30% for all other target analytes. Boundaries of ± 10% have been used in Figures 25 through 34 as a means to provide the reader with a better perspective in the actual distribution of the measured calibration points. The values for the acceptable ranges of each PCDD and PCDF compound based on the initial calibration are presented in Table 22. The data presented for the RRF values in Figures 25 through 34 are well within these established control limits. The average RRF values and corresponding standard deviations reported in each of these plots are calculated from the total 28 standard analyses. C. Blanks As specified in the quality assurance program plan, a laboratory method blank was prepared along with the 14 human adipose lipid samples. The method blank was taken through all procedures as if it were an actual sample, although no lipid matrix was introduced. The analysis of the method blank resulted in the data reported for each of the target analytes reported in Table 23. As noted in Table 23, 1,2,3,4,6,7,8-HpCDD and OCDD were detected at concentrations equivalent to 4.0 and 30 pg/g (equivalent to a 10-g lipid sample), respectively. The contribution of these PCDDs were not subtracted from the observed responses for the spiked and unspiked samples. These background levels accounted for less than 2% of the 1,2,3,4,6,7,8-HpCDD and less than 4% of the OCDD measured in the unspiked lipid samples. In addition to these compounds, responses that correspond to the elution of two TCDD isomers (1,3,6,8- and 1,3,7,9-) and a PeCDD (isomer not determined) were detected in the method blank. 65 �e.33?) (AREA/REF. AREA>/(AKT. /UEF.ANT. > 1.296 1T.CEU.i ST.DEU.6.331 X MOT I ? * -x, 8.9W X DATE 4/23/8S 4/13/85 V 0.897) ST.DPJ.» e.ew 7. ST.DEV.= 6.832 8.999 X Yx |l -T > » X 8.788 DATE 4/13/85 4/23/86 3/ 3/8S Figure 25. Control charts showing response factors by date for 2,3,7,8-TCDF and 2,3,7,8-TCDD. Dotted lines represent approximately ± 10% of the mean. 66 �Kfflfttift 9.971) <f*Efl/REF.ftKEA>/<AMT./REF.AMT.> (Ml 1.299 ST.OEU.» 9.083 7. ST.DEU.« 8.739 X 1 i ? I i 1.199 IK fl* II L1 1 — "tf. >f " 1.899 -JfC S « it ¥^ $ i^ '^ S * 1 0.999 1 ! V ^ Sjf ^ - TX % X 0.S00 DATE 4/13/86 4/23/86 5/ 3/86 CMF!2,3<4,7,8-FEI(TACHLORO-FyRrtN .?EFiC13-l,2,3,7,<)-PENTScHLORO-FlJKftH (flFEfl/REF.IWEft)/ ftMT./REF.AHT.) (flUl 1.499 1.963) ST.DEU.= 3.113 7. ST.OEU. 18.601 X 1.309 X — *•* V , 1 4 i « i) * N| 1 X 0.309 >— tt x — " * X ii >L) " * M M 1.998 ? S4- X M . " , * x ^ ix v • ! ' ^ __ X 0.809 X il.Tm DATE 4/13/86 4/23/85 3/ 3/86 Figure 26. Control charts showing response factors by date for 1,2,3,7,8-PeCDF and 2,3,4,7,8-PeCDF. Dotted lines represent approximately ± 10% of the mean. 67 �«K£A/FEF.«Efl>/<«HT./REF.flHT.> 3.800 - 1.375) ST.OEU." 8.233 7. ST.DEU.18.895 2.899 !x X MTE 4/13/86 4/23/86 3/ 3/85 Figure 27. Control chart showing response factors by date for 1,2,3,7,8-PeCDD. Dotted lines represent approximately ± 20% of the mean. 68 �Cffii 1,2.3,4,7,8-HEXACHLQRO-FUIMN SEFlC13-l,2..3,4,7,8-HEX«CHLORO-FURftH (AfiEA/REF.AREAVCAMT./REF.AMT.) <flUi MM 8.323) ST.DEU.« 9.816 !! ST.DEI'.' 4.899 .ew V V •xMTE 1 5/ 3/86 4/23/86 4/13^6 !l,2,3,6,7,8-HEXft 1013-1,2,3,4,778-1 <AREfl/REF.I»Efl)/<ftMT./l?EF.flMT. B.923) nee ST.OEU.= 8.048 X ST.DEU." 4.348 -¥• i.ese X i O m K * r 0.303 * x x 3.383 DATE 4/13/86 4/23/86 1 S/ 3/86 Figure 28. Control charts showing response factors by date for 1,2,3,4,7,8-HxCDF and 1,2,3,6,7,8-HxCDF. Dotted lines represent approximately ± IQ% of the mean. 69 �12,3,4,6,7,8 8.862) i.ew ST.DEU.0.838 7. ST. DP.'. 4.429 — — ,Jfr Y 8.309 X 1 i V * ' X V »• *• i ' Th T 1 " xf | x 0.880 'Y ' " A ' X x x x 8.788 DflTE 4/13/86 4/23/86 <flRE«/REF.<*B»/<ftMT./REF.«MT.> <(Wl 1.200 -| V 3/86 8.883) ST.OEV.9.882 Z ST.DEV.' X 1.180 9 327 ' ¥ x ^ >f H 8.900 x ? 'I 1* |5* ft fi ! i 0.388 ii x I ' u ' ' ' x X T X j ^ * •" X A. 7m DATE 4/13/86 4/23/86 !/ 3/86 Figure 29. Control charts showing response factors by date for 2,3,4,6,7,8-HxCDF and 1,2,3,7,8,9-HxCDF. Dotted lines represent approximately ± 10% of the mean. 70 �1.394) 1.580 ST.DEU.0.077 7. ST.DEU.= 5.919 V V 1.4P8 ft • X •^N '* Vfl' - fJL, k- * i RX 1 f; r II J1 I f ^ •SL i 1.399 V \j/ j^. /x ! i ! | | i ^ A I i sL * * 1 -uw A i ! 1.299 1 i 1 !W DATE 4/13/85 4/23/86 AREA/REF.AREA)/«W./REF.IVIT.) (AUt 1.70B 3/ 3/8S 1.433) ST.DEU.0.036 Z ST.DEU.» 6.01S 1.609 ? i I i ~ ' V X «< III 1.509 iii f in m £ * ¥ «i il 1.400 1.308 ( * i x • •!* i • x ii i /^ 1 ' l i X X I X I . f x >< X 1 . 7fl« 4/23/86 3/ 3/86 Figure 30. Control charts showing response factors by date for 1,2,3,4,7,8-HxCDD and 1,2,3,6,7,8-HxCDD. Dotted lines represent approximately ± 10% of the mean. 71 �1.657) 2.300 > 2.200 ST.DEU.= 0.156 7. ST.DEU. 9.384 2.1?? '} ( 2.000 1.909 1.800 X I 1 X 1.700 V u 1.600 1.590 * » f 1 ;xx 1 ; !x x1 ; ! * xx ¥ X 1 X 1 400 4/23/86 S/ 3/86 Figure 31. Control chart showing response factors by date for 1,2,3,7,8,9-HxCDD. Dotted lines represent approximately ± 10% of the mean. 72 �2.228) <(«BVREF.flREfl)/<ftMT./REF.«MT.) 2.600 •X * ST.DEU.0. 132 7. ST.OEU.» 8.163 2.508 X K X 2.488 in I * x "ft* & 2.309 * aI * i |i • i ' X t ' 2.288 A x i ' x X 2.100 X X X"1 2.088 X 1.960 X DrtTE 4/ 3/86 4/23/86 CMFil, 2,3,4, 7,8,3-HEPTflCHLORO-FURfiH *EFiC13-l, 2,3,4,6 .7,8-HEPTAa*J$J-DIOXIM i'flREfl/REF.AREfl)/< «1T./R£F.«1T.) (flUl 2. 180 ^ 1.T83) X ' ' 7 J _ * 2.088 5/ 3/86 ST.DEU.' 3. 169 7. ST.OEV.9.420 /jK i< T r> i 1.300 m ^11 ^ ; - i * 1.700 '' X x x — 1.608 —L r T X 1.598 | X X X" X ^ t.4B« DATE 4/13/8S 4/23/8S 5/ 3/96 Figure 32. Control charts showing response factors by date for 1,2,3,4,6,7,8-HpCDF and 1,2,3,4,7,8,9-HpCDF. Dotted lines represent approximately ± 10% of the mean. 73 �(AREA/REF.AREA)/(ANT./REF.MIT.) (AUl 1.486 1.175) ST.OEU.8.371 7. ST.DEV." 6.043 i.aee 1.2M *'! < !x M SAC X I I x, I X X . X 1.100 e.we DATE 4/13/86 4/23/86 5/ 3/36 Figure 33. Control chart showing response factors by date for 1,2,3,4,6,7,8-HpCDD. Dotted lines represent approximately ± 10% of the mean. 74 �lOCTflCHLORO-FUf WH :CI3-OCT«HLOR(M)tOXIH WT./FEF.flMT.) (flUl 1.238) 1.5W ST.DEU." 8.117 7. ST.OEU. 8.978 X 1 X *j (1 x U HI in •M 1.40B I' * ji >i M «l tl M M M >^f 1.390 ri 1.2W X X /x -iv -1- ^^ ^ < ; < i ina xX 3/8S 4/23/85 ;H IOXIN <AKEA/REF.ftfiEfl)/<AMT.-/R£F.flt1T.) 1.280 1.033) ST.DEU.8.943 7. ST.DEU.- 4.181 ^ X 1.180 V X L-9-ff 1 x ii ^ i —i 0.900 DATE X X 4/13/86 4/23/85 1 3/ 3/86 Figure 34. Control charts showing response factors by date for OCDF and OCDD. Dotted lines represent approximately ± 10% of the mean. 75 �Table 23. Summary of Results from the Analysis of a Laboratory Method Blank Concentration Compound3 (pg/g) 2,3,7,8-TCDD 2,3,7,8-TCDF 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,7,8-PeCDD 1,2,3,4,7,8-HxCDF ND ND ND ND ND ND ND ND ND ND 1,2,3,6,7,8-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF 1,2,3,4,6,7,8-HpCDD OCDF OCDD (0.50) (2.2) (0.5) (0.5) (1.2) (0.5) (0.5) (0.5) (0.5) (1.0) ND (0.9) ND (0.8) ND (0.5) ND (0.5) 4.0 ND (0.5) 30 a At least three other PCDD compounds were detected but not quantitated in the laboratory method blank. These included 1,3,6,8- and 1,3,7,9-TCDD and an unidentified PeCDD . isomer. Concentration based on assumption of 10.0 g equivalent lipid sample. The background concentration of 1,2,3,4,6,7,8-HpCDD and OCDD were not subtracted from the measured concentration for the spiked and unspiked lipid matrix. 76 �Further analysis of individual reagents used for preparation of the samples identified the activated acidic alumina as the source of the artifacts. Acidic alumina that had been cleaned by Soxhlet extraction but not activated at 190°C was analyzed, and the artifacts were not detected. This indicates that the artifacts are generated during activation of acidic alumina at elevated temperatures (190°C). Similar background problems from the same PCDD congeners have recently been reported by the Center for Disease Control.17'18 An experiment was designed to evaluate a procedure for cleaning the activated acidic alumina immediately prior to the fractionation of the sample extract. The acidic alumina (6.0 g) was packed in hexane. The packed column was eluted with 40 mL of methylene chloride/hexane (1:1) solution followed by 80 to 100 ml of hexane. The sample extract was added to the column and was eluted with 20 ml of hexane followed by 30 ml of 20% methylene chloride in hexane which was reserved for PCDD and PCDF analysis. The carbon-14 radiolabeled 2,3,7,8-TCDD, 1,2,3,4,7,8-HxCDD, and OCDD were used to evaluate recovery of PCDDs from the cleaned alumina. Recoveries of the radiolabeled PCDDs from the activated acidic alumina precleaned by the procedure described above are detailed in Table 24. These data demonstrate that the selected PCDDs are quantitatively (greater than 90%) recovered from the precleaned acidic alumina. This procedure for cleanup of activated acidic alumina was not initiated for the analysis of the lipid samples described in this report. However, it has been integrated into the analytical protocol (Appendix A) for routine application with sample preparation activities. D. Absolute Recoveries of the Internal Quantisation Standards The absolute recoveries of the carbon-13 labeled internal quantitation standards 13were determined by comparing responses with the internal recovery standard, C12-1,2,3,4-TCDD, which was added during final concentration prior to HRGC/MS analysis. A summary of the average and range of recoveries of the 8 internal quantisation standards from the 14 human adipose lipid samples is provided in Table 25. These data indicate that recoveries ranged from an average of 52.1% for 13C12-2,3,7,8-TCDD up to 88.9% for 13C12-OCDD. The average recoveries for the lower chlorinated internal standards were lower than the preliminary method studies with carbon-14 radiolabeled standards had indicated. This resulted in a closer evaluation of the final concentration step prior to mass spectrometry. The first extracts for the human adipose lipid extracts were concentrated with a nitrogen evaporation system equipped with a water bath at approximately 55°C. Final blowdown of the samples required addition of the internal recovery standard in 10 uL of tridecane as a keeper solution. However, it was noted at the elevated temperature final volumes from nitrogen evaporation were generally on the order of 2 to 5 uL. This required addition of another 10 uL of tridecane prior to HRGC/MS analysis. In an effort to assess the effect of reducing the final volume of tridecane at elevated temperatures on absolute recoveries of the internal quantisation standards, an experiment using the radiolabeled TCDD, HxCDD, and OCDD standards was conducted. 77 �Table 24. Recovery of Radio!abeled PCDDs from Precleaned Activated Alumina Spike level (pg) Compound Recovery (%) 14 100 300 300 92 96 97 14 1,000 3,000 3,000 103 101 100 2,500 7,500 7,500 102 99 97 C-2,3,7,8-TCDD C-l,2,3,4,7,8-HxCDD 14 C-OCDD 78 �Table 25. Absolute Recoveries of the Internal Quantisation Standards from the Human Adipose Lipid Matrix Average recovery ( ) % Standard deviation Relative standard deviation (%) 13 64.0 7.9 12.3 48-78 13 52.1 5.0 9.6 43-62 13 76.1 8.9 11.7 62-90 13 57.1 3.5 6.1 51-64 13 59.4 5.0 8.4 52-70 13 63.6 5.3 8.4 57-77 13 76.3 10.3 13.6 61-99 88.9 11.4 12.9 67-104 Internal quantitation standard C12-2,3,7,8-TCDF C12-2,3,7,8-TCDD C12-l,2,3,7,8-PeCDF C12-l,2,3,7,8-PeCDD C12-1,2,3,4,7,8HxCDF C12-1,2,3,6,7,8HxCDD C12-1,2,3,4,6,7,8HpCDD 13 C12-OCDD Values based on 14 analyses of human adipose lipid samples. 79 Range of recovery (%) �Four solutions of the same spike level were prepared with each radiolabeled compound in 1 mL of toluene. Two of the spiked solutions were heated at 55-60°C and the solvent was reduced under a gentle stream of prepurified nitrogen. The toluene solution was concentrated to 100 (jL, 500 pL of 1% toluene in methylene chloride was added, and the solution was concentrated to 200 uL. At this time 10 uL of the keeper tridecane was added and the solution was allowed to concentrate further. The remaining two solutions for each radiolabeled compound were taken through a similar solvent exchange and concentration procedure except the solution was allowed to concentrate at room temperature. One of the most obvious results was the observation that solutions held at elevated temperatures could be reduced to dryness even when tridecane had been added as a keeper. On the other hand, solutions for which tridecane had been added but remained at room temperature could only be concentrated to a 10-uL final volume. The recoveries of the radiolabeled standards from each of the solutions in this study are presented in Table 26. The results from this study indicate that the final concentration condition may have a pronounced effect on the absolute recoveries of the PCDDs and PCDFs, especially for the lower chlorinated congeners such as 2,3,7,8-TCDD. These conclusions are supported by an independent study in comparison of concentration techniques for 2,3,7,8-TCDD.19 However, it should be noted that the approach to target analyte quantisation based on the internal standard method (isotope dilution for 8 of the 17 target analytes) is not affected by absolute recoveries as low as 50%. The procedure for final concentration in the analytical protocol (Appendix A) for the analysis of the NHATS samples for the EPA/VA study has been modified to specify room temperature conditions. 80 �and OCDD as a Function of Final Concentration Conditions Spike level Compound (pg) Concentration conditions Observed final volume Observed recovery ( ) % 14 300 300 300 300 55-60°C 55-60°C 20°C 20°C 1-2 |jL dryness 10 uL 10 uL 78 54 98 93 14 3,000 3,000 3,000 3,000 55-60°C 55-60°C 20°C 20°C 1-2 uL 5 uL 10 M L 10 uL 94 102 105 107 7,500 7,500 7,500 7,500 55-60°C 55-60°C 20°C 20°C 1-2 uL 2-3 M L 10 uL 10 |jL 94 94 100 97 C-2,3,7,8-TCDD C-l,2,3,4,7,8-HxCDD 14 C-OCDD Each solution was concentrated under a gentle stream of flowing nitrogen. 81 �VII. GLOSSARY OF TERMS Accuracy - A measurement of the bias of a system, which for this study, is based on the agreement of the 2,3,7,8 substituted PCDD and PCDF to an accepted reference standard. Batch, sample - A sample batch consists of up to 10 human adipose tissue samples, one method blank, 2 internal quality control (QC) samples (spiked and unspiked), and an external performance audit sample (blind spike). Blank, laboratory method - This blank is prepared in the laboratory through performing all analytical procedures except addition of a sample aliquot to the extraction vessel. A minimum of one laboratory method blank will be analyzed with each batch of samples. Calibration standards (concentration calibration solutions) - Solutions containing known amounts of the native analytes (unlabeled 2,3,7,8-substituted PCDDs and PCDFs), the internal quantisation standards (carbon-13 labeled PCDDs and PCDFs), and the recovery standard, 13C12-1,2,3,4-TCDD. These calibration solutions are used to determine instrument response of the analytes relative to the internal quantisation standards and of the internal quantitation standards relative to the internal recovery standard. Lipid - The organic solvent extractable constituents of adipose tissue consisting of fatty oils, proteins, and carbohydrates. The concentrations of PCDDs and PCDFs are reported on the lipid content bases. Instrumental mass calibration - An internal instrumental systems check and tuning standard, perfluorokerosene (PFK), is introduced automatically by the instrument. The mass ion 380.976 is monitored by the analyst as an instrumental systems check and is also used to tune the instrument. Internal quantisation standards - Carbon-13 labeled PCDDs and PCDFs, which are added to every sample and are present at the same concentration in every method blank and quality control sample. These are added to the adipose tissue prior to extraction and are used to measure the concentration of each analyte. The concentration of each internal quantitation standard is measured in every sample, and percent recovery is determined using the internal recovery standard. Internal recovery standard - 13C12-1,2,3,4-TCDD and 13C12-l,2,3,7,8,9-HxCDD which is added to every sample extract just before the final concentration step and HRGC/MS-SIM analysis. Limit of detection (LOD) - A value, derived from the noise to signal response, which is equal to 2.5 times the average instrumental noise level is the limit of detection. Limit of quantitation (LOQ) - A value, derived from the noise to signal response, which is equal to 10 times the average instrumental noise level is the limit of quantitation. 82 �Mass resolution check - Standard method used to demonstrate static resolution of 10,000 minimum (10% valley definition). Not detected (ND) - A nonresponse or a response which is less than the limit of detection is reported as not detected. Precision - The results from analysis of replicate samples (spiked and unspiked) provide the measure of method precision. The precision of the method is reported as standard deviation or relative standard deviation. Performance check mixture, HRGC column - A mixture containing known amounts of selected TCDD standards; it is used to demonstrate continued acceptable performance of the capillary column, to separate ( 25 % valley on a 50-m CP ^ Sil 88 or 60-m SP-2330 HRGC column and 30 to 60% for a 60 m DB-5 HRGC column) 2,3,7,8-TCDD isomer from all other 21 TCDD isomers, and to define the TCDD retention time window. PCDD - Polychlorinated dibenzo-p_-dioxins. PCDF - Polychlorinated dibenzofurans. Relative response factor - Response of the mass spectrometer to a known amount of an analyte relative to a known amount of an internal standard (quantitation or recovery). Trace (TR) - A response which is greater than the limit of detection but less than the limit of quantitation is reported as a trace value. An estimated method detection limit is provided for trace value. 83 �VIII. REFERENCES 1. Stanley JS. 1984. Methods of analysis of polychlorinated dibenzo-£dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs) in biological matrices--!iterature review and recommendations. EPA-560/584-00. 2. Stanley JS, Going JE, Redford DP, Kutz KW, Young AL. 1985. A survey of analytical methods for measurement of polychlorinated dibenzo-£-dioxins (PCDD) and polychlorinated dibenzofurans (PCDF) in human adipose tissues. In: Chlorinated dioxins and dibenzofurans in the total environment II. Keith LH, Rappe C, Choudhary G, eds. Butterworth Publishers, pp. 181-195. 3. Stanley JS. 1984 (March 28). Proposed analytical method for analysis of PCDDs/PCDFs in human adipose tissue: special report. Washington, DC: Office of Pesticides and Toxic Substances, U.S. Environmental Protection Agency. Contract 68-02-3938, Work Assignment 8. 4. Stanley JS. 1986 (April 23). Broad scan analysis of human adipose tissue: polychlorinated dibenzo-£-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs). Draft final report. Washington, DC: Office of Pesticides and Toxic Substances, U.S. Environmental Protection Agency. Contract 68-023938, Work Assignment 8. 5. Albro et al. 1985. Methods for the quantitative determination of multiple, specific polychlorinated dibenzo-£-dioxin and dibenzofuran isomers in human adipose tissue in the parts-per-trillion range. An interlaboratory study. Anal Chem 57: 2717-2725. 6. Patterson DG et al. 1986. High-resolution gas chromatographic/highresolution mass spectrometric analysis of human adipose tissue for 2,3,7,8-tetrachlorodibenzo-£-dioxin. Anal Chem 58: 705-713. 7. Schecter A, Tiernan TO, Taylor ML, VanNess GF, Barrett JH, Wage! DJ, Gitlitz G, Bogdasarian M. 1985. Biological markers after exposure to polychlorinated dibenzo-p_-dioxins, dibenzofurans, biphenyls, and biphenylenes. Part I: Findings using fat biopsies to estimate exposure. In: Chlorinated dioxins and dibenzofurans in the total environment II. Keith LH, Rappe C, Choudhary G, eds. Butterworth Publishers, pp. 215-246. 8. Schecter A, Ryan JJ. 1985. Dioxin and furan levels in human adipose tissue from exposed and control populations. 189th National ACS Meeting Symposium on Chlorinated Dioxins and Dibenzofurans in the Total Environment III, Miami, Florida. Preprint Division of Environmental Chemistry, ACS 25:160-163, Paper No. 56. 9. Ryan JJ, Williams DT, Lau BPY, Sakuma T. 1985. Analysis of human fat tissue for 2,3,7,8-tetrachlorodibenzo-£-dioxin and chlorinated dibenzofuran residues. In: Chlorinated dioxins and dibenzofurans in the total environment II. Keith LH, Rappe C, Choudhary G, eds. Butterworth Publishers, pp. 205-214. 84 �10. Ryan JJ, Schecter A, Lizotte R, Sun W-F, Miller L. 1985. Tissue distribution of dioxins and furans in humans from the general population. Chemosphere 14: 929-932. 11. Nygren M, Hansson M, Rappe C, Domellof L, Hardel1 L. 1985. Analysis of polychlorinated dibenzo-p_-dioxins and dibenzofurans in adipose tissue from soft-tissue sarcoma patients and controls. 189th National ACS Meeting Symposium on Chlorinated Dioxins and Dibenzofurans in the Total Environment III, Miami, Florida, 1985. Preprint Division of Environmental Chemistry, ACS 25:160-163, Paper No. 55. 12. Smith LM, Stalling DL, Johnson JJ. 1984. Determination of part per trillion levels of polychlorinated dibenzofurans and dioxins in environmental samples. Anal Chem 58: 1830-1842. 13. Rappe C, Nygren M, Linstrom G, Hanson H. 1985. Dioxins and dibenzofurans in human tissues and milk of European origin. 5th International Symposium on Chlorinated Dioxins and Related Compounds, Bayreuth, FRG, September 16-19, 1985. 14. Ryan JJ. 1985. Variation of dioxins and furans in humans with age and organ by country. 5th International Symposium on Chlorinated Dioxins and Related Compounds, Bayreuth, FRG, September 16-19, 1985. 15. Graham M, Hileman FD, Wendling J, Wilson JD. 1985. Chlorocarbons in adipose tissue samples. 5th International Symposium on Chlorinated Dioxins and Related Compounds, Bayreuth, FRG, September 16-19, 1985. 16. Patterson DG, Holler JS, Smith SJ, Liddle JA, Sampson EJ, Needham LL. 1985. Human tissue data in certain U.S. populations. 5th International Symposium on Chlorinated Dioxins and Related Compounds, Bayreuth, FRG, September 16-19, 1985. 17. Patterson DG, Holler JS, Groce DF, Alexander LR, Lapeza CR, O'Conner RC, Liddle JA. 1986. Control of interferences in the analysis of human adipose tissue to 2,3,7,8-tetrachlorodibenzo-£-dioxin (TCDD). Environ Toxicol Chem 5: 355-360. 18. Holler JS, Patterson DG, Alexander LR, Groce OF, O'Connor RC, Lapeza CR. 1985. Control of artifacts and contamination in the development of a dioxin analytical program. 33rd Annual Conference on Mass Spectrometry and Allied Topics, San Diego, CA, May 26-31, 1985. 19. O'Keefe PN, Meyer C, Dillon K. 1982. Comparison of concentration techniques for 2,3,7,8-tetrachlorodibenzo-p_-dioxin. Anal Chem 54: 26232625. 85 �APPENDIX A ANALYTICAL PROTOCOL FOR DETERMINATION OF PCDDs AND PCDFs IN HUMAN ADIPOSE TISSUE A-l �TABLE OF CONTENTS Section Description Page 1 Scope and Application A-3 2 Summary of Method A-3 3 Definitions A-6 4 Interferences A-7 5 Safety A-7 6 Apparatus and Equipment A-8 7 Reagents and Standard Solutions A-ll 8 High Resolution Gas Chromatography/Mass Spectrometry Performance Criteria 9 A-13 Quality Control Procedures A-31 10 Sample Preservation and Handling A-33 11 Sample Extraction A-33 12 Cleanup Procedures A-35 13 Analytical Procedures A-38 14 Date Reduction A-43 15 Reporting and Documentation A-50 A-2 �ANALYTICAL PROTOCOL FOR DETERMINATION OF PCDDs AND PCDFs IN HUMAN ADIPOSE TISSUE 1. SCOPE AND APPLICATION 1.1 1.2 The minimum measurable concentration is estimated to range from 1 pg/g (1 part per trillion) for 2,3,7,8-TCDD and 2,3,7,8-TCDF up to 5 pg/g for OCDD and OCDF. However, these detection limits depend on the kinds and concentrations of interfering compounds in the sample matrix and the absolute method recovery. 1.3 2. This method provides procedures for the detection and quantitative measurement of polychlorinated dibenzo-p_-dioxins (PCDD) and polychlorinated dibenzofurans (PCDF) at concentrations ranging from 1 to 100 pg/g for the tetrachloro congeners up to 5 to 500 pg/g for the octachloro congeners in 10-g aliquots of human adipose tissue. The method will be used to determine PCDDs and PCDFs, particularly congeners with chlorine substitution in the 2,3,7,8 positions. Table 1 lists the specific PCDDs and PCDFs and target method detection limits. SUMMARY OF METHOD Figure 1 presents a schematic of the analytical procedures for determination of PCDDs and PCDFs in human adipose tissue. The analytical method requires extraction and isolation of lipid materials from human adipose samples. This is accomplished using sample sizes ranging up to 10 g. The tissue is spiked with known amounts of the carbon-13 labeled PCDDs and PCDFs (e.g., 500 pg of 13C12-TCDD/F to 2,500 pg of 13C12-OCDD/F) as internal quantitation standards. Extraction and homogenization are accomplished using methylene chloride and a Tekmar Tissuemizer®. The extract is filtered through anhydrous sodium sulfate to remove water. The extraction procedure is repeated (three to five times) until the tissue sample has been thoroughly homogenized. The final extract is adjusted to a known volume (100 ml) and the extractable lipid is determined using a minimum of 1% of the final volume. The methylene chloride in the remaining extract is concentrated until only an oily residue remains. The residue is diluted with hexane ( 200 mL), and ~ 100 g of sulfuric acid modified silica gel (40% w/w) is added to the solution with stirring. The mixture is stirred for approximately 2 h, and the supernatant is decanted and filtered through anhydrous sodium sulfate. The adsorbent is washed with at least two additional aliquots of hexane. The combined hexane extracts are eluted through a column consisting of a layer of sulfuric acid modified silica gel, and a layer of unmodified silica gel. The eluate is concentrated to approximately 1 ml and added to a column of acidic alumina. The PCDDs and PCDFs are eluted from the alumina using 20% methylene chloride/hexane. This eluate is concentrated to approximately 0.5 mL and is added to a 500-mg Carbopak C/Celite column. The PCDDs and PCDFs are eluted from the column using 20 mL of toluene. A-3 �Table 1. Target PCDD and PCDF Congeners and Target Method Detection Limits Compound 2,3,7,8-TCDD 2,3,7,8-TCDF 1,2,3,7,8-PeCDD 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,4,6,7,8-HpCDD 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF OCDD OCDF CAS no.3 1746-01-6 51207-31-9 40321-76-4 57117-41-6 57117-31-4 39227-28-6 57653-85-7 19408-74-3 70648-29-9 57117-44-9 72918-21-9 60851-34-5 35822-46-9 67562-39-4 55673-89-7 3268-87-9 39001-02-0 Chemical Abstract Services number. pg/g = parts per trillion. 3 A-4 Target method detection limit (pg/g) 1.0 1.0 1.0 1.0 1.0 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 5.0 5.0 �Add Internal Quantitation Standards Initial Sample Preparation Isolation of Extractable U'pid Materials (13c-PCDDs/PCDFs) Homogenization in Methylene Chloride I Lipid Determination Solvent Exchange Bulk U'pid Removal Acid Modified Silica Gel Slurry Technique Provides Cleanup of Oxidizable Compounds with Rapid Sample Turnaround, Improved Cleanup Efficiency and Recovery i Removal of Chemical Interferences Acidic Silica/Silica Acidic Alumina Carbopak C/Celite Provides Seperation of PCBs and Other Potential Interferences from PCDDs and PCDFs Selective Adsorption and Isolation of PCDDs/PCDFs Add Internal Recovery Standards HR'GC/MS-SIM Analysis 1 i LRMS Identification/Quantitation of Tetra-Octa PCDDs/PCDFs HRMS Confirmation of 2,3,7,8-TCDD Figure 1. Schematic of the sample preparation and instrumental analysis procedures for determination of PCDDs and PCDFs in human adipose tissue. A-5 �The toluene is concentrated to less than 1 ml and transferred to conical vials. Tridecane (10 uL) containing 500 pg of an internal recovery standard is added as a keeper, and the extract is concentrated to final volume. The HRGC/MS analysis is completed in the selected ion monitoring mode (SIM). Analysis of the tetra- through octachloro PCDD and PCDF congeners is achieved using low resolution mass spectrometry. Separation of the tetra- through octachloro PCDD and PCDF congeners is achieved using a 60-m DB-5 column. Verification of the 2,3,7,8-TCDD is achieved using either a 50-m CP Sil 88 column or 60-m SP-2330 column and HRGC/MS-SIM analysis in the high resolution mode (R = 10,000). 3. DEFINITIONS 3.1 Concentration calibration solutions — Solutions containing known amounts of the native analytes (unlabeled 2,3,7,8-substituted PCDDs and PCDFs), the internal quantisation standards (Carbon-13 labeled PCDDs and PCDFs), and the recovery standard, 13C121,2,3,4-TCDD. These calibration solutions are used to determine instrument response of the analytes relative to the internal quantitation standards and of the internal quantitation standards relative to the internal recovery standard. 3.2 Internal quantitation standards -- Carbon-13 labeled PCDDs and PCDFs, which are added to every sample and are present at the same concentration in every method blank and quality control sample. These are added to the adipose tissue and are used to measure the concentration of each analyte. The concentration of each internal quantitation standard is measured in every sample, and percent recovery is determined using the internal recovery standard. 3.3 Internal recovery standard — 13C12-1,2,3,4-TCDD and 13C121,2,3,7,8,9-HxCDD which is added to every sample extract just before the final concentration step and HRGC/MS-SIM analysis. 3.4 Laboratory method blank — This blank is prepared in the tory through performing all analytical procedures except of a sample aliquot to the extraction vessel. A minimum laboratory method blank will be analyzed with each batch ples. 3.5 HRGC column performance check mixture -- A mixture containing known amounts of selected TCDD standards; it is used to demonstrate continued acceptable performance of the capillary column, to separate (g 25% valley on a 50-m CP Sil 88 or 60-m SP-2330 HRGC column and 30 to 60% for a 60-m DB-5 HRGC column) 2,3,7,8TCDD isomer from all other 21 TCDD isomers, and to define the TCDD retention time window. 3.6 Relative response factor — Response of the mass spectrometer to a known amount of an analyte relative to a known amount of an internal standard (quantitation or recovery). A-6 laboraaddition of one of sam- �3.7 3.8 4. Mass resolution check — Standard method used to demonstrate static resolution of 10,000 minimum (10% valley definition). Sample batch -- A sample batch consists of up to 10 human adipose tissue samples, one method blank, 2 internal quality control (QC) samples (spiked and unspiked), and an external performance audit sample (blind spike). INTERFERENCES Chemicals which elute from the HRGC column with ± 10 scans of the internal and/or recovery standards and which produce within the retention time window ions at any of the masses used to detect or quantify PCDDs, PCDFs, or the internal quantitation and recovery standards are potential interferences. Most frequently encountered potential interferences are other sample components that are extracted along with the PCDDs and PCDFs, e.g., PCBs, chlorinated methoxybiphenyls, chlorinated hydroxydiphenyl ethers, chlorinated benzylphenyl ethers, chlorinated naphthalenes, DDE, DDT, etc. The actual incidence of interference by these chemicals depends also upon relative concentrations, mass spectrometric resolution, and chromatographic conditions. Because very low levels (pg/g) of PCDDs and PCDFs are anticipated, the elimination of interferences is essential. High purity reagents and solvents must be used and all equipment must be scrupulously cleaned. Laboratory method blanks must be analyzed to demonstrate absence of contamination that would interfere with measurement of the PCDDs and PCDFs. Column chromatographic procedures are used to remove coextracted sample components; these procedures must be performed carefully to minimize loss of PCDDs and PCDFs during attempts to increase their concentration relative to other sample components. 5. SAFETY 5.1 The toxicity or carcinogenicity of each reagent used in this method has not been precisely defined; however, each chemical compound should be treated as a potential health hazard. The 2,3,7,8-TCDD is a known teratogen, mutagen, and carcinogen. Ingestion of microgram quantities can result in toxic effects. The other 2,3,7,8-substituted PCDDs and PCDFs may exhibit teratogenic, mutagenic, and carcinogenic effects. From this viewpoint, exposure to these chemicals must be reduced to the lowest possible level by whatever means available. Only experienced personnel will be allowed to work with these chemicals. 5.2 All laboratory personnel will be required to wear laboratory coats or coveralls, gloves, and safety glasses. The neat standards, stock, and working solutions will be handled only in a Class A fume hood or glove box. When manipulating stock standards or working solutions, the analyst is advised to place the solution vials in a secure holder (sample block or glass beaker) to prevent accidental spills. A-7 �5.3 5.4 If handling of these compounds results in skin contact, immediately remove all contaminated clothing and wash the affected skin areas with soap and water for at least 15 min. 5.5 6. If these standards are spilled, absorb as much as possible with absorbent paper and place in a container clearly labeled as PCDD or PCDF waste. Solvent-wash all contaminated surfaces with toluene and absorbent paper followed by washing with a strong soap and water solution. Dispose of all contaminated materials in sealed steel containers labeled as contaminated with PCDD and/or PCDF residue and indicate the approximate level of contamination. As a final precaution, prepare a wipe sample of the exposed surface area and include the wipe as part of the sample analysis batch. This will be used to confirm that the work area is free of contamination. Disposal of laboratory wastes -- All laboratory wastes (solvents and absorbents) will be disposed of as hazardous wastes. The laboratory personnel should take care to dispose of the sodium sulfate, silica gel, and alumina in separate containers. Excess solvents should be disposed of in gallon polyethylene jugs containing a layer of activated charcoal. Excess solvent that is known to be contaminated with PCDDs or PCDFs should be kept at a minimum by evaporating the solvent with a stream of air. APPARATUS AND EQUIPMENT 6.1 High Resolution Gas Chromatograph/Mass Spectrometer/Data System (HRGC/HRMS/DS) 6.1.1 The GC must be equipped for temperature programming, and all required accessories must be available, such as syringes, gases, and a capillary column. The GC injection port must be designed for capillary columns. The use of splitless injection techniques is recommended. When using this method, a 1-pL injection volume is used. The injection volumes for all extracts, blanks, calibration solutions, and the performance check sample must be consistent. 6.1.2 High Resolution Gas Chromatograph-Mass Spectrometer Interface The HRGC/MS interface is directly coupled to the mass spectrometer ion source. All components of the interface should be glass or glass-lined stainless steel. The interface components should be compatible with 300°C temperatures. The HRGC/MS interface must be appropriately designed so that the separation of the PCDDs and PCDFs which is achieved in the gas chromatographic column is not appreciably degraded. Cold spots and/or active surfaces (adsorption sites) in the HRGC/MS A-8 �interface can cause peak tailing and peak broadening. It is recommended that the HRGC column be fitted directly into the MS ion source. Graphite ferrules should be avoided in the HRGC injection port since they may absorb PCDDs or PCDFs. Vespel or equivalent ferrules are recommended. 6.1.3 Mass Spectrometer The mass spectrometer must be capable of maintaining a minimum resolution of 10,000 (10% valley) for high resolution confirmation analysis. The mass spectrometer must be operated in a selected ion monitoring (SIM) mode with total cycle time (including voltage reset time) of 1 s or less. 6.1.4 Data System A dedicated hardware or data system is required to control the rapid multiple ion monitoring process and to acquire the data. Quantification data (peak areas or peak heights) and SIM traces (displays of intensities of each m/z (characteristic ion) being monitored as a function of time) must be acquired during the analyses. Quantifications may be reported based upon computergenerated peak areas or upon measured peak heights. 6.2 HRGC Columns For isomer-specific determinations of 2,3,7,8-TCDD, the following fused silica capillary columns are recommended: a 50-tn CP-Sil 88 column and a 60-m SP-2330 (SP-2331) column. However, any capillary column which separates 2,3,7,8-TCDD from all other TCDDs may be used for such analyses, provided that the minimum acceptance criteria in Section 8 are met. 6.3 Miscellaneous Equipment 6.3.1 Nitrogen evaporation apparatus with variable flow rate. 6.3.2 Balance capable of accurately weighing to ± 0.01 g. 6.3.3 Balance capable of accurately weighint to ± 0.0001 g. 6.3.4 Water bath — equipped with concentric ring cover and capable of being temperature-controlled. 6.3.5 Stainless steel spatulas or spoons. 6.3.6 Magnetic stirrers and stir bars. 6.3.7 High speed tissue homogenizer -- Tekmar Tissuemizer® equipped with an EN-8 probe or equivalent. 6.3.8 Vacuum dessicator. A-9 �6.4 Glassware 6.4.1 Erlenmeyer flask — 500 mL. 6.4.2 Kuderna-Danish apparatus — 500-mL evaporating flask, 15-mL graduated concentrator tubes with ground-glass stoppers, and three-ball macro Snyder column (Kontes K-570001-0500, K-503000-0121, and K-569001-0219 or equivalent). 6.4.3 Minivials -- 1-mL borosilicate glass with conical-shaped reservoir and screw caps lined with Teflon®-faced silicone disks. 6.4.4 Powder funnels -- glass. 6.4.5 Chromatographic columns for the silica and alumina chromatography -- 1 cm ID x 10 cm long and 1 cm ID x 30 cm long with 250-mL reservoir and equipped with TFE stopcocks. 6.4.6 Chromatographic column for the Carbopak cleanup -disposable 5-mL graduated glass pipets, 6 to 7 mm ID. 6.4.7 Glass rods. 6.4.8 Carborundum boiling chips -- Extracted for 6 hr in a Soxhlet apparatus with benzene and air dried. 6.4.9 Glass wool, silanized (Supelco) -- Extract with methylene chloride and hexane and air dry before use. 6.4.10 Glassware cleaning procedure -- All glassware used for these analyses will be cleaned via the following procedure. Wash the glassware in soap and water, rinse with copious amounts of tap water, distilled water, and distilled-in-glass acetone, in that order. Immediately prior to use, the glassware should be rinsed with distilled-in-glass quality solvents: methylene chloride, toluene, and hexane. The glassware should be allowed to dry fully. As an added precuation, all glassware will be marked with a unique code that should be noted in the extraction and cleanup procedures for each sample. This glassware tracking will allow background results from specific glassware to be documented. After use, each piece of glassware should be rinsed with the last solvent used in it, followed by a rinse with toluene, then acetone, before transferring it to the glassware washing facility. A-10 �7. REAGENTS AND STANDARD SOLUTIONS 7.1 Column Chromatography Reagents 7.1.1 Alumina, acidic (Biorad, AG-4) — Extract the alumina in a Soxhlet apparatus with methylene chloride for 18 h (minimum of two cycles per hour). Air dry and activate it by heating in a foil-covered glass container for 24 h at 190°C. 7.1.2 Silica gel -- High purity grade, type 60, 70-230 mesh; extract the silica gel in a Soxhlet apparatus with methylene chloride for 10 h (minimum of 2 cycles per hour). Air dry and activate it by heating in a foilcovered glass container for 24 h at 130°C. 7.1.3 Silica gel impregnated with 40% (by weight) sulfuric acid -- Add two parts (by weight) concentrated sulfuric acid to three parts (by weight) silica gel (extracted and activated) (e.g., 40 g of H2S04 plus 60 g of silica gel) in a glass screw-cap bottle. Tumble for 5 to 6 h, shaking occasionally until free of lumps. 7.1.4 Sulfuric acid, concentrated — ACS grade, specific gravity 1.84. 7.1.5 Graphitized carbon black (Carbopack C, Supelco), surface of approximately 12 m2/g, 80/100 mesh -- Mix thoroughly 3.6 g of Carbopack C and 16.4 g of Celite 545® in a 40-mL vial. Activate at 130°C for 6 h. Store in a desiccator. 7.1.6 Celite 545® (Fischer Scientific), reagent grade, or equivalent. 7.2 Desiccating agents — Sodium sulfate; granular, anhydrous. Before use extract with methylene chloride for 16 h (minimum of two cycles per hour), air dry and then muffle for ^ 4 h in a shallow tray at 400°C. Let it cool in a desiccator and store in oven at 130°C. 7.3 Solvents -- High purity, distilled in glass: methylene chloride, toluene, benzene, cyclohexane, methanol, acetone, hexane; reagent grade: tridecane. High purity solvents are dispensed from Teflon® squirt bottles. 7.4 Concentration Calibration Solutions (Table 2) Eight tridecane solutions containing native calibration standards, 13 C12-labeled internal quantisation standards, and two internal recovery standards are required. The complete compound list is A-11 �CO o Ln Ln CM l—I r H r H i —I T —( C \ J C \ J C N J C \ I C \ J C \ I C \ I C s J C \ J C \ J L T ) L n 1 CM CJ> Q . to o CM CM to o ooooiDLnLnmo to c o o to -p 3 c o O OO c. o rO s- u o rH rH rH rH CM O Ln o c o i—( T—1 rH rH CM to o rH rH rH rH CM CM O O O O O O O O O O O O O O O O O ooooLninLnLno 1—lrHrHrHrHCMCMCMCMCMCMCMCMCMCMLnLn rH i—I rH rH CM oo O) CD (J c o o o o C_J o ooooLnunLnLno LnmLnLncMCMCMCMLn T-HrHrHrHrHrHrHrHrHrHCMCM C_3 c. CM CM to o +J ro s- Ln CM i—Ir-lr-I T —I t —4 C \ J C S J C S I C X J C \ J C \ J C \ J C v J C \ I C \ I L n L n CO rO C_3 CM 1—I i—1 i— 1 i—I T —1 i— 1 r-1 i—l i —I r H C M C N J CM 00000000000000000 to ooooooooooooooooo O Ln ID CM CM CM CM C3 Ll_ (~*\ CD -Q rO a u_ u_ o oo QC3OLL.L1_L1_L1_C_3OO a a a o a o o. a. a. a. 3 O Q. r_3c_3c_5n:^x:3;3i^^cooocr> o u _ a > a ) 0 ) i i i i i i i > > > caoQ-Q-Q-oooocricoooo-ioor^f^-co . C-3 C_3 I I I o * . * . * * * . * . * ! * . * ! * litation 0 C3 O t_3 (— 3 in -a c c S*» C2 ' ' c CMCMrHi—ICMi—IrHrHrHi—IrHCMrHi—IrHOO 1—1 ^\ 0 C_3 0 Q o Q a. a. u_ 0 0 o X X 0 3C ^ CO ai o X 00 a i oo f~^ CT* •s r-~ in <J2 to ^> <d- *d- 00 CO r--. 0) r-. c S- v . * . ^ » . ^ . « . « ^ « t # . A ^ i ^ ^ u_ o 1— C_3 O) u_ 0 !. - r-~ CO CO CO ro CO CO 1— CO CO CM CM CM CM CM CM 1— CO 1 a> CO CO a i- o ro T3 CM CM i—t rH rH i-H rH rH 1 1 1 1 0J 0J 0 C1 CM CN IN <N <N c_> c_> CO O CO CO C_3 O CO CO c i. rH C ro to O 000 :0 "0 CO •o (—H *^ CM CM rH i—1 I 1 <N CM c_> o CO CO T-H TH �given in Table 2. The native 2,3,7,8-TCDD is supplied as a certified standard solution from the U.S. EPA QA Reference Materials Branch. All other native compounds were supplied in crystalline form by Cambridge Isotope Laboratories (Woburn, MA). 13C12~ Labeled internal quantitation standards were supplied in solution in n-nonane by Cambridge Isotope Laboratories. Portions of the native standards were accurately weighed to the nearest 0.001 mg with a Cahn 27 electrobalance and dissolved in toluene. 7.5 Column Performance Check Mixture The column performance check mixture consists of several TCDD isomers which will be used to document the separation of 2,3,7,8TCDD from all other isomers. This solution will contain TCDDs (A) eluting closely to 2,3,7,8-TCDD, and the first- (F) and lastel uting (L) TCDDs. Analyte Approximate amount per ampule 10 10 10 10 10 10 10 10 Unlabeled 2,3,7,8-TCDD 13 C12-2,3,7,8-TCDD 1,2,3,4-TCDD (A) 1,4,7,8-TCDD (A) 1,2,3,7-TCDD (A) 1,2,3,8-TCDD (A) 1,3,6,8-TCDD (F) 1,2,8,9-TCDD (L) 7.6 ng ng ng ng ng ng ng ng Spiking Solutions Three solutions are prepared using the same stock as in Section 7.4. A native standard solution and a 13C12 internal quantitation standard solution are prepared in isooctane (Tables 3 and 4). A recovery standard solution is prepared in tridecane (Table 4). Samples are spiked with 100 uL of internal quantitation standard solution and final sample extracts are spiked with 10 (jL of internal recovery standard solution. 8. HIGH RESOLUTION GAS CHROMATOGRAPHY/MASS SPECTROMETRY PERFORMANCE CRITERIA Samples and standards are analyzed by using a Carlo Erba MFC500 gas chromatography (GC) coupled to a Kratos MS50TC double-focusing mass spectrometer (MS) to be operated in the electron impact mode. The HRGC/MS interface is simply a direct connection of the fused silica HRGC column to the ion source of the MS via a heated interface oven. Data acquisition and processing are controlled by a Finnigan-MAT Incos 2300 data system. A-13 �Table 3. Native Spiking Solution3 Concentration (pg/jjL) Compound 2,3,7,8-TCDD 2,3,7,8-TCDF 1,2,3,7,8-PeCDD 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 5 5 5 5 5 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 12.5 12.5 12.5 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,4,6,7,8-HpCDD 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF OCDD OCDF 12.5 12.5 12.5 12.5 12.5 12.5 12.5 25 25 a Prepared in isooctane. A-14 �Table 4. Internal Standard Spiking Solutions Concentration (pg/uL) Compound Internal Quantitation Standards3 13 C12-2,3,7,8-TCDD 5 13 5 13 5 13 5 C12-2,3,7,8-TCDF C12-l,2,3,7,8-PeCDD C12-l,2,3,7,8-PeCDF 13 12.5 13 12.5 13 12.5 13 12.5 C12-l,2,3,6,7,8-HxCDD C12-l,2,3,4,7,8-HxCDF C12-l,2,3,4,6,7,8-HpCDD C12-l,2,3,4,6,7,8-HpCDF 13 C12-OCDD 25 Internal Recovery Standard 13 C12-1,2,3,4-TCDD 13 C12-l,2,3,7,8,9-HxCDD ^Prepared in isooctane. Prepared in tridecane. A-15 50 125 �8.1 HRGC/MS Analysis of PCDD/PCDF Single run selected ion monitoring (SIM) analysis of the tetrachloro through octachloro-dioxins and furans is carried out with the instrumental conditions and parameters outlined in Table 5. For each HRGC/MS run, five distinct groups of ions, which correspond to each chlorine level, are sequentially monitored. These ion descriptors are shown in Table 6. The masses of the two most abundant ions in the molecular ion cluster of each dioxin and furan and isotopically labeled standard are monitored. In addition, the masses corresponding to the molecular ions of the hexachloro through decachlorodiphenyl ethers (PCDEs) are monitored to aid in the confirmation of positive furan results. Interference from the presence of PCDE is noted by coincident response to the characteristic ions for PCDFs. A lock mass, m/z 381 from PFK (perfluorokerosene), is used to observe and correct any magnet/instrument drift during the analysis. 8.1.1 Tuning and Mass Calibration The mass spectrometer is tuned on a daily basis to yield optimum sensitivity and peak shape using an ion peak (m/z 381) from PFK. The resolution is visually monitored and maintained at S 3,000 (10% valley definition) to provide adequate noise rejection while maintaining good ion transmission. Mass calibration of the mass spectrometer for the HRGC/MS analysis of PCDD/PCDF is carried out on a daily basis. The magnetic field is adjusted to pass m/z 300 at full accelerating voltage. PFK is admitted to the MS and an accelerating voltage scan from 8,000 to 4,000 V is acquired by the data system. This corresponds to an effective mass range of 301 to 593 amu. Upon completion of a successful calibration step, the five ion descriptors shown in Table 6 are updated to reflect the new mass calibration. 8.1.2 Ion Descriptor Switching The ion descriptors shown in Table 6 are sequentially monitored during a PCDD/PCDF analysis to cover the retention windows of each chlorination level. The retention windows and hence the descriptor switch points are determined initially and whenever a new HRGC column is installed by injection of a mixture of PCDD and PCDF congeners. Daily adjustment of the descriptor switch times are performed when careful monitoring of the standard retention times shows this to be necessary. The descriptors are designed to ensure acquisition of all isomers of each homolog. A-16 �Table 5. HRGC/LRMS Operating Conditions for PCDD/PCDF Analysis Mass spectrometer 8,000 V 500 uA 70 eV -1,800 V Accelerating voltage: Trap current: Electron energy: Electron multiplier voltage: Source temperature: Resolution: Overall SIM cycle time: 280°C ^ 3,000 (10% valley definition) 1s Gas chromatograph Column coating: Film thickness: Column dimensions: He linear velocity: He head pressure: DB-5 0.25 urn 60 m x 0.25 mm ID ~ 25 cm/sec 1.75 kg/cm2 (25 psi) Injection type: Split flow: Purge flow: Injector temperature: Interface temperature: Injection size: Initial temperature: Initial time: Temperature program: Splitless, 45 s 30 mL/min 6 mL/min 270°C 300°C 1-2 uL 200°C 2 min 200°C to 330°C at 5°C/min A-17 �Table 6. Ions Monitored for HRGC/MS of PCDD/PCDF Descriptor Al Mass ID 303.902 305.899 315.942 317.939 319.897 321.894 331.937 333.934 373.840 380.976 TCDF 13 C12-TCDF TCDD 13 C12-TCDD HxCDPE PFK (lock mass) A2 TCDF TCDD PeCDF 13 C12-PeCDF PeCDD 13 C12-PeCDD PFK (lock mass) HpCDPE A3 Nominal dwell time (sec) HxCDF PFK (lock mass) 13 C12-HxCDF HxCDD 13 C12-HxCDD OCDPE A-18 0.090 0.090 0.090 0.090 0.090 0.090 0.090 0.090 0.090 0.090 303.902 305.899 319.897 321.894 337.863 339.860 349.903 351.900 353.858 355.855 365.898 367.895 380.976 407.801 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.035 0.035 373.821 375.818 380.976 385.861 387.858 389.816 391.813 401.856 403.853 443.759 0.080 0.080 0.080 0.080 0.080 0.080 0.080 0.080 0.080 0.080 �Table 6 (continued) Descriptor A4 Mass ID 380.976 389.816 391.813 407.782 409.779 419.822 421.819 423.777 425.774 435.817 437.814 429.768 431.765 477.720 HpCDF C12-HpCDF HpCDD 13 37 C12-HpCDD Cl4-HpCDD NCDPE A5 PFK (lock mass) OCDF 13 C12-OCDF OCDD 13 C12-OCDD DCDPE A-19 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 380.976 441.743 443.740 453.783 455.780 457.738 459.735 469.779 471.776 511.681 PFK (lock mass) HxCDD 13 Nominal dwell time (sec) 0.06 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.06 �8.1.3 HRGC Column Performance (60-m DB-5) The HRGC column performance must be demonstrated at the start of each 12-h analysis period. 8.1.3.1 Inject 1 uL of the column performance check solution (Section 7.5) and acquire selected ion monitoring (SIM) data for m/z 320, 322, 332, and 334. 8.1.3.2 The chromatographic peak separation between 2,3,7,8-TCDD and the peaks representing any other TCDD isomers should be resolved with a valley of 30-60%, where Valley % = ( / ) 1 0 xy(0) x = measured height of the valley between the chromatographic peak corresponding to 2,3,7,8-TCDD and the peak of the nearest TCDD isomer; and y = the peak height of 2,3,7,8-TCDD. Figure 2 is an example of the separation of a TCDD isomer mixture and the calculation of isomer resolution. It is the responsibility of the laboratory to verify the conditions suitable for the appropriate resolution of 2,3,7,8-TCDD from all other TCDD isomers. The column performance check solution also contains the TCDD isomers eluting first and last under the analytical conditions specified in this protocol, thus defining the retention time window for total TCDD determination. Any individual selected ion current profile or the reconstructed total ion current (m/z 320 + m/z 322) consititutes an acceptable form of data presentation. 8.1.4 Initial Calibration for PCDD/PCDF Analysis Initial calibration is required before any samples are analyzed for PCDD/PCDF. Initial calibration is also required if any routine calibration does not meet the required criteria listed in Section 8.1.7. 8.1.4.1 Tune and calibrate the instrument with PFK as outlined in Section 8.1.1. A-20 �TCDD Isomer Mixture 78.7 2, 3, 7, 8.-TCDD 2359290. 1, 2, 3, 4-/1, 2, 3, 7-/L 2, 3, 8- TCDD 319.856 <*= 0.500 320 100.0 2998270. •*- x 100% = 29% Y 322 y\ A 321.855 ± 0.500 X = 10 mm 59.6 1785850. Q2-2, 3, 7, 8,-TCDD f\> 331.851 ± 0.500 332 75.3 2256890. 334 333.850 ± 0.500 28:00 28:24 Figure 2. 28:48 29:12 29:36 30:00 30:24 30:48 Example of the separation of 2,3,7,8-TCDD from other TCDD isomers on a 60 m DB-5 column. Time �8.1.4.2 8.1.4.3 Using the HRGC and MS conditions in Table 5 and the SIM monitoring descriptors in Table 6, analyze a 1-uL aliquot of each of the six concentration calibration solutions in triplicate. 8.1.4.4 Compute the relative response factors (RRFs) for each analyte in the concentration calibration solution using the criteria for positive identification of PCDD/PCDF's given in Section 14.1 and the computational methods in Section 14.2. 8.1.4.5 Compute the means and their respective relative standard deviations (% RSD) for the RRFs from each triplicate analysis for each analyte in the standard. 8.1.4.6 8.1.5 Six of the eight concentration calibration solutions listed in Table 2 will be analyzed for the initial calibration phase. These must include solutions CS4 through CSS (Table 2). The analyst may select any of the remaining solutions for demonstrating calibration at the upper concentration range. Calculate the grand means (RRF) and their respective RSDs using the six mean RRFs for each analyte. Criteria for Acceptable Initial Calibration 8.1.5.1 The % RSD for the response factors for each triplicate analysis of a single concentration calibration standard for each analyte must be less than ± 30% except for the TCDD and TCDF, which must be less than ± 20%. 8.1.5.2 The variation of the mean RRFs for the six concentration calibrated standards (Section 8.1.5.1) must be less than 30% except for the TCDD and TCDF which must be less than 20%. 8.1.5.3 The SIM traces for all ions used for quantitation must present a signal-to-noise (S/N) ratio of ^ 2.5. This includes analytes and isotopically labeled standards. A-22 �8.1.5.4 Isotopic ratios must be within ± 20% of the theoretical values (see Table 7). NOTE: If the criteria for acceptable calibration listed above have been met, the RRF can be considered independent of the analyte quantity for the calibration concentration range. The grand mean RRF from the initial calibration for unlabeled PCDD/ PCDFs and for the isotopically labeled standards will be used for all calculations until routine calibration criteria (Section 8.1.7) are no longer met. At such time, new mean RRFs will be calculated from a new set of six triplicate determinations. 8.1.6 Routine Calibrations Routine calibrations must be performed at the beginning of every day before actual sample analyses are performed and as the last injection of every day. 8.1.6.1 8.1.6.2 8.1.7 Inject 1 uL of the concentration calibration solution CS 7 (see Table 2) as the initial calibration check on each analysis day. It is recommended that the analyst select a concentration calibration solution that brackets the sample concentrations observed on a single analysis date as the last injection of each analysis date. Compute the RRFs for each analyte in the concentration calibration solution using the criteria for positive identification of PCDD/Fs given in Section 14.1 and the computational methods in Section 14.2. Criteria for Acceptable Routine Calibration 8.1.7.1 The measured RRF for all analytes must be within ± 30% of the grand mean values established by triplicate analysis of the calibration concentration solutions, except for TCDD and TCDF, which must be within ± 20% of the mean values established in the initial calibration step. 8.1.7.2 Isotopic ratios must be within ± 20% of the theoretical value for each analyte and isotopically labeled standard (see Table 7). A-23 �Table 7. Ion Ratios for HRGC/LRMS Analysis of PCDD/PCDF Compound TCDF 13 C12-TCDF TCDD 13 C12-TCDD PeCDF 13 Cl2-PeCDF PeCDD 13 Cl2-PeCDD HxCDF 13 Cl2-HxCDF HxCDD 13 Cl2-HxCDD HpCDF 13 Cl2-HpCDF HpCDD 13 Cl2-HpCDD OCDF 13 C12-OCDF OCDD 13 C12-OCDD Ions monitored Theoretical ratio 304/306 316/318 320/322 332/334 338/340 350/352 354/356 366/368 374/376 386/388 390/392 402/404 408/410 420/422 424/426 436/438 442/444 454/456 458/460 470/472 0.76 0.76 0.76 0.76 0.61 0.61 0.61 0.61 1.22 1.22 1.22 1. 22 1.02 1.02 1.02 1.02 0.87 0.87 0.87 0.87 A-24 Acceptabl e range 0.61 0.61 0.61 0.61 0.49 0.49 0.49 0.49 0.98 0.98 - 0.91 0.91 0.91 0.91 0.73 0.73 0.73 0.73 1.46 1.46 0.98 - 1.46 0.98 - 1.46 0.82 - 1.22 0.82 - 1.22 0.82 - 1.22 0.82 0.70 0.70 0.70 0.70 - 1.22 1.04 1.04 1.04 1.04 �8.1.7.3 8.2 If any of the above criteria is not met, a second attempt may be made before repeating the entire initialization process. HRGC/HRMS Analysis (Isomer Specific TCDD Analysis) Isomer specific analysis for 2,3,7,8-TCDD is carried out with the instrumental conditions and parameters shown in Table 8. In addition to monitoring the masses of the most abundant molecular ions of TCDD, an ion corresponding to the loss of COC1 from the molecular ion is monitored for verification purposes. Mass spectrometer resolution is maintained at or above 10,000 (10% valley definition) in order to increase the specificity of the analysis. 8.2.1 Tuning and Mass Calibration 8.2.1.1 The mass spectrometer must be operated in the electron (impact) ionization mode. Static resolving power of at least 10,000 (10% valley) must be demonstrated before any analysis of a set of samples is performed. Static resolution checks must be performed at the beginning and at the end of each 12-h period of operation. However, it is recommended that a visual check (i.e., not documented) of the static resolution be made before and after each analysis. 8.2.1.2 The MS shall be tuned daily using PFK to yield a resolution of at least 10,000 (10% valley) and optimal response at m/z 254.986. This step is followed by calibration of an accelerating voltage scan of PFK beginning at m/z 254 (typical calibration range is 255 to 493 amu). Other voltage scans from the same data file are used to establish and document both the resolution at m/z 316.983 and the mass measurement accuracy at m/z 330.979. 8.2.1.3 Following calibration, the SIM experiment descriptor is updated to reflect the new calibration. Six masses (see Table 8) are monitored by scanning ^ m/10,000 amu (atomic mass units) over each mass. The total cycle time is kept to 1 s. The m/z 280.983 ion from PFK is used as a lock mass because it is the most abundant PFK ion within the range of m/z 255 to 334 and therefore permits the use of low partial pressures of PFK, which minimizes PFK interferences at the analytical masses. A-25 �Table 8. HRGC/HRMS Operating Conditions Mass spectrometer 8,000 V 500 MA 70 eV 2,000 V 280° C 10,000 (10% valley definition) Accelerating voltage: Trap current: Electron energy: Electron multiplier voltage: Source temperature: Resolution: SIM Parameters Identity TCDD-COC1 TCDD TCDD 13 Ci2-TCDD 13 C12-TCDD PFK (lock mass) Mass Nominal dwell times (s) 0.15 0.15 0.15 0.15 0.15 0.10 258.930 319.897 321.894 331.937 333.934 280.983 Overall SIM cycle time = 1 s Gas chromatograph Column coating: Film Thickness: Column dimensions: CP-Sil 88 0.2 Mm 50 m x 0.22 mm ID Helium linear velocity: Helium head pressure: ~ 25 cm/s 2 1.75 kg/cm (25 psi) Injection type: Split flow: Purge flow: Injector temperature: Interface temperature: Injection size: Initial temperature: Initial time: Temperature program: Splitless, 45 s 30 mL/min 6 mL/min 270°C 240°C 2 ML 200°C 1 min 200°C to 240°C at 4°C/min A-26 �8.2.2 Mass Measurement and Resolution Check Using a PFK molecular leak, tune the instrument to meet the minimum required resolving power of 10,000 (10% valley) at m/z 254.986 (or any other mass reasonably close to m/z 259). Calibrate the voltage sweep at least across the mass range m/z 259 to m/z 334 and verify that m/z 330.979 from PFK (or any other mass close to m/z 334) is measured within ± 5 ppm (i.e., 1.7 mmu, if m/z 331 is chosen) using m/z 254.986 as a reference. Documentation of the mass resolution must then be accomplished by recording the peak profile of the PFK reference peak m/z 318.979 (or any other reference peak at a mass close to m/z 320/322). The format of the peak profile representation must allow manual determination of the resolution; i.e., the horizontal axis must be a calibrated mass scale (amu or ppm per division). The results of the peak width measurement (performed at 5% of the maximum which corresponds to the 10% valley definition) must appear on the hard copy and cannot exceed 100 ppm (or 31.9 mmu if m/z 319 is the chosen reference ion). 8.2.3 HRGC Column Performance (50-m CP Sil 88/60-m SP-2330) Prior to any HRGC/HRMS analysis of calibration solutions or samples for 2,3,7,8-TCDD, the resolution of the HRGC columns must be documented to be within allowable limits in order to provide conditions adequate for unambiguous isomer-specific analysis of 2,3,7,8-TCDD. This column performance check must be demonstrated at the start of each 12-h analysis period. 8.2.3.1 Inject 2 uL of the column performance check solution and acquire selected ion monitoring (SIM) data for m/z 258.930, 319.897, 321.894, 331.937, and 333.934 within a total cycle time of ^ 1 s (Table 8). 8.2.3.2 The chromatographic peak separation between 2,3,7,8-TCDD and the peaks representing any other TCDD isomers must be resolved with a valley of ^ 25%, where Valley % = (x/y)(100) x = measured height of the valley between the chromatographic peak corresponding to 2,3,7,8-TCDD and the peak of the nearest TCDD isomer; and y = the peak height of 2,3,7,8-TCDD. A-27 �8.2.3.3 8.2.3.4 8.2.4 If the above resolution requirement is not met, corrective action must be taken and acceptable resolution documented prior to any further analyses. Corrective action may include removal of the first meter of the HRGC column, replacement or clearing of the injector port, or complete replacement of the GC column. The column performance check solution also contains the TCDD isomers eluting first and last under the analytical conditions specified in this protocol, thus defining the retention time window for total TCDD determination. The peaks representing 2,3,7,8-TCDD and the first and the last eluting TCDD isomer should be labeled and identified as such on the chromatograms (F and L, respectively). Any individual selected ion current profile or the reconstructed total ion current (m/z 259 + m/z 320 + m/z 322) constitutes an acceptable form of data presentation. Initial Calibration for HRGC/HRMS 2,3,7,8-TCDD Analysis Initial calibration is required before any samples are analyzed for 2,3,7,8-TCDD. Initial calibration is also required if any routine calibration does not meet the required criteria listed in Section 8.2.6. 8.2.4.1 At least six of the concentration calibration solutions listed in Table 2 must be utilized for the initial calibration. These must include solutions CS4 through CSS. The analyst may select any of the remaining solutions for demonstrating calibration at the upper concentration range. 8.2.4.2 Tune and calibrate the instrument with PFK as described in Section 8.2.1. 8.2.4.3 Inject 1 uL of the column performance check solution (Section 8.2.3) and acquire SIM mass spectra data for m/z 258.930, 319.897, 321.894, 331.937, and 333.934 using a total cycle time of ^ 1 s (see Table 8). The laboratory must not perform any further analysis until it has been demonstrated and documented that the criterion listed in Section 8.2.3.2 has been met. A-28 �8.2.4.4 8.2.4.5 Calculate the RRFs for unlabeled 2,3,7,8TCDD relative to 13C12-2,3,7,8-TCDD and the RRF for 13C12-2,3,7,8-TCDD relative to 13 C12-1,2,3,4-TCDD using the criteria for positive identification of TCDD by HRGC/ HRMS given in Section 14.1 and the computational methods in Section 14.2. 8.2.4.6 Calculate the six means (RRFs) and their respective relative standard deviations (% RSD) for the response factors from each of the triplicate analyses for both unlabeled and 13C12-2,3,7,8-TCDD. 8.2.4.7 8.2.5 Using the same GC and MS conditions (Table 8) that produced acceptable results with the column performance check solution, analyze a 1-uL aliquot of each of the six concentration calibration solutions in triplicate. Calculate the grand mean RRFs and their respective relative standard deviations ( RSD) using the six mean RRFs. % Criteria for Acceptable Initial Calibration The criteria listed below for acceptable calibration must be met before analysis of any sample is performed. 8.2.5.1 The percent relative standard deviation (RSD) for the response factors from each of the triplicate analyses of a single concentration calibration standard for both unlabeled and 13C12-2,3,7,8-TCDD must be less than 20%. 8.2.5.2 The variation of the mean RRFs from the six concentration calibration standards unlabeled and 13C12-2,3,7,8-TCDD must be less than 20% RSD. 8.2.5.3 SIM traces for 2,3,7,8-TCDD must present a signal-to-noise ratio of s 2.5 for m/z 258.930, m/z 319.897, and m/z 321.894. 8.2.5.4 SIM traces for 13C12-2,3,7,8-TCDD must present a signal-to-noise ratio ^2.5 for m/z 331.937 and m/z 333.934. 8.2.5.5 Isotopic ratios for 320/322 and 332/334 must be within the allowed range (0.61 to 0.91). A-29 �NOTE: If the criteria for acceptable calibration listed above have been met, the RRF can be considered independent of the analyte quantity for the calibration concentration range. The grand mean RRF from the initial calibration for unlabeled 2,3,7,8-TCDD and for 13C12-2,3,7,8-TCDD will be used for all calculations until routine calibration criteria (Section 8.2.6) are no longer met. At such time, new mean RRFs will be calculated from a new set of six triplicate determinations. 8.2.6 Routine Calibrations Routine calibrations must be performed at the beginning of a 12-h period after successful mass resolution and HRGC column performance check runs and before analysis of actual samples. The response factor calibration must also be verified at the end of each analysis date. 8.2.6.1 8.2.7 Inject 1 uL of the concentration calibration solution (CS7, Table 2) which contains 2.5 pg/uL of unlabeled 2,3,7,8-TCDD, 50.0 pg/uL of 13C12-2,3,7,8-TCDD, and 50 pg/uL of 13C12-1,2,3,4-TCDD. Using the same HRGC/ MS/DS conditions as used in Table 8, determine and document acceptable calibration as provided below. Criteria for Acceptable Routine Calibration The following criteria must be met before further analysis is performed. If these criteria are not met, corrective action must be taken and the instrument must be recalibrated. 8.2.7.1 The measured RRF for unlabeled 2,3,7,8-TCDD must be within 20% of the mean values established in the initial calibration by triplicate analyses of concentration calibration solutions. 8.2.7.2 The measured RRF for 13C12-2,3,7,8-TCDD must be within 20% of the mean value established by triplicate analysis of the concentration calibration solutions during the initial calibration. A-30 �8.2.7.3 Isotopic ratios must be within the allowed range (0.61 to 0.90). 8.2.7.4 If one of the above criteria is not satisfied, a second attempt can be made before repeating the entire initialization process. NOTE: An initial calibration must be carried out whenever the routine calibration solution is replaced by a new one from a different lot. 9. QUALITY CONTROL PROCEDURES 9.1 Summary of QC Analyses 9.1.1 Initial and routine calibration and instrument performance checks. 9.1.2 Analysis of a batch of samples with accompanying QC analyses: Sample batch -- 10 NHATS adipose tissue samples plus additional QC analyses including 1 method blank, a control tissue and a spiked tissue sample. "Blind" QC (external QC) samples may be submitted by an external source (quality assurance group or independent laboratory) and included among the batch of samples. Blind samples include spiked samples, unidentified duplicates, and performance evaluation samples. 9.2 Performance Evaluation Solutions -- Included among the samples in every third batch will be a solution provided by the quality control coordinator containing known amounts of unlabeled 2,3,7,8TCDD and/or other PCDD/PCDF isomers. The accuracy of measurements for performance evaluation samples should be in the range of 70-130%. 9.3 Column Performance Check Solutions 9.3.1 At the beginning of each 12-h period during which samples are to be analyzed, an aliquot of the HRGC column performance check solution shall be analyzed to demonstrate adequate HRGC resolution for selected TCDD isomers. A-31 �9.4 Method Blanks 9.4.1 A minimum of one method blank is generated with each batch of samples. A method blank is generated by performing all steps detailed in the analytical procedure using all reagents, standards, equipment, apparatus, glassware, and solvents that would be used for a sample analysis, but omit addition of the adipose tissue. 9.4.1.1 The method blank must contain the same amounts of Carbon-13 labeled internal quantitation standards that are added to samples before bulk lipid cleanup. 9.4.1.2 An acceptable method blank exhibits no positive response for any of the characteristic ions monitored. 9.4.1.2.1 If the above criterion is not met, solvents, reagents, spiking solutions, apparatus, and glassware are checked to locate and eliminate the source of contamination before any samples are extracted and analyzed. 9.4.1.2.2 If new batches of reagents or solvents contain interfering contaminants, purify or discard them. 9.5 Control Samples -- Control samples are prepared from a bulk sample(s) of human adipose tissue or similar matrix (e.g., porcine fat). This material is prepared by blending the tissue with methylene chloride, drying the extract by eluting through anhydrous sodium sulfate, and removing the methylene chloride using rotoevaporation at elevated temperatures (80°C). The evaporation process should be extended to ensure all traces of the extraction solvent have been removed. The resulting oily matrix (lipid) is subdivided into 10-g aliquots which are analyzed with each sample batch. The results of the individual analysis will be used to give a measure of precision from batch to batch over an entire program. Sufficient tissue should be extracted to provide a homogeneous lipid matrix that can be used over the total analysis program. Enough lipid matrix is necessary to prepare the spiked samples describe in Section 9.6. 9.6 Spiked Samples — Spiked lipid samples are prepared using a portion of the homogenized lipid described in Section 9.5. Sufficient spiked lipid matrix is prepared to provide a minimum of one spiked sample per sample batch. It is recommended that a minimum A-32 �of three spiked levels of the matrix are prepared ranging from 10 to 50 times the estimated limit of detection for each compound. Each analysis of spiked sample must be accompanied by analysis of a control sample in order to make the necessary corrections for background contribution before determining the accuracy of the method (Equation 9-1). A fw\ mn<v Cone, spiked sample-cone, control sample Eq 9 n1, ,. K Accuracy ( ) = 100% x % Spike level ' " 9.7 9.8 10. Duplicate Sample Analysis -- When possible a duplicate analysis of specific samples is included in the sample batch as an additional measure of method precision. It is suggested that the total tissue sample is extracted to isolate lipids material and then subdivided for duplicate analysis. Precision is calculated as relative percent difference (RPD) where the differences in the duplicate measurements (for each analyte) is divided by the average of the two measurements and multiplied by 100%. External Samples — Samples submitted as blinds to the analyst may consist of either performance solutions of PCDD and PCDF congeners or spiked sample matrices. These performance solutions or samples should be submitted by a source external to the analytical program (QA unit of analysis laboratory or independent laboratory). Performance audit solutions are intended to evaluate instrument calibration and quantisation procedures. Spiked blind samples must be accompanied by the corresponding unspiked samples to correct concentrations for background concentration. The blind spiked samples are intended to evaluate the total analytical procedure. The analyst must keep in mind that it is necessary to compare differences in standard sources for each type of external sample. SAMPLE PRESERVATION AND HANDLING All adipose tissue samples must be maintained at less than -20°C from time of collection. The analyst should instruct the collaborator collecting the sample(s) to avoid the use of chlorinated materials. Samples are handled using stainless steel forceps, spatulas, or scissors. Aliquots of samples removed from sample bottles not used for analysis are disposed rather than returned to the sample vial. All sample bottles (glass) are cleaned as specified in Section 6.4.10. Teflon®-lined caps should be used. As with any biological sample, the analyst should avoid any undue exposure. 11. SAMPLE EXTRACTION 11.1 Extraction of Adipose Tissue 11.1.1 Accurately weigh to the nearest 0.01 g a 10-g portion of a frozen adipose tissue sample into a culture tube (2.2 x 15 cm). Note: Sample size may be smaller, depending on availability. A-33 �11.1.2 Addition of internal quantitation standards Allow the adipose tissue specimen to reach room temperature and then add the carbon-13 internal quantitation spiking solution (Section 7.6) such that it delivers 500 to 2,500 pg of each of the surrogates specified in Table 4 in a 100-uL volume. 11.1.3 11.1.4 Allow the mixture to separate and decant the methylene chloride extract from the residual solid material using a disposable pipette. The methylene chloride is eluted through a filter funnel containing a plug of clean glass wool and 5 to 10 g of anhydrous sodium sulfate. The dried extract is collected in a 100-mL volumetric flask. 11.1.5 A second 10-mL aliquot of methylene chloride is added to the sample and homogenized for 1 min. The methylene chloride is decanted, dried, and transferred to the 100-mL volumetric flask as specified in Section 11.1.3 11.1.6 The culture tube is rinsed with at least two additional aliquots (10 mL each) of methylene chloride, and the entire contents are transferred to the filter funnel containing the anhydrous sodium sulfate. The filter funnel and contents are rinsed with additional methylene chloride (20 to 40 mL). The total eluent from the filter funnel is collected in the 100-mL volumetric flask. Discard the sodium sulfate. 11.1.7 11.2 Add 10 mL of methylene chloride and homogenize the mixture for approximately 1 min with a Tekmar Tissuemizer®. The final volume of the extract for each sample is adjusted to 100 mL in the volumetric flask using methylene chloride. Lipid Determination 11.2.1 Preweigh a clean 1-dram glass vial to the nearest 0.0001 g using an analytical balance tared to zero. 11.2.2 Accurately transfer 1.0 mL of the final extract (100 mL) from Section 11.1.7 to the 1-dram vial. Reduce the volume of methylene chloride from the extract using a water bath (50-60°C) gentle stream of purified nitrogen until an oil residue remains. A-34 �11.2.3 Accurately weigh the 1-dram vial and residue to the nearest 0.0001 g and calculate the weight of lipid present in the vial based on difference. Nitrogen blow-down is continued until a constant weight is achieved. 11.2.4 Calculate the percent lipid content of the original sample to the nearest 0.1% as shown in Equation 11-1. " I D * FYT Lipid content, LC ( ) = w % x 100% v AT AL Eq. 11-1 where: W.R = weight of the lipid residue to the nearest 0.0001 g calculated from Section 11.2.3; = total volume of the extract in mL from Section 11.1.6 (100.0 ml); WAT = weight of the original adipose tissue samples to the nearest 0.01 g from Section 11.1.1; and V.. 11.2.5 11.3 = volume of the aliquot of the final extract in ml used for the quantitative measure of the lipid residue (1.0 mL). Record the lipid residue measured in Section 11.2.3 and the percent lipid content calculated from Section 11.2.4. Extract Concentration 11.3.1 Quantitatively transfer the remaining extract volume (99.0 mL) to a 500-mL Erlenmeyer flask. Rinse the volumetric flask with 20 to 30 mL of additional methylene chloride to ensure quantitative transfer. 11.3.2 Place the Erlenmeyer flask on a hot plate at 40°C to remove solvent until an oily residue remains. 12. CLEANUP PROCEDURES 12.1 Bulk Lipid Removal 12.1.1 Add a total of 200 mL of rrhexane to the spiked lipid residue in the 500-mL Erlenmeyer flask. A-35 �12.1.2 Slowly add, with stirring, 100 g of the 40% w/w sulfuric acid impregnated silica gel (Section 7.1.3). Stir with a magnetic stir-plate for 2 h. 12.1.3 Allow solids to settle and decant liquid through a powder funnel containing 20 g of anhydrous sodium sulfate and collect in a 500-mL sample bottle. 12.1.4 Rinse solids with two 50-mL portions of hexane. Stir each rinse for 15 min, decant, and dry by elution through sodium sulfate combining the hexane extracts from Section 12.1.3. 12.1.5 After the rinses have gone through the sodium sulfate, rinse the sodium sulfate with an additional 25 ml of hexane and combine with the hexane extracts from Section 12.1.4. 12.1.6 Prepare an acidic silica column as follows: Pack a 1 cm x 10 cm chromatographic column with a glass wool plug, add approximately 25 ml of hexane, add 1.0 g of silica gel (Section 7.1.2) and allow to settle, then add 4.0 g of 40% w/w sulfuric acid impregnated silica gel (Section 7.1.3) and allow to settle. Pack a second chromatographic column (1 cm x 30 cm) with a glass wool plug, add approximately 25 mL of hexane, add 6.0 g of acidic alumina (Section 7.1.1), and allow to settle and then top with a 1-cm layer of sodium sulfate (Section 7.2). Elute the excess hexane solvent through the columns until the solvent level reaches the top of the chromatographic packing. Inspect columns to ensure they are free of channels and air bubbles. Wash the alumina column with 40 ml of 50% v/v methylene chloride/hexane. Remove the methylene chloride from the adsorbent by eluting the column with an additional 100 mL of hexane. Elute the excess solvent from the column until the solvent level reaches the top of the sodium sulfate layer. 12.1.7 Quantitatively transfer the hexane extract from the Erlenmeyer flask (Sections 12.1.3 through 12.1.5) to the silica gel column reservoir. Allow the hexane extract to percolate through the column and collect in a KD concentrator. 12.1.8 Complete the elution of the extract from the silica gel column with 50 ml of hexane in the KD concentrator. Concentrate the eluate to approximately 1.0 ml, using nitrogen blow-down as necessary. A-36 �Note: If the 40% sulfuric acid/silica gel is noted to be highly discolored throughout the length of the adsorbent bed it is necessary to repeat the cleaning procedure beginning with Section 12.1.1. 12.2 Separation of Chemical Interferences 12.2.1 Transfer the concentrate (1.0 mL) to the top of the alumina column. Rinse the K-D assembly with two 1.0-mL portions of hexane and transfer the rinses to the top of the alumina column. Elute the alumina column with 18 mL of hexane until the hexane level is just below the top of the sodium sulfate. Discard the eluate. Columns must not be allowed to reach dryness (i.e., a solvent "head" must be maintained). 12.2.2 Place 30 ml of 20% (v/v) methylene chloride in hexane on top of the alumina and elute the TCDDs from the column. Collect this fraction in a 50-mL culture tube. 12.2.3 Prepare an 18% Carbopak C/Celite 545® mixture by thoroughly mixing 3.6 g of Carbopak C (80/100 mesh) and 16.4 g of Celite 545® in a 40-mL vial. Activate at 130°C for 6 h. Store in a desiccator. Cut off a clean 5-mL disposable glass pipet (6 to 7 mm ID) at the 4-mL mark. Insert a plug of glass wool and push to the 2-mL mark. Add 500 mg of the activated Carbopak/Celite mixture followed by another glass wool plug. Using two glass rods, push both glass wool plugs simultaneously towards the Carbopak/Celite mixture and gently compress the Carbopak/Celite plug to a length of 3 to 3.5 cm. Pre-elute the column with 2 ml of toluene followed by 1 ml of 75:20:5 methylene chloride/methanol/ benzene, 1 ml of 1:1 cyclohexane in methylene chloride, and 2 mL of hexane. The flow rate should be less than 0.5 mL/min. While the column is still wet with hexane, add the entire eluate (30 ml) from the alumina column (Section 12.2.2) to the top of the column. Rinse the culture tube which contained the extract twice with 1 mL of hexane and add the rinsates to the top of the column. Elute the column sequentially with two 1-mL aliquots of hexane, 1 mL of 1:1 cyclohexane in methylene chloride, and I mL of 75:20:5 methylene chloride/methanol/benzene. Turn the column upside down and elute the PCDD/PCDF fraction with 20 mL of toluene into 6-dram vial. 12.2.4 Using a stream of nitrogen, reduce the toluene volume to approximately 1 mL. Carefully transfer the concentrate into a 1-mL mini vial and reduce the volume to about 200 (jL using a stream of nitrogen. A-37 �12.2.5 Rinse the concentrator tube with three washings using 500 uL of 1% toluene in methylene chloride. Concentrate to 200-500 |jL and add 10 |jL of the tridecane solution containing the internal recovery standard and store the sample in a refrigerator until HRGC/MS analysis. 12.2.6 Immediately prior to analysis, using a gentle stream of nitrogen at room temperature, remove toluene and methylene chloride. Submit sample to HRGC/MS once a stable 10 uL volume of tridecane is attained. 13. ANALYTICAL PROCEDURES 13.1 HRGC/MS Analysis for PCDD/PCDF 13.1.1 Once routine calibration criteria are met, the instrument is ready for sample analysis. Prior to the first sample, a blank injection of tridecane should be analyzed to document system cleanliness. If any evidence of system contamination is found, corrective action must be taken and another tridecane blank analyzed. The typical daily sequence of injections is shown in Table 9 and Figure 3. Note: Syringe Technique -- Congeners of PCDD/PCDF in the syringes used for HRGC/MS analysis can be problematic unless the syringes are properly handled between samples. The following procedure has been found to be very effective for PCDD/PCDF removal from contaminated syringes and will be used throughout these analyses. • Rinse the syringe 10 times with isooctane. Fill the syringe with toluene and sonicate syringe and plunger in toluene for 5 min and repeat at least twice. • Rinse the syringe 10 times with tridecane and pull up 1 pL of clean tridecane. • Syringe is ready for use. At no time should air be introduced into the HRGC column by using an air plug in the syringe. The oxygen present in the air plug will quickly degrade a nonbonded GC phase. 13.1.2 Inject a 1-uL aliquot of the extract into the GC, operated under the conditions previously used (Section 8.1) to produce acceptable results with the performance check solution. A-38 �Table 9. Typical Daily Sequence for PCDD/PCDF Analysis 1. Tune and calibrate mass scale versus perfluorokerosene (PFK). 2. Inject column performance mixture. 3. Inject concentration calibration solution 2.5 to 12.5 pg/nL (CS-7) solution. 4. Inject blank (tridecane). 5. Inject samples 1 through "N". 6 Inject concentration calibration solution 2.5 to 12.5 pg/pL (CS-7) solution or other concentration calibration solutions CS1 to CSS to bracket observed sample concentration. A-39 �INSTRUMENTAL ANALYSIS Instrument Mass Calibration vs PFK Mass Resolution Check 1 Column Performance Evaluation Does Column Performance Meet Minimum Resolution Requirements? No Adjust Column Length or Install New Column Calibration Standard Analysis Do Relative Response Factors Meet Criteria Based on Initial Calibrations ? No Reanalyze or Prepare Fresh Calibration Standards and Calibration Curve Yes Proceed with Sample Analysis Figure 3. Daily QA procedures for proceeding with sample analysis. A-40 �13.1.3 Acquire SIM data according to the same acquisition and MS operating conditions previously used (Section 8.1) to determine the relative response factors. 13.1.3.1 13.1.3.2 13.2 Acquire SIM data for the characteristic ions designated in Table 6. Instrument performance shall be monitored by examining and recording the peak areas for the recovery standard, 13C12-1,2,3,4-TCDD. If this area should decrease to less than 50% of the calibration standard, sample analyses shall be stopped until the problem is found and corrected. HRGC/HRMS Confirmation of 2,3,7,8-TCDD The presence of 2,3,7,8-TCDD observed through the general PCDD and PCDF procedure should be confirmed using HRGC/HRMS (resolution 10,000). 13.2.1 Once the daily criteria of mass calibration, mass resolution, HRGC performance, and routine calibration are met and documented, the instrument is ready for sample analysis. Prior to the first sample, a blank injection of tridecane will be made to document system cleanliness. The typical daily schedule for HRGC/HRMS analysis of TCDD is shown in Table 10 and Figure 3. 13.2.2 Inject a 1-uL aliquot of the extract into the GC, operated under the conditions previously used (Section 8.2) to produce acceptable results with the column performance check solution. 13.2.3 Acquire SIM data according to Section 8.2.4.3. Use the same acquisition and MS operating conditions previously used to determine the relative response factors. 13.2.3.1 Acquire SIM data for the following selected characteristic ions: m/z Compound 258.930 TCDD - COC1 319.897 Unlabeled TCDD 321.894 Unlabeled TCDD 331.937 13 C12-2,3,7,8-TCDD, 13 C12-1,2,3,4-TCDD 333.934 13 C12-2,3,7,8-TCDD, 13 C12-1,2,3,4-TCDD A-41 �Table 10. Typical Daily Schedule for HRGC/HRMS Analysis of TCDD 1. Tune and calibrate mass scale. 2. Perform mass measurement check and mass resolution check. 3. Inject column performance check solution. 4. Inject the routine concentration calibration solution (CS7) and confirm response factor consistency. 5. Inject tridecane blank. 6. Inject samples 1 through "N". 7. Inject concentration calibration solution and confirm response factor consistency. 8. Mass resolution check. A-42 �14. DATA REDUCTION In this section, the the analysis of data HRGC/HRMS method for qualitative criteria 14.1 procedures for the data reduction are outlined for from both the HRGC/MS method for PCDD/PCDF and the 2,3,7,8-TCDD. Figure 4 presents a schematic of the for identifying PCDDs and PCDFs. Qualitative Identification 14.1.1 14.1.2 The ion current intensities for a particular PCDD/PCDF must be ^ 2.5 times the noise level (S/N ^ 2.5) for positive identification of that isomer. 14.1.3 The integrated ion current ratios of the analytical masses for a particular PCDD/PCDF must fall within the ranges shown in Table 7. 14.1.4 14.2 The ion current responses for each mass for a particular PCDD/PCDF analyte must be within ± 1 s to attain positive identification of that analyte. For example, m/z 338 and m/z 340 must have maximum peak responses that are within ± 1 s to be positively identified as a pentachlorodibenzofuran. The recovery of the internal quantisation standards should be between 50 and 115%. Quantitative Calculations 14.2.1 Relative response factors for native PCDD and PCDF analytes (RRF). RRFs are calculated from the data obtained during the analysis of concentration calibration solutions using the following formula: A •r STD IS RRF = .b'U r i;> tt U IS ' STD where A... = the ,,, ion For sum m/z Eq. 14-1 sum of the areas of the integrated abundances for the analyte in question. example, for TCDD, ASTD would be the of the integrated ion abundances for 320 and 322; A,,, = the sum of the areas of the integrated ion abundances for the labeled PCDD/F used as the internal quantisation standard for the above analyte. For example, for 13C12~ 2,3,7,8-TCDD, A JS would be the sum of the integrated ion abundance for m/z 332 and 334. Cc-rn = concentration of the analyte in pg/(jL; A-43 �HRGC/MS-SIMData Response to Characteristic Molecular Ions within the Appropriate Homolog Retention Window? Report Compounds as Not Detected (ND) Calculate Sample LOP Characteristic Ion Ratios within ±20% Theoretical? Response Due to Coextracted Interference Response Corresponds to Specific Isomer Retention Time? Quantitate Compound as Per Protocol Report as Isomer Unknown Quantitate Specific Isomer as per Protocol Figure 4. Qualitative criteria for identifying PCDDs and PCDFs. A-44 �GIS = concentration of the internal quantisation standard in pg/uL; and Table 11 provides the pairing of target analytes to internal quantitation standards for determining RRF values for PCDD and PCDF compounds. 14.2.2 Relative response factors for the internal quantitation standards (RRF,<.)- The RRF TS values are calculated from data obtained auring the analysis of concentration calibration solutions using the following formula. C Eq. 14-2 ~ RS A XL x M RS IS where ATC. and CT<; are defined as given in Section 14.2.1 1:> i:> and CRS = concentrations of the internal recovery standard in pg/uL; and Apc; = the sum of the areas of the integrated ion 5 abundances for the labeled PCDD (13C121,2,3,4-TCDD or 13C12-l,2,3,7,8,9-HxCDD). For example, for 13C12-1,2,3,4-TCDD, A R$ would be the sum of the integrated ion abundance for m/z 332 and 334. Refer to Table 11 for pairing of the internal quantitation standards with the appropriate internal recovery standard. 14.2.3 Concentrations of sample components. Figure 5 presents a schematic for quantitation of PCDDs and PCDFs which meet the criteria specified in Section 14.1. Calculate the concentration of PCDD/Fs in sample extracts using the formula: A -.«nn~ " QT C . i r\n X Csampl e A sampl e » IS 100 Eq. 14-3 IS RRF wAT • LC • where C , = the lipid adjusted concentration of PCDD or sample congener 1n pg/g. A , = sum of the integrated ion abundances determined for the PCDD/PCDF in question; sample AIS = sum of the integrated ion abundances determined for the labeled PCDD/F used as the internal quantitation standard for the above analyte; A-45 �Table 11. Target Analyte/Internal Quantitation Standard and Internal Quantisation Standard/Internal Recovery Standard Pairs Internal standards Quantitation Target analyte 2,3,7,8-TCDD 13 13 2,3,7,8-TCDF 13 13 1,2,3,7,8-PeCDF 13 13 2,3,4,7,8-PeCDF 13 13 1,2,3,7,8-PeCDD 13 13 1,2,3,4,7,8-HxCDF 13 13 1,2,3,6,7,8-HxCDF 13 13 2,3,4,6,7,8-HxCDF 13 13 1,2,3,7,8,9-HxCDF 13 13 1,2,3,4,7,8-HxCDD 13 13 1,2,3,6,7,8-HxCDD 13 13 1,2,3,7,8,9-HxCDD 13 13 1,2,3,4,6,7,8-HpCDF 13 13 1,2,3,4,7,8,9-HpCDF 13 13 1,2,3,4,6,7,8-HpCDD 13 13 OCDF 13 13 OCDD 13 13 Recovery C12-2,3,7,8-TCDD C12-2,3,7,8-TCDF C12-l,2,3,7,8-PeCDF C12-l,2,3,7,8-PeCDF C12-l,2,3,7,8-PeCDD C12-l,2,3,4,7,8-HxCDF C12-l,2,3,4,7,8-HxCDF C12-l,2,3,4,7,8-HxCDF C12-l,2,3,4,7,8-HxCDF C12-l,2,3,6,7,8-HxCDD C12-l,2,3,6,7,8-HxCDD C12-l,2,3,6,7,8-HxCDD C12-l,2,3,4,6,7,8-HpCDF C12-l,2,3,4,6,7,8-HpCDF C12-l,2,3,4,6,7,8-HpCDD C12-OCDD C12-1,2,3,4-TCDD C12-1,2,3,4-TCDD C12-1,2,3,4-TCDD C12-1,2,3,4-TCDD C12-1,2,3,4-TCDD C12-l,2,3,7,8,9-HxCDD C12-l,2,3,7,8,9-HxCDD C12-l,2,3,7,8,9-HxCDD C12-l,2,3,7,8,9-HxCDD C12-l,2,3,7,8,9-HxCDD C12-l,2,3,7,8,9-HxCDD C12-l,2,3,7,8,9-HxCDD C12-l,2,3,7,8,9-HxCDD C12-l,2,3,7,8,9-HxCDD C12-l,2,3,7,8,9-HxCDD C12-l,2,3,7,8,9-HxCDD C12-OCDD C12-l,2,3,7,8,9-HxCDD A-46 �QUANTITATION HRGC/MS-SIMData Response Meets Al I Qualitative Criteria ? Report as Not Detected Calculate Sample LOD Response >2.5times S/N? Response >10 times S/N? Calculate as per Protocol Report as Trace (tr) Value Quantitate as per Protocol Report as Positive Quantifiable Value Figure 5. Procedure for quantitation of PCDDs and PCDFs in human adipose tissue. A-47 �QIS = the amount (total pg) of the labeled internal quantisation standard added to the sample prior to extraction; RRF = relative response factor of the above analyte relative to its labeled internal quantitation standard determined from the initial triplicate calibration; WAT = weight (g) of original adipose tissue sample; and LC = percent extractable lipid determined from Eq. 11-1. Refer to Table 11 for pairing of target analytes with the appropraite internal quantitation standard. Quantitative data should be classified to indicate the intensity of the signal response. Suggested qualifiers include: not detected, ND (signal-to-noise ratio is less than 2.5); trace, TR (signal-to-noise ratio is greater than or equal to 2.5 but less than 10); and positive quantifiable, PQ (signal-to-noise ratio is greater than or equal to 10). 14.2.4 Recovery of internal quantitation standards. Calculate the recovery of the labeled internal quantitation standards measured in the final extract using the formula: A IS • X0 Internal Quant. Std. _ RS . ,. . fl0 Percent Recovery AR$ - Q J$ • RRF 1UU f tq ' 1^4 where AT{. = sum of the integrated ion abundances determined for the labeled PCDD/PCDF internal quantitation standard in question; A0<- = sum of the integrated ion abundances deterKb mined for m/z 332 and m/z 334 of 13C121,2,3,4-TCDD or m/z 390 and m/z 392 of 13 C13-l,2,3,7,8,9-HxCDD (recovery standards) QR<- = amount (pg) of the respective recovery standard, added to the final extract; QTC. = amount (pg) the labeled internal quantitation standard added to the sample prior to extraction; and A-48 �= relative response factor for the labeled internal quantitation standard in question relative to the internal recovery standard. This value shall be the RRF determined from the initial calibration. Refer to Table 11 for pairing of the internal quantitation standards with the appropriate target analytes. Note: The result of calculations as presented in Section 14.2 may be off by as much as 1% due to the fact that 1 ml of the final 100 ml volume from the extraction was used for lipid determination. 14.3 Estimated Method Detection Limit Estimated where (1) sponse is sponse is method detection limits must be calculated in situations no response is noted for a specific congener; (2) a renoted but ion ratios are incorrect; and (3) where a requantitated as a trace value. 14.3.1 For samples in which no unlabeled PCDD or PCDF is detected, calculate the estimated minimum detectable concentration. The background area is determined by integrating the ion abundances for the characteristic ions in the appropriate region and relating the product area to an estimated concentration that would produce that product area. Use the formula: 2.5 sample RRF "IS Eq. 14-5 where C p = estimated concentration of unlabeled PCDD or PCDF required to produce A ,; sample 'IS sum of integrated ion abundances or peak heights for the characteristic ions of the unlabeled PCDD or PCDF isomer in the same group of ^ 5 scans used to measure A JS ; and sum of integrated ion abundances for the appropriate ions characteristic of the respective internal quantitation standard. Qyc> RRF, and W.T retain the definitions previously stated in Section 14.2. Alternatively, if peak height measurements are used for quantification, measure the estimated detection limit by the peak height of the noise in the 2,3,7,8-TCDD RT window. A-49 �14.3.2 14.3.3 15. For samples for which a response at the retention time of a specific PCDD or PCDF congener is noted, but the qualitative criteria for ion ratios are outside the acceptable range (Table 7), the estimated detection level is calculated as given in Eq. 14.3 except the values are qualified as not detected, ND, and the concentration is reported in parenthesis. If a response for a specific PCDD or PCDF congener is qualified as a trace, TR, value (signal to noise is greater than or equal to 2.5 but less than 10) the analyst must also provide an estimated method detection limit. This is accomplished by using the observed signal to noise on either side of the response and calculating as given in Eq. 14-5. REPORTING AND DOCUMENTATION All data should be reported on an individual sample basis using the data report format shown in Figure 6. The analyst is required to maintain all raw data, calculations, and control charts in a format as to allow a complete external data review. Suggested data formats for tracing calculations are provided in Figure 7. A-50 �P>g. 1 ot 1 U.S. ENVIRONMENTAL PROTECTION AGENCY OFFICE OF TOXIC SUBSTANCES EXPOSURE EVALUATION DIVISION (TS-798) WASHINGTON, DC 20460 NATIONAL HUMAN ADIPOSE TISSUE SURVEY ANALYSIS REPORT FORM EPA SAMPLE NUMBER. ANALYSIS DATE LAB NUMBER MS ANALYST BATCH NUMBER REPORT DATE _ REPORTED BY _ NATIVE COMPOUNDS CONCENTRATION DATA (pg/g)l/ QUALIFIER!/ INTERNAL OUANTITATION STANDARD 2.3.7.8-TCDD 1 . . . ! • ( , 1 1 2.3,7,8-TCOF 1 . . . 1*1 1 1 1: 1 ,2,3.7.8-PeCDD 1 ...!•!, 1 13 1 ,,.!•!, 1 13 2.3.4,7,8-PeCDF 1 i . i Ul . 1 13 1.2.3.4,7.8-HxCOD 1 ,..!•( i 1 13 1 ,2.3,6.7,8-HxCDD 1 , . , ! • ! , 1 13 1,2.3,7.8,9-HxCDD 1 , • , Ul iJ l3 . 1 1 1,2,3.7,8-PeCDF 1,2.3,4,7.8-HxCDF 1 , . , Ul 1 ,2.3.6,7.8-HxCOF 1 , ,.!•!, 1 1 2.3 7,8.9-HxCOF 1 ... 1,2,3,4.6.7,8-HpCDD 1 , . , Ul Ci2-'.2.3.7,8-PeCDF Ci2-'-2'3'6'7-8-HxCDD Ci2-l,2,3.4,7,8-HxCDF C(2-1,2.3,4,6.718-HpCOD C,2-'.2.3.4,a,7,8-HpCDF 30,2-0000 13 C,2-OCDF 1 . , , !•! , 1 1 ,2,3.4.6.7,8-HpCDF Ci2-'.2.3.7.8-PeCDD 1 , . , 1*1 , 1 2.3,4.6.7,8-HxCDF 3C,2-2.3.7.8-TCDD >C12-2.3,7,8-TCDF ULi_J . 1 1.2.3.4,7,8,9-HpCOF 1 . . , Ul . 1 OCDO 1 , , , Ul OCDF 1 , , , Ul , I . 1 REMARKS Jj Concentration reported is based on total axtractable iipid (g). 2j ND - Not Detected. TR - Trace. PG - Positive Quantifiable. Figure 6. Analysis report form. A-51 SPIKED LEVEL (pg) PERCENT (%) RECOVERY �RAW DATA SUMMARY FOR DETERMINATION OF 1.2.3.7.8-PeCDD IN HUMAN ADIPOSE TISSUE Sample no. Sample weight (xx.xx q) Extractable lipid content (XX. X % ) Analysis date Amount C12-PeCDD (pg) 13 13 C12-PeCDO n/z 332 13 Cl2-PeCDD m/z 334 Ion ratio 366/368 1,2,3,7,8PeCDD m/z 354 1,2,3,7,8 m/z 356 I tn ro Value reported as concentration in extractable lipid. Figure 7. Example of raw data summary format for the determination of 1,2,3,7,8-PeCDD in human adipose tissue. Ion ratio 354/356 1,2,3,7,8PeCOD cone. (pg/g) �TECHNICAL REPORT DATA (Please read Inunctions on the reverse before completing) 3. RECIPIENT'S ACCESSlON>NO. I a. NO. EPA 560/5-86-020 S. REPORT DATE 4. 7i7LS AND SUBTITLE Analysis for Polychlorinated Dibenzo-p_-dioxins (PCDD) and Dibenzofurans (PCDF) in Human Adipose Tissue: Method Evaluation Study September 17. 1986 6. PERFORMING ORGANIZATION CODE Midwest Research Institute 7.AUTHOR<s> js Stanley, RE Ayling, KM Bauer, MJ McGrath, TM Sack, and KR Thornberg 8. PERFORMING ORGANIZATION REPORT NO. a-JeRFOHMLNG ORGANIZATION NAME ANO ADDRESS 10. PROGRAM ELEMENT NO. Thawest Research Institute 425 Volker Boulevard Kansas City, MO 64110 11. CONTRACT/GRANT NO. 68-02-3938 68-02-4252 12. SPONSORING AGENCY NAME ANO ADDRESS Field Studies Branch (TS-798) Exposure Evaluation Division Office of Toxic Substances U.S. 8824-A(01) 13. TYPE OF REPORT ANO PERIOD COVERED Final Washinqton, DC 20460 14. SPONSORING AGENCY CODE Environmental Prntprt.inn Aqpnr.v 13. SUPPLEMENTARY NOTES J Remmers, Work Assignment Manager; J Breen, Project Officer 16. ABSTRACT This report focuses on the evaluation of an HRGC/MS analytical method for determination of 2,3,7,8-substituted polychlorinated dibenzo-p_-dioxins (PCDD) and dibenzofurans (PCDF) in human adipose tissue. This method will be used for analysis of samples from EPA's National Human Adipose Tissue Survey (NHATS) as part of a collaborative effort between EPA's Office of Toxic Substances and the Veterans Administration. The method was evaluated using aliquots of a bulk lipid matrix that was extracted from human adipose tissue. The results of the replicate analysis of spiked and unspiked homogenized human adipose tissue matrix demonstrate that the analytical method produces accurate and precise data for 17 specific 2,3,7,8-substituted PCDD and PCDF (tetra- through octachloro homologs) congeners. The endogenous or background levels of the PCDD and PCDF congeners in the homogenized adipose lipid matrix were estimated through regression analyses of measured versus spiked concentrations for each compound. This unspiked matrix will be used as a control sample with each batch of samples analyzed. KEY WORDS ANO DOCUMENT ANALYSIS 17. DESCRIPTORS b.lOENTIFIERS/OPSN ENDED TERMS C. 19. SECURITY CLASS < rhu Report/ 21. NO. OP PAGES COSATt Fteid'CfOUp D olychlorinated dibenzo-jp_-dioxin (PCDD) Polychlorinated dibenzofuran (PCDF) Human adipose tissue Method evaluation 2,3,7,8-Tetrachlorodibenzo-g-dioxin (2,3,7,8-TCDD) I «3.7.8-Tetrachl orodi benzofuran (2.3.7.f 13. ^ i S T P i a u T i O N STATEMENT Release unlimited Unclassified 20. SECURITY CLASS Unclassified EPA fttm 2220-1 (3-73) 145 22. � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Alvin L. Young Collection on Agent Orange Description An account of the resource <p style="margin-top: -1em; line-height: 1.2em;">The Alvin L. Young Collection on Agent Orange comprises 120 linear feet and spans the late 1800s to 2005; however, the bulk of the coverage is from the 1960s to the 1980s and there are many undated items. The collection was donated to Special Collections of the National Agricultural Library in 1985 by Dr. Alvin L. Young (1942- ). Dr. Young developed the collection as he conducted extensive research on the military defoliant Agent Orange. The collection is in good condition and includes letters, memoranda, books, reports, press releases, journal and newspaper clippings, field logs and notebooks, newsletters, maps, booklets and pamphlets, photographs, memorabilia, and audiotapes of an interview with Dr. Young.</p> <p>For more about this collection, <a href="/exhibits/speccoll/exhibits/show/alvin-l--young-collection-on-a">view the Agent Orange Exhibit.</a></p> Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Box The box containing the original item. 197 Folder The folder containing the original item. 5526 Series The series number of the original item. Series VIII Subseries I Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Creator An entity primarily responsible for making the resource Stanley, John S. et al. Description An account of the resource <strong>Corporate Author: </strong>Midwest Research Institute Date A point or period of time associated with an event in the lifecycle of the resource 1986-10-01 Title A name given to the resource Analysis for Polychlorinated Dibenzo-p-Dioxins (PCDD) and Dibenzofurans (PCDF) in Human Adipose Tissue: Method Evaluation Study: Final Report with attached letter transmitting the report to Alvin L. Young, from Janet C. Remmers, Field Studies Branch, Expo ao_seriesVIII https://www.nal.usda.gov/exhibits/speccoll/files/original/40acfd2554f623d0f152b8286123a60b.pdf e2abc03f964ecc5b150083b7f6d4a72e PDF Text Text Item D Number 0552 Author Stanley, John S. Corporate Author Midwest Research Institute Report/Article Title Analysis for Polychlorinated Dibenzo-p-Dioxins and Dibenzofurans in Human Adipose Tissue: Method Evaluation Study: Draft Final Report with attached letter transmitting the report to Alvin L. Young, from Janet C. Remmers, Field Studies Branch, Exposure Evaluation Division, United States Environmental Protection Agency 2 Not Scanned Journal/Book Title Year 1986 Month/Day Color D 143 DOSCrtotOU NOtOS EP/flk Prime Contract No. 68-02-3938, Work Assignment No. 46, MRI Project No. 8501 -A(46). See Item 5526 for final version Tuesday, March 19,2002 Page 5522 of 5611 �A\ Sfa | UNITED STATES ENVIRONMENTAL PROTECTION AGENCY ,/ WASHINGTON, D.C. 20460 JUNE 10, 1986 OFFICE OF PHT1CIDES AND TOXIC SUBSTANCES Alvin Young, Ph.D., Lt Col., USAF Senior Policy Analyst for Life Sciences Executive Office of the President Office of Science and Technology Policy Room 5005 New Executive Office Building Washington, DC 20506 Dear Dr . Young: Enclosed for your information is a report entitled, "Analysis for Polychlorinated Dibenzo-p_-dioxins and Dibenzofurans in Human Adipose Tissue: Method Evaluation Study", May 30, 1986. This report includes the analytical protocol which will be used in the EPA/VA dioxin study of adipose tissue from veterans and nonveterans. Also included is the method evaluation data from a single laboratory study. If you have any comments or questions, please feel f r e e to call me at 382-3583. Sincerely, fanet C. Remmers Field Studies Branch Exposure Evaluation Division (TS-798) Enclosure cc: M a r t i n Halper Joseph Breen (-11 �MIDWEST RESEARCH INSTITUTE ANALYSIS FOR POLYCHLORINATED DIBENZQ-£-DIOXINS AND DIBENZOFURANS IN HUMAN ADIPOSE TISSUE: METHOD EVALUATION STUDY DRAFT FINAL REPORT EPA Prime Contract No. 68-02-3938 Work Assignment No. 46 MRI Project No. 8501-A(46) For Field Studies Branch, TS-798 Office of Toxic Substances U.S. Environmental Protection Agency 401 M Street, S.W. Washington, DC 20460 Attn: Ms. Janet Remmers, Work Assignment Manager Dr. Joseph J. Breen, Project Officer MIDWEST RESEARCH INSTITUTE 425 VOLKER BOULEVARD, KANSAS CITY, MISSOURI 64110 • 816 753-7600 �ANALYSIS FOR POLYCHLORINATED DIBENZO-jD-DIOXINS AND DIBENZOFURANS IN HUMAN ADIPOSE TISSUE: METHOD EVALUATION STUDY by John S. Stanley, Randy E. Ayling, Karin M. Bauer, Michael J. McGrath, Thomas M. Sack, and Kelly R. Thornburg DRAFT FINAL REPORT EPA Prime Contract No. 68-02-3938 Work Assignment No. 46 MRI Project No. 8501-A(46) For Field Studies Branch, TS-798 Office of Toxic Substances U.S Environmental Protection Agency 401 M Street, S.W. Washington, DC 20460 Attn: Ms. Janet Remmers, Work Assignment Manager Dr. Joseph J. Breen, Project Officer MIDWEST RESEARCH INSTITUTE 425 VOLKER BOULEVARD, KANSAS CITY, MISSOURI 64110 • 816 753-7600 �DISCLAIMER This document is a preliminary draft. It has not been released formally by the Office of Toxic Substances, Office of Pesticides and Toxic Substances, U.S. Environmental Protection Agency. It is being circulated for comments on its technical merit and policy implications. �PREFACE This report provides a summary of the results from a method evaluation study for the determination of 2,3,7,8-substituted polychlorinated dibenzo-p_-dioxins (PCDDs) and dibenzofurans (PCDFs) in human adipose tissues. This method evaluation is an integral part of a collaborative program between the U.S. Environmental Protection Agency's Office of Toxic Substances and the Veterans Administration to determine if significant differences exist in the 2,3,7,8-substituted PCDD and/or PCDF levels in human adipose tissues for Vietnam veterans compared to the general adult male population. The study design will focus on specimens within EPA's National Human Adipose Tissue Survey (NHATS) repository. The method evaluation described in this report was necessary to establish method performance (accuracy and precision) before proceeding with actual sample analysis. This method evaluation study was completed under EPA Contract No. 68-02-3938, Work Assignment 46, "Analysis for Dioxins and Furans in Human Adipose Tissue," Ms. Janet Remmers, Work Assignment Manager, and Dr. Joseph Breen, Project Officer. MIDWEST RESEARCH INSTITUTE Paul C. Constant Program Manager Approved: Jack Balsinger QuaH\y Assurance Coordinator E. Going, Director' Chemical Sciences Department May 30, 1986 11 �TABLE OF CONTENTS Page I. Introduction II. Summary III. Recommendations 4 IV. Experimental 5 A. B. C. D. E. F. G. 5 5 10 12 17 20 20 V. 1 • Preparation of Homogenized Tissue Analytical Standards Analytical Procedure HRGC/MS Analysis Data Interpretation Quality Assurance/Quality Control (QA/QC) Preliminary Method Studies. Results 23 A.. Analytical Results B. Statistical Analysis VI. Quality Assurance/Quality Control (QA/QC) A. B. C. D. VII. 3 Initial Calibration Daily Verification of Response Factors Blanks Absolute Recoveries of the Internal Quantisation Standards References 23 23 62 62 64 64 76 81 Appendix A - Analytical Protocol for Determination of PCDDs and PCDFs in Human Adipose Tissue m A-l �LIST OF FIGURES Figure 1 2 Page Comparison of the HRGC/MS-SIM reconstructed ion chromatogram (RIC) from the analysis of unspiked homogenized human adipose tissue matrix and a calibration standard for PCDDs and PCDFs Example of the TCDF (m/z 304) and TCDD (m/z 320) HRGC/MS-SIM elution profiles in unspiked and spiked human adipose 3 4 5 6 7 8 9 10 11 12 30 31 Example of the PeCDF (m/z 338) and PeCDD (m/z 354) HRGC/MS-SIM elution profiles in unspiked and spiked human adipose 32 Example of the HxCDF (m/z 374) and HxCDD (m/z 390) HRGC/MS-SIM elusion profiles in unspiked and spiked human adipose 33 Example of the HpCDF (m/z 408) and HpCDD (m/z 424) HRGC/MS-SIM elution profiles in unspiked and spiked human adipose 34 Examples of the OCDF (m/z 442) and OCDD (m/z 458) HRGC/MS-SIM elution profiles in unspiked and spiked human adipose 35 Measured concentrations versus concentrations of 2,3,7,8TCDD spiked into the homogenized human adipose lipid matrix . 36 Measured concentrations'versus concentrations of 1,2,3,7,8-PeCDD spiked into the homogenized human adipose lipid matrix 37 Measured concentrations versus concentrations of 1,2,3,4,7,8-HxCDD spiked into the homogenized human adipose lipid matrix 38 Measured concentrations versus concentrations of 1,2,3,6,7,8-HxCDD spiked into the homogenized human adipose lipid matrix 39 Measured concentrations versus concentrations of 1,2,3,7,8,9-HxCDD spiked into the homogenized human adipose lipid matrix 40 Measured concentrations versus concentrations of 1,2,3,4,6,7,8-HpCDD spiked into the homogenized human adipose lipid matrix. 41 �LIST OF FIGURES (continued) Figure 13 14 15 16 17 18 19 20 21 22 23 24 25 Page Measured concentrations versus concentrations of OCDD spiked into the homogenized human adipose lipid matrix. . 42 Measured concentrations versus concentrations of 2,3,7,8-TCDF spiked into the homogenized human adipose lipid matrix 43 Measured concentrations versus concentrations of 1,2,3,7,8-PeCDF spiked into the homogenized human adipose lipid matrix 44 Measured concentrations versus concentrations of 2,3,4,7,8-PeCDF spiked into the homogenized human adipose lipid matrix 45 Measured concentrations versus concentrations of 1,2,3,4,7,8-HxCDF spiked into the homogenized human adipose lipid matrix. . . . . 46 Measured concentrations versus concentrations of 1,2,3,6,7,8-HxCDF spiked into the homogenized human adipose lipid matrix 47 Measured concentrations versus concentrations of 2,3,4,6,7,8-HxCDF spiked into the homogenized human adipose lipid matrix 48 Measured concentrations versus concentrations of 1,2,3,7,8,9-HxCDF spiked into the homogenized human adipose lipid matrix 49 Measured concentrations versus concentrations of 1,2,3,4,7,8,9-HpCDF spiked into the homogenized human adipose lipid matrix 50 Measured concentrations versus concentrations of 1,2,3,4,6,7,8-HpCDF spiked into the homogenized human adipose lipid matrix 51 Measured concentrations versus concentrations of OCDF spiked into the homogenized human adipose lipid matrix. . 52 Method accuracy estimates as determined from the slopes of the least squares regression lines for the 17 target PCDD and PCDF analytes 54 Control charts showing response factors by date for 2,3,7,8-TCDF and 2,3,7,8-TCDD . . 65 �LIST OF FIGURES (continued) Figure 26 27 28 29 30 31 Page Control charts showing response factors by date for 1,2,3,7,8-PeCDF and 2,3,4,7,8-PeCDF 66 Control chart showing response factors by date for 1,2,3,7,8-PeCDD 67 Control charts showing response factors by date for 1,2,3,4,7,8-HxCDF and 1,2,3,6,7,8-HxCDF 33 34 68 Control charts showing response factors by date for 2,3,4,6,7,8-HxCDF and 1,2,3,7,8,9-HxCDF 69 Control charts showing response factors by date for 1,2,3,4,7,8-HxCDD and 1,2,3,6,7,8-HxCDD 70 Control chart showing response factors by date for 1,2,3,7,8,9-HxCDD 32 , 71 Control charts showing response factors by date for 1,2,3,4,6,7,8-HpCDF and 1,2,3,4,7,8,9-HpCDF 72 Control chart showing response factors by date for 1,2,3,4,6,7,8-HpCDD 73 Control charts showing response factors by date for OCDF and OCDD 74 �LIST OF TABLES Table 1 Page Analytical Standards Available for the Method Evaluation Studies 7 2 Concentration Calibration Solutions 8 3 Native PCDD and PCDF Spiking Solution 9 4 Internal Standard Spiking Solutions 11 5 HRGC/LRMS Operating Conditions for PCDD/PCDF Analysis . . . 13 6 Ions Monitored for HRGC/MS Analysis of PCDD/PCDF 14 7 Typical Daily Sequence for PCDD/PCDF Analysis 16 8 Ion Ratios for HRGC/MS Analysis of PCDD/PCDF 18 9 Summary of the Results of the Sample Preparation Method Evaluation Using Carbon-14 PCDDs 22 Spiked Versus Measured Concentrations of 2,3,7,8-TCDF and 2,3,7,8-TCDD in Homogenized Human Adipose Lipid Samples 24 Spiked Versus Measured Concentrations of 1,2,3,7,8-PeCDF, 2,3,4,7,8-PeCDF, and 1,2,3,7,8-PeCDD in Homogenized Human Adipose Tissue Samples 25 Spiked Versus Measured Concentrations of 1,2,3,4,7,8-; 1,2,3,6,7,8-; 2,3,4,6,7,8-; and 1,2,3,7,8,9-HxCDF in Homogenized Human Adipose Lipid Matrix 26 Spiked Versus Measured Concentration of 1,2,3,4,7,8-; 1,2,3,6,7,8-; and 1,2,3,7,8,9-HxCDD in Homogenized Human Adipose Lipid Samples 27 10 11 12 13 14 Spiked Versus Measured Concentrations of 1,2,3,4,6,7,8HpCDF, 1,2,3,4,7,8,9-HpCDF, and 1,2,3,4,6,7,8-HpCDD in Homogenized Human Adipose Lipid Samples 15 28 Spiked Versus Measured Concentrations of OCDF and OCDD In Homogenized Human Adipose Lipid Samples 29 16 Regression Line Slopes with 95% Confidence Limits 55 17 Results of the Analysis of the Low and High Level Native Spike Solutions. 56 vii �LIST OF TABLES (continued) Table Page 18 Background Level Estimates with 95% Confidence Limits . . . 58 19 Day-to-Day Precision of Analysis of Specific Sample Extracts for Tetra- and Pentachloro PCDF and PCDD . . . . 59 Day-to-Day Precision of Analysis of Specific Sample Extracts for Hexa- and Heptachloro PCDF and PCDD 60 Day-to-Day Precision of Analysis of Specific Sample Extracts for OCDF and OCDD 61 Relative Response Factors (Grand Means) Determined from Multipoint Concentration Calibration Standards 63 Summary of Results from the Analysis of a Laboratory Method Blank 75 20 21 22 23 24 Recovery of Radiolabeled PCDDs from Precleaned Activated Alumina 25 26 77 Absolute Recoveries of the Internal Quantisation Standards from the Human Adipose Lipid Matrix 78 Recovery of Carbon-14 Labeled 2,3,7,8-TCDD, 1,2,3,4,7,8HxCDD, and OCDD as a Function of Final Concentration Conditions 80 �I. INTRODUCTION The Environmental Protection Agency Office of Toxic Substances (EPA/OTS) and the Veterans Administration (VA) have established an interagency agreement to study the level of polychlorinated dibenzo-p_-dioxins (PCDDs) and dibenzofurans (PCDFs) in human adipose tissues. The occurrence and levels of PCDDs and PCDFs with chlorine substitution in the 2,3,7,8 positions (especially 2,3,7,8-TCDD) of the parent molecules are of primary interest. As part of this interagency effort, it has been proposed to use selected adipose tissue samples that were collected for the Field Studies Branch (FSB) of EPA's Office of Toxic Substances (OTS) through the National Human Adipose Tissue Survey (NHATS) to determine exposure to PCDDs and PCDFs. The available adipose tissues include specimens obtained from young men whose age indicates that they could have served in Vietnam and could have been exposed to Agent Orange. The tissues were originally collected as part of a broadly based and statistical random sampling of the continental United States. The analysis of these tissues may provide information on the differences of exposure of the general adult male population and Vietnam veterans to the 2,3,7,8-substituted PCDDs and PCDFs. The overall objectives of the proposed EPA/VA collaborative studies are: 1. Evaluate the reliability, accuracy, precision, and sensitivity of a proposed method for the determination of 2,3,7,8substituted PCDDs and PCDFs (tetra- through octachloro homologs) in human adipose tissue; 2. Determine if these compounds can be detected in adipose tissues of the American male adult population; and 3. Determine if individuals with military service in Vietnam have significantly different levels of 2,3,7,8-substituted PCDDs and PCDFs (particularly 2,3,7,8-TCDD) than other American men. As a prelude to this work assignment, MRI conducted an extensive literature review of applicable analytical methods and conducted a meeting with recognized experts in this field to identify critical aspects of analytical methodology.1'2 Based on the information gathered through the literature review and the meeting with the recognized experts, a special report was prepared for OTS proposing 3 a framework for an analytical method for analysis of human adipose tissues. Several studies have been completed since the issuance of that report which reflect the advances in analytical techniques for adipose tissue analysis.4 16 The salient features of these methods have been combined into a single protocol for the routine analysis of tetra- through octachloro PCDDs and PCDFs at the low-parts-per-trillion level for the EPA/VA tissue study. This report focuses on a method evaluation study that was conducted to achieve the first objective of the interagency agreement. Clarification �of method performance is necessary before proceeding with the analysis of actual samples retrieved from the NHATS repository. This report includes a summary of the method evaluation study results (Section II). Recommendations to be implemented before proceeding with the actual tissue samples from the NHATS repository are presented in Section III. A description of the actual experimental procedures is provided in Section IV. Results of sample analyses are summarized in Section V, and quality assurance/quality control (QA/QC) aspects of the study are detailed in Section VI. Pertinent references are listed in Section VII. Appendix A contains the detailed analytical protocol that will be followed for the analysis of the NHATS specimens designated in the study design to be provided by EPA/VA. �II. SUMMARY The results of the replicate analysis of spiked and unspiked homogenized human adipose tissue matrix demonstrate that the analytical method produces accurate and precise data for 17 specific 2,3,7,8-substituted PCDD and PCDF (tetra- through octachloro homologs) compounds. Data are reported for three or four replicate analyses of samples spiked at three different concentration levels. The endogenous or background levels of the PCDD and PCDF congeners in the homogenized adipose lipid matrix were estimated through regression analyses of measured (found) versus spiked concentrations for each compound, The analytical method is capable of providing quantitative data for tetra- through octachloro PCDD and PCDF congeners to concentration levels as low as 1 pg/g (tetrachloro congeners). However, an interference was noted at m/z 304 which coeluted with 2,3,7,8-TCDF, resulting in a detection level of approximately 4 pg/g. Average absolute recoveries of the internal quantisation standards ranged from 52% for 13Ci2-TCDD up to 89% for 13C12-OCDD. The agreement of the measured concentrations versus the spiked concentrations for each PCDD and PCDF congener demonstrates that the internal standard quantisation procedure provides an accurate measure of concentration which is independent of the absolute recovery. Final concentration conditions were noted to have pronounced effect on the absolute recoveries of the lower chlorinated compounds, particularly 2,3,7,8-TCDD. Experiments with carbon-14 labeled 2,3,7,8-TCDD demonstrated that final concentration at temperatures of 55 to 60°C resulted in recoveries as low as 54% while the same procedure conducted at ambient conditions resulted in greater than 90% recovery. Analysis of method and reagent blanks provided information on potential artifacts in the sample preparation scheme. Additional experiments were conducted with carbon-14 labeled PCDDs to evaluate the cleanup efficiency and recovery of PCDDs from chromatographic materials, particularly acidic alumina. �III. RECOMMENDATIONS Some minor modifications have been made in the written protocol (Appendix A) that were not included in this phase of the method validation. These include: a cleanup procedure for activated acidic alumina prior to fractionation of sample extracts to remove artifacts; and final concentration of the sample extracts using nitrogen blowdown at room temperature rather than heating to 55-60°C. The spiking solutions used to prepare the spiked quality control samples should be submitted for replicate (minimum of three/per spike level) HRGC/MS analysis to assist the interpretation of positive or negative bias in the accuracy of QC sample data. The accuracy bounds should be extended to 50-130% from 50-115% as specified in the draft quality assurance program plan. The method should include additional internal quantisation standards to pair with the HpCDF 13 OCDF congeners. Also, an additional internal recovand ery standard, possibly C12-l,2,3,4,7,8-HxCDD, is required to provide better estimates of absolute method recovery. These additional compounds, if available, will be incorporated into the method before initiating sample analyses. �IV. EXPERIMENTAL A. Preparation of Homogeni zed Jissue A bulk lipid sample was prepared from the extracts of human adipose tissue samples collected through the NHATS program. The adipose tissue samples have been stored in a deep freezer at approximately -10°C since collection. The homogenized tissue extract or bulk lipid was used in this method evaluation study for preparation of replicate samples spiked with varying levels of specific PCDD and PCDF isomers. This homogenized matrix will also be used for preparing control and spiked quality control samples for the actual NHATS sample analysis phase of the program. A total of 2,465 g of adipose tissue was extracted, dried, and concentrated to yield 1,652 g (62% of original weight) of homogenized lipid. Specific procedures for preparing this matrix are described below. The adipose tissue samples were thawed at room temperature for 1 to 2 h. Portions of the samples were added to a blender cup of a Waring® blender and covered with methylene chloride. The volume of methylene chloride was approximately equal to the sample volume (100 to 200 ml). This mixture was blended at high speed for approximately 10 min, and the contents were transferred to a 500-mL Erlenmeyer flask and further blended with a Tekmar® Tissumizer, also at high speed for 10 min. A powder funnel was plugged with a wad of glass wool (si 1 anized, methylene chloride extracted) and filled with ~ 50 g of sodium sulfate (heated overnight to 600°C in a muffle furnace). The sodium sulfate was wetted with methylene chloride prior to elution of the sample extract. The dried effluents were refiltered in the same way using a fresh bed of sodium sulfate to remove particulate and residual water. The samples were transferred to 1-L round bottom flasks, and the solvent was removed by rotary evaporation. The water bath on the rotary evaporator was kept at 60°C using a thermostatted heating element. Once the solvent appeared to have been removed (constant volume in flask, no visible condensation in condenser), the heating and evaporation process was continued for at least 2 h. The flask and contents were removed and stored in a refrigerator. The extracted lipid solidified upon refrigeration and was visually checked for homogeneity. No precipitates or phase separation was observed. The lipid residue was allowed to liquify at room temperature and was transferred to a 4-L glass bottle with a Teflon®-lined lid. The lipid residue was brought to room temperature and heated just enough to allow the lipid to achieve an oily state prior to aliquotting portions for the method evaluation studies. B. Analytical Standards Analytical standards including native PCDD and PCDF congeners, stable isotope (carbon-13) labeled standards and radiolabeled (carbon-14) standards were purchased from Cambridge Isotope Laboratories, Woburn, Massachusetts, and Pathfinder Laboratories, St. Louis, Missouri. The 2,3,7,8TCDD was received from the EPA Reference Materials Branch as a solution. The �other native PCDD and PCOF congeners were received as 1-mg neat standards. The stable and radio!abeled isotopes were received as solutions. Table 1 provides a summary of the standards used for this study. Stock solutions of the individual PCDD and PCDF congeners were prepared from the neat standards. The neat materials were weighed using a Cahn 27 electrobalance calibrated versus a 1-mg (Class M) standard. The neat compounds were transferred to glass vials and were dissolved in 2.0 to 3.0 ml of toluene (Burdick and Jackson, distilled in glass). Toluene was added to each standard using volumetric pipettes (Class A). The OCDD required dilution to 10.0 ml using a 50:50 mixture of toluene and anisole. A working solution consisting of the 17 native PCDD and PCDF congeners was prepared at a concentration of 2 ug/mL for the TCDD, TCDF, PeCDD, and PeCDF congeners, 5 ug/ml for the HxCDD, HxCDF, HpCDD, and HpCDF congeners, and 10 |jg/mL for the OCDD and OCDF. The working solution was used to prepare both the lipid matrix spiking solution and the calibration standards. The stable isotope labeled internal standards were obtained as^solutions at 50 (jg/mL concentration with the exception of the 13Ci2-OCDD, which was provided at 10 |jg/mL. Separate working solutions containing mixtures of the carbon-13 labeled PCDDs and PCDFs were prepared for use in the calibration standards and the sample spiking solutions. The carbon-14 radiolabeled PCDDs were used for preliminary method evaluation studies. The specific activity of the 14C-2,3,7,8-TCDD (117.56 mCi/mmole) was high enough to allow recovery studies at spike levels equivalent to 10 pg/g for a 10^-g sample. 1. Calibration Standards Eight concentration calibration standards containing the 17 native and the 9 carbon-13 labeled internal standards were prepared for determining the consistency of response factors for the native PCDDs and PCDFs versus the corresponding carbon-13 congeners. Table 2 presents a summary of the calibration standards prepared for the method calibration study. 2. Spi king Solutions a. Native PCDD and PCDF A solution containing the 2,3,7,8-substituted PCDD and PCDF congeners was prepared for spiking the homogenized lipid materials for the method evaluation study. Table 3 specifies the levels of each of the native PCDD and PCDF congeners present in this solution. �Table 1. Analytical Standards Available for the Method Evaluation Studies Compound Source Native 2,3,7,8-TCDD EPA QA Reference Materials Branch 2,3,7,8-TCDF Cambridge Isotope Laboratories 1,2,3,7,8-PeCDD Cambridge Isotope Laboratories 1,2,3,7,8-PeCDF Cambridge Isotope Laboratories 2,3,4,7,8-PeCDF Cambridge Isotope Laboratories 1,2,3,4,7,8-HxCDD Cambridge Isotope Laboratories 1,2,3,6,7,8-HxCDD Cambridge Isotope Laboratories 1,2,3,7,8,9-HxCDD Cambridge Isotope Laboratories 1,2,3,4,7,8-HxCDF Cambridge Isotope Laboratories 1,2,3,6,7,8-HxCDF Cambridge Isotope Laboratories 1,2,3,7,8,9-HxCDF Cambridge Isotope Laboratories 2,3,4,6,7,8-HxCDF Cambridge Isotope Laboratories 1,2,3,4,6,7,8-HpCDD Cambridge Isotope Laboratories 1,2,3, 4,6, 7,8-HpCDF Cambridge Isotope Laboratories 1,2,3,4,7,8,9-HpCDF Cambridge Isotope Laboratories OCDD Cambridge Isotope Laboratories OCDF Cambridge Isotope Laboratories Lot/Code 20603 AWN 1203-74/EF-903C MLB-706-53/ED-950C AWN-729-21/EF-953C AWN-729-45/EF-956C 830244/ED-961C MLB-706-47/ED-960C MLB-706-73/ED-969C AWN-729-20/EF-964C MB 13106-7/EF-962-C MB 13106-47/EF-967-C MB 13106-3/EF-968-C MLB-706-21/ED-971C AWN-729-22/EF-973C MB-13-106-77/EF-975C 8465-F-982-C/EF-982C F2832/ED-980C 13 Ci2-Internal standards 2,3,7,8-TCDD Cambridge 2,3,7,8-TCDF Cambridge 1,2,3,7,8-PeCDD Cambridge 1,2,3,7,8-PeCDF Cambridge 1,2,3,6,7,8-HxCDD Cambridge 1,2,3,4,7,8-HxCDF Cambridge 1,2,3,4,6,7,8-HpCDD Cambridge OCDD Cambridge Isotope Isotope Isotope Isotope Isotope Isotope Isotope Isotope Laboratories Laboratories Laboratories Laboratories Laboratories Laboratories Laboratories Laboratories R00208/ED-900 R00236/EF-904 R00241/ED-955 R00221/EF-952 R00249/ED-966C R00234/EF-963C R00248/ED-972 R00263/ED-981 14 Ci2"Radiolabe1ed standards 2,3,7,8-TCDD Pathfinder Laboratories 1,2,3,4,7,8-HxCDD Pathfinder Laboratories OCDD Pathfinder Laboratories S.A. = specific activity. S.A.a = 117.56 mCi/mmole S.A. = 24.16 mCi/mmole S.A. = 20.50 mCi/mmole �ininminininininmin o ,E o co c o o in o in inininininininininin m inininincMCMCMCMCMCMCM CM CM CM in in CMCMCMCMCMeDeoeoeDeoeovDeoeoeocMCM o o o o in in in in in in in CM CM CM CD in o m o ooooinLnLninininininininocD in mininincMCMCMCMCMCMCMCMCMCMinin o o o o in in in in in in m CM CM CM o in o m oCDOOOOOOOOCDOOOOOO o ooooinLnininininininminoo m S- ,-. ,f co fU CJ L) c C O «0 5-P jjfc u e o o o o CD in in in in m in m CM CM CM o o o o in in in in in un m CM CM CM o in o m -a O O O CD in in in c. in in in m CM CM CM o in o •o in rH rH «o i. CO co o CM CM rH r-4 rH rH r H r H r H r H e M C M C M C M C M C M C M C M C M C M i n i n CM r-4 rH rH CM V) T3 S- • oCDOOOOOOOOCDOOOOOCD CD O O C D O O O O O O O O O O C D O O CM cMCMCMCMinininininininmininoo •* «t CM rH rH CJ CM rH rH rH T*H r~( i~H i~l t~4 i~~f r~f r~f r^l rH Csl CSJ c co 0) CJ (J c o CM rH rH 1-4 -M •i— c Q) o m rH rH .rH o ooocDinininininininininmocD c in o CO r-p- o o in m CM 000 rHt—f r H r H r H r H r H r H r H r H C M C M •r- CJ (0 O o in in m in in in m CM CM CM rH r H r H r H i - H C M C M C M C M C M C M C M C M C M C M i n i n ^_) ~J CJ ^— O >r_ JQ CD in r-4 rH inininininininininin in inininincMCMCMCMCMCMCMCMCMCMinin eo eo *.p s_ 0 CM C M C M C M C M C D C Q C D C O C D e D C D C D l O t O C M C M rH rH co C •£ O CD O o in in m in in m m CM CM TH in m in in CM CM CM ininininininininLnm m inininincMCMCMCMCMCMCMCMCMCMinin •r- CJ in c CM o m O OO o IX. C CD in rH r H r H r H r H C M C M C M C M C M C M C M C M C M C M i n i n eo —' ^x CD in in m rH rH rH oo r-4 CO CJ QJ («- .Q to a u. u. o oa QQQU_U_LL.U_OOCJ Q Q Q Q Q Q C3 O.O.O. QU_LJ_CJCJOC_)CJOOnia:ni r- •o SSS^miini^zoooocn a a) CO rH rH r-4 V) a u_ aQ +3 c. o +3 to 4«> •r- a U_ oQ 0 O CJ H- U X X 00 > -P m Is**. ooi^r^r^^vor^^-^r^vD^j-^'* f*^« CO CO ^ CO CO CO CO CO CO ^ CO CO CO S- Z CM C M r H r H C M r H r H i - H r H r H r H C M r H r H r H O O 0) * I I I en M ev CO r—i ^» i- co oo r^ r*~ eo ^~ •3r**^ r*^ co co co co co CO CO o <u o o ^— » 1 L_] N IN CV I tNO CJ CJ CJ CJ 0 CJ CO C O C M C M C O C M C M C M C M C M C M C O C M C M C M O a . * UJ «v CO CO CM CM CM CM CM o CO rH CO rH CO rH <0 ro -M CJ CJ . o. i. -a c 1 CJ CJ 3: IT 00 U. 0) <J) 1 1 Q 0- 0. 00 00 rO 1 1 - ' h- 00 oo r*- r^- CD I I C. CM I M CJ 31 O M *A Vi rH rH o o o i— i co CM «* rH i. st\ -P t— LH 1—4 1 CJ CO 00 �Table 3. Native PCDD and PCDF Spiking Solution Compound Concentration (pg/nl_) 2,3,7,8-TCDD 5 2,3,7,8-TCDF 5 1,2,3,7,8-PeCDD 1,2,3,7,8-PeCDF 5 5 2,3,4,7,8-PeCDF 1,2,3,4,7,8-HxCDD 5 12.5 1,2,3,6,7,8-HxCDD 12.5 1,2,3,7,8,9-HxCDD 1,2,3,4,7,8-HxCDF 12.5 12.5 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 12.5 12.5 2,3,4,6,7,8-HxCDF 12.5 1,2,3,4,6,7,8-HpCDD .. 12.5 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF 12.5 12.5 OCDD OCDF 25 25 �b. Internal Standards Two different internal standard spiking solutions were prepared for quantisation of native PCDD and PCDF congeners. The compositions of each of the spiking solutions are presented in Table 4. The internal quantisation standards were spiked into the lipid aliquots prior to any cleanup procedures and hence were carried throughout the method exactly as the corresponding native congeners. The internal recovery standard was added in 10 uL of a keeper solution (tridecane) during final extract concentration prior to analysis. The recovery standard was used to measure the absolute method recoveries of the internal quantisation standards. C. Analytical Procedure The homogenized human adipose lipid matrix was allowed to come to room temperature and then warmed in a water bath until the matrix changed to an oily state. Approximately 10.0 g of the oily material was transferred by pipette to preweighed glass vials, and the actual weight of the lipid was determined to the nearest 0.01 g by difference using an analytical balance. Four 10.00-g aliquots were spiked with 20 uL of the native spiking solution presented in Table 3, another four aliquots were spiked with 50 uL of the same solution, and three additional aliquots were spiked with 100 uL of native PCDD and PCDF solution. These spikes were equivalent to concentrations ranging from 10, 25, and 50 pg/g in the lipid matrix for the tetra- and pentachloro PCDD and PCDF congeners up to 50, 125, and 250 pg/g for the OCDD and OCDF for the low, medium, and high level spikes. In addition to the spiked samples, three aliquots of the lipid material were transferred for determining the endogenous levels of each of the PCDD and PCDF congeners in the control matrix. Each of the sample aliquots was fortified with 100 pL of the internal quantisation standard spiking solution (Table 4). The spiked samples were each quantitatively transferred to 500-mL Erlenmeyer flasks using hexane. The residues were diluted with a total of 200 ml of hexane, and 100 g of sulfuric acid (H2S04) modified silica gel (40% w/w) was added to each solution with stirring. The mixtures were stirred for approximately 2 h, and the supernatants were decanted and filtered through filter funnels packed with anhydrous sodium sulfate (Na2S04). The H2S04 modified silica adsorbents were washed with at least two additional aliquots of hexane and dried by elution through Na2S04. The combined hexane extracts for each sample were eluted through columns consisting of the 40% H2S04 modified silica gel (4.0 g) and silica gel (1.0 g). The eluates were concentrated to approximately 15 ml and added to columns of acidic alumina (Bio-Rad, AG-4, 6.0 g). The acidic alumina columns were eluted first with 20 ml of hexane, which was collected but not analyzed, followed by elution with 30 ml of 20% methylene chloride in hexane. The PCDDs and PCDFs were eluted from the acidic alumina using the 20% methylene chloride in hexane. The PCDDs and PCDFs in the eluates were isolated from other chlorinated planar aromatics using columns (5-mL disposable pipettes containing 500 mg of 18% Carbopak C and Celite-545). The Carbopak C/Celite 10 �Table 4. Internal Standard Spiking Solutions Concentration (pg/(jL) Compound Internal quantitation standard8 13 5 13 5 13 5 13 5 C12-2,3,7,8-TCDD C12-2,3,7,8-TCDF C12-l,2,3,7,8-PeCDD C12-l,2,3,7,8-PeCDF 13 12.5 13 12.5 13 12.5 C12-l,2,3,6,7,8-HxCDD C12-l,2,3,4,7,8-HxCDF C12-1,2,3,4,6,7,8-HpCDD 13 C12-OCDD 25 Internal recovery standard 13 C12-1,2,3,4-TCDD 5 uSolution prepared in isooctane. Solution prepared in tridecane. 11 �columns were pre-eluted with 2 ml of toluene, I mi of 75:20:5 methylene chloride/methanol/benzene, 1 mL of 1:1 methylene chloride/cyclohexane, and 2 ml of hexane. The sample extracts (30 ml) were added to the columns, which were eluted with 2 ml of hexane, 1 ml of 1:1 methylene chloride/cyclohexane, and 1 ml of the 75:20:5 methylene chloride/methanol/benzene. These eluents were collected and combined but were not analyzed. The Carbopak C/Celite columns were turned upside down, and the PCDDs and PCDFs were eluted with 20 ml of toluene. The toluene was concentrated to less than 1 ml using flowing nitrogen and a heated water bath (55-60°C) and transferred to 1.0-mL conical vials using a solution of 1% toluene in methylene chloride. Tridecane (10 nD containing 500 pg of the internal recovery standard 13C12-1,2,3,4-TCDD was added as a keeper when the solution had concentrated to approximately 200 pL. The extracts were concentrated to final volume using nitrogen and the heated water bath. D. HRGC/MS Analysis The analyses of the spiked and unspiked lipid samples were completed ,using a Kratos MS50TC double-focusing magnetic sector mass spectrometer. The determination for the tetra- through octachloro homologs was achieved in a single analysis using the conditions described in Table 5. Table 6 provides the characteristic ions monitored for each PCDD and PCDF homolog. As noted from Table 6, the analysis requires five different parameter descriptions that were switched automatically during the course of the analysis. Parameters monitored included two characteristic molecular ions for each PCDD and PCDF homolog and the corresponding carbon-13 labeled internal standard. In addition, a fragment ion of perfluorokerosene (PFK) was monitored throughout each analysis to ensure that proper mass calibration was maintained. The parameter descriptors also included an ion characteristic of specific homologs of chlorinated diphenyl ethers to demonstrate that responses meeting the qualitative criteria for specific PCDF congeners were not due to these potential interferences. Triplicate analyses of six of the eight calibration solutions (Table 2) were completed, and the variability in relative response factors across this range was calculated. The analyst was required to demonstrate on a daily basis that the relative response factors (RRF) were in agreement within ± 20% of the established averages for 2,3,7,8-TCDD and 2,3,7,8-TCDF and within ± 30% of the average RRF values for the other congeners. The analyst was also required to determine column performance by analyzing a mixture of TCDD isomers before proceeding with sample analysis. Table 7 gives an example of the typical daily sequence for PCDD/PCDF analysis. 12 �Table 5. HRGC/LRMS Operating Conditions for PCDD/PCDF Analysis Mass spectrometer Accelerating voltage: Trap current: Electron energy: Electron multiplier voltage: Source temperature: Resolution: Overall SIM cycle time: 8,000 V 500 MA 70 eV -1,800 V 280°C £ 3,000 (10% valley definition) 1s Gas chromatograph Column coating: Film thickness: Column dimensions: He linear velocity: He head pressure: DB-5 0.25 Mm 60 m x 0.25 mm ID * 25 cm/sec 1.75 kg/cm2 (25 psi) Injection type: Split flow: Purge flow: Injector temperature: Interface temperature: Injection size: Initial temperature: Initial time: Temperature program: Splitless, 45 s 30 mL/min 6 mL/min 270°C 300°C 1-2 ML 200°C 2 min 200°C to 330°C at 5°C/min 13 �Table 6. Ions Monitored for HRGC/MS Analysis of PCDD/PCDF Descriptor Al ID Mass TCDF 303.902 305.899 315.942 317.939 319.896 321.894 331.937 333.934 373.840 380.976 0.090 0.090 0.090 0.090 0.090 0.090 0.090 0.090 0.090 0.090 303.902 305.899 319.896 321.894 337.863 337.860 349.903 351.900 353.858 355.855 365.898 367.895 380.976 407.801 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.035 0.035 HxCDF 373.821 375.819 PFK (lock mass) 13 C12-HxCDF 380.976 385.861 387.859 389.816 391.813 401.856 0.080 0.080 0.080 0.080 0.080 13 C12-TCDF TCDD 13 C12-TCDD HxDPE PFK (lock mass) A2 TCDF TCDD PeCDF 13 C12-PeCDF PeCDD 13 C12-PeCDD PFK (lock mass) HpCDPE A3 Nominal dwell time (sec) HxCDD 13 C12-HxCDD 403.853 443.759 OCDPE 14 0.080 0.080 0-. 080 0.080 0.080 �Table 6 (continued) Descriptor A4 ID Mass 380.976 389.816 391.813 407.782 409.779 419.822 421.819 423.777 425.774 435.817 437.814 429.768 431.765 477.720 13 C12-HpCDF HpCDD 13 C12-HpCDD 37 Cl4-HpCDD NCDPE PFK (lock mass) OCDF 13 C12-OCDF OCDD 13 C12-OCDD DCDPE 15 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 380.976 441.743 443.740 453.784 455.781 457.738 459.735 469.779 471.776 511.681 PFK (lock mass) HxCDD HpCDF A5 Nominal dwell time (sec) 0.060 0.070 0.070 0.070 0.070 0.070 0.070 0.070 0.070 0.060 �Table 7. Typical Daily Sequence for PCDD/PCDF Analysis 1. Tune and calibrate mass scale versus perfluorokerosene (PFK). 2. Determine column performance by injecting the TCDD isomer mixture. 3. Inject concentration calibration solution 2.5 to 12.5 pg/uL (CS-7) solution. 4. Inject blank (tridecane). 5. Inject samples 1 through "n". 6 Inject concentration calibration solution 2.5 to 12.5 pg/uL (CS-7) solution. 16 �E. Data Interpretation 1. Qualitative The HRGC/MS elution profiles of the tetra- through octachloro PCDD and PCDF homologs were established through the analysis of environmental sample extract (fly ash from a municipal waste incinerator). The characteristic ions for each homolog were plotted within the retention window established using this mixture. The criteria for identification of a response as a PCDD or PCDF were the coincidental response of the characteristic ions monitored within the established retention window, and within ± 20% of the theoretical ion ratio. Table 8 presents the range of ion ratios used for the qualitative criteria for the specific PCDD and PCDF homologs and internal standards. 2. Quantitation Quantisation of the specific PCDD and PCDF congeners was achieved using the respective internal quantisation standards. For example, TCDD was quantitated versus the 13C12-2,3,7,8-TCDD; PeCDD versus the 13C12-1,2,3,7,8PeCDD, etc. The HpCDF and OCDF responses were quantitated versus the carbon-13 labeled hepta- and octachloro dibenzo-p_-dioxin internal standards since the corresponding dibenzofuran internal standards were not available for this study. The absolute recovery of the internal quantisation standards was achieved using 13C12-1,2,3,4-TCDD. A second internal recovery standard, 37 Cl4-l,2,3,4,6,7,8-HpCDD, was evaluated but was not used for recovery measurements due to interference arising from the corresponding native HpCDD. Relative response factors (RRF) were calculated for each compound listed in Table. 2. The RRF values were calculated as shown in Equation 1. "crn x Cjr STD IS A H Xc IS x ^STD where A Eq. 1 = the sum of the area responses for the two characteristic ions of the standard compound; = the sum of the area responses for the two characteristic ions of the corresponding internal quantisation standard; = concentration of the internal quantisation standard (pg/|jL); and = concentratlon °f tne standard compound (pg/uL). A calibration curve was established using six concentration levels of standards; for example, the calibration curve for 2,3,7,8-TCDD was initially established with standards at concentrations of 1, 2.5, 5, 10, 50, and 100 pg/uL The 2.5 pg/uL standard was analyzed daily to verify response factors and instrument sensitivity. The RRF values for each of the internal quantisation standards were calculated versus the internal recovery standard, 13 C12-1,2,3,4-TCDD, using Equation 1. 17 �Table 8. Ion Ratios for HRGC/MS Analysis of PCDD/PCDF Compound TCDF 13 C12-TCDF TCDD 13 C12-TCDD PCDF 13 C12-PeCDF PeCDD 13 C12-PeCDD HxCDF 13 C12-HxCDF HxCDD 13 C12-HxCDD HpCDF 13 C12-HpCDF HpCDD 13 C12-HpCDD OCDF 13 C12-OCDF OCDD 13 C12-OCDD Ions monitored Theoretical ratio 304/306 316/318 320/322 332/334 338/340 350/352 0.76 0.76 0.76 0.76 0.61 0.61 0.61 0.61 1.22 1.22 1.22 1.22 1.02 1.02 1.02 1.02 0.87 0.87 0.87 0.87 354/356 366/368 374/376 386/388 390/392 402/404 408/410 420/422 424/426 436/438 442/444 454/456 458/460 470/472 18 Acceptabl e range 0.61 0.61 0.61 0.61 0.49 0.49 0.49 0.49 0.96 0.96 0.96 0.96 0.82 0.82 0.82 0.82 0.70 0.70 0.70 0.70 - 0.91 0.91 0.91 0.91 0.73 0.73 0.73 0.73 1.46 1.46 1.46 1.46 1.22. 1.22 1.22 1.22 1.04 1.04 1.04 1.04 �The concentration of a PCDD or PCDF congener in a composite sample was calculated as shown in Equation 2. r - Asample x QIS L WT " A x RRF x Wt where Fn 9 L tq< CWT = wet tissue concentration of the PCDD or PCDF congener in each tissue (pg/g); A sample = sum of the area responses for the two characteristic ions of i pcD[) Qy pCDp congerier; A,o = sum of the area responses for the two characteristic ions of the respective internal quant i tat ion standard; QT<: = amount of the internal quantisation standard added to the i:> sample (500 pg of 13C12-TCDD or 2,500 pg of 13C12-OCDD); RRF = the relative response factor for the PCDD or PCDF congener from Equation 1; and Wt = mass of the sample (grams). The absolute recovery of the internal quantisation standard was calculated using Equation 3. Recovery ( ) = ,-/* % RS IS where K ; 4 , x 100 Eq. 3 IS ARS = sum of the area responses for the two characteristic ions of . the internal recovery standard, 13C12-1,2,3,4-TCDD; QR<. = amount of the internal recovery standard added to the final extract (500 pg); and RRF jg = response factor of the internal quantisation standard relative to the internal recovery standard. All data were qualified to reflect whether the compound was a positive quantifiable parameter, present as a trace value only, or was not detected. Positive quantifiable values were identified for responses greater than 10 times background signal to noise. Trace (TR) values were assigned to responses which were in the range of 2.5 to 10 times background signal to noise. A value of not detected (ND) was used to reflect that a response was not detected at greater than 2.5 times signal to noise. A limit of detection (LOD) was calculated for all trace and not detected values using the peak 19 �height response of the respective internal standard and the average measured signal to noise for the characteristic ions of the PCDDs and PCDFs. F. Quality Assurance/Quality Control (QA/QC) The QA/QC procedures included analysis of multipoint calibration concentration standards to establish relative response factor (RRF) 'curves for each of the 17 native PCDD and PCDF congeners. The mean RRF (RRF) values and instrument sensitivity were verified daily by bracketing the sample analyses with an injection of a standard that ranged from 2.5 pg/uL for 2,3,7,8TCDD and 2,3,7,8-TCDF up to 12.5 pg/uL. for OCDD and OCDF. The criterion for continuing with the sample analysis was agreement of the measured RRF value with the mean RRF within ± 20% for 2,3,7,8-TCDD and TCDF and ± 30% for all other PCDD and PCDF congeners. Other activities included the analysis of laboratory method blanks and reagent blanks and measurement of the absolute recoveries of the internal quantisation standards. G. Preliminary Method Studies Prior to analysis of the homogenized human adipose lipid matrix by HRGC/MS, several experiments were conducted to confirm that the sample preparation scheme was feasible. 1. Gravimetric Studies The first concern was the efficient removal of up to 10 g of lipid matrix from extracted adipose tissue. A series of experiments was conducted with 10-g lipid aliquots to demonstrate removal of lipid using the H2S04-Si02 slurry technique. Initially, 50 g of the acid modified silica was added to the lipid extract in 100 ml of hexane. The acid modified silica was noted to turn dark brown immediately on contact with the lipid solution. The hexane was recovered and the adsorbent was extracted with additional hexane. The extracts were combined and concentrated to 5 ml with Kuderna-Danish evaporators. The extract was eluted through a column of 4.0 g of acid modified silica and 1.0 g of silica with 45 mL of hexane. The acid modified silica was noted to be highly discolored throughout, and the extract required a second slurry treatment of the eluent with an additional 50 g of acid modified silica gel. The adsorbent from the second slurry procedure was noted to discolor significantly, indicating that lipid materials had not been efficiently removed from the first step of the procedure. The hexane supernatant from the second slurry cleanup was reduced in volume and taken to dryness in a preweighed glass vial. The final residue was measured at approximately 10 mg for duplicate samples, which translates into a removal efficiency of 99.9% based on the initial 10-g aliquot. This lipid cleanup procedure was modified such that 100 g of acid modified silica gel was used in the initial slurry cleanup, followed by elution of the resulting extract through a column containing 4.0 g of acid modified silica and 1.0 g of silica gel. The lipid removal efficiency of duplicate samples through the cleanup procedure was determined to average 99.8% 20 �(20 to 30 mg of the initial 10-g lipid remained after cleanup). The column cleanup step in this procedure did not exhibit any significant color change. Thus this step of the procedure was incorporated into the method as a check of the efficiency of lipid removal to prevent overloading of the acidic alumina fractionation column. 2. Carbon-14 Recovery Studies The carbon-14 radiolabeled PCDD standards listed in Table 1 were used to estimate overall method recoveries for the tetra- through octachloro homologs prior to proceeding with the HRGC/MS evaluation. Triplicate analyses (10-g aliquots of lipid materials) were conducted with each of the three radiolabeled standards. The first experiment addressed the recovery of the compounds from bulk lipid cleanup. Triplicate analyses using 10-g aliquots were completed for the three compounds at the following concentrations: 14 C-2,3,7,8-TCDD, 10 pg/g; 14C-l,2,3,4,7,8-HxCDD, 100 pg/g; and 14C-OCDD, 250 pg/g. The results of these analyses indicated that all compounds were recovered in the range of approximately 70 to 80%. Following this ^experiment, the total sample preparation procedure described earlier in this re*port was evaluated using triplicate analysis of 10-g lipid samples. Table 9 provides a summary of the results from this study. These data indicate that overall method recovery is limited by the initial bulk lipid removal procedure. This assumption is based on the similar recoveries of the carbon-14 labeled compounds noted for evaluation of the bulk lipid removal step as compared to the total sample preparation scheme. 21 �Table 9. Summary of the Results of the Sample Preparation Method Evaluation Using Carbon-14 PCDDs Spike levels Bulk lipid removal, recovery Analytes (pg/g) Total method recovery ( ) % 14 C-2,3,7,8-TCDD 10 68 75 14 100 79 66 250 82 76 C-l,2,3,4,7,8-HxCDD 14 C-OCDD * Average value for triplicate analyses taken through the total sample preparation scheme. Precision of measurements varied by less than ± 10% .(relative standard deviation). Average value for triplicate analyses taken through bulk lipid cleanup only. Precision of measurement varied by less than ± 6% (relative standard deviation). 22 �V. RESULTS A. Analytical Results The analytical results for the quantisation of the 17 target PCDD and PCDF 2,3,7,8-substituted congeners in the spiked and unspiked homogenized human adipose lipid samples are presented in Tables 10 to 15. These data demonstrate that 13 of the 17 congeners were definitely detected in the unspiked lipid matrix. Although 2,3,7,8-TCDF is reported as not detected, responses for the characteristic ions (m/z 304 and 306) greater than 10 times signal to noise were noted to be coincident with the internal standard, 13C12~ 2,3,7,8-TCDF. The ratio of the responses (m/z 304/306) in each of the triplicate analyses of the unspiked matrix were well outside the acceptable ratio of 0.90 to 0.61 established in Table 8. Figure 1 provides a comparison of the HRGC/MS-SIM responses noted for the unpsiked human adipose lipid matrix as compared to a concentration calibration standard. Figures 2 through 6 provide examples of the individual PCDD and PCDF responses observed for the unspiked lipid samples as compared to fortified matrices. In general, the precision of the replicate measurements at each spike level is good (relative standard deviations typically less than 10%) for PCDD and PCDF congeners that were detected with responses greater than 10 times signal to noise. The precision for estimated detection limits for 1,2,3,7,8-PeCDF (Table 11), 1,2,3,7,8,9-HxCDF (Table 12), and OCDF (Table 15) ranges from 21.6% to 43.1% as a result of little or no response at the specified retention window. B. Statistical Analysis The regression results for each of the 17 specific PCDD and PCDF congeners are plotted separately in Figures 7 to 23. These plots provide the results for the individual sample analyses, a line defining the results of a least squares regression analysis, and boundaries that depict the confidence limits for the range of spiked concentrations. The regression lines were obtained by the method of least squares using the sample measurements at the three spiking levels and the unspiked level. Two types of upper and lower 95% confidence limits or bounds were calculated for the least square regressions of measured (found) concentrations versus spiked levels. The first set of confidence limits (defined by the inner pair of curves closest to the regression line) is the 95% confidence bounds for the regression line. These bounds are interpreted as follows: The true regression line (as would be determined if the experiment were repeated a countless number of times at the same spiked levels) lies within these confidence limits unless the analytical results are sufficiently unusual to be among those expected to occur less than 5% of the time. The second set of confidence bounds, depicted by the outer pair of lines, constitutes the 95% confidence limits for the result of a single analysis at a particular spiking level. The interpretation is as follows: the result (reported value) of a single analysis of a sample spiked at a given level can be predicted to fall between these 95% confidence bounds unless the analytical result is among those sufficiently unusual to be expected less than 5% of the time. 23 �Table 10. Spiked Versus Measured Concentrations of 2,3,7,8-TCDF and 2,3,7,8-TCDD in Homogenized Human Adipose Lipid Samples 2,3,7,8-TCDF spike level 2,3,7,8-TCDF concentration (pg/g) (pg/g) 13 C12-TCDF absolute recovery ( ) % 13 2,3,7,8-TCDD spike level 2,3,7,8-TCDD concentration (pg/g) (pg/g) C12-TCDD absolute recovery (%) (.) 41a a (.) 41a (.) 40 (.) 41 0.1 1.8 59 63 59 60.3 2.3 3.8 0 0 0 10.7 11.4 13.1 11.7 1.2 10.5 53 52 46 50.3 3.8 7.5 10 10 10 10 14.3 14.8 13.6 13.8 14.1 0.5 3.9 61 67 71 78 69.3 7.1 10.3 10 10 10 10 23.4 22.8 24.7 22.5 23.3 1.0 4.1 53 53 53 62 55.3 4.5 8.1 25 25 25 25 30.8 30.8 30.7 28.7 30.2 1.1 3.5 59 75 62 72 67.0 7.7 11.5 25 25 25 25 40.8 40.6 40.3 38.3 40.0 1.2 2.9 48 53 51 60 53.0 5.1 9.6 50 50 50 57.7 59.4 55.8 57.6 1.8 3.1 64 48 58 56.7 8.1 14.3 50 50 50 65.8 72.1 67.6 68.5 3.3 4.8 55 43 48 48.7 6.0 12.4 0 0 0 Mean STD RSD (%) ro Mean STD RSD (%) ND ND ND ND -pi Mean STD RSD ( ) % Mean STD RSD ( ) % ND = not detected. Value in parentheses is the estimated limit of detection. A response of greater than 10 times signal-to-noise was noted for both characteristic ions (m/z 304 and 306) at the appropriate retention time for 2,3,7,8-TCDF. However, the ion ratio was considerably greater than the acceptable range of 0.61 to 0.90. �Table 11. Spiked Versus Measured Concentrations of 1,2,3,7,8-PeCDF, 2,3,4,7,8-PeCDF, and 1,2,3,7,8-PeCDD In Homogenized Human Adipose Tissue Samples 1.2,3,7,8-PeCDF spike level (pg/g) 0 0 0 10 10 10 10 Mean STD RSO ( ) % 2,3,4,7,8-PeCDF concentration (pg/g) (pg/g) 0 0 0 20.8 21.6 19.0 75 76 80 20.5 1.3 6.5 77.0 2.6 3.4 12.2 11.5 11.9 11.4 10 10 10 10 Mean STO RSO (X) 29.2 30.1 28.5 25.5 Mean STD RSD ( ) % ND = not detected. 51.3 55.9 55.5 54.2 2.5 4.7 . 90 63 84 75 50 50 50 54.9 48.5 41.7 48.2 62 81 87 84 64 74 70 69.8 6.1 8.8 69.3 5.0 7.3 The value in parentheses is the estimated limit of detection. 1,2,3,7,8-PeCDD concentration »3C,2-PeCDO absolute recovery ( ) % (pg/g) (pg/g) 0 0 0 20.2 19.9 18.1 51 54 57 19.4 1.1 5.7 54.0 3.0 5.6 32.1 37.6 30.8 30.4 60 57 55 62 30.2 1.9 6.3 58.5 3.1 5.3 48.0 46.1 49.3 43.5 55 60 57 64 46.7 2.5 5.4 59.0 3.9 6.6 69.7 72.2 72.8 58 54 56 71.6 1.6 2.3 56.0 2.0 3.6 10 10 10 10 78.5 11.3 14.4 74.6 62.9 71.9 1.2,3,7,8-PeCDD spike level 78.0 11.7 15.1 48.4 5.4 11.1 25 25 25 25 28.3 2.0 7.0 50 50 50 27.6 37.0 28.7 36.3 C12-PeCOF absolute recovery ( ) % 32.4 4.9 15.2 11.8 0.4 3.1 25 25 25 25 CJ1 NO ( . ) 11" NO ( . ) 08 NO (0.8) 13 2,3,4,7,8-PeCDF spike level ND ( . ) 09 0.2 21.6 Mean STD RSO ( ) % ro 1,2,3,7,8-PeCDF concentration (P9/g) 25 25 25 25 50 50 50 �Table 12. Spiked Versus Measured Concentrations of 1,2,3,4,7,8-; 1,2,3,6,7,8-; 2,3,4,6,7,8-; and 1,2,3,7,8,9-HxCDF in Homogenized Human Adipose Lipld Matrix 1,2,3,4,7,8HxCOF spike level (pg/g) 0 0 0 Mean STO 25 25 25 25 Mean STO RSD ( ) % 22.0 22.1 22.4 (pg/g) 0 0 0 46.7 52.0 48.8 49.0 Mean STD RSD (X) 125 125 125 90.2 89.1 90.3 90.7 25 25 25 25 156.3 0.9 0.6 (pg/g) 12.3 12.4 12.7 2.3,4.6,7,8HxCDF spike level (pg/g) 0 0 0 36.6 38.6 40.8 39.8 62.5 62.5 62.5 62.5 83.7 80.5 79.6 76.0 25 25 25 25 151.0 157.7 149.1 (pg/g) 4.9 4.2 4.2 1,2.3,7,8,9HxCOF spike level (pg/g) 0 0 0 32.3 33.6 32.0 32.5 62.5 62.5 62.5 62.5 74.7 75.5 74.2 71.1 25 25 25 25 152.6 4.5 2.9 HO = not detected. The value in parentheses reflects the estimated limit of detection. 141.9 141.6 151.0 144.9 5.3 3.7 (pg/g) a 13 Ct2-HxCDF absolute recovery () « NO ( . ) 05 NO ( . ) 07 NO (0.9) 55 57 52 55.7 2.5 4.6 34.8 28.5 30.9 29.2 30.9 2.8 • 9.1 59 57 60 67 60.8 4.3 7.2 125 125 125 82.7 89.4 82.2 76.3 60 63 63 70 82.7 5.4 6.5 62.5 62.5 62.5 62.5 73.9 2.0 2.6 125 125 125 1,2,3,7.8,9HxCOF concentration NO ( . ) 07 0.2 25.3 32.6 0.7 2.1 79.9 3.2 3.9 125 125 125 2,3,4,6,7,8HxCDF concentration 4.4 0.4 8.9 39.0 1.8 4.7 90.1 0.7 0.7 157.2 155.4 156.4 1,2,3,6,7,8HxCDF concentration 12.4 0.2 1.7 49.1 2.2 4.4 62.5 62.5 62.5 62.5 Mean STD RSD (X) (pg/g) 1,2,3,6,7.8HxCDF spike level 22.2 0.2 1.1 RSD (X) ro 1,2,3.4,7,8HxCDF concentration 64.0 4.2 6.6 143.0 144.1 163.4 57 54 57 150.1 11.4 7.6 56.0 1.7 3.1 �in Homogenized Human Adipose Lipid Samples 1,2,3,4,7,8-HxCDD spike level (pg/9) 0 0 0 25 25 25 25 (pg/g) 0 0 0 4. 78 52.0 53.7 52.7 62.5 62.5 62.5 62.5 Mean STO RSD (X) 8. 29 96.7 90.3 81.9 25 25 25 25 141.1 150.8 146.0 145.9 4.9 3.3 (pg/g) 157.0 162.0 154.0 1,2,3,7,8,9-HxCDD spike level (pg/g) 0 0 0 184.0 165.0 198.0 193.0 62.5 62.5 62.5 62.5 220.0 239.0 220.0 220.0 280 8. 299.0 266.0 284.3 16.8 5.9 (pg/g) 13 Ct2-HxCDD absolute recovery (X) 19.1 26.0 24.7 58 60 58 58.7 1.2 2.0 63.5 46.1 40.8 57.3 65 61 65 70 51.9 10.4 20.0 25 25 25 25 65.3 3.7 5.6 125 125 125 114.0 99.9 101.2 107.4 64 66 66 77 105.6 6.5 6.1 62.5 62.5 62.5 62.5 224.8 9.5 4.2 125 125 125 1,2,3,7,8,9-HxCDD concentration 23.2 3.7 15.8 185.0 1' 5 4. 7.9 8. 79 7.0 7.9 125 125 125 1.2,3,6,7,8-HxCDD concentration 157.7 4.0 2.6 51.6 2.6 5.1 Mean STD RSD (X) Mean STD RSD (X) 21.6 22.7 20.3 1,2,3,6,7,8-HxCDD spike level 21.5 1.2 5.5 Mean STD RSD (X) ro 1.2,3,4.7,8-HxCDD concentration (pg/9) 68.3 5.9 8.7 141.3 152.7 189.4 62 57 61 161.1 .25.2 15.6 60.0 2.6 4.4 �Table 14. Spiked Versus Measured Concentrations of 1,2,3,4,6,7,8-HpCDF, 1,2,3,4,7.8,9-HpCOF, and 1,2,3,4,6,7,8-HpCOO in Homogenized Hunan Adipose Lipid Samples 1 ?,3,4,6,7,8-HpCDF spike level (99 P/) 0 0 0 Mean STD RSD ( ) % (pg/g) 30.6 27.3 28.6 Mean STD RSO (X) (pg/g) 0 0 0 48.6 48.6 51.0 56.3 Mean STD RSD ( ) % 86.3 85.7 80.9 83.4 25 25 25 25 154.5 157.3 154.4 155.4 1.6 1.0 (pg/g) NO ( . ) 13* ND ( . )a 11* HD ( . ) 10 1,2,3,4,6,7,8-HpCDD spike level 26 23 23 25 62.5 62.5 62.5 62.5 60 60 59 58 119 126 121 122.4 3.7 3.0 ND = not detected. The value in parentheses is the estimated limit of detection. 13 C,z-HpCDD absolute recovery ( ) % (pa/g) 0 0 0 214.7 210.8 215.5 71 74 69 213.7 2.5 1.2 71.3 2.5 3.5 214.7 239.0 243.9 248.5 83 75 76 78 243.3 40 . 1.7 78.0 3.6 4.6 281.5 288.1 269.9 274.9 77 84 90 99 278.6 8.0 2.9 87.5 9.3 10.7 353.2 355.4 343.7 62 61 69 350.8 6.2 1.8 64 4.4 6.8 25 25 25 25 62.5 62.5 62.5 62.5 59.0 0.8 1.3 125 125 125 1,2,3,4.6,7,8-HpCDD concentration (pg/g) 24.1 1.4 5.7 84.1 2.5 2.9 125 125 125 1,2,3,4,7,8,9-HpCDF concentration ND ( . ) lla 0.2 13 51.1 3.6 7.0 62.5 62.5 62.5 62.5 a 1,2,3,4,7,8,9-HpCDF spike level 28.9 1.6 5.7 25 25 25 25 ro CO 1,2,3,4,6,7,8-HpCDF concentration 125 125 125 �Table 15. Spiked Versus Measured Concentrations of OCDF and OCDD in Homogenized Human Adipose Lipid Samples OCDF spike level (pg/g) 0 0 0 Mean STD RSD ( ) % 50 50 50 50 Mean STD RSD ( ) % 125 125 125 125 Mean STD RSD ( ) % 250 250 250 Mean STD RSD ( ) % OCDF concentration (pg/g) 4.9 2.3 2.6 (pg/g) (pg/g) 0 0 0 804 833 781 91 88 94 806.1 26.1 3.2 91.0 3.0 3.3 3.2 1.4 43.1 44.2 45.0 45.9 49.6 860.4 11.4 1.3 125 125 125 125 932 934 944 907 110.7 2.2 2.0 227.8 231.0 228.6 OCDD concentration 849 856 876 860 50 50 50 50 46.2 2.4 5.2 111.1 107.8 110.7 113.1 13 C12-OCDD abolute recovery () % OCDD spike level 929.1 15.7 1.7 250 250 250 100 87 91 91 92.3 5.5 6.0 90 96 102 104 98.0 6.3 6.5 1,080 1,140 1,070 1,096 34.7 3.2 229.1 1.7 0.7 29 69 67 74 70.0 3.6 5.2 �Umplked Hutmn Adipo» Tillue 1. 2. 3. . . . . . . t . 2.3.7.8-ICOf '3C|J-2.3.7.»-FCnf "CI2-I.J.3.4-TCDO 7.3.7.J-FCOD "CI2-2.3.7.MCDD 1,2.], 7,1.F.CDF "ClJ-1.2 3.7,1 -P.CDF 2.3,4. 7.». FrCDF l,2,3.7,t-NCOD '3C|}>l.2.3,7.a-r«CDD 1 . '3Cn-l.2.3.4,7.«-H.Crif Df 1 . 1.2.3.6.7 RIC 4. 2.J.4.i.7.e-M.CDr S. l,2.3.4,7.e-ll<CDD «. 1,2,3. 6,7. »-H»COD 7. '3Cl2-l.2.3.6.7.t-H.CDD I. l.!.3.7.B.»-HrCDl> '. 1,2.3,7,1,9-HiCDF 20. 1, 2,3.4. *.7.»-HpCDF 21. l,2,3,4.«.7.«-HpCDI> 'K. I3CI2-1. 2.3.4. 4.7,1-HpCOD 23. 37ci4.|.2,3,4(«,7,«.HpCDO ,24 " 25. OCOD 26. "CI2-OCOO 27. OCDF i 21.:2,23 1 14.17 1 2 7 4,5 \{ li 9, 10 I i r-.fl 880 IBM IS 12 18 19 1488 1288 20 1 2888 1688 SCAN Calibration Standard 11,12 1,2 P.IC 15,14,17 6,7 19 21,22,23 9.10 1888 1288 1488 1888 1988 2888 SCAN Figure 1. Comparison of the HRGC/MS-SIM reconstructed ion chromatogram(RIC) from the analysis of unspiked homogenized human adipose tissue matrix and a calibration standard for PCDDs and PCDFs. 30 �Uraplk.d Human Adipou 384. Spiked Human AdipoM |2,3.7.8-TCDF 384 350 1888 last 1188 1158 1200 1290 SCAN Umpilnd Human Adipax r '3C|2-lCDF /-2.3,7,8-ICDD AJL lee.eSpikwi Human Ad!po» 320 . I3C 12-2.3,7.8-ICDF-i - A A me y-2.3,7,8-TCDO A UN lisa i2w 1230 sew Figure 2. Example of the TCDF (m/z 304) and TCDD (m/z 320) HRGC/MS-SIM elution profiles in unspiked and spiked human adipose. 31 �Unpllwd Hunan AdlpoM 2.3,4,7.8-PeCDF 1.2,3.7,8-f.CDF \ Spilwd Hunan Adlpou 2.3.4.7,8-P.CDF 1.2.3,7,8-PeCDF 1 I'"'1 ' ' '"I ' 1859 nee USe 1288 12S8 1308 1358 1488 SCAN 1458 SCAN Uiapikld Hunan Adlpoi. 354 . .2,3.7,8-P«COO Spiked Human AdipoM l3C|2-l,2.3.7.8-PeCOF 334. I.2.3.7.8-P.COD USB 1288 USB 1388 1310 1468 Figure 3. Example of the PeCDF (m/z 338) and PeCDD (m/z 354) HRGC/MS-SIM elution profiles in unspiked and spiked human adipose. 32 �Unipikod Human Adipose |l,2.3,.4.7,8-HxCDF ll.2.3.6,7,8-HxCDF 374 . .8-H»CDF 1 ' i I '' Spik.d Human Adipou 1.2.3,4,7,8-HxCDF | 11.2.3.6.7.8-HxCDF [2.3,4,4.7.8-. \H»CDF 374 . 1.2.3,7,8,9-HxCDF 1343 1258 1398 1983 1358 H93 1480 J498 USB I \. ISM 1.556 1550 1609 SCAN Uraplked Human Adipose 1.2,3,6,7,8-HxCDD 1,2,3.4,7.8-HxCDD- 339 . 13C]2-l,2,3.4.7,8-HxCDF A A 1.2.3,7,8,9-H.CDD Spik.d Human Adipou 1,2.3.6,7.8-HxCDD 399 . 1.2.3,4.7,8-HxCDD 1.2.3.7,8.9-HxCDD l3 C|2-l.2,3,4,7,8-HxCDF|l 1 1358 1488 1438 1589 1559 1689 1658 SCAN Figure 4. Example of the HxCDF (m/z 374) and HxCDD (m/z 390) HRGC/MS-SIM elution profiles in unspiked and spiked human adipose. 33 �Urapilwd Human AdlpaM 11,2,3.4.6,7.8-HpCDF 498. Spllud Human Adipon 1.2,3,4.6.7,8-HpCDF 1,2,3,4,7,8,9-HpCDF 1458 15B8 1556 1669 1658 1766 1758 1809 SCAN Umplk.d Human Adlpou l,2.3.4.6,7.B-HpCDD 424 . Spilwd Human Adipou 1.2,3,4.6,7,8-HpCDD 424. •I 1586 1559 1686 ' ' "> '! '•" 1658 I 1706 • ' ' I 1730 "I"' 1806 I1 1838 SCAN Figure 5. Example of the HpCDF (m/z 408) and HpCDD (m/z 424) HRGC/MS-SIM elution profiles in unspiked and spiked human adipose. 34 �Urapilw) Hunan Adtpou 442. Spiked Human Mlpow OCDF 442 . '3C|2-OCDD| A) me (798 teee test ts«e 1950 2eee SCAN isee ISM zeea sow Umplk.d Hunan Adlpon lOCDD 438 . Spilwd Hunan Adipoj. lOCDD 4S8 . 1766 ITM teee IBM Figure 6. Examples of the OCDF (m/z 442) and OCDD (m/z 458) HRGC/MS-SIM elution profiles in unspiked and spiked human adipose. 35 �80 70 - Regression Line95% Confidence Limits for Regression Line 60 - d 0 50 - d CO en <D 0 d 0 u 30 - •a • 95 % Confide nee Li mi rs for Individual Anafyses d 10 0 0 10 I 30 30 I 40 Spiked Concentration (pg/g) lipid sample meas. Figure 7. Measured concentrations versus concentrations of 2,3,7,8-TCDD spiked into the homogenized human adipose lipid matrix. 50 �80 Regression Line 70 95% Confidence Limits for Regression Line 60 - d 0 50 d 0 0 CO & 0 o S3 d 3 o 30 '95% Confidence for individual Analyses 10 - "T ~ 0 10 20 D 30 i 40 Spiked Concentration (pg/'g) lipid sample meas. Figure 8. Measured concentrations versus concentrations of 1,2,3,7,8-PeCDD spiked into the homogenized human adipose lipid matrix. 50 �160 150 Regression Line' 140 130 95% Confidence Limits for Regression Line W a Vx' 120 110 100 90 80 - co CO G i3 0 o •a 70 60 50 •95% Confidence Limits for Individual Analyses 40 30 30 -t 10 - 0 20 40 n 60 80 100 Spiked Concentration (pg/g) lipid sample meas. Figure 9. Measured concentrations versus concentrations of 1,2,3,4,7,8-HxCDD spiked into the homogenized human adipose lipid matrix. 120 �Regression Line \ 95% Confidence Limits for Regression Line v.,/ d 0 0 co d o o •a d ^ 95% Confidence Limits for Individual Analyses 0 60 0 n 80 100 Spiked Concentration (pg/g) lipid sample meas. Figure 10. Measured concentrations versus concentrations of 1,2,3,6,7,8-HxCDD spiked into the homogenized human adipose lipid matrix. 120 �200 ~r a d o <sJ I, 4> 0 d 0 u •d d 3 o 190 180 170 160 150 140 H 130 130 110 H 100 90 H 80 70 60 50 40 30 ~ 20 4 10 0 -10 0 Regression Line 95% Confidence Li mil's • for Regression Line 95% Confidence Limits for Individual Analyses '"I" 20 40 D Figure 11. "T 60 80 100 Spiked Concentration (pg/'g) lipid sample meas. Measured concentrations versus concentrations of 1,2,3,7,8,9-HxCDD spiked into the homogenized human adipose lipid matrix. 120 �370 Regression Line /•""N W <0! a 95 % Confidence Limits for Regression Line v-> d 0 •H +» flj d a) o d o o -d d 95% Confidence Limits for Individual Analyses 200 0 100 20 n Spiked Concentration (pg/g) lipid sample meas. Figure 12. Measured concentrations versus concentrations of 1,2,3,4,6,7,8-HpCDD spiked into the homogenized human adipose lipid matrix. 120 �1.19 Regression Line- 1.1 95% Confidence Limitsfor Regression Line 1.05 - s~ 1- d 0.95 - 0 a rf & :; o o ro o O •0 •95% Confidence Limits for individual Analyses 0.75 0.7 - 0 r- j r_ 40 80 n j j 120 __ r j 160 300 Spiked Concentration (pg/g) lipid sample meas. Figure 13. Measured concentrations versus concentrations of OCDD spiked into the homogenized human adipose lipid matrix. 240 �60 Wi \ fcfi Regression Line- 50 - 95% Confidence^ Limits for Regression Line a 0 •H •P 30 0 co ti o o 20 - •0 •95% Confide nee Li mits for Individual Analyses d 10 0 0 i 10 I 20 30 Spiked Concentration (pg/g) lipid sample meas. Figure 14. Measured concentrations versus concentrations of 2,3,7,8-TCDF spiked into the homogenized human adipose lipid matrix. 50 �60 Regression Line- 50 U 95% Confidence Limitsfor Regression Line 40 - d o 30 - o d o o •d • 95 % Confidence Limits for Individual Analyses 10 - -10 0 I 10 T 20 30 i 40 Spiked Concentration (pg/g) D lipid sample meas. Figure 15. Measured concentrations versus concentrations of 1,2,3,7,8-PeCDF spiked into the homogenized human adipose lipid matrix. 50 �90 Regression Line- 80 - W) ft d 0 95%JTonfidence Limitsfor Regression Line 70 - 60 - 50 <D 0 d 0 o 40 - •a 30 95% Confidence Limits for Individual Analyses 20 10 I 0 10 I 20 i 30 i 40 Spiked Concentration (pg/g) D lipid sample meas. Figure 16. Measured concentrations versus concentrations of 2,3,4,7,8-PeCDF spiked into the homogenized human adipose lipid matrix. 50 �160 ~r 150 r~\ \ 140 130 - A 110 - •p 100 - t CTl 95% Confidence Limitsfor Regression Line 130 - d o •pa Regression Line- 90 - d 0 0 d 0 80 70 - u 60 - d 50 - 0 •95% Confidence Limits for Individual Analyses 40 30 20J 10 - "T 0 Figure 17. 20 40 60 80 100 Spiked Concentration (pgXg) n lipid sample meas. Measured concentrations versus concentrations of 1,2,3,4,7,8-HxCDF spiked into the homogenized human adipose lipid matrix. 120 �__ 160 150 140 Regression Line- 130 0. d 0 rf 130 95% Confidence Limitsfor Regression Line 110 100 90 80 - 0 ti 0 u Tl d 3 0 70 60 50 95 % Confidence Limits for Individual Analyses 40 30 20 ^ 10 0 0 i 40 i 20 o Figure 18. i 60 80 100 Spiked Concentration (pg/g) lipid sample meas. Measured concentrations versus concentrations of 1,2,3,6,7,8-HxCDF spiked into the homogenized human adipose lipid matrix. 130 �160 150 W 0. vx {3 0 >r4 +J at 140 130 120 Regression Line95% Confidence Limitsfor Regression Line 110 100 90 80 70 0 CO d 60 o 50 0 95% Confidence Limits for Individual Analyses 30 0 20 10 0 10 0 20 40 60 80 100 Spiked Concentration (pg/g) n lipid sample meas. Figure 19. Measured concentrations versus concentrations of 2,3,4,6,7,8-HxCDF spiked into the homogenized human adipose lipid matrix. 120 �bi' P, s*> d o IO d fi) 0 d 0 u •d 170 160 150 140 130 120 110 100 90 Regression Line- 95% Confidence LimiHfor Regression Line 80 70 60 50 40 30 20 10 0 -10 -20 •95% Confidence Limits for Individual Analyses 0 20 40 n Figure 20. 60 80 100 Spiked Concentration (pg/g) lipid sample ineas. Measured concentrations versus concentrations of 1,2,3,7,8,9-HxCDF spiked into the homogenized human adipose lipid matrix. 120 �\ 95 for Regression Line ti 0 0 0 01 o d 0 95% Confidence Limits for Individual Analyses u •a d -10 80 0 D 100 Spiked Concentration (pg/g) lipid sample meas. Figure 21. Measured concentrations versus concentrations of 1,2,3,4,7,8,9-HpCDF spiked into the homogenized human adipose lipid matrix. 120 �170 160 150 140 W P. d 0 Regression Line- 130 95% Confidence Limitsfor Regression Line 130 110 100 - d 0) 0 d 0 u •d d 3 0 90 80 70 - 60 - _ _ _ for Individual Analyses 50 40 - fcl 30 20 H 10 0 20 40 D Figure 22. 60 1 80 100 Spiked Concentration (pg/g) lipid sample meas. Measured concentrations versus concentrations of 1,2,3,4,6,7,8-HpCDF spiked into the homogenized human adipose lipid matrix. 120 �240 220 200 Regression Line 180 H 95% Confidence Limits for Regression Line 160 d o 140 120 H d 0 100 - 0 Ol IN3 d 0 u -a d 95% Confidence Limits fpr Individual Analyses 80 60 -20 0 —, 80 n Figure 23. 120 160 200 Spiked Concentration (pg/g) lipid sample meas. Measured concentrations versus concentrations of OCDF spiked into the homogenized human adipose lipid matrix. 240 �The slopes of the calculated regression lines from the data points in each of the 14 analyses can be used as an indication of the accuracy of the analytical method for the 17 target analytes. Figure 24 is a plot of the slope of regression lines versus the 17 individual compounds. Table 16 provides a key to specific compounds associated with a number on the x-axis of this plot. The plot presents the estimated slope from each least squares regression line as well as the upper and lower 95% confidence limits for the slope. The slope of the regression line can be interpreted as a measure of accuracy with a value of 1.00 equivalent to 100% agreement of the measured concentration with the theoretical values (background plus spike level). The plot of the 95% confidence limits presents some confirmation on the precision of measurements across the four spike levels. These confidence bounds can also be used to determine whether the accuracy of the measurements (slope of regression line) is significantly different from 100% (or 1.00). If the vertical line connecting the lower and upper 95% confidence limits intersects with the horizontal line at 1, then the accuracy of the method (as determined from the regression line) is not significantly different from 100% (slope = 1.00). The results plotted in Figure 24 demonstrate that the method accuracies for 7 of the 17 analytes are not significantly different from 100%. On the other hand, if the upper and lower confidence limits are both greater than or both less than 1.00, then the accuracy of the method is significantly different from 100%. The data presented graphically in Figure 24 indicate that some positive bias (greater than 100%) is associated with the method accuracies for 9 of the 17 analytes while the measurements for a single analyte (OCDF) result in a slightly negative (less than 100%) bias. Table 16 provides a key to the compound identification in Figure 24 and tabulates the slope of the regression lines and the upper and lower 95% confidence limits for each of the 2,3,7,8-substituted PCDD and PCDF analytes. As noted from Table 16, method accuracy (as defined by the regression line slope) ranges from 90% for OCDF up to 121% for 1,2,3,7,8,9-HxCDF. The accuracies for all other measurements fall within a range of 97 to 115% with the exception of the 1,2,3,7,8,9-HxCDF (compound no. 9). The overall method accuracies meet the initial accuracy objective of 50-115% identified in the project quality assurance program plan. However, the predicted accuracy results for individual analysis as defined by the 95% upper confidence limits indicate that this range should be adjusted to 50-130%. The bias in the accuracy of the measurements may be a result of slight differences in the concentration calibration standards and the internal quantisation standard and native PCDD and PCDF spiking solutions. As a preliminary check on these differences, solutions of the low level and of the high level native spike combined with the internal quantisation standards were analyzed. The results of these analyses are provided in Table 17. Accuracy was calculated as measured/spiked x 100. The results of these analyses suggest that bias observed in overall method accuracy is attributed to the differences in the spiking solutions versus the calibration standards. For instance, the four HxCDF isomers demonstrated a consistent positive bias to method accuracy based on the least squares regression analysis. The analysis of the spiking solutions, submitted as samples, also indicates a definite positive bias for the same four HxCDF isomers. 53 �ACCURACY ESTIMATES 1.4 1.3 <D o 1.2 H V) Oi £ II 1.1 t) + n o 0) J! i 09 . OJ T—r _j j j r—r j.. -T r #01 #Q2#03#04#05#06#Q7#08#09#10#11#12#13#14#15#16# J 7 a slope estSmote Compound Number -tlower 95% CL o upp«r 35% GL Figure 24. Method accuracy estimates as determined from the slopes of the least squares regression lines for the 17 target PCDD and PCDF analytes. Refer to Table 16 for the key to compound number. 54 �Table 16. Regression Line Slopes with 95% Confidence Limits Compound no. Compound 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 2,3,7,8-TCDF 2,3,7,8-TCDD 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,7,8-PeCDD 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,8-HxCDD 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF 1,2,3,4,6,7,8-HpCDD OCDF OCDD Significantly Lower 95% Upper 95% confidence confidence different Slope from 1.00? limit limit 1.08 1.13 1.07 0.98 1.04 1.07 1.12 1.12 1.21 0.98 1.01 1.12 1.01 0.97 1.08 0.90 1.15 yes yes yes no no yes yes yes yes no no no no no yes yes yes 55 1.04 1.08 1.02 0.82 0.98 1.06 1.09 1.09 1.12 0.92 0.86 0.94 0.95 0.95 1.01 0.88 1.00 1.11 1.19 1.12 1.13 1,11 1.09 1.16 1.15 1.29 1.05 1.16 1.30 1.07 1.00 1.16 0.92 1.30 �Table 17. Results of the Analysis of the Low and High Level Native Spike Solutions Low level spike Compound Spi ke concentration (pg/uL) Measured concentration (pg/uL) Accuracy () % High level spike Spike Measured concentration concentration (pg/ML) (pg/ML) 50 Accuracy () % 2,3,7,8-TCDF 2,3,7,8-TCDD 13 12 130 120 50 54 57 108 114 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,7,8-PeCDD en 10 10 10 10 10 11 10 12 110 100 120 50 50 50 52 49 53 104 98 106 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 25 25 25 25 25 25 25 27 32 35 40 27 24 39 108 128 140 160 108 96 156 125 125 125 125 125 125 125 134 137 152 185 123 132 148 107 110 122 148 98 106 118 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF 1,2,3,4,6,7,8-HpCDD 25 25 25 22 25 26 88 100 104 125 125 125 116 121 130 93 97 104 OCDF OCDD 50 50 44 48 88 96 250 250 247 250 99 100 �Similar trends are noted for other compounds in Table 17 compared to the data presented in Figure 24 and Table 16. The limited number of analyses of the spiking solutions does not provide an adequate comparison with the sample data to confirm the bias. However, it is recommended that at least triplicate measurements of the spiking solutions at each fortification level should be analyzed at the outset of the actual NHATS sample analysis program. This will be necessary to account for any biases that will be observed from the determination of PCDD and PCDF residue levels in spiked QC samples. It should be noted that additional homogenized spiked samples will be prepared prior to initiation of the NHATS sample analyses. 1. Recovery of Internal Quantitation Standards The absolute recoveries for the carbon-13 labeled internal quantitation standards were determined for each sample by comparing the responses to the internal quantisation standard, 13C12-1,2,3,4-TCDD. The 13 average recoveries of the compounds in Tables 10 to 15 range from 52.1% for C12-2,3,7,8TCDD up to 88.9% for 13C12-OCDD. The results for the absolute recoveries compared to the overall method accuracy for each compound indicate the importance of the internal standard quantisation technique for analysis of the PCDDs and PCDFs in human adipose. 2. Estimation of Background Levels of PCDDs and PCDFs The estimated background levels of the various PCDD and PCDF congeners were determined as the intercept obtained from the least squares linear regression analyses. Table 18 summarizes the estimated background levels along with their upper and lower 95% confidence limits. These background levels and confidence limits can be viewed as the intersections of the regression line and its upper and lower 95% confidence bounds, respectively, with the y-axis (measured or found concentration). These values will be used as the initial data points for developing control charts of the unspiked lipid matrix which will be analyzed with each batch of samples throughout the EPA/VA study. 3. Day-to-Day HRGC/MS Analysis Precision In addition to the analysis of the replicate spiked samples, four extracts were analyzed by HRGC/MS on two different dates. The results of the duplicate HRGC/MS analyses of these four samples for the 17 target compounds are presented in Tables 19 to 21. Concentration values from the second analysis date were included in the statistical analysis of data presented earlier in this section. 57 �Table 18. Background Level Estimates with 95% Confidence Limits Compound no. 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 Compound 2,3,7,8-TCDF 2,3,7,8-TCDD 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,7,8-PeCDD 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,6,7,8HpCDF 1,2,3,4,7,8,9HpCDF 1,2,3,4,6,7,8HpCDD OCDF OCDD Lower 95% Upper 95% Estimated Level background significantly confidence confidence limit level limit different from zero? (pg/g) (pg/g) (pg/g) ND (3.6)a 11.8 ND (1.1) 22.2 19.8 11.2 4.3 ND (2.3) 24.9 159.4 26.5 25.7 yes yes no yes yes yes yes yes no yes yes yes yes 2.7 10.4 -0.2 17.9 18.2 21.3 8.6 2.1 -3.6 20.5 149.3 14.3 21.9 4.5 13.3 2.5 26.4 21.5 23.5 13.7 6.5 8.2 29.3 169.5 38.7 29.6 ND (0.3) no -2.1 1.5 214.0 yes 208.9 219.1 ND (1.0) 799.7 no yes -1.7 779.4 3.8 820.0 22.4 ND = not detected, The value in parentheses reflects the estimated method detection limit. 58 �Table 19. Day-to-Day Precision of Analysis of Specific Sample Extracts for Tetra- and Pentachloro PCDF and PCDD Spike 2,3,7,8Analysis level TCDF date (pg/g) (pg/g) 4/22/86 4/28/86 0 0 RPD ( ) %b 4/22/86 4/28/86 25 0 0 RPD ( ) % 4/22/86 4/28/86 RPD ( ) % ND (3.1) ND ( . ) 41 28 0 0 ND (3.3) ND ( . ) 40 1,2,3,7,8PeCDD 1,2,3,7,8PeCDF (pg/g) (pg/g) 10 11 ND ( . 4 08) ND (1.12) 24 21 18 20 14 12 23 22 17 20 5 14 16 19 17 18 (pg/g) 4 29 ND (0.78) ND (0.75) 10 11 4 13 10 13 ND (0.81) ND (0.75) (pg/g) 19 RPD ( ) % 4/22/86 4/28/86 ND (3.0)a ND ( . ) 41 2,3,4,7,8PeCDF 2,3,7,8TCDD 10 10 31 8 15 5 12 14 21 23 11 12 26 28 27 19 14 12 13 7 36 a .ND = not detected. Value in parentheses is the estimated limit of detection. Relative percent difference. Calculated as the difference of the two values divided by the mean of the two values times 100%. 59 �Table 20. Day-to-Day Precision of Analysis of Specific Sample Extracts for Hexa- and Heptachloro PCOF and PCDO 4/22/86 42/6 /88 0 0 RPO ( ) Xb 0 0 RPO (%) 4/22/86 4/28/86 0 0 RPD (%) 42/6 /28 4/28/86 RPO (X) 25 25 20 22 16 19 28 31 10 13 20 8 NO ( . 2 03) ND ( . 4 07) 21 23 150 178 18 26 25 27 79 9 17 38 9 21 20 176 171 18 25 27 29 102 5 3 32 8 116 3 28 32 30 35 43 49 171 181 46 63 46 49 23 25 235 241 13 15 15 6 31 5 12 3 12 12 1 22 22 12 12 5 20 23 12 13 9 10 44 47 36 37 6 1 (pg/g) 1,2,3,4, 7,8,9HpCDF 149 170 21 22 (pg/g) 1.2,3, 6.7,8HxCDD 1.2,3,4, 6,7.8HpCDF 10 (pg/g) (pg/g) 4.2 4.9 15 4.0 4.2 5 3.9 4.3 1.2,3, 7,8,9HxCDF 1,2,3, 4,7,8HxCDO 1,2,3, 7,8,9HxCDD 1,2,3, 6,7,8HxCOF 1 Analysis date 4/22/86 4/28/86 2,3,4, 6,7,8HxCOF 1,2,3, 4,7,8HxCDF 6 Spike level (pg/g) NO ( . 3 a 03) ND ( . 1 05) 43 ND ( . 1 04) ND ( . 6 08) (pg/g) (pg/g) (pg/g) (pg/g) rND = not detected. The value in parentheses is the estimated limit of detection. Relative percent difference. Calculated as the difference of the two values divided by the mean of the two values times 100%. (pg/g) ND ( . 3 08) NO ( . 6 03) 79 ND ( . 9 10) ND ( . 4 11) 4 NO ( . 6 10) ND ( . 8 02) 1,2,3,4, 6,7,8HpCDO (pg/g) 210 216 3 207 211 2 223 216 �Table 21. Day-to-Day Precision of Analysis of Specific Sample Extracts for OCDF and OCDD Analysis date 4/22/86 4/28/86 OCDF concentration OCDD concentration (pg/g) (pg/g) (pg/g) 0 0 4.9 3.5 811 810 Spike level RPD ( ) %a 4/22/86 4/28/86 0 0 2.2 2.3 RPD ( ) % 4/22/86 4/28/86 0 0 1.9 2.6 2 788 784 31 1 43 44 834 848 3 50 50 RPD ( ) % a 819 836 4 RPD (%) 4/22/86 4/28/86 0 33 2 Relative percent difference. Calculated as the difference of the two values divided by the mean of the two values times 100%. 61 �VI. QUALITY ASSURANCE/QUALITY CONTROL (QA/QC) As discussed in the experimental section of this report, the QA/QC activities included the analysis of a multipoint calibration curve, daily verification of relative response factors for each analyte, analysis of a method blank and reagent blanks along with the samples, and determining the absolute recoveries of each of the internal quantisation standards for every sample. Each of these QA/QC activities is discussed below. A. Initial Calibration At the outset of sample analysis activity, six calibration concentration standards containing each of the target PCDDs and PCDFs at varying levels and constant concentrations of the internal quantisation and recovery standards were analyzed in triplicate. The relative response factors (RRF) for each native compound and internal quantitation standard were determined for each standard analysis. An average RRF and relative percent standard deviation (RSD) were determined for each concentration level. The average RRF values from each of the six concentration calibration standards were then used to calculate a grand mean RRF value for each compound in the calibration solution. Table 22 presents a summary of the grand mean RRF values for each component in the standards. As noted from Table 22, the average RRF values for native PCDDs and PCDFs generally varied by less than ± 10% (RSD) with the exception of the pentachloro congeners. These results fall well within the criteria established in the draft quality assurance program plan which required the variability of RRF values for the tetrachloro homologs to be within ± 20% (RSD) while the RRF criterion on all other compounds was set at ± 30% (RDS). The variability of the RRF values for the internal quantisation standards, on the other hand, was noted to increase with the degree of chlorination. This is a result of the measurement of all internal quantisation standards versus the single internal 37 recovery standard, 13C12-1,2,3,4-TCDD. A second internal recovery standard, Cl4-l,2,3,4,6,7,8-HpCDD, was evaluated. However, problems resulting from contribution of native HpCDD to the characteristic ions of this internal standard resulted in variabilities in the RRF value up to 50%. Hence, this internal standard was not used for any calculations. It is anticipated that an additional internal recovery standard, such as 13C12-l,2,3,4,7,8-HxCDD, will improve the variability in the RRF values of the higher chlorinated internal quantisation standards. This compound will be incorporated into the method if available. The sensitivity of the Kratos MS-50TC to the tetra- through octachloro PCDDs and PCDFs was demonstrated through the triplicate analysis of the low level standard (CS-8, Table 2) that ranged in concentration from 1 pg/uL for the tetra- and pentachloro congeners up to 5 pg/uL for the octachloro congeners. Table 22 provides an indication of the observed signal-tonoise ratio for each of the native PCDD and PCDF congeners. These data demonstrate that the low level standard is well above the instrument detection limit, which is defined as the amount of a particular compound necessary to give a signal 2.5 times the background signal to noise for each of the characteristic ions while meeting the qualitative criteria for ion ratios. 62 �Table 22. Relative Response Factors (Grand Means) Determined from Multipoint Concentration Calibration Standards Compound RRFa 1.00 0.80 0.98 1.06 1.33 0.94 0.93 0.86 0.86 1.31 1.44 1.61 2.33 1.89 1.19 1.38 1.04 5.7 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,7,8-PeCDD 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,6,7,8-HpCDD 1,2,3,4,7,8,9-HpCDD 1,2,3,4,6, 7, 8-HpCDD OCDF OCDD 13 C12-l,2,3,4-TCDDb'c 13 C -2,3,7,8-TCDF 13 12 1.00 1.98 C -2,3,7,8-TCDD 1.73 13 12 1.36 C12-l,2,3,7,8-PeCDF 13 C12-l,2,3,7,8-PeCDD 0.70 13 C12-l,2,3,4,7,8-HxCDF 1.28 13 C -l,2,3,6,7,8-HxCDD 0.41 13 12 C12-1,2,3,4,6,7,80.33 HpCDD 37 C14-1,2,3,4,6,7,80.12 HpCDD 13 C12-OCDD 0.24 6.2 5.2 10.1 11.3 3.1 2.4 2.7 6.9 4.9 3.0 1.0 4.0 3.6 4.7 3.3 2.5 _ Calibration range (pg/MD 12 6.5 11 8.9 5.7 32 30 29 17 13 14 14 35 26 13 31 21 _ RSD ( ) % 2,3,7,8-TCDF 2,3,7,8-TCDD Signal-to- noise ratio for low. level standard 1-100 1-100 1-100 1-100 1-100 2.5-250 2.5-250 2.5-250 2.5-250 2.5-250 2.5-250 2.5-250 2.5-250 2.5-250 2.5-250 5-500 5-500 - 50 50 50 50 50 125 125 125 51.8 - 125 28.0 ™ 250 7.6 4.7 4.5 7.7 15.8 19.6 25.8 . - JIJRRF = grand mean RRF. characteristic ion for each native PCDD or PCDF congener (Data File 8501D17X02). Internal recovery standard. 63 �B. Daily Verification of Response Factors Before proceeding with analysis of samples, the analyst was required to verify the existing response factor calibration through the analysis of a calibration standard (CS-7, Table 2). Criteria for proceeding with sample analysis required that the measured RRF value for 2,3,7,8-TCDD and 2,3,7,8-TCDF were within ± 20% (and all other congeners within ± 30%) of the mean RRF established from the calibration curve. This standard was also analyzed at the end of each working day to demonstrate that the calibration had been maintained. All RRF values were tabulated to generate RRF control charts for each specific PCDD and PCDF congener. Figures 25 through 34 are plots (control charts) of the RRF values established for the 17 individual target analytes. The RRF data are plotted versus time of analysis. These plots contain 28 individual data points, 18 of which were generated for triplicate analysis of 6 concentration calibration solutions from initial calibration and 10 analyses of solution CS-7 (Table 2) injected over the 5 days for which actual samples were analyzed. The upper and lower boundaries (dashed lines) represent a relative standard deviation of approximately ± 10% with the exception of the plot for 1,2,3,7,8-PeCDD, for which the boundaries are plotted as ± 20%. It should be noted that the actual control limits as specified in the project QAPP were set at ± 20% for 2,3,7,8-TCDD and 2,3,7,8-TCDF and ± 30% for all other target analytes. The average RRF values and corresponding standard deviations reported in each of these plots are calculated from the total 28 standard analyses. C. Blanks As specified in the quality assurance program plan, a laboratory method blank was prepared along with the 14 human adipose lipid samples. The method blank was taken through all procedures as if it were an actual sample, although no lipid matrix was introduced. The analysis of the method blank resulted in the data reported for each of the target analytes reported in Table 23. As noted in Table 23, 1,2,3,4,6,7,8-HpCDD and OCDD were detected at concentrations equivalent to 4.0 and 30 pg/g (equivalent to a 10-g lipid sample), respectively. In addition to these compounds, responses that correspond to the elution of two TCDD isomers (1,3,6,8- and 1,3,7,9-) and a PeCDD (isomer not determined) were detected in the method blank. Further analysis of individual reagents used for preparation of the samples identified the activated acidic alumina as the source of the artifacts. Acidic alumina that had been cleaned by Soxhlet extraction but not activated at 190°C was analyzed, and the artifacts were not detected. This indicates that the artifacts are generated during activation of acidic alumina at elevated temperatures (190°C). Similar background problems from the same PCDD 18 congeners have recently been reported by the Center for Disease Control.17' 64 �(flRBVREF.flKeW/CflMT./REF.flHT.) 1.208 9.337) ¥ i ST.DEV.9.865 Z ST.OEU.« S.5S1 X 1.189 Jl • . y T if * i • DATE " n ii ii y 1 x 3.390 t i X" i TI •If if 'j v _ i «! X _ S/ 4/23/85 4/13/SS 9.897) ST.OPJ.8.035 2 ST.OEU.5.932 X 1 9.909 . 1 v • ? , t | <! J 1 tl 1 Jl ^X 11 11 n^ 1 1 .** \ J ^ V 1 1J II H 1 X X 9.789 DftTE 5/3/86 Figure 25. Control charts showing response factors by date for 2,3,7,8-TCDF and 2,3,7,8-TCDD. Dotted lines represent approximately ± 10% of the mean. 65 �F<C13-l/2(3.7/ MKEft/KEF.ARBWAANT./REF.flHT.) 1.268 1.188 _____ 9.971) T [ x f r _ j^_ L. Tr 1 ( ST.OEU.' 9.835 7. ? ST.OEU.8.739 _ _ \(. -T n • n /£ n v( ii w * * | 1 '*lC 9.966 _ w 1 * \L! 7W ^ y X* -f Tl x r 1 w A *i ~~Tx X 8.866 x 9.709 DATE 4/13/86 4/23/86 v'fiKEft/REF.flfi£ft)/<fW.--REF.ft«T.) (AUt 1.469 1.353) ST.DEU.» 9.113 r. ST.OEU. 18.691 X 1.366 1 1 1 1 1.290 i i " a n S V MM 1.668 i^ x 8.369 *r ^ " 1111 >1 -uA x ' w i I X /\ ^f x X 8.789 DATE 3/ 3/86 4/13/86 4/23^5 5/ 3/86 Figure 26. Control charts showing response factors by date for 1,2,3,7,8-PeCDF and 2,3,4,7,8-PeCDF. Dotted lines represent approximately ± 10% of the mean. 66 �.AHT.) (flUi 1.375) 3.000 ST.OEU.0.233 Z ST.DEU.' 18.8135 ¥ i i 2.889 i i i x X DATE 4/13/86 4/23/86 5/ 3/85 Figure 27. Control chart showing response factors by date for 1,2,3,7,8-PeCDD. Dotted lines represent approximately ± 20% of the mean. 67 �<f*£ft/R£F.f*BV/««T./REF.ftl1T.) (flVi 9.329) 1.188 ST.DEU.» 9.046 2 ST.OEU.» 4.899 it ii 1.999 ft II ii II 4» y» 11 ' (ft jj AL A 0.309 v v 1 A^ y T v v A1 i t ' '' ! ^v A 1 X v ^ -^ 9.390 4/ 13/86 MTE 4/23/86 (flREft/REF.fWEft>/<ftWT./ReF.fl«T.) «Wi 1.168 -| S/ 3/85 8.923) ST.OEU.0.S48 ?. ST.OEU.' 4.348 • 1.808 ' x 1 ¥ * k J ! v 1. , K v X 1. f 0.388 If y ! j! X X x x 9.399 DATE 4/13/86 4/23/86 V 3/86 Figure 28. Control charts showing response factors by date for lj2,3,4,7,8-HxCDF and 1,2,3,6,7,8-HxCDF. Dotted lines represent approximately ± 10% of the mean. 68 �((WEfl/R£F.ftR£rt)/(AnT./REF.ArtT.) (flUi 0.862) l.VW ST.DEU." 0.038 7. ST. OP.'.4.429 i i i 0.300 ? 1 % $ Jf i* ii !* ' 9.899 x i 0.760 VI 3/86 DOTE 4/23/86 5/3/86 « 3 M U <ftR£fl/REF.AREA>/«OT./REF.ftMT.) <(Wl 1.299 8.883) ST.OE«.« 9.982 Z ST.OEV.- x 9 327 - 1.199 ¥ 1.999 ¥ * 0.308 ! H ! ' ' r w ' JJ « e.309 X X 1 i X ¥ ' ! x x 1 x X a. ran 4/23/86 3/ 3/86 Figure 29. Control charts showing response factors by date for 2,3,4,6,7,8-HxCDF and 1,2,3,7,8,9-HxCDF. Dotted lines represent approximately ± 10% of the mean. 69 �PiCl3-l,2;376,7,^flBSiCHLOR(M)IOXIH <ARBVR£F.flRe»/<flNT./REF.««r.) ((Wt 1.589 1.394) ST.OEU.« 9.977 Z ST.DEV.3.310 1.4TO x T 1.399 1 i ' X w A 1.299 X 1.199 DATE 4/13/8S V 3/BS 4/23/85 ./REF.AHT.) <W| 1.433) 1.796 ST.DEU.» 9.936 Z ST.OEV.' 6.015 1.680 -Ti i HI ni in HI in in 1.569 ,. x 1.409 x « i X i 1.399 ,_*.- x 1.299 DATE 4/13/86 4/23/86 S/ 3/86 Figure 30. Control charts showing response factors by date for 1,2,3,4,7,8-HxCDD and 1,2,3,6,7,8-HxCDD. Dotted lines represent approximately ± 10% of the mean. 70 �1.637) 2.399 s h 2.299 e ST.DEU.0.135 Z ? ST.OEU.9.384 2.199 < ) 2.999 1.899 X 1 1 1.799 1 1 V& sx ^ i 1.S99 X i v "T i X 1.48B WTE V13/8S 4/23/8S Figure 31. Control chart showing response factors by date for 1,2,3,7,8,9-HxCDD. Dotted lines represent approximately ± 10% of the mean. 71 �mmm <ftREft/REF.AR£ft>/< AHT./R£F.fl«T.) (flUl 2.660 2.228) « * J i 2.500 ST.DEV." 0.182 2 ST.OEU.8.163 . k * xi 2.460 H a 2.309 *x • * » i * 2.260 ! x x ;; A i •! x X 2.160 " W A x 1.309 X x OrtTE 4/13/85 4/23/86 S/ 3/86 S:-fa§i^I;3ji^aiBWKirsra««.(fKEA/f£F.AREA>/( WT./REF.filfT.) (flU« 2.160 1.783) X i ,1 $ 2.669 ii ST.OEU." 0.169 Z ST.OE1^.9.426 i 1.360 1! 1.3W !i !j ! X' 1.760 x 1.660 j ii M *J ! • X x 1.500 f x fb «•«••» X ;: X I.46A DATE V13^S 4/23/SS Figure 32. Control charts showing response factors by date for 1,2,3,4,6,7,8-HpCDF and 1,2,3,4,7,8,9-HpCDF. Dotted lines represent approximately ± 10% of the mean. 72 �«*Eft/R£F.ftRB»/<AMT./REF.<*T.) (AUi 1.460 1.175) ST.OEU.8.071 2 ST.OEU.' S.043 ¥ 1.369 _ *• . . _ . » .. — — j *» _ M * W M _ t a «...-,~._VyJ__-.« iI • •H* K* M x Y PS A • ' '!• A « if I * X 1.106 -M • , ' ' X i x ii ' lv 'i y t • * « « . • '1 i! ' X| i x .'!-•««,«,. • . X e.see DATE 4/13/86 4/23/8S 5/ 3/8S Figure 33. Control chart showing response factors by date for 1,2,3,4,6,7,8-HpCDD. Dotted lines represent approximately + 10% of the mean. 73 �0IOXIN <«R»/R£F.f*EA>/<«1T./P.EF.fl«T.> (AUi 1.238) 1.500 X ' 5fi it V i i i i i 1.480 w III ST.DEV.9.117 * ST.DEW.» 8.378 Jt I III t «. - l«Zvv M • « M M - i 'I { X ~ i : X x 4 ^.j.i_ _ ___________ X 4/23/86 4/ 3/86 DATE w AJy X 1.189 5/3/86 ffiS»K»?«M 1.033) (AREft/KEF.rtRBWAAHT.v'REF.AMT.) <AUi 1.283 -i ST.OEU.3.843 Z ST.OEU.. 4.181 1.183 i i X X X m m J\ 1.309 , i * ^ 1 x ' X Xx —1-»— • ^ X X X i i i X 8.300 4/23/86 5/ 3/86 Figure 34. Control charts showing response factors by date for OCDF and OCDD. Dotted lines represent approximately ± 10% of the mean. 74 �Table 23. Summary of Results from the Analysis of a Laboratory Method Blank Concentration3 (pg/g) Compound 2,3,7,8-TCDD 2,3,7,8-TCDF 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,7,8-PeCDD 1,2,3,4,7,8-HxCOF 1,2,3,6,7,8-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD NO ND ND ND ND ND ND ND ND ND ND ND ND ND (0.50) (2.2) (0.5) (0.5) (1.2) (0.5) (0.5) (0.5) (0.5) (1.0) (0.9) (0.8) (0.5) (0.5) 4.0 ND (0.5) 30 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF 1,2,3,4,6,7,8-HpCDD OCDF OCDD 75 �An experiment was designed to evaluate a procedure for cleaning the activated acidic alumina immediately prior to the fractionation of the sample extract. The acidic alumina (6.0 g) was packed in hexane. The packed column was eluted with 40 ml of methylene chloride/hexane (1:1) solution followed by 80 to 100 ml of hexane. The sample extract was added to the column and was eluted with 20 ml of hexane followed by 30 ml of 20% methylene chloride in hexane which was reserved for PCDD and PCDF analysis. The carbon-14 radiolabeled 2,3,7,8-TCDD, 1,2,3,6,7,8-HxCDD, and OCDD were used to evaluate recovery of PCDDs from the cleaned alumina. Recoveries of the radiolabeled PCDDs from the activated acidic alumina precleaned by the procedure described above are detailed in Table 24. These data demonstrate that the selected PCDDs are quantitatively (greater than 90%) recovered from the precleaned acidic alumina. This procedure for cleanup of activated acidic alumina has been integrated into the analytical protocol (Appendix A) for routine application with sample preparation activities. D. Absolute Recoveries of the Internal Quantisation Standards The absolute recoveries of the carbon-13 labeled internal quantitation standards 13were determined by comparing responses with the internal recovery standard, C12-1,2,3,4-TCDD, which was added during final concentration prior to HRGC/MS analysis. A summary of the average and range of recoveries of the 8 internal quantisation standards from the 14 human adipose lipid samples is provided in Table 25. These data indicate that recoveries ranged from an average of 52.1% for 13C12-2,3,7,8-TCDD up to 88.9% for 13C12-OCDD. The average recoveries for the lower chlorinated internal standards were lower than the preliminary method studies with carbon-14 radiolabeled standards had indicated. This resulted in a closer evaluation of the final concentration step prior to mass spectrometry. The first extracts for the human adipose lipid extracts were concentrated with a nitrogen evaporation system equipped with a water bath at approximately 55°C. Final blowdown of the samples required addition of the internal recovery standard in 10 uL of tridecane as a keeper solution. However, it was noted at the elevated temperature final volumes from nitrogen evaporation were generally on the order of 2 to 5 uL. This required addition of another 10 uL of tridecane prior to HRGC/MS analysis. In an effort to assess the effect of reducing the final volume of tridecane at elevated temperatures on absolute recoveries of the internal quantisation standards, an experiment using the radiolabeled TCDD, HxCDD, and OCDD standards was conducted. Four solutions of the same spike level were prepared with each radiolabeled compound in 1 ml of toluene. Two of the spiked solutions were heated at 55-60°C and the solvent was reduced under a gentle stream of prepurified nitrogen. The toluene solution was concentrated to 100 uL, 500 uL of 1% toluene in methylene chloride was added, and the solution was concentrated to 200 uL. At this time 10 pL of the keeper tridecane was added and the solution was allowed to concentrate further. The remaining two solutions for each radiolabeled compound were taken through a similar solvent exchange 76 �Table 24. Recovery of Radiolabeled PCDDs from Precleaned Activated Alumina Spike level (pg) Compound Recovery 14 100 300 300 92 96 97 14 1,000 3,000 3,000 103 101 100 14 2,500 7,500 7,500 102 99 97 C-2,3,7,8-TCDD C-l,2,3,4,7,8-HxCDD C-OCDD 77 �Table 25. Absolute Recoveries of the Internal Quantitation Standards from the Human Adipose Lipid Matrix Internal quantisation standard Average recovery ( ) % Standard deviation Relative standard deviation (%) 13 64.4 7.9 12.3 46-78 13 52.1 5.0 9.6 43-62 13 76.1 8.9 11.7 62-90 13 57.1 3.5 6.1 54-64 13 59.4 5.0 8.4 54-70 13 63.6 5.3 8.4 57-77 76.3 10.3 13.6 61-99 88.9 11.4 12.9 67-104 C12-2,3,7,8-TCDF C12-2,3,7,8-TCDD C12-l,2,3,7,8-PeCDF C12-l,2,3,7,8-PeCDD C12-1,2,3,4,7,8HxCDF C12-1,2,3,6,7,8HxCDD 13 c12-i,2,3,4,6,7,8- Range of recovery (%) HpCDD 13 C12-OCDD Values based on 14 analyses of human adipose lipid samples. 78 �and concentration procedure except the solution was allowed to concentrate at room temperature. One of the most obvious results was the observation that solutions held at elevated temperatures could be reduced to dryness even when tridecane had been added as a keeper. On the other hand, solutions for which tridecane had been added but remained at room temperature could only be concentrated to a 10-uL final volume. The recoveries of the radiolabeled standards from each of the solutions in this study are presented in Table 26. The results from this study indicate that the final concentration condition may have a pronounced effect on the absolute recoveries of the PCDDs and PCDFs, especially for the lower chlorinated congeners such as 2,3,7,8-TCDD. However, it should be noted that the approach to target analyte quantisation based on the internal standard method (isotope dilution for 8 of the 17 target analytes) is not affected by absolute recoveries as low as 50%. The procedure for final concentration in the analytical protocol (Appendix A) for the analysis of the NHATS samples for the EPA/VA study has been modified to specify room temperature conditions. 79 �Table 26. Recovery of Carbon-14 Labeled 2,3,7,8-TCDD, 1,2,3,4,7,8-HxCDD, and OCDD as a Function of Final Concentration Conditions Spike level Compound (pg) Concentration conditions Observed final volume Observed recovery ( ) % 14 300 300 300 300 55-60°C 55-60°C 20°C 20°C 1-2 uL dry ness 10 |jL 10 ML 78 54 98 93 14 3,000 3,000 3,000 3,000 55-60°C 55-60°C 20°C 20°C 1-2 ML 94 102 105 7,500 7,500 7,500 7,500 55-60°C 55-60°C 20°C 20°C 1-2 2-3 10 10 C-2,3,7,8-TCDD C-l,2,3,4,7,8-HxCDD 14 C-OCDD 5 ML 10 ML 10 ML ML ML ML ML 107 94 94 100 97 Each solution was concentrated under a gentle stream of flowing nitrogen. 80 �VII. REFERENCES 1. Stanley JS. 1984. Methods of analysis of polychlorinated dibenzo-pdioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs) in biological matrices—literature review and recommendations. EPA-560/584-00. 2. Stanley JS, Going JE, Redford DP, Kutz KW, Young AL. 1985. A survey of analytical methods for measurement of polychlorinated dibenzo-p_-dioxins (PCDD) and polychlorinated dibenzofurans (PCDF) in human adipose tissues. In: Chlorinated dioxins and dibenzofurans in the total environment II. Keith LH, Rappe C, Choudhary G, eds. Butterworth Publishers, pp. 181-195. 3. Stanley JS. 1984 (March 28). Proposed analytical method for analysis of PCDDs/PCDFs in human adipose tissue: special report. Washington, DC: Office of Pesticides and Toxic Substances, U.S. Environmental Protection Agency. Contract 68-02-3938, Work Assignment 8. 4. Stanley JS. 1986 (April 23). Broad scan analysis of human adipose tissue: polychlorinated dibenzo-jD-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs). Draft final report. Washington, DC: Office of Pesticides and Toxic Substances, U.S. Environmental Protection Agency. Contract 68-023938, Work Assignment 8. 5. Albro et al. 1985. Methods for the quantitative determination of multiple, specific polychlorinated dibenzo-p_-dioxin and dibenzofuran isomers in human adipose tissue in the parts-per-trillion range. An interlaboratory study. Anal Chem 57: 2717-2725. 6. Patterson DG et al. 1986. High-resolution gas chromatographic/highresolution mass spectrometric analysis of human adipose tissue for 2,3,7,8-tetrachlorodibenzo-p_-dioxin. Anal Chem 58: 705-713. 7. Schecter A, Tiernan TO, Taylor ML, VanNess GF, Barrett JH, Wagel DJ, Gitlitz G, Bogdasarian M. 1985. Biological markers after exposure to polychlorinated dibenzo-p_-dioxins, dibenzofurans, biphenyls, and biphenylenes. Part I: Findings using fat biopsies to estimate exposure. In: Chlorinated dioxins and dibenzofurans in the total environment II. Keith LH, Rappe C, Choudhary G, eds. Butterworth Publishers, pp. 215-246. 8. Schecter A, Ryan JJ. 1985. Dioxin and furan levels in human adipose tissue from exposed and control populations. 189th National ACS Meeting Symposium on Chlorinated Dioxins and Dibenzofurans in the Total Environment III, Miami, Florida. Preprint Division of Environmental Chemistry, ACS 25:160-163, Paper No. 56. 9. Ryan JJ, Williams DT, Lau BPY, Sakuma T. 1985. Analysis of human fat tissue for 2,3,7,8-tetrachlorodibenzo-jD-dioxin and chlorinated dibenzofuran residues. In: Chlorinated dioxins and dibenzofurans in the total environment II. Keith LH, Rappe C, Choudhary G, eds. Butterworth Publishers, pp. 205-214. 81 �10. Ryan JJ, Schecter A, Lizotte R, Sun W-F, Miller L. 1985. Tissue distribution of dioxins and furans in humans from the general population. Chemosphere 14: 929-932. 11. Nygren M, Hansson M, Rappe C, Domellof L, Hardell L. 1985. Analysis of polychlorinated dibenzo-p_-dioxins and dibenzofurans in adipose tissue from soft-tissue sarcoma patients and controls. 189th National ACS Meeting Symposium on Chlorinated Dioxins and Dibenzofurans in the Total Environment III, Miami, Florida, 1985. Preprint Division of Environmental Chemistry, ACS 25:160-163, Paper No. 55. 12. Smith LM, Stalling DL, Johnson JJ. 1984. Determination of part per trillion levels of polychlorinated dibenzofurans and dioxins in environmental samples. Anal Chem 58: 1830-1842. 13. Rappe C, Nygren M, Linstrom G, Hanson H. 1985. Dioxins and dibenzofurans in human tissues and milk of European origin. 5th International Symposium on Chlorinated Dioxins and Related Compounds, Bayreuth, FRG, September 16-19, 1985. 14. Ryan JJ. 1985. Variation of dioxins and furans in humans with age and organ by country. 5th International Symposium on Chlorinated Dioxins and Related Compounds, Bayreuth, FRG, September 16-19, 1985. 15. Graham M, Hileman FD, Wendling J, Wilson JD. 1985. Chlorocarbons in adipose tissue samples. 5th International Symposium on Chlorinated Dioxins and Related Compounds, Bayreuth, FRG, September 16-19, 1985. 16. Patterson DG, Holler JS, Smith SJ, Liddle JA, Sampson EJ, Needham LL 1985. Human tissue data in certain U.S. populations. 5th International Symposium on Chlorinated Dioxins and Related Compounds, Bayreuth, FRG, September 16-19, 1985. 17. Patterson DG, Holler JS, Groce DF, Alexander LR, Lapeza CR, 0'Conner RC, Liddle JA. 1986. Control of interferences in the analysis of human adipose tissue to 2,3,7,8-tetrachlorodibenzo-p_-dioxin (TCDD). Environ Toxicol Chem 5: 355-360. 18. Holler JS, Patterson DG, Alexander LR, Groce OF, O'Connor RC, Lapeza CR. 1985. Control of artifacts and contamination in the development of a dioxin analytical program. 33rd Annual Conference on Mass Spectrometry and Allied Topics, San Diego, CA, May 26-31, 1985. 82 �APPENDIX A ANALYTICAL PROTOCOL FOR DETERMINATION OF PCDDs AND PCDFs IN HUMAN ADIPOSE TISSUE A-l �TABLE OF CONTENTS Section Description Page 1 Scope and Application A-3 2 Summary of Method A-3 3 Definitions 4 Interferences A-7 5 Safety A-7 6 Apparatus and Equipment A-8 7 Reagents and Standard Solutions A-ll 8 High Resolution Gas Chromatography/Mass Spectrometry Performance Criteria. A-13 Quality Control Procedures A-30 10 Sample Preservation and Handling A-32 11 Sample Extraction A-33 12 Cleanup Procedures A-35 13 Analytical Procedures A-37 14 Date Reduction A-42 15 Reporting and Documentation A-47 9 . A-2 A-6 �ANALYTICAL PROTOCOL FOR DETERMINATION OF PCDDs AND PCDFs IN HUMAN ADIPOSE TISSUE 1. SCOPE AND APPLICATION 1.1 1.2 The minimum measurable concentration is estimated to range from 1 pg/g (1 part per trillion) for 2,3,7,8-TCDD and 2,3,7,8-TCDF up to 5 pg/g for OCDD and OCDF. However, these detection limits depend on the kinds and concentrations of interfering compounds in the sample matrix and the absolute method recovery. 1.3 2. This method provides procedures for the detection and quantitative measurement of polychlorinated dibenzo-p_-dioxins (PCDD) and polychlorinated dibenzofurans (PCDF) at concentrations ranging from 1 to 100 pg/g for the tetrachloro congeners up to 5 to 500 pg/g for the octachloro congeners in 10-g aliquots of human adipose tissue. The method will be used to determine PCDDs and PCDFs, particularly congeners with chlorine substitution in the 2,3,7,8 positions. Table 1 lists the specific PCDDs and PCDFs and target method detection limits. SUMMARY OF METHOD Figure 1 presents a schematic of the analtyical procedures for determination of PCDDs and PCDFs in human adipose tissue. The analytical method requires extraction and isolation of lipid materials from human adipose samples. This is accomplished using sample sizes ranging up to 10 g. Extraction and homogenization are accomplished using methylene chloride and a Tekmar Tissuemizer®. The extract is filtered through anhydrous sodium sulfate to remove water. The extraction procedure is repeated (three to five times) until the tissue sample has been thoroughly homogenized. The final extract is adjusted to a known volume (100 ml) and the extractable lipid is determined using a minimum of 1% of the final volume. The methylene chloride in the remaining extract is concentrated until only an oily residue remains. The residue is spiked with known amounts of the carbon-13 labeled PCDDs and PCDFs (e.g., 500 pg of 13C12-TCDD/F to 2,500 pg of 13C12-OCDD/F) as internal quantisation standards. The residue is diluted with hexane ( 200 ml), and 100 g of ^ sulfuric acid modified silica gel (40% w/w) is added to the solution with stirring. The mixture is stirred for approximately 2 h, and the supernatant is decanted and filtered through anhydrous sodium sulfate. The adsorbent is washed with at least two additional aliquots of hexane. The combined hexane extracts are eluted through a column consisting of a layer of sulfuric acid modified silica gel, and a layer of unmodified silica gel. The eluate is concentrated to approximately 1 ml and added to a column of acidic alumina. The PCDDs and PCDFs are eluted from the alumina using 20% methylene chloride/hexane. This eluate is concentrated to approximately 0.5 ml and is added to a 500-mg Carbopak C/Celite column. The PCDDs and PCDFs are eluted from the column using 20 ml of toluene. A-3 �Table 1. Target PCDD and PCDF Congeners and Target Method Detection Limits Compound 2,3,7,8-TCDD 2,3,7,8-TCDF 1,2,3, 7,8- PeCDD 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,4, 7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 1, 2,3,7,8, 9-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,4,6,7,8-HpCDD 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF OCDD OCDF CAS no. Target method detection limit (pg/g)D 1746-01-6 51207-31-9 40321-76-4 57117-41-6 57117-31-4 39227-28-6 57653-85-7 19408-74-3 70648-29-9 57117-44-9 72918-21-9 60851-34-5 35822-46-9 67562-39-4 55673-89-7 3268-87-9 39001-02-0 1.0 1.0 1.0 1.0 1.0 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 5.0 5.0 Chemical Abstract Services number. = parts per trillion. A-4 �Initial Sample Preparation Isolation of Extractable U'pid Materials Homogenization in Methylene Chloride I Llpid Determination Solvent Exchange Add Internal Quantitation Standards (13C-PCDDs/PCDFs) Bulk Lipid Removal Acid Modified Silica Gel Slurry Technique Provides Cleanup of Oxidizable Compounds with Rapid Sample Turnaround, Improved Cleanup Efficiency and Recovery Removal of Chemical InterferencesAcidic Silica/Silica Acidic Alumina Provides Seperation of PCBs and Other Potential Interferences from PCDDs and PCDFs. I Carbopak C/Cel 7 te Selective Adsorption and Isolation of PCDDs/PCDFs Add Internal Recovery Standards HRGC/MS-S1M Analysis I 1 LRMS I de nti fi cat i on/Quant? tati on of Terra-6'cta PCDDs/PCDFs HRMS Confirmation of 2,3,7,8-TCDD Figure 1. Schematic of the sample preparation and instrumental analysis procedures for determination of PCDDs and PCDFs in human adipose tissue. A-5 �The toluene is concentrated to less than 1 mL and transferred to conical vials. Tridecane (10 jjL) containing 500 pg of an internal recovery standard is added as a keeper, and the extract is concentrated to a final volume. The HRGC/MS analysis is completed in the selected ion monitoring mode (SIM). Analysis of the tetra- through octachloro PCDD and PCDF congeners is achieved using low resolution mass spectrometry. Separation of the tetra- through octachloro PGDD and PCDF congeners is achieved using a 60-m DB-5 column. Verification of the 2,3,7,8-TCDD is achieved using a 50-m CP Sil 88 column and HRGC/MS-SIM analysis in the high resolution mode (R = 10,000). 3. DEFINITIONS 3.1 Concentration calibration solutions ~ Solutions containing known amounts of the native analytes (unlabeled 2,3,7,8-substituted PCDDs and PCDFs), the internal quantisation standards13(Carbon-13 labeled PCDDs and PCDFs), and the recovery standard, C121,2,3,4-TCDD. These calibration solutions are used to determine instrument response of the analytes relative to the internal quantisation standards and of the internal quantisation standards relative to the internal recovery standard. 3.2 Internal quantitation standards — Carbon-13 labeled PCDDs and PCDFs, which are added to every sample and are present at the same concentration in every method blank and quality control sample. These are added to the lipid residue extracted from the adipose tissue and are used to measure the concentration of each analyte. The concentration of each internal quantitation standard is measured in every sample, and percent recovery is determined using the internal recovery standard. 3.3 Internal recovery standard -- 13C12-1,2,3,4-TCDD which is added to every sample extract just before the final concentration step and HRGC/MS-SIM analysis. 3.4 Laboratory method blank — This blank is prepared in the laboratory through performing all analytical procedures except addition of a sample aliquot to the extraction vessel. A minimum of one laboratory method blank will be analyzed with each batch of samples. 3.5 HRGC column performance check mixture ~ A mixture containing known amounts of selected TCDD standards; it is used to demonstrate continued acceptable performance of the capillary column, i.e., separation ( 25% valley on a 50-m CP Sil 88 or 60-m SP-2330 ^ HRGC column) of 2,3,7,8-TCDD isomer from all other 21 TCDD isomers, and to define the TCDD retention time window. 3.6 Relative response factor -- Response of the mass spectrometer to a known amount of an analyte relative to a known amount of an internal standard (quantitation or recovery). A-6 �3.7 3.8 4. Mass resolution check — Standard method used to demonstrate static resolution of 10,000 minimum (10% valley definition). Sample batch -- A sample batch consists of up to 10 human adipose tissue samples, one method blank, 2 internal quality control (QC) samples (spiked and unspiked), and an external performance audit sample (blind spike). INTERFERENCES Chemicals which elute from the HRGC column with ± 10 scans of the internal and/or recovery standards and which produce within the retention time window ions at any of the masses used to detect or quantify PCDDs, PCDFs, or the internal quantisation and recovery standards are potential interferences. Most frequently encountered potential interferences are other sample components that are extracted along with the PCDDs and PCDFs, e.g., PCBs, chlorinated methoxybiphenyls, chlorinated hydroxydiphenyl ethers, chlorinated benzylphenyl ethers, chlorinated naphthalenes, DDE, DDT, etc. The actual incidence of interference by these chemicals depends also upon relative concentrations, mass spectrometric resolution, and chromatographic conditions. Because very low levels (pg/g) of PCDDs and PCDFs are anticipated, the elimination of interferences is essential. High purity reagents and solvents must be used and all equipment must be scrupulously cleaned. Laboratory method blanks must be analyzed to demonstrate absence of contamination that would interfere with measurement of the PCDDs and PCDFs. Column chromatographic procedures are used to remove coextracted sample components; these procedures must be performed carefully to minimize loss of PCDDs and PCDFs during attempts to increase their concentration relative to other sample components. 5. SAFETY 5.1 The toxicity or carcinogenicity of each reagent used in this method has not been precisely defined; however, each chemical compound should be treated as a potential health hazard. The 2,3,7,8-TCDD is a known teratogen, mutagen, and carcinogen. Ingestion of microgram quantities can result in toxic effects. The other 2,3,7,8-substituted PCDDs and PCDFs may exhibit teratogenic, mutagenic, and carcinogenic effects. From this viewpoint, exposure to these chemicals must be reduced to the lowest possible level by whatever means available. Only experienced personnel will be allowed to work with these chemicals. 5.2 All laboratory personnel will be required to wear laboratory coats or coveralls, gloves, and safety glasses. The neat standards, stock, and working solutions will be handled only in a Class A fume hood or glove box. When manipulating stock standards or working solutions, the analyst is advised to place the solution vials in a secure holder (sample block or glass beaker) to prevent accidental spills. A-7 �5.3 5.4 If handling of these compounds result's in skin contact, immediately remove all contaminated clothing and wash the affected skin areas with soap and water for at least 15 min. 5.5 6. If these standards are spilled, absorb as much as possible with absorbent paper and place in a container clearly labeled as PCDD or PCDF waste. Solvent-wash all contaminated surfaces with toluene and absorbent paper followed by washing with a strong soap and water solution. Dispose of all contaminated materials in sealed steel containers labeled as contaminated with PCDD and/or PCDF residue and indicate the approximate level of contamination. As a final precaution, prepare a wipe sample of the exposed surface area and include the wipe as part of the sample analysis batch. This will be used to confirm that the work area is free of contamination. Disposal of laboratory wastes — All laboratory wastes (solvents and absorbents) will be disposed of as hazardous wastes. The laboratory personnel should take care to dispose of the sodium sulfate, silica gel, and alumina in separate metal containers. Excess solvents should be disposed of in gallon polyethylene jugs containing a layer of activated charcoal. Excess solvent that is known to be contaminated with PCDDs or PCDFs should be kept at a minimum by evaporating the solvent with a stream of air. APPARATUS AND EQUIPMENT 6.1 High Resolution Gas Chromatograph/Mass Spectrometer/Data System (HRGC/HRMS/DS) 6.1.1 The GC must be equipped for temperature programming, and all required accessories must be available, such as syringes, gases, and a capillary column. The GC injection port must be designed for capillary columns. The use of splitless injection techniques is recommended. When using this method, a l-pL injection volume is used. The injection volumes for all extracts, blanks, calibration solutions, and the performance check sample must be consistent. 6.1.2 High Resolution Gas Chromatograph-Mass Spectrometer Interface The HRGC/MS interface is directly coupled to the mass spectrometer ion source. All components of the interface should be glass or glass-lined stainless steel. The interface components should be compatible with 300°C temperatures. The HRGC/MS interface must be appropriately designed so that the separation of the PCDDs and PCDFs which is achieved in the gas chromatographic column is not appreciably degraded. Cold spots and/or active surfaces (adsorption sites) in the HRGC/MS A-8 �interface can cause peak tailing and peak broadening. It is recommended that the HRGC column be fitted directly into the MS ion source. Graphite ferrules should be avoided in the HRGC injection port since they may absorb PCDDs or PCDFs. Vespel or equivalent ferrules are recommended. 6.1.3 Mass Spectrometer The mass spectrometer must be capable of maintaining a minimum resolution of 10,000 (10% valley) for high resolution confirmation analysis. The mass spectrometer must be operated in a selected ion monitoring (SIM) mode with total cycle time (including voltage reset time) of 1 s or less. 6.1.4 Data System A dedicated hardware or data system is required to control the rapid multiple ion monitoring process and to acquire the data. Quantification data (peak areas or peak heights) and SIM traces (displays of intensities of each m/z (characteristic ion) being monitored as a function of time) must be acquired during the analyses. Quantifications may be reported based upon computergenerated peak areas or upon measured peak heights. 6.2 HRGC Columns For isomer-specific determinations of 2,3,7,8-TCDD, the following fused silica capillary columns are recommended: a 50-m CP-Sil 88 column and a 60-m SP-2330 (SP-2331) column. However, any capillary column which separates 2,3,7,8-TCDD from all other TCDDs may be used for such analyses, provided that the minimum acceptance criteria in Section 8 are met. 6.3 Miscellaneous Equipment 6.3.1 Nitrogen evaporation apparatus with variable flow rate. 6.3.2 Balance capable of accurately weighing to ± 0.01 g. 6.3.3 Water bath — equipped with concentric ring cover and capable of being temperature-controlled. 6.3.4 Stainless steel spatulas or spoons. 6.3.5 Magnetic stirrers and stir bars. 6.3.6 High speed tissue homogenizer -- Tekmar Tissuemizer® equipped with an EN-8 probe or equivalent. 6.3.7 Vacuum dessicator. A-9 �6.4 Glassware 6.4.1 Erlenmeyer flask — 500 ml. 6.4.2 Kuderna-Danish apparatus -- 500-mL evaporating flask, 15-mL graduated concentrator tubes with ground-glass stoppers, and three-ball macro Snyder column (Kontes K-570001-0500, K-503000-0121, and K-569001-0219 or equivalent). 6.4.3 Minivials -- 1-mL borosilicate glass with conical-shaped reservoir and screw caps lined with Teflon®-faced silicone disks. 6.4.4 Powder funnels -- glass. 6.4.5 Chromatographic columns for the silica and alumina chromatography -- 1 cm ID x 10 cm long and 1 cm ID x 30 cm long with 250-mL reservoir and equipped with TFE stopcocks. 6.4.6 Chromatographic column for the Carbopak cleanup — disposable 5-mL graduated glass pi pets, 6 to 7 mm ID. 6.4.7 Glass rods. 6.4.8 Carborundum boiling chips -- Extracted for 6 hr in a Soxhlet apparatus with benzene and air dried. 6.4.9 Glass wool, silanized (Supelco) — Extract with methylene chloride and hexane and air dry before use. 6.4.10 Glassware cleaning procedure -- All glassware used for these analyses will be cleaned via the following procedure. Wash the glassware in soap and water, rinse with copious amounts of tap water, distilled water, and distilled-in-glass acetone, in that order. Immediately prior to use, the glassware should be rinsed with distilled-in-glass quality solvents: methylene chloride, toluene, and hexane. The glassware should be allowed to dry fully. As an added precuation, all glassware will be marked with a unique code that should be noted in the extraction and cleanup procedures for each sample. This glassware tracking will allow background results from specific glassware to be documented. After use, each piece of glassware should be rinsed with the last solvent used in it, followed by a rinse with toluene, then acetone, before transferring it to the glassware washing facility. A-10 �7. REAGENTS AND STANDARD SOLUTIONS 7.1 Column Chromatography Reagents 7.1.1 Alumina, acidic (Biorad, AG-4) -- Extract the alumina in a Soxhlet apparatus with methylene chloride for 18 h (minimum of two cycles per hour). Air dry and activate it by heating in a foil-covered glass container for 24 h at 190°C. 7.1.2 Silica gel -- High purity grade, type 60, 70-230 mesh; extract the silica gel in a Soxhlet apparatus with methylene chloride for 10 h (minimum of 2 cycles per hour). Air dry and activate it by heating in a foilcovered glass container for 24 h at 130°C. 7.1.3 Silica gel impregnated with 40% (by weight) sulfuric acid -- Add two parts (by weight) concentrated sulfuric acid to three parts (by weight) silica gel (extracted and activated) (e.g., 40 g of H2S04 plus 60 g of silica gel) in a glass screw-cap bottle. Tumble for 5 to 6 h, shaking occasionally until free of lumps. 7.1.4 Sulfuric acid, concentrated -- ACS grade, specific gravity 1.84. 7.1.5 Graphitized carbon black (Carbopack C, Supelco), surface of approximately 12 mVg, 80/100 mesh -- Mix thoroughly 3.6 g of Carbopack C and 16.4 g of Celite 545® in a 40-mL vial. Activate at 130°C for 6 h. Store in a desiccator. 7.1.6 Celite 545® (Fischer Scientific), reagent grade, or equivalent. 7.2 Desiccating agents -- Sodium sulfate; granular, anhydrous. Before use extract with methylene chloride for 16 h (minimum of two cycles per hour), air dry and then muffle for ^ 4 h in a shallow tray at 400°C. Let it cool in a desiccator and store in oven at 130°C. 7.3 Solvents -- High purity, distilled in glass: methylene chloride, toluene, benzene, cyclohexane, methanol, acetone, hexane; reagent grade: tridecane. High purity solvents are dispensed from Teflon® squirt bottles. 7.4 Concentration Calibration Solutions (Table 2) Eight tridecane solutions containing native calibration standards, 13 C12-labeled internal quantisation standards, and two internal recovery standards are required. The complete compound list is A-11 �Table 2. Concentration Calibration Solutions Compound Native Concentration in cal ibration solutions i n pg/uL CS1 CS2 CS3 CS4 CSS CS6 CS7 CSS 2,3,7,8-TCDD 2,3,7,8-TCDF 1,2,3,7,8-PeCDD 1,2,3, 7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,4, 7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,4,6,7,8-HpCDD 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF OCDD OCDF Internal Quantisation Standards 13 C12-2,3,7,8-TCDD 13 C12-2,3,7,8-TCDF 13 C12-l,2,3,7,8-PeCDD 13 C12-l,2,3,7,8-PeCDF 13 C12-l,2,3,6,7,8-HxCDD 13 C12-l,2,3,4,7,8-HxCDF 13 C12-l,2,3,4,6,7,8-HpCDD 13 C12-OCDD Internal Recovery Standard 13 C12-1,2,3,4-TCDD 200 200 200 200 200 500 500 500 500 500 500 500 500 500 500 1,000 1,000 100 100 100 100 100 250 250 250 250 250 250 250 250 50 50 50 50 50 125 125 125 125 25 25 25 25 25 62. 5 250 500 500 250 62. 5 62. 5 62. 5 62. 5 62. 5 62. 5 62. 5 62. 5 62. 5 125 250 125 250 125 125 125 125 125 125 10 10 10 10 10 25 25 25 25 25 25 25 25 25 25 50 50 50 50 50 50 125 125 125 250 50 50 50 50 50 50 50 50 50 125 125 125 250 125 125 125 250 50 50 50 125 125 125 250 50 50 50 50 125 125 125 250 50 50 50 50 50 A-12 5 5 5 5 5 12. 5 12. 5 12. 5 12. 5 12. 5 12. 5 12. 5 12. 5 12. 5 12. 5 25 25 50 50 2. 5 2. 5 2. 5 2. 5 2. 5 6. 25 6. 25 6. 25 6. 25 6. 25 6. 25 6. 25 6. 25 6. 25 6. 25 12. 5 12. 5 1 1 1 1 1 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 5 5 50 50 50 50 125 125 250 50 50 50 50 125 125 125 250 50 50 50 50 50 125 125 125 125 250 �given in Table 1. The native 2,3,7,8-TCDD is supplied as a certified standard solution from the U.S. EPA QA Reference Materials Branch. All other native compounds were supplied in crystalline form by Cambridge Isotope Laboratories (Woburn, MA). 13C12~ Labeled internal quantisation standards were supplied in solution in ji-nonane by Cambridge Isotope Laboratories. Portions of the native standards were accurately weighed to the nearest 0.001 mg with a Cahn 27 electrobalance and dissolved in toluene. 7.5 Column Performance Check Mixture The column performance check mixture consists of several TCDD isomers which will be used to document the separation of 2,3,7,8TCDD from all other isomers. This solution will contain TCDDs (A) eluting closely to 2,3,7,8-TCDD, and the first- (F) and lasteluting (L) TCDDs. Analyte Approximate amount per ampule Unlabeled 2,3,7,8-TCDD 13 C12-2,3,7,8-TCDD 1,2,3,4-TCDD (A) 1,4,7,8-TCDD (A) 1,2,3,7-TCDD (A) 1,2,3,8-TCDD (A) 1,3,6,8-TCDD (F) 1,2,8,9-TCDD (L) 7.6 10 10 10 10 10 10 10 10 ng ng ng ng ng ng ng ng Spiking Solutions Three solutions are prepared using the 13 same stock as in Section 7.4. A native standard solution and a C12 internal quantitation standard solution are prepared in isooctane (Tables 3 and 4). A recovery standard solution is prepared in tridecane (Table 4). Samples are spiked with 100 (jL of internal quantisation standard solution and final sample extracts are spiked with 10 |jL of internal recovery standard solution. 8. HIGH RESOLUTION GAS CHROMATOGRAPHY/MASS SPECTROMETRY PERFORMANCE CRITERIA Samples and standards are analyzed by using a Carlo Erba MFC500 gas chromatography (GC) coupled to a Kratos MS50TC double-focusing mass spectrometer (MS) to be operated in the electron impact mode. The HRGC/MS interface is simply a direct connection of the fused silica HRGC column to the ion source of the MS via a heated interface oven. Data acquisition and processing are controlled by a Finnigan-MAT Incos 2300 data system. A-13 �Table 3. Native Spiking Solution Concentration (pg/uO Compound 2,3,7,8-TCDD 2,3,7,8-TCDF 1,2,3,7,8-PeCDD 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,4,6,7,8-HpCDD 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF OCDD OCDF 5 5 5 5 5 12.5 12.5 12.5 12.5 12.5 12.5 12.5 12.5 12.5 12.5 25 25 A-14 �Table 4. Internal Standard Spiking Solutions Concentration (pg/^L) Compound Internal Quanti tati on Standards C12-2,3,7,8-TCDD 13 5 13 5 ^Cia-l^.SJ.S-PeCDD 5 13 5 Ci2-2,3,7,8-TCDF C12-l,2,3,7,8-PeCDF 13 12.5 13 12.5 13 12.5 C12-l,2,3,6,7,8-HxCDD C12-l,2,3,4,7,8-HxCDF C12-l,2,3,4,6,7,8-HpCDD 13 C12-OCDD 25 Internal Recovery Standard 13 C12-1,2,3,4-TCDD A-15 5 �8.1 HRGC/MS Analysis of PCDD/PCDF Single run selected ion monitoring (SIM) analysis of the tetrachloro through octachloro-dioxins and furans is carried out with the instrumental conditions and parameters outlined in Table 5. For each HRGC/MS run, five distinct groups of ions, which correspond to each chlorine level, are sequentially monitored. These ion descriptors are shown in Table 6. The masses of the two most abundant ions in the molecular ion cluster of each dioxin and furan and isotopically labeled standard are monitored. In addition, the masses corresponding to the molecular ions of the hexachloro through decachlorodiphenyl ethers (PCDEs) are monitored to aid in the confirmation of positive furan results. A lock mass, m/z 381 from PFK (perfluorokerosene), is used to observe and correct any magnet/instrument drift during the analysis. 8.1.1 Tuning and Mass Calibration The mass spectrometer is tuned on a daily basis to yield optimum sensitivity and peak shape using an ion peak (m/z 381) from PFK. The resolution is visually monitored and maintained at £ 3,000 (10% valley definition) to provide adequate noise rejection while maintaining good ion transmission. Mass calibration of the mass spectrometer for the HRGC/MS • analysis of PCDD/PCDF is carried out on a daily basis. The magnetic field is adjusted to pass m/z 300 at full accelerating voltage. PFK is admitted to the MS and an accelerating voltage scan from 8,000 to 4,000 V is acquired by the data system. This corresponds to an effective mass range of 301 to 593 amu. Upon completion of a successful calibration step, the five ion descriptors shown in Table 6 are updated to reflect the new mass calibration. 8.1.2 Ion Descriptor Switching The ion descriptors shown in Table 6 are sequentially monitored during a PCDD/PCDF analysis to cover the retention windows of each chlorination level. The retention windows and hence the descriptor switch points are determined initially and whenever a new HRGC column is installed by injection of a mixture of PCDD and PCDF congeners. Daily adjustment of the descriptor switch times are performed when careful monitoring of the standard retention times shows this to be necessary. The descriptors are designed to overlap to ensure acquisition of all isomers of each homolog. A-16 �Table 5. HRGC/LRMS Operating Conditions for PCDD/PCDF Analysis Mass spectrometer Accelerating voltage: Trap current: Electron energy: Electron multiplier voltage: Source temperature: Resolution: Overall SIM cycle time: 8,000 V 500 |jA 70 eV -1,800 V 280°C £ 3,000 (10% valley definition) 1s Gas chromatograph Column coating: Film thickness: Column dimensions: He linear velocity: He head pressure: DB-5 0.25 pm 60 m x 0.25 mm ID * 25 cm/sec 1.75 kg/cm2 (25 psi) Injection type: Split flow: Purge flow: Injector temperature: Interface temperature: Injection size: Initial temperature: Initial time: Temperature program: Splitless, 45 s 30 mL/min 6 mL/min 270°C 300°C 1-2 (JL 200°C 2 min 200°C to 330°C at 5°C/min A-17 �Table 6. Ions Monitored for HRGC/MS of PCDD/PCDF Descriptor Al ID Mass TCDF 303.902 305.899 315.942 317.939 319.896 321.894 331.937 333.934 373.840 380.976 TCDD 13 C12-TCDD HxDPE PFK (lock mass) TCDF TCDD PeCDF 13 C12-PeCDF PeCDD 13 C12-PeCDD PFK (lock mass) HpCDPE A3 HxCDF PFK (lock mass) 13 C12-HxCDF HxCDD 13 C12-HxCDD OCDPE A-18 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.035 0.035 373.821 375.818 380.976 385.861 387.858 389.816 391.813 401.856 403.853 443.759 C12-TCDF 0.090 0.090 0.090 0.090 0.090 0.090 0.090 0.090 0.090 0.090 303.902 305.899 319.896 321.894 337.863 337.860 349.903 351.900 353.858 355.855 365.898 367.895 380.976 407.801 13 A2 Nominal dwell time (sec) 0.080 0.080 0.080 0.080 0.080 0.080 0.080 0.080 0.080 0.080 �Table 6 (continued) Descriptor A4 ID Mass PFK (lock mass) HxCDD HpCDF 13 C12-HpCDF HpCDD 13 C12-HpCDD 37 Cl4-HpCDD NCDPE A5 PFK (lock mass) OCDF 13 C12-OCDF OCDD 13 C12-OCDD DCDPE A-19 Nominal dwell time (sec) 380.976 389.816 391.813 407.782 409.779 419.822 421.819 423.777 425.774 435.817 437.814 429.768 431.765 477.720 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 380.976 441. 743 443.740 453.783 455.780 457.738 459.735 469.779 471.776 511.681 0.06 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.06 �8.1.3 HRGC Column Performance (60-m DB-5) 8.1.3.1 Inject 1 uL of the column performance check solution (Section 7.7) and acquire selected ion monitoring (SIM) data for m/z 320, 322, 332, and 334. 8.1.3.2 The chromatographic peak separation between 2,3,7,8-TCDD and the peaks representing any other TCDD isomers should be resolved with a valley of 30-60%, where Valley % = (x/y)(100) x = measured height of the valley between the chromatographic peak corresponding to 2,3,7,8-TCDD and the peak of the nearest TCDD isomer; and y '= the peak height of 2,3,7,8-TCDD. Figure 2 is an example of the separation of a TCDD isomer mixture and the calculation of isomer resolution. It is the responsibility of the laboratory to verify the conditions suitable for the appropriate resolution of 2,3,7,8-TCDD from all other TCDD isomers. The column performance check solution also contains the TCDD isomers eluting first and last under the analytical conditions specified in this protocol, thus defining the retention time window for total TCDD determination. Any individual selected ion current profile or the reconstructed total ion current (m/z 320 + m/z 322) consititutes and acceptable form of data presentation. 8.1.4 Initial Calibration for PCDD/PCDF Analysis Initial calibration is required before any samples are analyzed for PCDD/PCDF. Initial calibration is also required if any routine calibration does not meet the required criteria listed in Section 8.1.6. 8.1.4.1 Tune and calibrate the instrument with PFK as outlined in Section 8.1.1. 8.1.4.2 Six concentration calibration solutions listed in Table 2 will be analyzed for the initial calibration phase. A-20 �TCDD Isomer Mixture 78.7 2. 3. 7, 8,-TCDD 1, 2, 3, 4-/1, 2359290. 2, 3, 7-/L 2. 3, 8- TCDI 319.856 ±0.500 A 320 100.0, 2998270. A x 100% = 29% 322 3> 321.855 ± 0.500 A 59.6 1785850. Q2-2. 3. 7, 8.-TCDD INS 331.851 ± 0.500 332 75.3 2256890. 333.850 ± 0.500 334 28:00 28:24 28:48 29:12 29:36 30:00 30:24 30:48 Figure 2. Example of the separation of 2,3,7,8-TCDD from other TCDD isomers on a 60 m DB-5 column. Time �8.1.4.3 8.1.4.4 Compute the relative response factors (RRFs) for each analyte in the concentration calibration solution using the criteria for positive identification of PCDD/PCDF's given in Section 14.1 and the computational methods in Section 14.2. 8.1.4.5 Compute the means and their respective relative standard deviations (% RSD) for the RRFs from each triplicate analysis for each analyte in the standard. 8.1.4.6 8.1.5 Using the HRGC and MS conditions in Table 5 and the SIM monitoring descriptors in Table 6, analyze a 1-jjL aliquot of each of the six concentration calibration solutions in triplicate. Calculate the grand means (RRF) and their respective RSDs using the six mean RRFs for each analyte. Criteria for Acceptable Initial Calibration 8.1.5.1 The % RSD for the response factors for each triplicate analysis for each analyte must be less than 30% except for the TCDD and TCDF, which must be less than 20%. 8.1.5.2 The variation of the mean RRFs obtained from the triplicate analysis must be less than 30% except for the TCDD and TCDF which must be less than 20%. 8.1.5.3 The SIM traces for all ions used for quantitation must present a signal-to-noise (S/N) ratio of S 2.5. This includes analytes and isotopically labeled standards. 8.1.5.4 Isotopic ratios must be within ±20% of the theoretical values (see Table 7). NOTE: If the criteria for acceptable calibration listed abo've have been met, the RRF can be considered independent of the analyte quantity for the calibration concentration range. The mean RRF from triplicate determinations for unlabeled PCDD/ PCDFs and for the isotopically labeled standards will be used for all calculations until routine calibration criteria (Section 8.1.7) are no longer met. At such time, new mean RRFs will be calculated from a new set of six triplicate determinations. A-22 �Table 7. Ion Ratios for HRGC/LRMS Analysis of PCDD/PCDF Compound TCDF 13 C12-TCDF TCDD 13 C12-TCDD PeCDF 13 C12-PeCDF PeCDD 13 C12-PeCDD HxCDF 13 C12-HxCDF HxCDD 13 C12-HxCDD HpCDF 13 C12-HpCDF HpCDD 13 C12-HpCDD OCDF 13 C12-OCDF OCDD 13 C12-OCDD Ions monitored Theoretical ratio 304/306 316/318 320/322 332/334 338/340 350/352 354/356 366/368 374/376 386/388 390/392 402/404 408/410 420/422 424/426 436/438 442/444 454/456 458/460 470/472 0.76 0.76 0.76 0.76 0.61 0.61 0.61 0.61 1.22 1.22 1.22 1.22 1.02 1.02 1.02 1.02 0.87 0.87 0.87 0.87 A-23 Acceptabl e range 0.61 0.61 0.61 0.61 0.49 0.49 0.49 0.49 0.96 0.96 0.96 0.96 0.82 0.82 0.82 0.82 0.70 0.70 0.70 0.70 - 0.91 0.91 0.91 0.91 0.73 0.73 0.73 0.73 1.46 1.46 1.46 1.46 1.22 1.22 1.22 1.22 1.04 1.04 1.04 1.04 �8.1.6 Routine Calibrations Routine calibrations must be performed at the beginning of every day before actual sample analyses are performed and as the last injection of every day. 8.1.6.1 8.1.6.2 8.1.7 Inject 1 uL of the concentration calibration solution CS 7 (see Table 2). Compute the RRFs for each analyte in the concentration calibration solution using the criteria for positive identification of PCDD/Fs given in Section 14.1 and the computational methods in Section 14.2. Criteria for Acceptable Routine Calibration 8.1.7.1 8.1.7.2 Isotopic ratios must be within 20% of the theoretical value for each analyte and isotopically labeled standard (see Table 8). 8.1.7.3 8.2 The measured RRF for all analytes must be within 30% of the mean values established by triplicate analysis of the calibration concentration solutions, except for TCDD and TCDF, which must be within 20% of the mean values established in the initial calibration step. If the above criteria are not met, a second attempt may be made before repeating the entire initialization process. HRGC/HRMS Analysis (Isomer Specific TCDD Analysis) Isomer specific analysis for 2,3,7,8-TCDD is carried out with the instrumental conditions and parameters shown in Table 8. In addition to monitoring the masses of the most abundant molecular ions of TCDD, an ion corresponding to the loss of COC1 from the molecular ion is monitored for verification purposes. Mass spectrometer resolution is maintained at or above 10,000 (10% valley definition) in order to increase the specificity of the analysis. 8.2.1 Tuning and Mass Calibration 8.2.1.1 The mass spectrometer must be operated in the electron (impact) ionization mode. Static resolving power of at least 10,000 (10% valley) must be demonstrated before any analysis of a set of samples is performed. Static resolution checks must be performed at the beginning and at the end A-24 �Table 8. HRGC/HRMS Operating Conditions Mass spectrometer Accelerating voltage: Trap current: Electron energy: Electron multiplier voltage: Source temperature: Resolution: 8,000 V 500 (jA 70 eV 2,000 V 280°C 10,000 (10% valley definition) Sim Parameters Identity Mass TCDD-COC1 TCDD TCDD 13 C12-TCDD 13 C12-TCDD 0.15 0.15 0.15 0.15 0.15 0.10 258.930 319.897 321.894 331.937 333.934 PFK (lock mass) Nominal dwel1 times (s) 280.983 Overall SIM cycle time = 1 s Gas chromatqgraph Column coating: Film Thickness: Column dimensions: CP-Sil 88 0.2 pm 50 m x 0.22 mm ID Helium linear velocity: Helium head pressure: •v 25 cm/s 2 1.75 kg/cm (25 psi) Injection type: Split flow: Purge flow: Injector temperature: Interface temperature: Injection size: Initial temperature: Initial time: Temperature program: Split!ess, 45 s 30 mL/min 6 mL/min 270°C 240°C 2 ML 200°C 1 min 200°C to 240°C at 4°C/min A-25 �of each 12-h period of operation. However, it is recommended that a visual check.(i.e., not documented) of the static resolution be made before and after each analysis. 8.2.1.2 8.2.1.3 8.2.2 The MS shall be tuned daily using PFK to yield a resolution of at least 10,000 (10% valley) and optimal response at m/z 254.986. This step is followed by calibration of an accelerating voltage scan of PFK beginning at m/z 254 (typical calibration range is 255 to 493 amu). Other voltage scans from the same data file are used to establish and document both the resolution at m/z 316.983 and the mass measurement accuracy at m/z 330.979. Following calibration, the SIM experiment descriptor is updated to reflect the new calibration. Six masses (see Table 8) are monitored by scanning ^ m/10,000 amu (atomic mass units) over each mass. The total cycle time is kept to 1 s. The m/z 280.983 ion from PFK is used as a lock mass because it is the most abundant PFK ion within the range of m/z 255 to 334 and therefore permits the use of low partial pressures of PFK, which minimizes PFK interferences at the analytical masses. Mass Measurement and Resolution Check Using a PFK molecular leak, tune the instrument to meet the minimum required resolving power of 10,000 (10% valley) at m/z 254.986 (or any other mass reasonably close to m/z 259). Calibrate the voltage sweep at least across the mass range m/z 259 to m/z 334 and verify that m/z 330.979 from PFK (or any other mass close to m/z 334) is measured within ± 5 ppm (i.e., 1.7 mmu, if m/z 331 is chosen) using m/z 254.986 as a reference. Documentation of the mass resolution must then be accomplished by recording the peak profile of the PFK reference peak m/z 318.979 (or any other reference peak at a mass close to m/z 320/322). The format of the peak profile representation must allow manual determination of the resolution; i.e., the horizontal axis must be a calibrated mass scale (amu or ppm per division). The results of the peak width measurement (performed at 5% of the maximum which corresponds to the 10% valley definition) must appear on the hard copy and cannot exceed 100 ppm (or 31.9 mmu if m/z 319 is the chosen reference ion). A-26 �8.2.3 HRGC Column Performance (50-m CP Sil 88/60-m SP-2330) Prior to any HRGC/HRMS analysis of calibration solutions on samples for 2,3,7,8-TCDD, the resolution of the HRGC columns must be documented to be within allowable limits in order to provide conditions adequate for unambiguous isomer-specific analysis of 2,3,7,8-TCDD. 8.2.3.1 Inject 2 pL of the column performance check solution and acquire selected ion monitoring (SIM) data for m/z 258.930, 319.897, 321.894, 331.937, and 333.934 within a total cycle time of < 1 s (Table 8). 8.2.3.2 The chromatographic peak separation between 2,3,7,8-TCDD and the peaks representing any other TCDD isomers must be resolved with a valley of ^ 25%, where Valley % = (x/y)(100) x = measured height of the valley between the chromatographic peak corresponding to 2,3,7,8-TCDD and the peak of the nearest TCDD isomer; and y = the peak height of 2,3,7,8-TCDD. 8.2.3.3 If the above resolution requirement is not met, corrective action must be taken and acceptable resolution documented prior to any further analyses. Corrective action may include removal of the first meter of the HRGC column, replacement or clearing of the injector port, or complete replacement of the GC column. 8.2.3.4 The column performance check solution also contains the TCDD isomers eluting first and last under the analytical conditions specified in this protocol, thus defining the retention time window for total TCDD determination. The peaks representing 2,3,7,8-TCDD and the first and the last eluting TCDD isomer should be labeled and identified as such on the chromatograms (F and L, respectively). Any individual selected ion current profile or the reconstructed total ion current (m/z 259 + m/z 320 + m/z 322) constitutes an acceptable form of data presentation. A-27 �8.2.4 Initial Calibration for HRGC/HRMS 2,3,7,8-TCDD Analysis Initial calibration is required before any samples are analyzed for 2,3,7,8-TCDD. Initial calibration is also required if any routine calibration does not meet the required criteria listed in Section 8.2.6. 8.2.4.1 8.2.4.2 Tune and calibrate the instrument with PFK as described in Section 8.2.1.- 8.2.4.3 Inject 1 uL of the column performance check solution (Section 8.2.3) and acquire SIM mass spectra data for m/z 258.930, 319.897, 321.894, 331.937, and 333.934 using a total cycle time of S 1 s (see Table 8). The laboratory must not perform any further analysis until it has been demonstrated and documented that the criterion listed in Section 8.2.3.2 has been met. 8.2.4.4 Using the same GC and MS conditions (Table 8) that produced acceptable results with the column performance check solution, analyze a l-(jl_ aliquot of each of the six concentration calibration solutions in triplicate. 8.2.4.5 Calculate the RRFs for unlabeled 2,3,7,8TCDD relative to 13C12-2,3,7,8-TCDD and the RRF for 13C12-2,3,7,8-TCDD relative to 13 C12-1,2,3,4-TCDD using the criteria for positive identification of TCDD by HRGC/ HRMS given in Section 14.1 and the computational methods in Section 14.2. 8.2.4.6 Calculate the six means (RRFs) and their respective relative standard deviations ( RSD) for the response factors from each % of the triplicate analyses for both unlabeled and 13C12-2,3,7,8-TCDD. 8.2.4.7 8.2.5 At least six of the concentration calibration solutions listed in Table 2 must be utilized for the initial calibration. Calculate the grand mean RRFs and their respective relative standard deviations ( RSD) using the six mean RRFs. % Criteria for Acceptable Initial Calibration The criteria listed below for acceptable calibration must be met before analysis of any sample is performed. A-28 �8.2.5.1 The percent relative standard deviation (RSD) for the response factors from each of the triplicate analyses for both unlabeled and 13C12-2,3,7,8-TCDD must be less than 20%. 8.2.5.2 The variation of the six mean RRFs for unlabeled and 13C12-2,3,7,8-TCDD obtained from the triplicate analyses must be less than 20% RSD. 8.2.5.3 SIM traces for 2,3,7,8-TCDD must present a signal-to-noise ratio of § 2.5 for m/z 258.930, m/z 319.897, and m/z 321.894. 8.2.5.4 SIM traces for 13C12-2,3,7,8-TCDD must present a signal-to-noise ratio 5 2.5 for m/z 331.937 and m/z 333.934. 8.2.5.5 Isotopic ratios for 320/322 and 332/334 must be within the allowed range (0.67 to 0.90). NOTE: If the criteria for acceptable calibration listed above have been met, the RRF can be considered independent of the analyte quantity for the calibration concentration range. The mean RRF from six triplicate determinations for unlabeled 2,3,7,8-TCDD and for 13C12-2,3,7,8-TCDD will be used for all calculations until routine calibration criteria (Section 8.2.6) are no longer met. At such time, new mean RRFs will be calculated from a new set of four triplicate determinations. 8.2.6 Routine Calibrations Routine calibrations must be performed at the beginning of a 12-h period after successful mass resolution and HRGC column performance check runs. 8.2.6.1 Inject 1 yL of the concentration calibration solution (CS5, Table 2) which contains 10 pg/|Jl of unlabeled 2,3,7,8-TCDD, 50.0 pg/HL of 13C12-2,3,7,8-TCDD, and 50 pg/|jL of 13C12-1,2,3,4-TCDD. Using the same GC/ MS/DS conditions as used in Table 8, determine and document acceptable calibration as provided below. A-29 �8.2.7 Criteria for Acceptable Routine Calibration The following criteria must be met before further analysis is performed. If these criteria are not met, corrective action must be taken and the instrument must be recalibrated. 8.2.7.1 The measured RRF for unlabeled 2,3,7,8-TCDD must be within 20% of the mean values established in the initial calibration by triplicate analyses of concentration calibration solutions. 8.2.7.2 The measured RRF for 13C12-2,3,7,8-TCDD must be within 20% of the mean value established by triplicate analysis of the concentration calibration solutions during the initial calibration. 8.2.7.3 Isotopic ratios must be within the allowed range (0.61 to 0.90). 8.2.7.4 If one of the above criteria is not satisfied, a second attempt can be made before repeating the entire initialization process. NOTE: An initial calibration must be carried out whenever the routine calibration solution is replaced by a new one from a different lot. 9. QUALITY CONTROL PROCEDURES 9.1 Summary of QC Analyses 9.1.1 Initial and routine calibration and instrument performance checks. 9.1.2 Analysis of a batch of samples with accompanying QC analyses: Sample batch -- 10 NHATS adipose tissue samples plus additional QC analyses including 1 method blank, a control tissue and a spiked tissue sample. "Blind" QC (external QC) samples may be submitted by an external source (quality assurance group or independent laboratory) and included among the batch of samples. Blind samples include spiked samples, unidentified duplicates, and performance evaluation samples. A-30 �9.2 Performance Evaluation Samples -- Included among the samples in every third batch will be a solution provided by the quality control coordinator containing known amounts of unlabeled 2,3,7,8TCDD and/or other PCDD/PCDF isomers. The accuracy of measurements for performance evaluation samples should be in the range of 70-130%. 9.3 Performance Check Solutions 9.3.1 9.4 At the beginning of each 12-h period during which samples are to be analyzed, an aliquot of the HRGC column performance check solution shall be analyzed to demonstrate adequate HRGC resolution for selected TCDD isomers. Method Blanks 9.4.1 A minimum of one method blank is generated with each batch of samples. A method blank is generated by performing all steps detailed in the analytical procedure using all reagents, standards, equipment, apparatus, glassware, and solvents that would be used for a sample analysis, but omit addition of the adipose tissue. 9.4.1.1 The method blank must contain the same amounts of Carbon-13 labeled internal quantitation standards that are added to samples before bulk lipid cleanup. 9.4.1.2 An acceptable method blank exhibits no positive response for any of the characteristic ions monitored. 9.4.1.2.1 9.4.1.2.2 9.5 If the above criterion is not met, solvents, reagents, spiking solutions, apparatus, and glassware are checked to locate and eliminate the source of contamination before any samples are extracted and analyzed. If new batches of reagents or solvents contain interfering contaminants, purify or discard them. Control Samples -- Control samples are prepared from a bulk sample(s) of human adipose tissue or similar matrix (e.g., porcine fat). This material is prepared by blending the tissue with methylene chloride, drying the extract by eluting through anhydrous sodium sulfate, and removing the methylene chloride using rotoevaporation at elevated temperatures (80°C). The evaporation process should be extended to ensure all traces of the extraction A-31 �solvent have been removed. The resulting oily matrix (lipid) is subdivided into 10-g aliquots which are analyzed with each sample batch. The results of the individual analysis will be used to give a measure of precision from batch to batch over an entire program. Sufficient tissue should be extracted to provide a homogeneous lipid matrix that can be used over the total analysis program. Enough lipid matrix is necessary to prepare the spiked samples describe in Section 9.6. 9.6 Spiked Samples — Spiked lipid samples are prepared using a portion of the homogenized lipid described in Section 9.5. Sufficient spiked lipid matrix is prepared to provide a minimum of one spiked sample per sample batch. It is recommended that a minimum of three spiked levels of the matrix are prepared ranging from 10 to 50 times the estimated limit of detection for each compound. Each analysis of spiked sample must be accompanied by analysis of a control sample in order to make the necessary corrections for background contribution before determining the accuracy of the method (Equation 9-1). Accuracy (%) = 100% x Cone, spiked samp^-conc. control sample Eq ^ 9.7 Duplicate Sample Analysis -- When possible a duplicate analysis of specific samples is included in the sample batch as an additional measure of method precision. It is suggested that the total tissue sample is extracted to isolate lipids material and then subdivided for duplicate analysis. Precision is calculated as relative percent difference (RPD) where the differences in the duplicate measurements (for each analyte) is divided by the average of the two measurements and multiplied by 100%. 9.8 External Samples ~ Samples submitted as blinds to the analyst may consist of either performance solutions of PCDD and PCDF congeners or spiked sample matrices. These performance solutions or samples should be submitted by a source external to the analytical program (QA unit of analysis laboratory or independent laboratory). Performance audit solutions are intended to evaluate instrument calibration and quantisation procedures. Spiked blind samples must be accompanied by the corresponding unspiked samples to correct concentrations for background concentration. The blind spiked samples are intended to evaluate the total analytical procedure. The analyst must keep in mind that it is necessary to compare differences in standard sources for each type of external sample. 10. SAMPLE PRESERVATION AND HANDLING All adipose tissue samples must be maintained at less than -20°C from time of collection. The analyst should instruct the collaborator collecting the sample(s) to avoid the use of chlorinated materials. Samples are handled using stainless steel forceps, spatulas, or scissors. A-32 �Aliquots of samples removed from sample bottles not used for analysis are disposed rather than returned to the sample vial. All sample bottles (glass) are cleaned as specified in Section 6.4.10. Teflon®-!ined caps should be used. As with any biological sample, the analyst should avoid any undue exposure. 11. SAMPLE EXTRACTION 11.1 Extraction of Adipose Tissue 11.1.1 Accurately weigh to the nearest 0.01 g a 10-g portion of a frozen adipose tissue sample into a culture tube (2.2 x 15 cm). Note: Sample size may be smaller, depending on availability. 11.1.2 11.1.3 Allow the mixture to separate and decant the methylene chloride extract from the residual solid material using a disposable pipette. The methylene chloride is eluted through a filter funnel containing a plug of clean glass wool and 5 to 10 g of anhydrous sodium sulfate. The dried extract is collected in a 100-mL volumetric flask. 11.1.4 A second 10-mL aliquot of methylene chloride is added to the sample and homogenized for 1 min. The methylene chloride is decanted, dried, and transferred to the 100-mL volumetric flask as specified in Section 11.1.3 11.1.5 The culture tube is rinsed with at least two additional aliquots (10 mL each) of methylene chloride, and the entire contents are transferred to the filter funnel containing the anhydrous sodium sulfate. The filter funnel and contents are rinsed with additional methylene chloride (20 to 40 mL). The total eluent from the filter funnel is collected in the 100-mL volumetric flask. Discard the sodium sulfate. 11.1.6 11.2 Allow the adipose tissue specimen to reach room temperature. Add 10 ml of methylene chloride and homogenize the mixture for approximately 1 min with a Tekmar Tissuemizer®. The final volume of the extract for each sample is adjusted to 100 mL in the volumetric flask using methylene chloride. Lipid Determination 11.2.1 Preweigh a clean 1-dram glass vial to the nearest 0.0001 g using an analytical balance tared to zero. A-33 �11.2.2 Accurately transfer 1.0 ml of the final extract (100 ml) from Section 11.1.6 to the 1-dram vial. Reduce the volume of methylene chloride from the extract using a water bath (50-60°C) gentle stream of purified nitrogen until an oil residue remains. 11.2.3 Accurately weigh the 1-dram vial and residue to the nearest 0.0001 g and calculate the weight of lipid present in the vial based on difference. Nitrogen blow-down is continued until a constant weight is achieved. 11.2.4 Calculate the percent lipid content of the original sample to the nearest 0.1% as shown in Equation 11-1. I D ^ PYT Lipid content, LC (%) = r^ X V r x 100% ^ W AT Eq. 11-1 AL where: W.R = weight of the lipid residue to the nearest 0.0001 g calculated from Section 11.2.3 (100.0 ml); VtAI = total volume of the extract in mL from FYT Section 11.1.6; W.,. = weight of the original adipose tissue samples to the nearest 0.01 g from Section 11.1.1; and V.. = volume of the aliquot of the final extract in ml used for the quantitative measure of the lipid residue (1.0 mL). 11.2.5 11.3 Record the lipid residue measured in Section 11.2.3 and the percent lipid content calculated from Section 11.2.4. Extract Concentration 11.3.1 11.3.2 11.4 Quantitatively transfer the remaining extract volume (99.0 mL) to a 500-mL Erlenmeyer flask. Rinse the volumetric flask with 20 to 30 mL of additional methylene chloride to ensure quantitative transfer. Place the Erlenmeyer flask on a hot plate at 40°C to remove solvent until an oily residue remains. Addition of Internal Quantisation Standards To each lipid residue add the carbon-13 internal quantisation spiking solution (Section 7.8) such that it delivers 500 to 2,500 pg of each of the surrogates specified in Table 4 in a 100-(jL volume. A-34 �12. CLEANUP PROCEDURES 12.1 Bulk Lipid Removal 12.1.1 Add a total of 200 ml of n-hexane to the spiked lipid residue in the 500-mL Erlenmeyer flask. 12.1.2 Slowly add, with stirring, 100 g of the 40% w/w sulfuric acid impregnated silica gel (Section 7.1.3). Stir with a magnetic stir-plate for 2 h. 12.1.3 Allow solids to settle and decant liquid through a powder funnel containing 20 g of anhydrous sodium sulfate and collect in a 500-mL sample bottle. 12.1.4 Rinse solids with two 50-mL portions of hexane. Stir each rinse for 15 min, decant, and dry by elution through sodium sulfate combining the hexane extracts from Section 12.1.3. 12.1.5 After the rinses have gone through the sodium sulfate, rinse the sodium sulfate with an additional 25 mL of hexane and combine with the hexane extracts from Section 12.1.4. 12.1.6 Prepare an acidic silica column as follows: Pack a 1 cm x 10 cm chromatographic column with a glass wool plug, add approximately 25 mL of hexane, 1.0 g of silica gel (Section 7.1.2), and 4.0 g of 40% w/w sulfuric acid impregnated silica gel (Section 7.1.3). Pack a second chromatographic column (1 cm x 30 cm) with a glass wool plug, approximately 25 mL of hexane, 6.0 g of acidic alumina (Section 7.1.1), and top with a 1-cm layer of sodium sulfate (Section 7.4). Elute the combined hexane solutions (Section 12.1.5) through the columns until the solvent level reaches the top of the chromatographic packing. Inspect columns to ensure they are free of channels and air bubbles. Wash the alumina column with 40 mL of 50% v/v methylene chloride/hexane. Remove methylene chloride from the adsorbent by eluting the column with an additional 100 mL of hexane. 12.1.7 Quantitatively transfer the hexane extract from the Erlenmeyer flask (Sections 12.1.3 through 12.1.5) to the silica gel column reservoir. Allow the hexane extract to percolate through the column and collect in a KD concentrator. 12.1.8 Complete the elution of the extract from the silica gel column with 50 mL of hexane in the KD concentrator. Concentrate the eluate to approximately mL, using nitrogen blow-down as necessary. A-35 �Note: If the 40% sulfuric acid/silica gel is noted to be highly discolored throughout the length of the adsorbent bed it is necessary to repeat the cleaning procedure beginning with Section 12.1.1. 12.2 Separation of Chemical Interferences 12.2.1 Transfer the concentrate (1.0 ml) to the top of the alumina column. Rinse the K-D assembly with two 1.0-mL portions of hexane and transfer the rinses to the top of the alumina column. Elute the alumina column with 18 ml of hexane until the hexane level is just below the top of the sodium sulfate. Discard the eluate. Columns must not be allowed to reach dryness (i.e., a solvent "head" must be maintained). 12.2.2 Place 30 ml of 20% (v/v) methylene chloride in hexane on top of the alumina and elute the TCDDs from the column. Collect this fraction in a 50-mL culture tube. 12.2.3 Prepare an 18% Carbopak C/Celite 545® mixture by thoroughly mixing 3.6 g of Carbopak C (80/100 mesh) and 16.4 g of Celite 545® in a 40-mL vial. Activate at 130°C for 6 h. Store in a desiccator. Cut off a clean 5-mL disposable glass pipet (6 to 7 mm ID) at the 4-mL mark. Insert a plug of glass wool (Section 7.3) and push to the 2-mL mark. Add 500 mg of the activated Carbopak/Celite mixture followed by another glass wool plug. Using two glass rods, push both glass wool plugs simultaneously towards the Carbopak/Celite mixture and gently compress the Carbopak/Celite plug to a length of 2 to 2.5 cm. Pre-elute the column with 2 ml of toluene followed by 1 ml of 75:20:5 methylene chloride/methanol/ benzene, 1 ml of 1:1 cyclohexane in methylene chloride, and 2 ml of hexane. The flow rate should be less than 0.5 mL/min. While the column is still wet with hexane, add the entire eluate (30 ml) from the alumina column (Section 12.2.2) to the top of the column. Rinse the culture tube which contained the extract twice with 1 ml of hexane and add the rinsates to the top of the column. Elute the column sequentially with two 1-mL aliquots of hexane, 1 ml of 1:1 cyclohexane in methylene chloride, and I ml of 75:20:5 methylene chloride/methanol/benzene. Turn the column upside down and elute the PCDD/PCDF fraction with 20 ml of toluene into 6-dram vial. 12.2A Warm the vial to approximately 60°C and reduce the toluene volume to approximately 1 ml using a stream of nitrogen. Carefully transfer the concentrate into a 1-mL minivial and reduce the volume to about 200 nL using a stream of nitrogen. A-36 �12.2.5 Rinse the concentrator tube with three washings using 500 ML of 1% toluene in CH2C12 (Section 12.2.5) concentrated to 200-500 ML and add 10 ML of the tridecane solution containing the internal recovery standard and store the sample in a refrigerator until HRGC/MS analysis. 12.2.6 Immediately prior to analysis, using a gentle stream of nitrogen at room temperature, remove toluene and methylene chloride. Submit sample to HRGC/MS once a stable 10 ML volume of tridecane is attained. 13. ANALYTICAL 13.1 PROCEDURES HRGC/MS Analysis for PCDD/PCDF 13.1.1 Once routine calibration criteria are met, the instrument is ready for sample analysis. Prior to the first sample, a blank injection of tridecane should be analyzed to document system cleanliness. If any evidence of system contamination is found, corrective action must be taken and another tridecane blank analyzed. The typical daily sequence of injections is shown in Table 9 and Figure 3. Note: Syringe Technique — Congeners of PCDD/PCDF in the syringes used for HRGC/MS analysis can be problematic unless the syringes are properly handled between samples. The following procedure has been found to be very effective for PCDD/PCDF removal from contaminated syringes and will be used throughout these analyses. • Rinse the syringe 10 times with isooctane. • Fill the syringe with toluene and sonicate syringe and plunger in toluene for 5 min and repeat at least twice. • Rinse the syringe 10 times with tridecane and pull up 1 ML of clean tridecane. • Syringe is ready for use. At no time should air be introduced into the HRGC column by using an air plug in the syringe. The oxygen present in the air plug will quickly degrade a nonbonded GC phase. 13.1.2 Inject a 1-ML aliquot of the extract into the GC, operated under the conditions previously used (Section 8.1) to produce acceptable results with the performance check solution. A-37 �Table 9. Typical Daily Sequence for PCDD/PCDF Analysis 1. Tune and calibrate mass scale versus perfluorokerosene (PFK). 2. Inject concentration calibration solution 2.5 to 12.5 pg/(jL (CS-7) solution. 3. Inject blank (tridecane). 4. Inject samples 1 through "N". 5 Inject concentration calibration solution 2.5 to 12.5 pg/uL (CS-7) solution. A-38 �INSTRUMENTAL ANALYSIS Instrument Mass Calibration vs PFK I Column Performance Evaluation Does Column Performance Meet Minimum Resolution Requirements? No Adjust Column Length or Install New Column No Reanalyze or Prepare Fresh Calibration Standards and Calibration Curve Yes Calibration Standard Analysis Do Relative Response Factors Meet Criteria Based on Initial Calibrations ? Yes Proceed with Sample Analysis Figure 3. Daily QA procedures for proceeding with sample analysis. A-39 �13.1.3 Acquire SIM data according to the same acquisition and MS operating conditions previously used (Section 8.1) to determine the relative response factors. 13.1.3.1 13.1.3.2 13.2 Acquire SIM data for the characteristic ions designated in Table 6. Instrument performance shall be monitored by examining and recording 13 peak areas the for the recovery standard, C12-1,2,3,4-TCDD. If this area should decrease to less than 50% of the previous run, sample analyses shall be stopped until the problem is found and corrected. HRGC/HRMS Confirmation of 2,3,7,8-TCDD 13.2.1 Once the daily criteria of mass calibration, mass resolution, HRGC performance, and routine calibration are met and documented, the instrument is ready for sample analysis. Prior to the first sample, a blank injection of tridecane will be made to document system cleanliness. The typical daily schedule for HRGC/HRMS analysis of TCDD is shown in Table 10 and Figure 3. 13.2.2 Inject a 1-uL aliquot of the extract into the GC, operated under the conditions previously used (Section 8.1) to produce acceptable results with the performance check solution. 13.2.3 Acquire SIM data according to Section 12.3.1. Use the same acquisition and MS operating conditions previously used (Section 8.3.4) to determine the relative response factors. 13.2.3.1 Acquire SIM data for the following selected characteristic ions: m/z Compound 258.930 TCDD - COC1 319.897 Unlabeled TCDD 321.894 Unlabeled TCDD 331.937 13 C12-2,3,7,8-TCDD, 13 C12-1,2,3,4-TCDD 333.934 13 C12-2,3,7,8-TCDD, 13 C12-1S2,3,4-TCDD A-40 �Table 10. Typical Daily Schedule for HRGC/HRMS Analysis of TCDD 1. Tune and calibrate mass scale. 2. Perform mass measurement check and mass resolution check. 3. Inject column performance check solution. 4. Inject the routine concentration calibration solution. 5. Inject tridecane blank. 6. Inject samples I through "N". 7. Inject column performance check solution. 8. Mass resolution check. A-41 �14. DATA REDUCTION In this section, the the analysis of data HRGC/HRMS method for qualitative criteria 14.1 procedures for the data reduction are outlined for from both the HRGC/MS method for PCDD/PCDF and the 2,3,7,8-TCDD. Figure 4 presents a schematic of the for identifying PCDDs and PCDFs. Qualitative Identification 14.1.1 14.1.2 The ion current intensities for a particular PCDD/PCDF must be S 2.5 times the noise level (S/N S 2.5) for positive identification of that isomer. 14.1.3 The integrated ion current ratios of the analytical masses for a particular PCDD/PCDF must fall within the ranges shown in Table 7. 14.1.4 14.2 The ion current responses for each mass for a particular PCDD/PCDF analyte must be within ± 1 s to attain positive identification of that analyte. For example, m/z 338 and m/z 340 must have maximum peak responses that are within ± 1 s to be positively identified as a pentachlorodibenzofuran. The recovery of the internal quantisation standards should be between 50 and 115%. Quantitative Calculations 14.2.1 Relative response factors (RRF). RRFs are calculated from the data obtained during the analysis of concentration calibration solutions using the following formula: C A x IS RRF = -^ ' j£ A C Eq. 14-1 IS ' x where A = the sum of the integrated ion for the analyte in question. for TCDD, A would be the sum grated ion abundances for m/z abundances For example, of the inte320 and 322; A,s = the sum of the integrated ion abundances for the labeled PCDD/F used as the internal quantisation standard for the above analyte. For example, for 13C12-TCDD, A,s would be the sum of the integrated ion abundance for m/z 332 and 334. C = concentration of the analyte in A-42 �HRGC/MS-SIM Data Response to Characteristic Molecular Ions within the Appropriate Homolog Retention Window? Report Compounds as Not Detected (ND) Calculate Sample LOD Characteristic Ion Ratios within "±20% Theoretical? Response Due to Coextracted Interference Response Correspondsito Specific Isomerj Retention Time? 'Quantitate Compound as Per Protocol Report as Isomer Unknown Quantitate Specific Isomer as per Protocol Figure 4. Qualitative criteria for identifying PCDDs and PCDFs. A-43 �CjS = concentration of the internal quantisation standard in pg/uL; and RRF = relative response factor. NOTE: The above formula is also used to compute the RRFs for the various internal standards relative to the recovery standard, 13C12-1,2,3,4-TCDD. 14.2.2 Concentrations of sample components. Figure 5 presents a schematic for quantisation of PCDDs and PCDFs which meet the criteria specified in Section 14.1. Calculate the concentration of PCDD/Fs in sample extracts using the formula: y TC A,. - Qis X C x= A IS' RRF •' W AT • LC Eq. 14-2 where Cx = the lipid adjusted concentration of PCDD or PCDF congener in pg/g; Ax = sum of the integrated ion abundances determined for the PCDD/PCDF in question; AIS = sum of the integrated ion abundances determined for the labeled PCDD/F used as the internal quantisation standard for the above analyte; QjS = the amount (total pg) of the labeled internal quantisation standard added to the sample prior to extraction; RRF = relative response factor of the above analyte relative to its labeled internal quantisation standard determined from the initial triplicate calibration; WAT = weight of original adipose tissue sample; Al and LC = percent extractable lipid determined from Eq. 11-1. Quantitative data should be classified to indicate the intensity of the signal response. Suggested qualifiers include: not detected, NO (signal-to-noise ratio is less than 2.5); trace, TR (signal-to-noise ratio is greater than or equal to 2.5 but less than 10); and positive quantifiable, PQ (signal-to-noise ratio is greater than or equal to 10. A-44 �QUANTITATION HRGC/MS-SIMData Response Meets All Qualitative Criteria? Report as Not Detected Calculate Sample LOD Response >2.5t[mes S/N? Response > 10 times S/N? Calculate as per Protocol Report as Trace (tr) Value Quantitate as per Protocol Report as Positive Quantifiable Value Figure 5. Procedure for quantitation of PCDDs and PCDFs in human adipose tissue. A-45 �14.2.3 Recovery of internal quantisation standards. Calculate the recovery of the labeled internal quantisation standards measured in the final extract-using the formula: QRS Internal Quant. Std. _ Eq. 14.3 Percent Recovery ~ ARS RRF 100 where AT<; = sum of the integrated ion abundances deteri:> mined for the labeled PCDD/PCDF internal quantisation standard in question; A Q RS sum of the integrated ion abundances determined for m/z 332 and m/z 334 of 13C121,2,3,4-TCDD (recovery standard); recovery standard, RS ~ amount (pg) of the added to the final 13 C12-1,2,3,4-TCDD extract; QIS = amount (pg) the labeled internal quantitation standard added to the sample prior to extraction; and RRF = relative response factor for the labeled internal quantitation standard in question relative to the internal recovery standard. This value shall be the RRF determined from the initial calibration. 14.3 Estimated Method Detection Limit Estimated where (1) sponse is sponse is method detection limits must be calculated in situations no response is noted for a specific congener; (2) a renoted but ion ratios are incorrect; and (3) where a requantitated as a trace value. 14.3.1 For samples in which no unlabeled PCDD or PCDF is detected, calculate the estimated minimum detectable concentration. The background area is determined by integrating the ion abundances for the characteristic ions in the appropriate region and relating the product area to an estimated concentration that would produce that product area. Use the formula: (2.5) Eq. 14-4 < A IS> A-46 (RRF) • (W) �where C^ = estimated concentration of unlabeled PCDD or PCDF required to produce AX; A = sum of integrated ion abundances or peak heights for the characteristic ions of the unlabeled PCDD or PCDF isomer in the same group of ^ 5 scans used to measure A,s; and AjS = sum of integrated ion abundances for the appropriate ions characteristic of the respective internal quantitation standard. Qjc, RRF, and W retain the definitions previously stated in Section 14.2. Alternatively, if peak height measurements are used for quantification, measure the estimated detection limit by the peak height of the noise in the 2,3,7,8-TCDD RT window. 14.3.2 For samples for which a response at the retention time of a specific PCDD or PCDF congener is noted, but the quantitative criteria for ion ratios is outside the acceptable range (Table 7), the estimated detection level is calculated as given in Eq. 14.3 except the values are qualified as not detected, ND, and the concentration is reported in parenthesis. 14.3.3 If a response for a specific PCDD or PCDF congener is qualified as a trace, TR, value (signal to noise is greater than or equal to 2.5 but less than 10) the analyst must also provide an estimated method detection limit. This is accomplished by using the observed signal to noise on either side of the response and calculating as given in Eq. 14-4. 15. REPORTING AND DOCUMENTATION All data should be reported on an individual sample basis using the data report format shown in Figure 6. The analyst is required to maintain all raw data, calculations, and control charts in a format as to allow a complete external data review. Suggested data formats for tracing calculationsare provided in Figure 7. A-47 �Pig* 1 oil U.S. ENVIRONMENTAL PflOTECTION AGENCY OFFICE OF TOXIC SUBSTANCES EXPOSURE EVALUATION DIVISION (TS-7M) WASHINGTON, DC 20460 NATIONAL HUMAN ADIPOSE TISSUE SURVEY ANALYSIS REPORT FORM EPA SAMPLE NUMBER. ANALYSIS DATE LAB NUMBER MS ANALYST BATCH NUMBER REPORT DATE _ REPORTED BY _ NATIVE COMPOUNDS CONCENTRATION (pg/gLl/ DATA INTERNAL QUANTITATION STANDARD QUALIFIER!/ 2.3,7.8-TCDD 1 ,,.!•!, 1 1 2.3,7,8-TCOF 1 . . ... |*l , 1 13 1,2.3,7,8-PeCDO 1 , , . 1*1 , r 1 3C12-2.3,7.8-TCDO Ci8-2,3,7.8-TCDF 3C12-1.2,3.7,8-P«COD 1,2.3,7,8-PeODF 1 . . . |*)l , 1 13 2.3,4,7.8-PeCDF 1 , . , Ul jlT l 1 <,2.3,4.7,8-HxCDO 1 ... 1 1,2.3,8,7.8-HxCDD 1 ...!•( i 1 1,2.3,7.8.9-HxCDO 1 , i , 1*1 . 1 '3C12-1,2,3,4,8.7,a-HpCDF 1,2.3,4.7.8-HxCDF 1 ...!•), 1 13 1,2.3,8,7,8-HxCDF 1 . . . Ul , 1 'SCu-OODF 1,2.3,7.8.9-HxCDF 1 . . . I.I . 1 2.3.4,8,7,8-HxCDF 1 , i , |«4 , 1 t.2,3,4,8,7,8-HpCDD 1 ... 1,2,3,4,8,7,8-HpCDF 1 , , , 1*1 , 1 1*1 , 1 C12-1.2,3.7,8-PeCDF 3Ct2-1.2.3.6.'.a-H»CDD 3C12-t.2,3,4.7.8-HxCDF '3012-1.2.3.4,8,7,8^0000 Cia-OCDO 1*1 , 1 1.2.3.4,7.8,9-HpCDF 1 , . . Ul OCOO I , . . !•( . 1 . 1 OCOF 1 . . . 1*1 . 1 REMARKS Jj Concantratlon reported la based on total extractable lipld (a). I NO - Not Detected, TR - Traoe. PQ - Positive Quantifiable. / Figure 6. Analysis report form. A-48 SPIKED LEVEL (P9) PERCENT (%) RECOVERY �RAW DATA SUMMARY FOR DETERMINATION OF 1.2.3.7,8-PeCDD IN HUMAN ADIPOSE TISSUE Sample no. 3 Sample weight (xx.xx g) Extractable lipid content (xx.x %) Analysis date Amount C12-PeCDD (pg) 13 13 C12-PeCDD m/z 332 3 > C12-PeCOD m/z 334 Ion ratio 366/368 1,2,3,7,8PeCDD m/z 354 Value reported as concentration in extractable lipid. Figure 7. Example of raw data summary format for the determination of 1,2,3,7,8-PeCDD in human adipose tissue. 1,2,3,7,8 m/2 356 Ion ratio 354/356 1,2,3,7,8PeCDD cone. (pg/g) � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Alvin L. Young Collection on Agent Orange Description An account of the resource <p style="margin-top: -1em; line-height: 1.2em;">The Alvin L. Young Collection on Agent Orange comprises 120 linear feet and spans the late 1800s to 2005; however, the bulk of the coverage is from the 1960s to the 1980s and there are many undated items. The collection was donated to Special Collections of the National Agricultural Library in 1985 by Dr. Alvin L. Young (1942- ). Dr. Young developed the collection as he conducted extensive research on the military defoliant Agent Orange. The collection is in good condition and includes letters, memoranda, books, reports, press releases, journal and newspaper clippings, field logs and notebooks, newsletters, maps, booklets and pamphlets, photographs, memorabilia, and audiotapes of an interview with Dr. Young.</p> <p>For more about this collection, <a href="/exhibits/speccoll/exhibits/show/alvin-l--young-collection-on-a">view the Agent Orange Exhibit.</a></p> Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Box The box containing the original item. 197 Folder The folder containing the original item. 5522 Series The series number of the original item. Series VIII Subseries I Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Creator An entity primarily responsible for making the resource Stanley, John S. et al. Description An account of the resource <strong>Corporate Author: </strong>Midwest Research Institute Date A point or period of time associated with an event in the lifecycle of the resource 1986 Title A name given to the resource Analysis for Polychlorinated Dibenzo-p-Dioxins and Dibenzofurans in Human Adipose Tissue: Method Evaluation Study: Draft Final Report with attached letter transmitting the report to Alvin L. Young, from Janet C. Remmers, Field Studies Branch, Exposure Eva ao_seriesVIII https://www.nal.usda.gov/exhibits/speccoll/files/original/62f7399c340eba5e215718304bfbac19.pdf e3f2b4b55454ed73c4ece6477b153742 PDF Text Text Ram D Number °5385 Author Erickson, Mitchell D, D ^ot Scanned Midwest Research Institute Report/Article TitlB Analytical Methods for By-Product RGBs - Preliminary Validation and Interim Methods Journal/Book TitlB Year 1982 Month/Day October 11 Color D Number of Images 246 UBSCriDtOn NOtBS Task 51, Interim Report No. 4, EPA Contract No. 68-01-5915, MRI Project No. 4901 -A(51) Friday, March 08, 2002 Page 5385 of 5427 �Office of Toxic Substances Washington DC 20460 United States Environmental Protection Agency EPA-560/5-82-006 October, 1982 Toxic Substances v>EPA Analytical Methods for By-Produet PCBsPreliminary Validation and Interim Methods 100.0 T —T"—T 100.0 100.0 l3c 12 H 6 Ci 4 T 280 285 290 T-—r 295 300 305 Da I tons 3tO 315 320 325 �DISCLAIMER This document has been reviewed and approved for publication by the Office of Toxic Substances, Office of Pesticides and Toxic Substances, U.S. Environmental Protection Agency. Approval does not signify that the contents necessarily reflect the views and policies of the Environmental Protection Agency, nor does the mention of trade names or commercial products constitute endorsement or recommendation for use. �ANALYTICAL METHODS FOR BY-PRODUCTS PCBs—PRELIMINARY VALIDATION AND INTERIM METHODS By Mitchell D. Erickson, John S. Stanley, Kay Turman, Gil Radolovich, Karin Bauer, Jon Onstot, Donna Rose, and Margaret Wickham Midwest Research Institute 425 Volker Boulevard Kansas City, MO 64110 TASK 51 INTERIM REPORT NO. 4 EPA Contract No. 68-01-5915 MRI Project No. 4901-A(51) October 11, 1982 For U.S. Environmental Protection Agency Office of Toxic Substances Field Studies Branch TS-798 Washington, D.C. 20460 Attn: Dr. Frederick W. Kutz, Project Officer Mr. David P. Redford, Task Manager �PREFACE This report presents the results of a preliminary method validation accomplished on MRI Project No. 4901-A, Task 51, "PCB Analytical Methodology Task," for the Environmental Protection Agency (EPA Prime Contract No. 6801-5915) during the period April 24 to August 31, 1982. The document was prepared by Drs. Mitchell D. Erickson (Task Leader) and John S. Stanley and Ms. Kay Turman, with assistance from Kathy Funk, Cindy Melenson, and Gloria Sultanik. The laboratory work was conducted by Kay Turman, and Donna Rose, with assistance from Steven Turner. The gas chromatography/mass spectrometry analysis was performed by Gil Radolovich, Margaret Wickham, Jon Onstot, and Arbor Drinkwine. Statistical analysis of the data was provided by Karin Bauer. Editorial comments were provided by Rudena Mallory and Jeanne Robson. The EPA Task Manager, David Redford, has been especially helpful and encouraging. The helpful comments of Ann Carey, Frederick W. Kutz, and John Smith, all of EPA, are also appreciated. SEARCH INSTITUTE C J/mn E. Going, Head Environmental Analysis Section Approved: James L. Spigarelli, Director Analytical Chemistry Department ill �CONTENTS Preface Figures Tables 1. 2. 3. 4. 5. iii vii ix Introduction Summary Experimental Preparation of PCB stock solutions and working standards. Gas chromatography/electron impact mass spectrometry. . . Determination of PCB response factors (GC/EIMS) Validation of method steps Validation with product and product waste samples . . . . Method Validation Preparation of analytical methods Gas chromatography/mass spectrometry of PCBs Validation of selected method steps Validation of product and product waste method with samples Discussion References ' i 1 2 3 3 8 8 19 20 24 24 26 38 45 56 70 Appendix A - Supplementary GC/EIMS Data on PCB Congeners A-l Appendix B - Analytical Method: The Analysis of Incidentally Generated Chlorinated Biphenyls in Commercial Products and Product Wastes B-l Appendix C - Analytical Method: The Analysis of Incidentally Generated Chlorinated Biphenyls in Air C-l Appendix D - Analytical Method: The Analysis of Incidentally Generated Chlorinated Biphenyls in Industrial Wastewater D-l �FIGURES Number 1 2 3 4 5 6 1 8 9 Plot of average response factor versus homolog for 77 PCB congeners 27 Plot of response factor per isomer versus homolog for 77 PCB congeners 28 Plot of response factor per isomer versus homolog for 77 PCB congeners, determined on a single day 32 Retention times of 77 PCB congeners relative to 3,3*4,4'tetrachlorobiphenyl-de (RRT of 1.00) 36 Capillary gas chromatography/electron impact ionization mass spectrometry (CGC/EIMS) chromatogram or the calibration standard solution required for quantitation of PCBs by homolog 39 Reconstructed ion chromatogram for SIM analysis of the CMA-A sample no. 2110 59 SIM ion plots for monochlorobiphenyls (188 and 190 Daltons) and the 13C6-monochlorobiphenyl surrogate (194 Daltons) in CMA-A sample no. 2110 60 SIM ion plots for dichlorobiphenyls (222 and 224 Daltons) in CMA-A sample no. 2110 61 SIM ion plots for trichlorobiphenyls (256 and 258 Daltons) in CMA-A sample no. 2110 62 10 SIM ion plots for tetrachlorobiphenyls (290 and 292 Daltons), 3,3',4,4'-tetrachlorobiphenyl-d6 (298 Daltons), and the 13C12-tetrachlorobiphenyl surrogate (304 Daltons) in CMA-A sample no. 2110 63 11 SIM ion plots for pentachlorobiphenyls (326 and 328 Daltons) in CMA-A sample no. 2110 64 SIM ion plots of hexachlorobiphenyls (360 and 362 Daltons) in CMA-A sample no. 2110 65 12 13 SIM ion plots of heptachlorobiphenyls (394 and 396 Daltons) in CMA-A sample no. 2110 66 vii �FIGURES (continued) 14 15 16 SIM ion plots of octachlorobiphenyls (428 and 430 Daltons) and the ^Cj^-octachlorobiphenyl surrogate (442 Daltons) in CMA-A sample no. 2110 67 SIM ion plots of nonachlorobiphenyl (464 and 466 Daltons) in CMA-A sample no. 2110 68 SIM ion plots of decachlorobiphenyl (498 and 500 Daltons) and the 13C12-decachlorbiphenyl (510 Daltons) in CMA-A sample no. 2110 69 Vlll �TABLES (continued) Number 34 35 Pagj PCB Concentration (pg/g) of CMA-A Samples Treated With Various Cleanup Procedures (Surrogate Compound Correceted) 55 Recovery ( ) of Carbon-13 Labeled Surrogate Compounds % From Diarylide Yellow and Phthalocyanine Blue and Green Pigments 57 XI �TABLES Number Page 1 Numbering of PCB Congeners 5 2 Working Solutions for PCB Response Factors 6 3 Approximate Concentration of Individual PCB Congeners in Dilute Working Standards 7 Concentrations of Congeners in PCB Calibration Standards (ng/ml) 9 4 5 6 7 8 9 Composition of Surrogate Spiking Solution (SS100) Containing 13C-Labeled PCBs 10 Operating Parameters for Capillary Column Gas Chromatographic System 11 DFTPP Key Ions and Ion Abundance Criteria for Quadrupole Calibration 12 Operating Parameters for Quadrupole Mass Spectrometer System 13 Operating Parameters for Magnetic Sector Mass Spectrometer System 14 10 Characteristic Single lion Monitoring (SIM) Ions for PCBs . 15 11 Limited Mass Scanning (LMS) Ranges for PCBs 16 12 Characteristic Ions for 13C-Labeled PCB Surrogates 17 13 Pairings of Analyte, Calibration, and Surrogate Compounds . 18 14 Commercial Product and Product Waste Stream Samples Received for Preliminary Method Validation Studies. . . . 21 15 Preliminary Method Validation Samples 22 16 Comparison of Average Relative Response Factors (RRF) for 77 Commercially Available PCB Congeners Measured Over Several Days as Four Replicates Each Versus Single Measurements of All Congeners in a Single Day 30 IX �TABLES (continued) Number 17 18 Page Average Relative Response Factors (RRF) for PCB Congeners in Solution 1 Measured as Replicates on a Single Day and as Single Measurements for Day-to-Day Basis Measured Average Response Factor (RRF) and Corresponding Upper and Lower 95% Confidence Limits 19 20 31 Relative Response Factors Measured Versus 3,3',4,4'-Tetrachlorobiphenyl-de by Electron Impact Mass Spectrometry Quadrupole (Finnigan 4023) and Magnetic Sector (Varian (MAT 311A) Instruments 34 35 Relative Retention Time (RRT) Ranges of PCB Homologs Versus d6-3,3' ,4,4'-Tetrachlorobiphenyl 37 21 Recovery Data for Acid Cleanup 40 22 Recovery Data for Florisil Column Protocol Cleanup 41 23 Recovery Data for Florisil Slurry Protocol Cleanup 42 24 Recovery Data for KOH Protocol Cleanup 43 25 Recovery Data for Alumina Protocol Cleanup 44 26 Uncorrected PCB Concentrations (pg/g) in CMA-A Samples. . . 46 27 Corrected PCB Concentrations (pg/g) in CMA-A Samples. . . . 47 28 Uncorrected and Corrected PCB Concentrations (|Jg/g) in CMA-E Sample (Dilution Preparation) 49 Uncorrected PCB Concentration (|Jg/g) in the CMA-A Sample Matrix (Internal Standard Calculation) 50 Corrected PCB Concentration ((Jg/g) in the CMA-A Sample Matrix 51 Uncorrected PCB Concentration (|Jg/g) of Spiked CMA-A Samples Determined by the Internal Standard Quantitation Method 52 Corrected PCB Concentration (|Jg/g) of Spiked CMA-A Samples Determined by Surrogate Recovery Correction 53 29 30 31 32 33 PCB Concentration (|Jg/g) of CMA-A Samples Heated With Different Cleanup Procedures (Internal Standard Quantitation) 54 �SECTION 1 INTRODUCTION The Environmental Protection Agency (EPA) is in the process of preparing rules for regulation of certain polychlorinated biphenyls (PCB) which are generated as by-products in the manufacture of commercial products (U.S. EPA, 1982). This regulation is under the Toxic Substances Control Act (PL 94-469), and EPA's Office of Toxic Substances has been assigned the task of preparing the rule. As part of the rule, EPA is suggesting analytical methods for PCBs in air (stack gas and fugitive emissions), wastewater, product waste streams, and final products to assist organizations seeking an exclusion under this rule. To assist EPA in this mission, Midwest Research Institute (MRI) was asked to prepare appropriate analytical methodologies. A literature review and recommendation of general analytical approaches (Erickson and Stanley, 1982; Stanley and Erickson, 1982) constituted the first phase. The second phase, reported here, covers initial method validation and preparation of interim methods. As part of the method validation, four 13C-PCB surrogates were synthesized and are reported separately (Roth et al., 1982). The third phase will involve interlaboratory validation and method refinement. This report presents the initial results of method validation for analysis of by-product PCBs in product and product waste samples. Specifically, gas chromatography/electron impact mass spectrometry retention time and response factor data for 77 PCB congeners for two different gas chromatography/ mass spectrometry systems, recoveries from several proposed cleanup steps, and recoveries from industrial samples using a variety of the method options are presented. �SECTION 2 SUMMARY The objective of this study was to present EPA with appropriate methodologies for the analysis of by-product PCBs in commercial products, product waste streams, wastewaters, and air. In addition, EPA requested preliminary analytical studies to provide data in support of the proposed methods. This document presents proposed analytical methods for the analysis of by-product polychlorinated biphenyls in commercial products and product waste streams (Appendix B), wastewater (Appendix C), and air (Appendix D). The proposed methods are based on determination of PCBs using gas chromatography/ electron impact mass spectrometry (GC/EIMS). Capillary column gas chromatography (CGC) and packed column gas chromatography (PGC) are presented as alternate approaches. The 13C-labeled PCB surrogates are added to samples prior to any sample preparation to allow method flexibility for a wide spectrum of matrices. Recovery of the surrogates will allow determination of the quality of analytical data. This method is valid only if the surrogates are thoroughly incorporated into the matrix. The analytical method for commercial products and product waste streams relies heavily on a strong quality assurance program consisting of use of four 13C-labeled surrogate PCBs, blanks, duplicates, spiked samples, and quality control samples. The analytical methods for water and wastewater are based on EPA Methods 608 and 625, revised to include the use of the 13Clabeled surrogates. Likewise, the air method is a revision of a proposed method for PCBs in air and flue gas emissions. This document presents relative response factors (RRF) of 77 PCB congeners which were used to determine the average RRF for PCBs by homolog. Statistical analysis of the data was performed to check the validity of the response factor data and to extrapolate RRFs for the unavailable congeners. Relative retention time (RRT) data for the 77 PCB congeners are also presented. The RRF and RRT data were determined on both magnetic sector and quadrupole mass spectrometer systems. Preliminary studies were undertaken to check the validity of the proposed methods for the analysis of PCBs in commercial products and product waste streams. Data are presented for analysis of individual cleanup procedures as well as for analysis of product and product waste samples. The data indicate that the proposed method is applicable and useful for analysis of the matrices studied. However, these studies are preliminary and additional validation is necessary and ongoing. �SECTION 3 EXPERIMENTAL The method validation was conducted in three stages: (a) determination of GC/EIMS parameters for 77 PCB congeners; (b) validation of individual method steps with clean matrices; and (c) validation of selected method options with real samples. PREPARATION OF PCB STOCK SOLUTIONS AND WORKING STANDARDS Source of Standards Seventy-seven PCB congeners were acquired from Ultra Scientific, Inc., Hope, Rhode Island, and Analabs, North Haven, Connecticut. Quality control gas chromatography/flame ionization detection (GC/FID) data for the specific isomers were requested to verify the 99% purity assigned to these compounds. The GC/FID data supported the reported purity. In addition, all available nuclear magnetic resonance spectra used for specific isomer identification were requested but not supplied. Weighing Procedures Accurate mass measurement required calibration of a Cahn microbalance with National Bureau of Standards (NBS) certified masses of 5 and 10 mg. The balance was calibrated with the NBS standards followed by calibration of an in-house working standard mass. The calibration of the microbalance with the NBS certified masses was witnessed by a representative of the MRI quality assurance office. The mass of the working standard was measured between all measurements of individual PCB isomers to ensure that the balance was operating accurately. A record of the measured working standard mass was kept in a laboratory notebook. The mean value for the working standard was 10.037 ± 0.002 mg (0.02% relative standard deviation). When all measurements were completed, the mass of the NBS certified standards was determined as a final measure of the accuracy of the Cahn microbalance. Preparation of Solutions Preparation of PCB standard stocks began after accurate performance of the Cahn balance was demonstrated with the certified NBS and daily working standard. An aluminum weighing pan was preshaped such that complete transfer of the weighing pan plus sample could be made directly into the appropriate dilution vessel. The Cahn balance was tared to compensate for the weight of the aluminum boat, and the PCB standards were added via a micro spatula. The mass of the particular PCB was determined with the Cahn balance. �The aluminum pan containing the PCB standard was transferred to the dilution vessel using clean forceps, taking care not to spill any of the sample. The dilution vessel was capped tightly until solvent was added. All PCB congeners were dissolved in toluene (Burdick and Jackson, distilled in glass). Masses of 0.1 to 5 mg were dissolved in a total of 1.0 ml toluene while masses of approximately 10 mg and greater were dissolved in 5.0 ml toluene. The solvent was delivered volumetrically by pipette. Room temperature and solvent temperature were recorded at the time of standard dissolution. Volumetric pipettes used for solvent delivery were calibrated so that the most accurate determination of analyte concentration could be calculated. Toluene was pipetted into a tared vessel, and the total mass was measured. Density of the solvent at the specific room temperature was used to calculate the actual volume dispensed. This calibration was performed for all pipettes used for volumetric delivery of solvent. The stock solutions were sonicated in an ultrasonic bath for at least 15 sec after the volumetric addition of toluene to ensure complete dissolution of the PCBs. The solution level was etched on the side of the dilution vessel as a means of detecting losses by evaporation. The individual PCB congeners were referred to by the congener number indicated in Table 1. The stable labeled PCBs, 3,3',4,4'-tetrachlorobiphenyl-d6, 4-chlorobiphenyl-13Cg, 3,3',4,4'-tetrachlorobiphenyl-13C12, 2,2',3,3',5,5',6,6,'• octachlorobiphenyl-13Ci2> and decachlorobiphenyl-13Ci2 were assigned congener numbers of 210 to 214, respectively, for the purpose of this work. Sample labels were generated in duplicate to identify the specific PCB isomer stock solution and to document entries in the laboratory notebook. Table 2 presents the dilute working solutions that were prepared for determination of the response factors for the PCB congeners. The working solutions were prepared as 10 ml total volume. Table 3 presents the approximate concentration of each congener that was in the dilute working standard used for response factor determination. Tetrachlorobiphenyl-dg was added to 1.0 ml of each solution as the internal standard. All stocks were added to the working solutions in volumes of 20, 200, 250, 400, 500, or 1,000 | l The syringes were calibrated at j. these volumes. Calibration of the 10-ml volumetric flasks used for working standards was accomplished by measuring the difference between the mass of the empty flask and the mass of the flask plus toluene added to the appropriate dilution mark. The density of toluene at the correct solvent temperature was used to calculate the final volume of each solution. The dilute working solutions were divided into multiple aliquots. One hundred micrograms of tetrachlorobiphenyl-de was added to each of the 1.0-ml aliquots of the solutions that were used to establish CGC/EIMS response factors. The remaining dilute working solutions were stored in at least four crimp seal vials and refrigerated. The solvent meniscus was marked in permanent form to note losses of solvents from evaporation or spills. All solutions, stock standards and working solutions, were stored in a refrigerator. All vials removed from storage were first brought to room temperature and then sonicated for at least 15 to 30 sec before removing any of the solution. �Mb. Structure No. Honoetilorobiohenyli 1 2 3 2 3 4 D<eh1arob1ph«ny!s 4 5 6 7 8 9 10 11 12 13 1415 2.2' 2.3 2,3' 2,4 2,4' 2,5 2,6 3,3' 3,4 3,4' 3,5 4,4' TrlehlaroMehtnyls 16 17 13 19 20 21 22 23 24 25 26 27 23 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 15 46 47 48 49 50 51 2,2', 3 2,2', 4 2,2', S 2,2', 6 2,3,3' 2,3,4 2,3,4' 2,3,5 2,3,6 2,3', 4 2, 3', 5 2, 3', 6 2,4,4' 2,4,5 2,4,6 2, 4 ' , 5 2, 4'. 5 2 ' , 3, 4 2', 3,5 3,3', 4 3,3',S 3,4,4' 3,4,5 3.4', S NTJMBERING OF PCB CONGENERS a Structure Ha, Tttnehl orofal phtnyl s 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 2,2'.5.5' 2,2'.5,6' 2,2', 5,6' 2.3.3',4 2,3,3'. 4' 2,3,3', S 2,3,3'. 5' 2.3.3'. 6 2.3,4,4' 2,3,4,5 2,3,4.6 2, 3, 4', 5 2, 3,4', 6 2,3,5,6 2, 3', 4, 4' 2. 3'. 4,5 2,3'. 4,5' 2,3'. 4, 6 2,3', 4 ' , 5 2,3' ,4'. 6 2,3' ,5, 5' 2,3',5'.6 2,4,4', 5 2,4,4' .6 2'. 3 4,5 3,3' 4,4' 3,3' 4,5 3,3' 4,5' 3,3' 5,5' 3. 4, 4 ' , 5 Pentachl orobl cheny 1 s 82 83 84 85 86 37 88 39 90 91 92 93 94 TttracMoromehwyli 95 96 97 2, 2', 3,3' 98 2 2' 3 4 2;2','3,4' 99 100 2. 2'. 3.5 101 2,2', 3, 5' 2,2',3,6 102 2.2' .3.6' 103 2,2',4,4' 104 2,2' ,4, 5 2,2', 4,5' 2,2', 4, 6 2,2', 4, 6' TABLE 1. Structure 2,2',3.3',4 2.2'. 3. 3', S 2,2', 3, 3' .6 2,2', 3, 4, 4' 2,2'.3.4.5 2, 2'. 3, 4, 5' 2,2', 3, 4, 6 2,2', 3, 4, 5' 2,2', 3, 4 ' , 5 2, 2', 3, 4'. 5 2.2* .3,5,5' 2,2', 3,5, 6 2,2'.3,5,5' 2, 2', 3.5', 5 2,2',3,6,6' 2,2'. 3'. 4.5 2,2',3'.4,5 2,2' ,4. 4'. 5 2 2' .4,4'. 6 2,2'. 4, 5, 5' 2,2' ,4, 5, 6' 2,2', 4,5'. 5 2,2' .4. 6,5' NO. Structure 161 162 163 164 165 166 167 168 169 2,3.3 I ,4,5',6 2, 3, 3'. 4 ' , 5, 5' 2, 3,3'. 4 ' , 5, 6 2, 3, 3', 4 ' , S ' , 6 2, 3,3', 5. 5 ' , 5 2. 3. 4, 4 ' , 5, 6 2,3', 4, 4 ' . 5. 5' 2,3',4,4',5'.S 3,3',4,4',5.S' Pentaehl orobi oheny 1 s 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 2,3, 3', 4,4' 2,3,3',4,5 2, 3, 3'. 4'. 5 2,3, 3'. 4, 5' 2.3, 3', 4, 6 2.3, 3'. 4' ,6 2.3.3' ,5,5' 2,3,3',5,6 2,3.3', 5', 6 2,3,4. 4'.5 2,3,4,4'. 6 2,3,4,5,6 2,3, 4 ' , 5, 6 2,3',4,4',5 2, 3', 4, 4 ' , 6 2,3' ,4, 5,5' 2,3', 4, 5 ' , 5 2' ,3, 3', 4,5 2' .3. 4, 4 ' . 5 2' .3. 4, 5, 5' 2'. 3, 4.5, 6' 3, 3' ,4, 4 ' , 5 3,3',4,5,5' Hexaehlorobiofienyls 128 129 130 131 132 133 134 135 '136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 2,2', 3,3', 4, 4 ' 2,2', 3, 3', 4, 5 2,2', 3,3', 4, 5' 2,2'. 3, 3'. 4,6 2,2', 3,3', 4, 6' 2,2'.3,3',5,5' 2,2'. 3, 3', 5, 6 2,2', 3, 3' ,5, 5' 2, 2'. 3,3'. 6, 6' 2,2* ,3,4,4', 5 2,2'.3,4,4 I .S 1 2, 2 ' , 3, 4, 4', 6 2, 2 ' , 3, 4, 4 ' . 6' 2,2', 3, 4, 5, 5' 2,2', 3, 4, 5, 6 2,2' ,3, 4. 5,6' 2,2', 3, 4, 5 ' , 6 2,2', 3. 4, 5, 5' 2,2'. 3. 4 ' , 5, 5' 2,2', 3. 4 ' , 5,5 2,2', 3, 4 ' . 5, 5' 2, 2' ,3, 4', 5 ' , 5 2,2',3,4',6,6' 2,2' .3,5, 5'. 5 2,2', 3, 5,6, 6' 2, 2 ' , 4, 4 ' , 5, 5' 2.2', 4, 4 ' , 5, 5' 2,2', 4 , 4 ' , 6, 5' 2,3,3', 4, 4 ' , 5 2,3, 3', 4, 4 ' , 5' 2,3, 3', 4, 4 ' . 5 2,3, 3'. 4, 5, 5' 2,3,3'. 4, 5.5 HexachlorobiBnefiyls Heotactilorobiohenyl s 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 2, 2', 3, 3' , 4 , 4 ' ,5 2, 2 ' , 3, 3', 4 , 1 ' , 5 2, 2', 3, 3', 1,5,5' 2, 2'. 3, 3 ' , 1,5, 5 2, 2 ' , 3, 3 ' , 1,5, 5' 2. 2'. 3, 3', 4 , 5 ' , 5 2, 2', 3, 3 ' , 4, 5, 5' 2.2',3,3',4',5,a 2, 2 ' , 3, 3', 5, 5 ' , 5 2, 2 ' , 3, 3 ' , 5, 5, 5' 2. 2'. 3, 4. 4 ' , 5, 5' 2, 2', 3, 4, 4 ' , 5,6 2, 2 ' , 3. 4, 4 ' , 5, 5' 2, 2'. 3, 4, 4 ' , = ' , 6 2, 2', 3, 4, 4 ' , i, 5' 2, 2 ' , 3, 4, 5, 5' ,6 2, 2 ' , 3, 4, 5. 6, 6' 2, 2 ' , 3, 4 ' , 5,5' ,5 2. 2 ' , 3, 4 ' , 5, 5, 6' 2, 3, 3', 4, 4 ' . 5,5' 2, 3, 3', 4, 4 ' , 5, 5 2, 3, 3' ,4, 4 ' , 5 ' , 5 2,3, 3', 4, S, 5 ' , a 2,3,3'.4',S,5',5 Octacnl orjb • cne«y 1 s 194 195 196 197 198 199 200 201 202 203 204 205 2 , 2 ' , 3, 3', 4 , 4' , 5 , 5 ' 2, 2', 3, 3' , 4 , 4 ' ,5,5 2 , 2 ' , 3 , 3 ' ,4,1', = , 5 ' 2, 2', 3, 3 ' , 4,1', 5, 5' 2, 2 ' , 3 , 3 ' , 1,5,5' ,5 2, 2 ' , 3, 3 ' , 4, 5, 5, 5' 2, 2 ' , 3,3', 4, 5' , 5 , 5 ' 2, 2 ' , 3, 3'. 4, 5. 5 ' . 5' 2, 2'. 3, 3', 5, 5' ,5,5' 2, 2 ' , 3, 4, 4 ' , 5, 5 ' . 6 2, 2 ' , 3, 4, 4' , 5 , 6 , 5 ' 2,3. 3', 4, 4 ' . 5, 5 ' , 5 Monsehtcrobi;ns''yl s 206 207 208 2. 2 ' , 3, 3 ' , 1,1', 5, s ' , 5 2. 2', 3,3', 4 , 1 ' , 5, 5, 5' 2, 2 ' , 3, 3'. 4, 5, 5 ' , 5,5' DecachloHjOionenyi 209 2,2',3,3'4,4',5,5'.5,5' Adapted from Ballschmiter K, Zell M. 1980. Analysis of polychlorinated biphenyls (PCB) by glass capillary gas chromatography. Composition of technical Aroclorand Clophen-PCB mixtures, Fresenius Z. Anal Chera 302:20-31. �TABLE 2. PCB Soln. homolog no. 1 Soln. no. 2 Soln. no. 3 Soln. no. 4 WORKING SOLUTIONS FOR PCB RESPONSE FACTORS Soln. no. 5 PCB congener no. Soln. Soln. Soln. Soln. no. 6 no. 7 no. 8 no. 9 Soln. Soln. Soln. Soln. no. 10 no. 11 no. 12 no. 13 1 2 3 Dichloro- 11 5 7 8 9 10 4 12 14 Trichloro- 29 21 31 26 24 28 18 33 30 Tetrachloro- 47 44 40 49 50 52 53 54 66 61 65 69 72 Pentachloro- 121 97 88 93 101 103 100 104 a 115 87 116 119 Hexachloro- 136 129 128 137 138 141 143 151 139 153 154 155 156 Heptachloro- 181 171 183 185 Octachloro- 195 194 198 200 202 204 Nonachloro- 207 208 206 Decachloro- 209 9 9 Soln. no. 14 Monochloro- 15 70,75,77 Total congeners a 10 7 6 6 5 5 4 4 3 3 3 3 Congener no. 112 was added to this solution but, on analysis, was determined to have a mass of 286 and appeared to be a diaminotrichlorobiphenyl. This congener was omitted from any further consideration. �TABLE 3. APPROXIMATE CONCENTRATION OF INDIVIDUAL PCB CONGENERS IN DILUTE WORKING STANDARDS5 PCB horaolog Concentration (pg/ml) Monochlorobiphenyl 50 Dichlorobiphenyl 50 Trichlorobiphenyl 50 Tetrachlorobiphenyl Pentachlorobiphenyl 100 Hexachlorobiphenyl 100 Heptachlorobiphenyl 100 Octachlorobiphenyl 200 Nonachlorobiphenyl 200 Decachlorobiphenyl a 100 200 Tetrachlorobiphenyl-de was added to all solutions as an internal standard at *> 100 |Jg/ml. �Preparation of Calibration Standard and Spiking Mixtures A mixture of 11 congeners was used for calibration. This solution was spiked into solvent for protocol step validation experiments and into product and product waste samples for standard addition experiments. These congeners were determined to be the best standards for quantitation calibration based on the average relative response factor for each PCB homolog, as will be discussed in Section 5. Table 4 presents the composition of the 11-component solutions that are specified as the calibration standards, CSxxx, where the xxx is used to encode the nominal concentration in nanograms per milliliter. A more concentrated solution was diluted as necessary to prepare spiked samples and the appropriate standards for GC/EIMS analysis. The internal standard, tetrachlorobiphenyl-dg, was added to all standards and final extracts before GC/ EIMS analysis. The standards contained the four 13C-labeled PCBs that were added from the spiking solution shown in Table 5. GAS CHROMATOGRAPHY/ELECTRON IMPACT MASS SPECTROMETRY The capillary gas chromatography parameters used are shown in Table 6. The quadrupole and magnetic sector mass spectrometer parameters used are shown in Tables 7 through 9. The characteristic ions for single ion monitoring and limited mass scanning are presented in Tables 10 through 12. All data generated for relative response factors and concentration levels of PCBs in sample extracts were calculated based on the area of the primary quantitation ion specified in Table 10. The quantitation ions for the 13Clabeled monochloro-, tetrachloro-, octachloro-, and decachlorobiphenyl were 194, 304, 442, and 510 Daltons, respectively. The pairings of analyte, calibration, and surrogate compounds are presented in Table 13. DETERMINATION OF PCB RESPONSE FACTORS (GC/EIMS) The response factors for 77 PCB isomers were determined by GC/EIMS using the working standards prepared as described in Tables 2 and 3. A high resolution capillary column (J&W Scientific Durabond DB-5, 15 m, 0.25 |Jm film thickness) was used for the separation of the PCB mixtures. Scanning mass spectrometry was used to calculate response factors for the PCB isomers present in each solution versus a known quantity of tetrachlorobiphenyl-dgThe quadrupole GC/EIMS system was tuned daily prior to any acquisition of data for PCB response factor calculations. The system was brought to operating temperature for at least 15 min. The fluorocarbon FC-43 was introduced to the ion source, and 176 and 502 Daltons were manually adjusted 'to a two-to-one ratio. This was accomplished by adjusting the multiplier voltage to 300 mV while monitoring 176 Daltons. A selected ion monitor acquisition was set up for 176 and 502 Daltons with a variance of 1 Dalton. The ratio of the two values was tuned to the two-to-one ratio as described above. The mass spectrometer was operated in the normal full scan acquisition mode after tuning with the FC-43. Approximately 100 ng of decafluorotriphenylphosphine was injected and the ratio of the values of 198/442 was monitored. �TABLE 4. CONCENTRATIONS OF CONGENERS IN PCS CALIBRATION STANDARDS (ng/ml)a Homolog Congener no. CS1000 CS100 CS050 CS010 1 1 1,040 104 52 10 1 3 1,000 100 50 10 2 7 1,040 104 52 10 3 30 1,040 104 52 10 4 50 1,520 152 76 15 5 97 1,740 174 87 17 6 143 1,920 192 96 19 7 183 2,600 260 130 26 8 202 4,640 464 232 46 9 207 5,060 506 253 51 10 209 4,240 424 212 42 4 255 255 255 255 1 211 (RS) 104 104 104 104 4 212 (RS) 257 257 257 257 8 213 (RS) 407 407 407 407 10 a 210 (IS) 214 (RS) 502 502 502 502 Concentrations given as examples only. �TABLE 5. COMPOSITION OF SURROGATE SPIKING SOLUTION (SS100) CONTAINING 13C-LABELED PCBs3 Congener no. Compound Concentration (|jg/ml) 211 104 212 ,3' (13C12)3 ,4,4'-tetrachlorobiphenyl 257 213 (13C12)2 ,2', 3, 3', 5, 5' ,6,6'-octachlorobiphenyl 214 a 1 13 (I1, 2', 3 ,4',5',6'- C6)4-chlorobiphenyl (13C12)decachlorobiphenyl Concentrations given as examples only. 10 395 502 �TABLE 6. OPERATING PARAMETERS FOR CAPILLARY COLUMN GAS CHROMATOGRAPHIC SYSTEM Parameter Value Gas chromatograph Finnigan 9610 Column 15 m x 0.255 mm ID Fused silica Liquid phase DB-5 Liquid phase thickness 0.25 urn Carrier gas Helium Carrier gas velocity 45 cm/sec Injector On-column (J&W) Injector temperature Optimum performance Injection volume 1.0 Mlb Initial column temperature 110°C (2 min)c Column temperature program . 110° to 325°C at 10°C/min d Separator None Transfer line temperature 280°C (J&W) a Measured by injection of air or methane at 270°C oven temperature. b For on-column injection, follow J&W instructions regarding injection technique. c With on-column injection, the initial temperature equals the boiling point of the solvent; in this instance toluene. d C12Clio elutes at 270°C. Programming above this temperature ensures a clean column and lower background on subsequent runs. e Fused silica columns may be routed directly into the ion source to prevent separator discrimination and losses. 11 �TABLE 7. DFTPP KEY IONS AND ION ABUNDANCE CRITERIA FOR QUADRUPOLE CALIBRATION Mass Ion abundance criteria 197 198 199 Less than 1% of mass 198 100% relative abundance 5-9% of mass 198 275 10-30% of mass 198 365 Greater than 1% of mass 198 441 442 443 Present but less than mass 443 Greater than 40% of mass 198 17-23% of mass 442 12 �TABLE 8. OPERATING PARAMETERS FOR QUADRUPOLE MASS SPECTROMETER SYSTEM Parameter Value Mass spectrometer Finnigan 4023 Data system Incos 2400 Scan range 95-550 Scan time 1 sec Resolution Unit Ion source temperature 280°C Electron energy3 70 eV Trap current 0.2 mA Multiplier voltage -1,600 V Preamplier sensitivity 106 A/V a Filaments should be shut off during solvent elution to improve instrument stability and prolong filament life, especially if no separator is used. 13 �TABLE 9. OPERATING PARAMETERS FOR MAGNETIC SECTOR MASS SPECTROMETER Parameter SYSTEM Value Mass spectrometer Finnigan MAT 311A Data system Incos 2400 Scan range 98-550 Scan mode Exponential Cycle time 1.2 sec Resolution 1,000 Ion source temperature 280°C pt Electron energy 70 eV Emission current 1-2 mA Filament current Optimum Multiplier -1,600 V a Filaments should be shut off during solvent elution to improve instrument stability and prolong filament life, especially if no separator is used. 14 �TABLE 10. CHARACTERISTIC SINGLE ION MONITORING (SIM) IONS FOR PCBs Ion (relative intensity) Homo log Primary Secondary Ca2H9Cl 188 (100) 190 (33) CigHgCla 222 (100) 224 (66) 226 (11) C12H7C13 256 (100) 258 ( 9 9) 260 (33) Ci2H6Cl4 292 (100) 290 (76) 294 (49) Ci2HsCl5 326 (100) 328 ( 6 6) 324 (61) C12H4C16 360 (100) 362 (82) 364 (36) Ci2H3Cl7 394 (100) 396 ( 8 9) 398 ( 4 5) £12^2^-8 430 (100) 432 ( 6 6) 428 (87) Ci2HClg 464 (100) 466 (76) 462 (76) C 498 (100) 500 ( 7 8) 496 ( 8 6) 12CllO Source: a Tertiary _a Rote JW, Morris WJ. 1973. Use of isotopic abundance ratios in identification of polychlorinated biphenyls by mass spectrometry. J Assoc Offie Anal Chem 56(1):188-199. / None available. 15 �TABLE 11. LIMITED MASS SCANNING (LMS) RANGES FOR PCBs ft Mass range (Daltons) Compound C^Cl, 186-190 C12H8C12 220-226 C12H7C13 254-260 C12H6C14 288-294 C12H5C15 322-328 C12H4C16 356-364 Lj *i oXloL^J-V 386-400 \_, -i OJlOvjJ^Q 426-434 \j i O-H-W-L o 460-468 1210 494-504 C12D6C14 294-300 13 192-196 13 300-306 13 438-446 C612C6H9C1 C12H6C14 C12H2C18 13 506-516 C12C110 a Adapted from Tindall GW, Wininger PE. 1980. Gas chromatography-mass spectrometry method for identifying and determining polychlorinated biphenyls. J Chromatogr 196:109-119. 16 �TABLE 12. CHARACTERISTIC IONS FOR 13C-LABELED PCB SURROGATES Primary Ion (relative intensity) Secondary Tertiary 13 194 (100) 196 (33) a 13 304 (100) 306 (49) 302 (78) 13 442 (100) 444 (65) 440 (89) 13 510 (100) 512 (87) 514 (50) Compound C612C6H9C1 C12H6C14 C12H2C18 Ci2Cl10 a None available. 17 �TABLE 13. PAIRINGS OF ANALYTE, CALIBRATION, AND SURROGATE Analyte Congener no. I 2,3 4-15 16-39 40-81 82-127 128-169 170-193 194-205 206-208 209 Calibration standard Compound 2-C12H9Cl 3- and 4-C12H9Cl C12HgCl2 C12H7C13 C^HsCls Cl^EUClg C12H3C17 C12H2Clg Ci2HCla C12C110 Congener no. 1 3 7 30 50 97 143 183 202 207 209 Compound 2 4 2,4 2,4, 6 2,2' ,4,6 2,2' ,3', 4, 5 2,2' ,3,4,5 ,6' 2,2' ,3', 4, 4',5',6 2,2' Q Q t 5,5' ,6,6' 2,2' 3 3' 4, 4', 5, 6, 6' r L Lior COMPOUNDS Surrogate Congener no. 211 211 211 212 212 212 212 213 213 213 214 Compound 13 C6-4 C6-4 13 C6-4 13 C12-3 ,3' ,4,4' 13 C12-3 ,3' ,4,4' 13 C12-3 ,3' ,4,4' 13 C12-3 ,3' ,4,4' 13 Ci2-2 ,2' ,3,3' ,5,5' ,6,6' 13 3 3* Ci2-2 ,2' ,0,0 ,5,5' ,6,6' 13 C12-2 ?' ,3,3' 5 5 r ,6,6' 1ft 13 �The response of 198 Daltons was 100% full scale and 442 Daltons was adjusted from 40 to 45% of the base peak. These criteria were met daily before data acquisition for response factor calculations was initiated. All working standards were brought to room temperature and sonicated before injection into the GC/MS system. Solution No. 1 was analyzed daily as a means of normalizing response factors calculated from day to day. This allowed some compensation for differences in sensitivity due to subtle changes in the mass spectrometer operation from day to day. Also, a solution of tetrachlorobiphenyl-de (internal standard) was analyzed separately. Four replicates of each working standard were analyzed to calculate variances of the response factors. The solutions were sonicated at least 15 sec prior to removal of sample for injection. The syringe and needle were rinsed with 200- to 300-|Jl of toluene between injections. The gas chromatograph was operated at 110°C for 2 min, and programmed at 10°C/min to 325°C. One microliter injections were made with a J&W on-column injection system. Helium carrier flow was adjusted to 45 cm/sec. The peak shape of the eluting PCBs was monitored. If excessive tailing was noted, the injection end of the fused silica capillary column was removed and shortened by at least 10 cm. Tables 6, 7, and 8 present the instrument and operating parameters that were used to measure the response factors for the individual PCB isomers in the working solutions. Response factors (RF) were calculated using the area of the peaks for these ions according to the equation: A M _ ~r PCB rj IS OT — Kr A IS WPCB A where MPCB = .IS = ..IS = PCB = Area of the quantitation peak of the specific PCB, Mass (in nanograms) of the internal standard injected, Area of the quantitation peak of the internal standard, and Mass (nanograms) of the specific PCB injected. All relative response factor data were subjected to Student's t-test at the 95% confidence level to test for significant differences for day-to-day and solution-to-solution variances. VALIDATION OF METHOD STEPS A limited number of experiments were completed as preliminary validation steps for the proposed method presented in Appendices B through D. The experiment included evaluation of several of the cleanup procedures using solvent spiked with the 13C-labeled surrogates and a mixture of PCB congeners representing each of the possible homologs. The laboratory cleanup procedures followed the protocol steps except where noted. One hexane solvent blank was analyzed by each procedure with the samples to monitor interferences and contamination. 19 �All samples were analyzed by CGC/EIMS in the full scan mode using the Finnigan 4023 system. Tables 6, 7, and 8 present the instrumental parameters. VALIDATION WITH PRODUCT AND PRODUCT WASTE SAMPLES Sources of Samples Product waste samples were received from Dow Chemical Company (Kent Hodges) and Vulcan Materials Company (Thomas Robinson) through the cooperation of the Chemical Manufacturers Association (Robert Fensterheim). These samples are aliquots of the materials used for the Chemical Manufacturers Association (CMA) round robin study (CMA, 1982). The CMA and associates supplied samples of chlorinated benzene waste streams, mixtures of chlorinated benzenes, composite waste streams from a chlorinated aliphatic process and a benzene column bottom sample. Table 14 presents an inventory of all the samples received. Product samples were received from the Dry Color Manufacturers Association (J. Lawrence Robinson and Maria DaRoche). These samples included diarylide yellow, phthalocyanine green, and phthalocyanine blue pigments that were used in the Dry Color Manufacturers Association (DCMA) round robin study of an analytical method, reported by the DCMA (1981). These samples are also included in the inventory in Table 14. The samples supplied by industry are examples of the samples which will be analyzed using the method in Appendix B. However, since no attempt was made to span the range of products and product wastes, the samples analyzed do not include all matrices which an analyst could encounter. Experimental Design Table 15 presents an overview of the preliminary method validation samples. The samples from Table 14 that were used for these studies included the chlorinated benzene waste stream, CMA-A; the benzene column bottom sample, CMA-E; and the yellow, blue, and green pigment samples, DCMA-1, DCMA-4, and DCMA-8, respectively. Blind quantitation standards and quality control samples were prepared by the MRI quality control staff either through spiked addition or by dilution of particular sample matrices. Other quality control procedures included the analysis of duplicate samples and blanks and the validation of cleanup steps. Two sets of samples were prepared and run at separate times. This first sample set us designated by numbers 10 through 110 and the second sample set is designated by numbers 2001 through 2210Q in Table 15. The sample preparations ranged from addition of the 13C-labeled surrogates followed by dilution and injection, to preparation of pigment samples via sulfuric acid dissolution and hexane extraction or methylene chloride extraction with Florisil cleanup. 20 �TABLE 14. COMMERCIAL PRODUCT AND PRODUCT WASTE STREAM SAMPLES RECEIVED FOR PRELIMINARY METHOD VALIDATION STUDIES3 Sample no. Quantity Sample description Sample source CMA-A 100 ml Chlorinated benzene waste stream CMA-B 100 ml Mixture of chlorinated benzenes Dow Chemical Co. with Aroclor 1254 spike CMA-C 100 ml Blind spike of CMA-B with the addition of 64 ppm of PCB isomers Dow Chemical Co. CMA-A 5 ml Vulcan Materials Co. CMA-B 5 ml CMA-C 5 ml CMA-D 5 ml CMA-E 5 ml Chlorinated benzene waste stream Mixture of chlorinated benzenes with Aroclor 1254 spike Blind spike of CMA-B with the addition of 64 ppm of PCB isomers Composite waste stream sample from a chlorinated aliphatic process Benzene column bottoms sample Diarylide yellow pigment Phthalocyanine green pigment Phthalocyanine blue pigment Phthalocyanine blue pigment Phthalocyanine green pigment DCMA DCMA DCMA DCMA DCMA DCMA-1 DCMA-4 DCMA-6 DCMA-8 DCMA-9 100 100 100 100 100 g g g g g Dow Chemical Co. Vulcan Materials Co. Vulcan Materials Co. Vulcan Materials Co. Vulcan Materials Co. a Aliquots of CMA-A, CMA-B, and CMA-C were received from two sources, who indicated that they were identical. MRI has assumed that both aliquots are the same. 21 �TABLE 15. PRELIMINARY METHOD VALIDATION SAMPLES Sample no. Description Preparation 10 20A 20B 60 110 2001 2005 2010 CMA-A CMA-A CMA-A Hexane blank CMA-E Hexane blank CMA-A3 CMA-A 0.1 g/10 ml hexane 0.1 g/10 ml hexane 0.1 g/10 ml hexane None None None • 0.1 g/1 ml hexane 0.1 g/1 ml hexane 0.1 g/1 ml hexane 0.5-0.2 g/1 ml hexane 0.1 g/1 ml hexane 0.1 g/1 ml hexane 0.1 g/1 ml hexane 0.1 g/1 ml hexane None DCMA-A DCMA-A (0.1 g) DCMA-B DCMA-B (0.1 g) Base Base (0.1 g) DCMA-B (1.0 g) DCMA-B (1.0 g) DCMA-B (1.0 g) DCMA-B (1.0 g) DCMA-B (1.0 g) DCMA-B DCMA-B DCMA-B DCMA-A DCMA-A DCMA-A None 2020 CMA-A 2025Q 2030 2040 2050 2060Q 2070Q 2080 2090 2100 2110 2120 2130 2135 2140 2150 2160 2170Q 2175 2180 2185 2190 2195 2200Q 2210Q CMA-A CMA-A + CS002 CMA-A + CS005 CMA-A + CS010, CMA-A + CSXXX CSxxx Blank, CMA-AD Blank, CMA-AD Blank CMA-A DCMA-13 DCMA-1 DCMA-1 DCMA-1 + no. 11 (50 ppm) DCMA-1 + no. 11 (20-80 ppm) DCMA-4 DCMA-4 DCMA-4' DCMA-8 DCMA-8* DCMA-8 CSxxx Dilution factor 1/100 1/100 1/100 None None None 1/10 1/10 1/10 1/10 1/10 1/10 1/10 1/10 None 1/10 1/10 1/10 1/10 1/10 1/10 1/100 1/100 1/100 1/200 1/200 1/100 1/100 1/100 1/50 1/50 1/50 None a No surrogates added to assess any background interferences for these compounds. b Prepared from aliquot received from Dow Chemical Company; all other CMA-A samples prepared from aliquot received from Vulcan Materials Company. 22 �The CMA-A and CMA-E samples were each analyzed after 1/10 or 1/100 dilution, depending on the operating sensitivity of the mass spectrometer. The CMA-A chlorinated benzene waste was the most extensively studied matrix of the available samples. Sample preparation included the simple dilution described above with and without the addition of the four surrogates. The samples prepared without surrogates allowed measurement of the background that might interfere with the four surrogate compounds. Duplicate samples of the CMA-A were analyzed at the same dilution in two separate experiments. The CMA-A matrix was also analyzed by standard addition methods with total spiked PCS levels of the 11-compound spiking solution (CS050) at approximately 70, 140, and 270 ng/sample. The CMA-A matrix was also prepared using the sulfuric acid and ethanolic KOH procedures discussed in Section 9.3.2 of Appendix D, Cleanup of the Analytical Method: The Analysis of By-Product Chlorinated Biphenyls in Commercial Product and Product Wastes (Appendix B). Variations of the analytical procedures used by the Dry Color Manufacturers Association (1981) for the analysis of PCBs in various pigments were also applied to the CMA-A matrix. The DCMA procedures included acid dissolution followed by hexane extraction from the acid (DCMA Preparation A) and Florisil treatment of the concentrated sample matrix (DCMA Preparation B). The homogenization and centrifugation steps required by the DCMA-B procedure were not included for the CMA-A matrix. All samples except those representing blanks were spiked with the surrogates at levels of 100 to 500 ng and were mixed thoroughly before beginning the sample preparation. The typical CMA-A sample size was 0.1 g. The diarylide yellow (DCMA-1), phthalocyanine green (DCMA-4), and pthalocyanine blue (DCMA-8) pigments were also studied in these preliminary validations. The yellow pigment was prepared according to the recommended DCMA-B procedure, while the green and blue pigments were analyzed following the DCMA-A procedure. The preparation of the pigments followed the DCMA procedures except that the preparation was scaled to 1 g of the yellow pigment instead of the recommended 5 g. Blanks, duplicates, and spiked samples were also analyzed with the set of DCMA samples. Sample Analysis All extracts were analyzed by capillary column gas chromatography/electron impact mass spectrometry (CGC/EIMS). Limited mass scanning (LMS) or selected ion monitoring (SIM) mass spectrometry methods were used for extract analysis, depending on the level of PCBs in the sample extracts and the complexity of the matrix. The parameters for analysis via CGC/LMS and CGC/ MS-SIM are presented in Tables 6 through 13. 23 �SECTION 4 METHOD VALIDATION PREPARATION OF ANALYTICAL METHODS Analytical methods were prepared for the analysis of by-product PCBs in: * Commercial products and product wastes (Appendix B). * Air (Appendix C). * Industrial wastewater (Appendix D). The analysis of commercial products and product wastes was covered in one method since the diversity of matrices in both categories dictates the same generalized approach. Air was defined to include stack gases, fugitive emissions, and static (room, other container, or outside) air. Commercial Products and Product Wastes Method The objective was to devise an analytical method suitable for enforcement of the regulation concerning by-product PCBs in commercial products and product wastes. A detailed rationale for selection of the techniques used in the method may be found in a separate report (Erickson and Stanley, 1982). Sample Workup-The general approach taken with sample preparation (collection, preservation, extraction, and cleanup) was to provide a framework within which any reasonable technique could be used. This is the only acceptable approach to a method designed to cover "any" matrix. , The use of 13C-labeled recovery surrogates in conjunction with GC/EIMS was judged to be the most suitable approach (Erickson and Stanley, 1982; Stanley and Erickson, 1982; Roth et al., 1982). Using the recovery surrogates, any losses of PCBs would be detected and could be corrected for in the calculation of the PCB concentration. 24 �When surrogates are not fully incorporated into the matrix, their recovery will not be representative of the analyte PCB recoveries and recovery assessment will not be possible. It is incumbent upon the analyst to recognize this problem and use good scientific judgment with samples that present a potential problem. Nonextractable solid polymers may be an example of a matrix presenting incorporation problems. PCB Determination-As discussed elsewhere (Erickson and Stanley, 1982; Stanley and Erickson, 1982) , GC/EIMS appears to be the only acceptable general technique for determining PCBs in commercial products and product wastes. The use of either capillary or packed column GC is permitted. While strong arguments are presented for both techniques (Stanley and Erickson, 1982), the analytical results should be comparable for both techniques provided proper instrument calibration and operation, analytical, and quality control procedures are followed as described in the analytical methods. Quantitation-The analytical objective of these methods is to determine if the sample contains quantifiable PCBs and, if so, at what concentration. On the assumption that a general knowledge of the congener distribution is important, reporting of the concentration by homolog is proposed in the reporting form. Since a "total PCB" value is also important for summary and comparative purposes, space for this value is also provided on the reporting form. Other reporting formats, including "largest isomer or resolvable peak" or "all peaks greater than a regulatory value," may be easily accommodated using different tabulations and reporting procedures. The PCB concentrations found may be lower than the actual value due to nonquantitative recovery during extraction or cleanup. The measured recoveries of the surrogates may be used to derive a corrected concentration. The analyst must take care that the surrogates are thoroughly incorporated into the matrix prior to extraction, as discussed above. The analyst must also guard against improper corrections because of errors in surrogate quantitation. These errors may arise from background interferences. A more thorough discussion of quantitation options is presented in a previous report (Erickson and Stanley, 1982). Air Method The sample collection, preservation, extraction, and cleanup aspects were taken from the work of Haile and Baladi (1977). The determination, using GC/ EIMS, is identical to that in the commercial products and product wastes method except that recovery surrogates are not used. Wastewater Method The water method is a direct modification of the commercial products and product wastes method. As noted in this method, the cleanup and extraction procedures for EPA Methods 608 (U.S. EPA, 1979b) and 625 (U.S. EPA, 1979a) may be used. It is anticipated that, unless conditions dictate otherwise, most analysts will choose this option. 25 �Quality Control Each method includes a strong quality control (QC) section. Given the complexity of the matrices and complexity of the analyte (209 compounds), the need for QC is evident. The various aspects of the QC section were designed assuming a reasonably large (10 to 100) batch of samples. For small batches of samples, the percentage of effort spent on QC can become sizeable. Alternate Methods The methods presented here are intended to be primary methods capable of generating the best quality data technologically feasible. The development and acceptability of secondary (alternate, equivalent, or screening) methods is not addressed in this report. GAS CHROMATOGRAPHY/MASS SPECTROMETRY OF PCBs Analysis for PCBs requires the use of selected representative standard compounds since all 209 congeners are not available. One of the major disadvantages of many instrumental methods for PCB analysis is the large variance of the instrumental response factors for PCB congeners, both within a homolog and between homologs. These large differences in response factors create problems in selecting representative compounds for quantitation purposes. The response factors of 77 of the possible 209 PCB congeners measured by GC/EIMS are presented in Tables 16 and 17. The data suggests that the EIMS response factor variance among PCB congeners is small relative to other detectors such as the electron capture detector or negative chemical ionization mass spectrometry. Relative Response Factors Quadrupole Mass Spectrometer-The relative response factors (RRF) of the 77 PCB congeners were determined with the Finnigan 4023 quadrupole mass spectrometer as discussed in the experimental section. The RRFs were determined two ways to assess the effects of instrumental variability. The replicate RRF determinations are the average of four replicate analyses for each of the PCB congeners, all determined on a single day to assess the variability of the measurement. The single RRF determinations are single values from an experiment in which all 14 solutions containing all 77 congeners were run on one day to minimize instrumental variability with time. The data are presented in Appendix A. The RRFs vary from approximately 0.2 for decachlorobiphenyl to 4.1 for 2-chlorobiphenyl. Figures 1 and 2 present a visual comparison of average replicate and single RRFs of PCB congeners determined as replicate measurements and as single measurements . 26 �4.5 4.0 Quadrupole Mass Spectrometer 3.5 3.0 ! — 2 o 2.5 cd M-l <1> CO 0 2.0 . CO 0) to CO eu | c I. O »PH 1.0 1 2 7 3 0.5 J_ 3 I 4 5 6 7 Homolog (degree of chlorination) 10 Figure 1. Plot of average response factor versus homolog for 77 PCB congeners. Each average is the mean response per congener, i.e., mean of four replicates with the Finnigan 4023 quadrupole mass spectrometer. This plot indicates the number of data points that overlap for specific isomers. �3.Or— 1 Quadrupole Mass Spectrometer 2.5 2.0 o 0} 1.5 N5 00 1.0 4 2 1 2 1 2 2 1 7 0.5 - _L _1_ _L JL X 3 4 5 6 7 10 Homolog (degree of chlorination) Figure 2. Plot of response factor per isomer versus homolog for 77 PCB congeners, determined on a single day. Each value is representative of single measurements of each congener with the Finnigan 4023 quadrupole mass spectrometer. This plot indicates the number of data points that overlap for specific isomers. �Table 16 is a summary of the RRF data, where the replicate and single measurements are averaged over all measured isomers for a homolog. The relative standard deviation (Table 16) for the replicate measurements reflects the variance of the average RRF for each isomer within a homolog. The absolute area of the internal standard, Congener No. 210, varied by only 4.4% for all solutions during the single day experiment, as compared to 9.9% for the 7 days required to complete the replicate analyses. The relative standard deviations based on the four replicate analyses for each of the PCB congeners, ranged from 0.4 to 9.1%, indicating the reproducibility of the injection for each solution. The average response factors from replicate determinations and single measurements were subjected to a Student's t-test to determine if there were any significant differences in measured response factors. No significant difference was found for the average response factor values for any of the PCB homologs except the heptachlorobiphenyl isomers. A more detailed presentation of the Student's t-test for these values is presented in Table A-2 of Appendix A. A solution of 3,3',4,4'-tetrachlorobiphenyl-de (Congener No. 210) and Solution No. 1 (Table 2) were both analyzed daily. The solution of Isomer No. 210 was used to tune the quadrupole mass spectrometer to the desired working conditions. Solution No. 1 was used to determine fluctuations of response factors from day to day due to differences in instrumental operating parameters. Table 17 presents the data for single day replicate measurements and day-to-day determination of the response factors for the PCB congeners in Solution No. 1. The relative standard deviations calculated for the single day measurements are considerably lower than the relative standard deviations from day-to-day analyses. This is a reflection of the reproducibility on the part of the operator as well as of the stability of the quadrapole mass spectrometer system on a given day. The relative standard deviation calculated for day-to-day analyses is indicative of the variation that might be expected for routine analysis of PCBs. A Student's t-test of the Solution No. 1 data (Table 17) indicated that there are significant differences in response factors from day to day compared to single day measurements for PCB Congener Nos. 1, 11, 29, and 207. A more detailed presentation of this t-test is presented in Table A-3 of Appendix A. Magnetic Sector Mass Spectrometer— The RRFs for the 77 PCB congeners were also determined with a Varian MAT 311A double focusing magnetic sector mass spectrometer. The RRF values were determined by single measurements of all congeners on a single day. The data are presented in Appendix A and summarized in Figure 3. Extrapolation of Response Factor Data to All Congeners-Since all 209 PCB congeners were not available for determination of RRFs, it was necessary to extrapolate the average RRF data to project the range of response factors that might be encountered. This extrapolation was based on the assumption that the number of measured isomers (n) are a representative sample of the entire set of the possible isomers (N). Thus it was assumed that the mean for the measured isomers (n) is an unbiased estimate of the mean for the possible isomers (N). 29 �TABLE 16. AVERAGE RELATIVE RESPONSE FACTORS (RRF) FOR 77 COMMERCIALLY AVAILABLE PCB CONGENERS MEASURED OVER SEVERAL DAYS AS FOUR REPLICATES EACH AND RRF FOR SINGLE MEASUREMENTS OF ALL CONGENERS IN A SINGLE DAY PCB homolog No. of isomers RRF from replicate measurements Relative standard deviation ( ) % RRF from single b measurement Relative standard deviation ( ) % 3 3.331 19.3 2.739 9.3 Dichloro- 10 2.027 22.0 2.048 15.7 Trichloro- 9 1.573 21.7 1.592 18.1 Tetrachloro- 16 0.950 18.4 0.946 20.0 Pentachloro- 12 0.720 16.7 0.725 17.6 Hexachloro- 13 0.513 15.1 0.500 19.1 Heptachloro- 4 0.361 6.6 0.308 8.0 Octachloro- 6 0.253 11,9 0.224 17.3 Nonachloro- 3 0.229 14.7 0.188 16.2 Decachloro- 1 0.213 2.8 0.179 Monochloro- - a Four replicate measurements of the RRF were made for each isomer. For example, the three monochlorobiphenyl isomers were measured four times each. Hence, the RRF and relative standard deviation ( ) were calculated from 12 distinct % values. b A single measurement for each of the 77 PCB congeners was completed in a single day. Hence, the RRF reported is the average of one measured RRF for each isomer within a homolog. For example, the RRF and relative standard deviation ( ) reported for the monochlorobiphenyls were calculated from three distinct % values. 30 �TABLE 17. AVERAGE RELATIVE RESPONSE FACTORS (ERF) FOR PCB CONGENERS IN SOLUTION 1 MEASURED AS REPLICATES ON A SINGLE DAY AND AS SINGLE MEASUREMENTS FOR DAY-TO-DAY BASIS3 Congener no. Single day measurements Relative std. Std. deviation ( ) % RRF deviation £ Day-to-day measurements Std. Relative std. RRF deviation deviation ( ) % 1 0.118 2.905 3.544 0.452 12.767 11 3.073 0.073 2.363 2.733 0.300 10.977 29 2.195 0.048 2.188 2.005 0.171 8.535 47 1.062 0.059 5.591 1.032 0.061 5.876 121 0.948 0.020 2.127 0.955 0.036 3.747 136 0.689 0.016 2.336 0.685 0.046 6.688 181 0.383 0.009 2.379 0.377 0.028 7.347 195 0.263 0.003 1.184 0.270 0.022 8.304 207 0.237 0.008 3.547 0.257 0.030 11.757 209 a 4.073 0.213 0.006 2.837 0.223 0.023 10.352 See Tables 6 and 8 for CGC/EIMS operating conditions. b These values calculated from four replicates. c These values calculated from 11 separate analyses. 31 �2.50 Magnetic Sector Mass Spectrometer 2.00 1.50 to C N3 O p. CO 0) 1.00 - 0.50 - 10 Homolog (degree of chlorination) Figure 3. Plot of response factor per isomer versus homolog for 77 PCB congeners, determined on a single day. Each value is representative of single measurements of each congener with the Varian Fat 311A magnetic sector mass spectrometer. This plot indicates the number of data noints that overlap for specific isomers. �Table 18 presents the upper and lower 95% confidence limits for the measured average RRFs. The extrapolation was necessary for the dichloro- through octachlorobiphenyl homologs. The projected upper and lower limits of the average RRF ranged from 13% for each PCB homolog for trichlorobiphenyls to approximately 6.5% for the dichlorobiphenyls. The projected ranges for the tetrachloro- to octachlorobiphenyls were between these values. Comparison of Magnetic Sector and Quadrupole RRF Data-The two instruments used operate on entirely different principles, so the results may represent the range of RRFs to be expected from these compounds on different instruments. Table 19 presents a summary of the data. As expected, the RRF trends are much different. Since quadrupole spectrometers discriminate at the high masses, the RRFs for high homologs (higer masses) are much lower than corresponding values for the magnetic detector spectrometer. A statistical analysis of the data (Student's t-test presented in Table 4 of Appendix A) confirmed that the average RRFs are significantly different for many of the homologs. However, the relative standard deviations for the average RRF of each homolog are not significantly different. Thus, the extrapolation from a single calibration isomer to all isomers of a homolog should have similar precision for the two instrument types. Relative Retention Times Relative retention times (KRT) were also calculated from the data generated for relative response factor measurements with both the quadrupole and magnetic sector mass spectrometer instruments. All RRTs for each PCB congener were calculated versus 3,3',4,4'-tetrachlorobiphenyl-de- Figure 4 is a plot of the RRT data versus PCB homolog. All data points for the 77 PCB congeners measured with the quadrupole mass spectrometer are presented. This plot also indicates that the relative retention window for the dichloro- to octachlorobiphenyl homologs may be larger than that actually measured if more of the possible congeners were present. Table 20 presents the observed range of RRTs for the 77 PCB congeners and additional congeners, identified only by homolog, in an Aroclor mixture (1016, 1254, 1260). These RRTs were established using a 15-m fused silica DB-5 capillary column. It must be recognized that the RRT windows on other columns may be substantially different. Table 20 also presents a projected RRT window for PCB anaysis. The overlap of the retention windows of each homolog must be considered in establishing an instrumental analysis approach to quantitation of the specific PCB homologs. This consideration has been accounted for in the GC/MS requirements for PCB analysis in Appendices B to D. The relative retention times of the 77 PCB congeners as determined with both the quadrupole and magnetic sector mass spectrometers are presented in tabular form in Appendix A. 33 �TABLE 18. MEASURED AVERAGE RELATIVE RESPONSE FACTOR (RRF) AND CORRESPONDING UPPER AND LOWER 95% CONFIDENCE LIMITS PCB homolog Monochloro- No. of possible isomers (N) Average measured response RRF No. of available isomers (n) Sample std. deviation (S) T Lower a limit T, Upperb limit - 3 3 3.331 0.643 Dichloro- 12 10 2.027 0,447 1.896 2.158 Trichloro- 24 9 1.573 0.341 1.366 1.780 Tetrachloro- 42 16 0.950 0.175 0.877 1.023 Pentachloro- 46 12 0.720 0.120 0.654 0.786 Hexachloro- 42 13 0.513 0.078 0.474 0.552 Heptachloro- 24 4 0.361 0.024 0.326 0.396 Octachloro- 12 6 0.253 0.030 0.231 0.275 Nonachloro- 3 3 0.229 0.034 Decachloro- 1 1 0.213 a Lower 95% limit = RRF -S("'j1* +-c 1 n \ \. N 34 - - �TABLE 19. RELATIVE RESPONSE FACTORS MEASURED VERSUS 3,3',4,4'-TETRACHLORO BIPHENYL-d6 BY ELECTRON IMPACT MASS SPECTROMETRY QUADRUPOLE (FINNIGAN 4023) AND MAGNETIC SECTOR (VARIAN MAT 311A) INSTRUMENTS No. of isomers measured RRF Quadrupole_ Mean RSD" ( ) % Magnetic sector RSDa ( ) % Mean 3 2.739 9.3 2.329 8.5 Dichloro- 10 2.048 15.7 1.663 13.8 Trichloro- 9 1.592 18.1 1.167 21.3 Tetrachloro- 16 0.946 20.0 0.902 14.0 Pentachloro- 12 0.725 17.6 0.780 17.4 Hexachloro- 13 0.500 19.1 0.640 19.4 Heptachloro- 4 0.308 8.0 0.497 12.1 Octachloro- 6 0.224 17.3 0.463 15.3 Nonachloro- 3 0.188 16.2' 0.467 22.5 Decachloro- 1 0.179 - 0.586 PCB homolog Monochloro- a Relative standard deviation. 35 - �H2<-'10 Relative Retention Times of PCB Congeners by Homolog Versus 3. 3'. 4. 4' Tetrachlorobiphenyl-d, C]2H,C19 2M2 7 ° 204 202 2OO C12H2CI8 195 194 185 171 183 181 C12H3CI7 PCB homologs 198 ** 155 C12H4CI6 154 139 153 138 129 136 15t 143 141 137 128 156 97 . „ 121 II6 104 103I009388 101119 115 C12H5CI5 15 545053 Cl2H6CI4 30 C12H7CI3 10 4 C12H8CI2 1 C,2H9CI, 9 7 8 5 14 5247 44 72 -«061 66 6577 2631 33 29 28 21 18 24 lt'^15 2 3 1 1 0.40 0.50 1 0.60 1 1 1 1 0.70 0.80 0.90 1.00 1 1.10 1 1.20 1 1.30 1 1.40 Relative retention time Figure 4. Retention times of 77 PCB congeners relative to 3,3',4,4'-tetrachlorobiphenyl-d6 (RRT of 1.00) The dashed line indicates that not all of the possible isomers of a particular homolog were measured. Relative retention times were determined on a J&W DB-5, 15-m fused silica column in a Finnigan 4023 GC/EIMS system. Temperature program: 110°C for 2 min,. then 10°C/min to 325°C. �TABLE 20. RELATIVE RETENTION TIME (RRT) RANGES OF PCB HOMOLOGS VERSUS d6-3,3',4,4'-TETRACHLOROBIPHENYL PCB homo log No. of isomers measured Observed range of RRTsa Calibration solution Congener Observed no. RRT3 Projected range of RRTsD 3 0.40-0.50 1 3 0.43 0.50 0.35-0.55 Dichlorobiphenyl 10 0.52-0.69 7 0.58 0.35-0.80 Trichlorobiphenyl 9 0.62-0.79 30 0.65 0.35-1.10 Tetrachlorobiphenyl 16 0.72-1.01 50 0.75 0.55-1.05 Pentachlorobiphenyl 12 0.82-1.08 97 0.98 0.80-1.10 Hexachlorobiphenyl 13 0.93-1.20 143 1.05 0.90-1.25 Heptachlorobiphenyl 4 1.09-1.31 183 1.15 1.05-1.35 Octachlorobiphenyl 6 1.19-1.36 202 1.19 1.10-1.50 Nonachlorobiphenyl 3 1.31-1.42 207 1.33 1.25-1.50 Decachlorobiphenyl 1 1.44-1.45 209 1.44 1.35-1.50 Monochlorobiphenyl a The RRTs of the 77 congeners and a mixture of Aroclor 1016/1254/1260 were measured versus de-3,3',4,4'-tetrachlorobiphenyl (internal standard) using a 15-m J&W DB-5 fused silica column with a temperature program of 110°C for 2 min, then 10°C/min to 325°C, helium carrier at 45 cm/sec, and an on-column injector. A Finnigan 4023 Incos quadrupole mass spectrometer operating with a scan range of 95-550 Daltons was used to detect each PCB congener. b The projected relative retention windows account for overlap of eluting homologs and take into consideration differences in operating systems and lack of all possible 209 PCB congeners. 37 �Selection of Congeners for a Calibration Standard The data generated from the RRF and RRT measurements were used to select the PCB congeners for an analytical quantitation/calibration standard for GC/EIMS analysis of PCBs. Selection of the standard compounds was based primarily on the ratio of the measured response factor to the average response factor for a particular homolog. The PCBs with KRFs closest to the average values were selected as standard compounds. In addition, the RRT was considered to assure that the selected PCB congeners did not coelute. Two monochlorobiphenyls were selected for the calibration standard because the average RRF and RRT did not clearly coincide with any of the three possible isomers. One isomer (2-chlorobiphenyl) had a substantially different RRF. This isomer was quantitated separately. 4-Chlorobiphenyl was selected as the calibration isomer for the two remaining isomers. Figure 5 is a CGC/EIMS chromatogram of the 11-component PCB calibration standard. The composition of this solution is identified in Tables 4 and 20 along with the observed RRT of each of the 11 congeners. VALIDATION OF SELECTED CLEANUP STEPS As part of the overall method validation, several of the cleanup techniques were validated. A mixture of the 11 calibration standard congeners and three recovery surrogates (the 13C-octachlorobiphenyl was unavailable for these experiments) was diluted in an appropriate solvent and then subjected to the cleanup procedures as described in Appendix B. After the cleanup, the internal standard was added and the volume adjusted. The samples were analyzed by CGC/EIMS using a quadrupole spectrometer operated under the condition listed in Tables 6 through 8. Data were collected in the full scan mode and quantitated using the primary ions listed in Table 10 and the congener pairs listed in Table 13. A blank was run through the procedure alongside the recovery spikes. As expected, no PCBs except the internal standard were observed in the blanks. The results for the 11 calibration congeners were calculated as percentage recovery. Tables 21 through 25 present the uncorrected recoveries, calculated using Equation 12-1 of Appendix B, using the internal standard (Congener No. 210); the actual percentage recoveries of the 13C-labeled recovery surrogates, calculated using Equation 12-2 of Appendix B; and the corrected recoveries of the calibration congeners, calculated using Equation 12-3 of Appendix B. Inspection of Tables 21 to 25 reveals that the accuracy of the corrected recoveries is higher than for the uncorrected recoveries (104% versus 77% average). On the other hand, the precision of the uncorrected recoveries is slightly higher than for the corrected recoveries (11% versus 9% relative standard deviation average). This is the expected trend since the uncorrected recovery relies on two GC/MS measurements (area of the PCB congener peak and area of the internal standard peak) and the corrected recovery relies on those two values and the area of the surrogate peak. Thus, these results indicate that accuracy is improved by recovery correction, at a sacrifice of precision. 38 �T3 4-1 ^-v rH Cfl cn oi cd C 00 0 Ol rJ •U G s v 100.0 U co -' ^£3 O rH rH C C 01 X! D. •H XI 01 XI tX 'H XI 1 i—1 C C Ol Ol XI XI a ex •rH XI Xi ^-s O O 0) M rJ 0 0 CO rH rH X! XI U 0 O 0 4J rH XI U 00 0 CO rJ rJ rJ cx -H ,Q 0 t-i o rH XI O 0 C £ T CM H ^ c rH ^*> c Ol XI ex •H XI o o Ol XI ex •H XI o ^1 o 01 01 H X! rH <- ^j- XI K**I rj Ol XI o o <J- o o X •H }-l x; o cfl a o C oi ex •H Xi o c Ol ex ^ •H XI O r4 0) •U CO 00 o o x; o cfl o Ol p rH )-l O 3 Cfl CD O "--' Ol P " CO CO CO vO LO cfl CM •V c • ^ CM ^ fc XI 33 1 4_) rH rH vD vO cfl X X! O r, s^- o t-l o vO r^ CO ex -H XI O 0) rO •\ CO 0 1 4-> O a r, rH 01 ex •H x> o X! O cfl 01 XI a cfl 4-1 m ex ^ ai m r4 O rH XI <t" ». }-t H C Ol XI •rH •\ r> •. ex ro XI 0 cfl VJ ex »i •\ •H XI CO rH rH o •H Q o H 1 vD 1 •s I. CM ^ <J- " ^ CM 1316 w~ rv i CM CO m CN „ -o- 0> CN 4- CM J I CO l I(V1 i ivi 1178 CM 798 936 , » 1 6K3 c 4J X! CN 618 iH ^» ex iH 01 ,£ rH XI U Cfl r4 4J 01 H 1 •H C75 ^ c x; ex •H Xi o C CD <r sr u C c 01 rH C C 3 o o w S S s-' me 1 o o o'f & "rH 112592. rH rH rH 1101 —' T3 CO LO VO 3 to •H C'JO 13:20 [ ^^. 1030 16:49 1 — 12^0 20 :&) SCA'l Figure 5. Capillary gas chromatography/electron impact ionization mass spectrometry (CGC/EIMS) chromatogram or the calibration standard solution required for quantitation of PCBs by homolog, This chromatogram includes PCBs representative of each homolog, three ^C-iabeled surrogates, and the deuterated internal standard. The concentration of all components and the CGC/EIl^S parameters are presented in Tables 4, 5, 6 and 9. �TABLE 21. RECOVERY DATA FOR ACID CLEANUP' Congener no. -P- o PCB homolog 1 3 7 30 50 97 143 183 202 207 209 X Standard deviation Relative standard deviation ( ) % Monochlorobiphenyl Monochlorobiphenyl Dichlorobiphenyl Trichlorobiphenyl Tetrachlorobiphenyl Pentachlorobiphenyl Hexa chlo rob ipheny 1 Heptachlorobiphenyl Octachlorobiphenyl Nona chlo rob ipheny 1 De ca chlo rob ipheny 1 211 212 214 X Standard deviation Relative standard deviation ( ) % 13 Ce-nionochlorobiphenyl Ci2~tetrachlorobiphenyl 13 Ci2~o!ecachlorobiphenyl 13 a 0.52 0.50 0.52 0.52 0.76 0.87 0.96 1.30 2.30 2.50 2.10 2.60 5.30 10.20 Spike No. 1 not analyzed. b Total spike level (pg) Corrected via surrogate response. c Not detected. d Large background signal prevented quantitation of the compound. e Not applicable. Spike 2 ( recovery) % Corrected Uncorrected 100.0 83.4 82. £ NQ3 78.0 99.5 81.2 85.9 80.7 83.2 87.3 86.2 7.5 9 70.2 87.1 91.3 82.9 11.2 13 142.4 118.8 117.5 NQ 89.6 114.2 93.2 98.5 88.4 91.1 95.7 104.9 17.7 17 _ — - Blank NDC ND ND ND ND ND ND ND ND ND ND ND ND ND - �TABLE 22. Congener no. RECOVERY DATA FOR FLORISIL COLUMN CLEANUP PCB homo log 1 3 7 30 50 97 143 183 202 207 209 X Standard deviation Relative standard deviation ( ) % Monochlorobiphenyl Monochlorobiphenyl Dichlorobiphenyl Trichlorobiphenyl Tetrachlorobiphenyl Pentachlorobiphenyl Hexa chlo r ob ipheny 1 Heptachlorobiphenyl Octachlorobiphenyl Nonachlorobiphenyl Decachlorobiphenyl 211 212 214 X Standard deviation Relative standard deviation ( ) % 13 Total spike level (pg) 0.52 0.50 0.52 0.52 0.76 0.87 0.96 1.30 2.30 2.50 2.10 Spike 1 ( recovery) % Corrected" Uncorrected 57.9 63.0 66.0 69.4 70.7 73.4 72.6 76.6 77.8 78.1 77.7 71.2 6.7 9 90.6 98.6 103.2 160.5 163.6 169.7 168.1 177.2 102.5 102.9 102.4 130.9 35.8 27 Spike 2 ( recovery) % Corrected Uncorrected 54.9 58.3 60.0 62.3 62.4 66.1 67.0 72.3 72.3 70.5 72.8 65.4 6.2 10 _ C6-monochlorobiphenyl 13 C12-tetrachlorob ipheny 1 13 C12~decachlorobiphenyl a Corrected via surrogate response, b Not detected, c Not applicable. 2.60 5.30 10.20 63.9 43.2 75.9 61.0 16.5 27 - 57.6 47.9 69.6 58.4 10.9 19 95.4 101.2 104.4 130.0 130.3 138.1 140.1 151.0 103.8 101.3 104.5 118.2 19.8 17 - Blank NDb ND ND ND ND ND ND ND ND ND ND c ND ND ND - �TABLE 23. RECOVERY DATA FOR FLORISIL SLURRY CLEANUP Congener no. PCB homolog 1 3 7 30 50 97 143 183 202 207 209 X Standard deviation Relative standard deviation ( ) % Mpnochlorobiphenyl Monochlorobiphenyl Dichlorobiphenyl Trichlorobiphenyl Tetrachlorobiphenyl Pentachlorobiphenyl Hexachlorobiphenyl Hep ta chlorob iphenyl Octachlorobiphenyl Nona chlo r ob iphenyl Decachlorobiphenyl 211 212 214 X Standard deviation Relative standard deviation ( ) % 13 Total spike level (pg) 0.52 0.50 0.52 0.52 0.76 0.87 0.96 1.30 2.30 2.50 2.10 Spike 1 ( recovery) % Corrected" Uncorrected 80.5 81.2 87.5 NQC 90.0 96.0 95.5 95.1 97.2 95.1 96.2 91.4 6.3 7 96.0 96.8 104.4 NQ 91.6 97.6 97.2 96.8 91.2 89.4 90.4 95.1 4.5 5 Spike 2 ( recovery) % Corrected Uncorrected 71.1 72.7 75.0 76.4 80.1 83.5 82.0 79.8 88.8 87.6 83.7 80.1 5.8 7 _ C6-monochlorob iphenyl 13 C12-tetrachlorobiphenyl 13 C12~decachlorobiphenyl 2.60 5.30 10.20 83.9 98.3 106.5 92.5 7.5 8 a Corrected via surrogate response. b Not detected. c Large background signal prevented quantitation of this compound. d Not applicable. - 92.9 94.8 98.1 85.5 89.6 93.5 91.6 89.3 101.0 99.5 95.2 93.7 4.7 5 Blank NDb ND ND ND ND ND ND ND ND ND NDd - _ 76.7 89.4 87.9 84.7 6.9 8 - ND ND ND - �TABLE 24. RECOVERY DATA FOR KOH CLEANUP Congener no. 1 3 7 30 50 97 143 183 202 207 209 X PCB homolog Monochlorobiphenyl Mono chlo rob iphenyl Dichlorobiphenyl Trichlorobiphenyl Tetrachlorob iphenyl Pentachlorob iphenyl Hexachlorob iphenyl Heptachlorob iphenyl Octachlorob iphenyl Nona chl o rob iphenyl Decachlorob iphenyl Total spike level ( j ) |g 0.52 0.50 0.52 0.52 0.76 0.87 0.96 1.30 2.30 2.50 2.10 Standard deviation Relative standard Spike 1 ( recovery) % Uncorrected 60.2 69.0 73.5 75.0 79.7 85.8 84.0 81.2 89.2 88.2 69.9 77.8 9.1 12 Corrected" 82.6 94.6 100.8 83.5 88.7 95.4 93.4 90.3 113.0 111.8 88.6 94.8 10.2 11 Spike 2 ( recovery) % Uncorrected Corrected 67.7 73.6 77.5 77.6 80.7 85.0 85.0 81.3 89.3 87.6 71.9 79.7 6.8 9 90.1 98.0 103.3 89.2 92.9 97.7 97.7 93.5 117.5 115.3 94.6 99.1 9.5 10 Blank NDb ND ND ND ND ND ND ND ND ND NDC - deviation ( ) % _ 211 212 214 X Standard deviation 13 C6-monochlorobiphenyl 13 C12-tetrachlorob iphenyl 13 C12-decachlorob iphenyl Relative standard deviation ( ) % a Corrected via surrogate response, b Not detected, c Not applicable. 2.60 5.30 10.20 72.9 89.9 78.9 80.6 8.6 11 - _ 75.1 87.0 76.0 79.4 6.6 8 - ND ND ND - �TABLE 25. RECOVERY DATA FOR ALUMINA CLEANUP Congener no. PCB homolog 1 3 7 30 50 97 143 183 202 207 209 X Standard deviation Relative standard deviation ( ) % Mono chlo rob iphenyl 211 212 214 X Standard deviation 13 Monochlorobiphenyl Dichlorob iphenyl Trichlorob iphenyl Tetrachlorob iphenyl Pentachlorobiphenyl Hexachlorob iphenyl Heptachlorobiphenyl Octachlorob iphenyl Nona chlorob iphenyl Decachlorob iphenyl Total spike level (|Jg) 0.52 0.50 0.52 0.52 0.76 0.87 0.96 1.30 2.30 2.50 2.10 Spike 1 ( recovery) % Corrected Uncorrected 63.1 60.0 67.9 NQC 67.2 70.4 69.4 75.8 76.8 77.3 74.0 70.2 5.9 8 97.1 92.2 104.8 NQ 97.2 101.9 100.4 109.7 92.2 92.9 88.9 97.8 6.5 7 Spike 2 ( recovery) % Uncorrected 61.1 58.4 66.4 NQ 66.3 68.3 67.5 75.1 75.3 76.8 78.3 70.1 6.9 10 13 Cj2~tetrachlorob iphenyl 13 C12~decachlorobiphenyl Relative standard 2.60 5.30 10.20 64.8 69.1 83.2 72.4 9.6 13 deviation ( ) % a Corrected via surrogate response. b Not detected. c Large background signal prevented quantitation of this compound. d Not applicable. - Blank 101.0 96.2 109.4 NDb ND ND ND ND ND ND ND ND ND NDd NQ 102.2 105.4 104.2 115.8 89.5 91.2 93.0 100.8 8.4 8 - _ _ C6-monochlorob iphenyl Corrected 60.7 64.9 84.2 69.9 12.5 18 - ND ND ND - �The preliminary data presented here contain an apparent anomaly: the low recovery of the 13C-tetrachlorobiphenyl surrogate (Congener No. 212) from the Florisil column cleanup. These two data points contribute substantially to the imprecision of the surrogate recoveries and induce some very high (130 to 177%) corrected recoveries for the tri- through hepta- compounds. The experiment should be repeated. VALIDATION OF THE PRODUCT AND PRODUCT WASTE METHOD WITH INDUSTRIAL SAMPLES Strategy Selected samples, obtained from industrial sources, were subjected to a variety of sample preparations as listed in Table 15 and then analyzed by CGC/EIMS. This section presents the results of this preliminary validation and, where possible, compares our values with those of previous analyses of the same sample. The results for quality control samples are also reported. The most extensively studied matrix was the CMA-A chlorinated benzene waste stream sample. This particular sample was chosen because of the wide distribution of PCB homologs (mono- through decachlorobiphenyls). Sample preparation with this matrix included simple dilution, treatment with sulfuric acid, Florisil, and saponification with ethanolic potassium hydroxide. The CMA-A samples were analyzed in duplicate in two sets of experiments. The 11 PCB congeners used for calibration purposes were spiked into the CMA-A matrix for standard addition experiments. Blind spiked samples and quantitation standards, prepared by the MRI quality control personnel as analytical performance checks, were analyzed along with the other samples. First Sample Set Tables 26 and 27 present the uncorrected and corrected concentrations found for CMA-A samples in preliminary studies of the application of the proposed methods for commercial products and product wastes. Sample 10 was analyzed without surrogates to approximate the analytical procedure used by most other laboratories. As anticipated, the uncorrected values compare well with 20A and 20B, while the corrected values are slightly lower than the values for 10. Both corrected and uncorrected values for the duplicate samples 20A and 20B are in agreement. The values for samples 10, 20A, and 20B average about 400 pg/g. These values are higher than the mean of 280 JJg/g reported in the CMA round robin but are in good agreement with the values (402 |Jg/g) reported by the sample supplier (Appendix E of Pittaway and Horner, 1982). The homolog distribution of our data agrees in general with the CMA data and the data that accompanied the samples. Sample 110 (CMA-E) was determined to contain about 18 (Jg/g PCB (Table 28) mostly as the dichloro homolog. These results are slightly higher than the CMA round robin data, which had a mean reported value of 9 pg/g. The isomer distribution agrees with most of the CMA round robin data (Pittaway and Horner, 1982). 45 �TABLE 26. Congener no. 1 3 7 30 50 97 143 183 202 207 209 UNCORRECTED PCB CONCENTRATIONS (Mg/g) IN CMA-A SAMPLES PCB homolog 10 Dilution, no surrog. 20A Dilution 20B Dilution 1 1 2 3 4 5 6 7 8 9 10 9 19 64 55 60 50 56 60 0 0.9 9.3 11 21 70 52 63 40 48 84 0 0 20 10 19 64 49 55 36 38 68 0 0 20 408 358 96b 108 154 94 97 152 414 Total 211 212 214 1 4 10 a No surrogates added. b NSa NS NS Surrogate recovery (percent). 46 �TABLE 27. CORRECTED PCS CONCENTRATIONS (iig/g) IN CMA-A SAMPLES Congener no. 1 3 7 30 50 97 143 183 202 207 209 PCB homo log 10 Dilution, no surrog. 20A Dilution 20B Dilution 1 1 2 3 4 5 6 7 8 9 10 NSa NS NS NS NS NS NS NS NS NS NS 11 22 73 49 58 37 44 78 0 0 13 11 21 68 50 57 37 39 70 0 0 13 385 366 Total a No surrogates added. 47 �Second Sample Set CMA Product Waste Samples— The corrected and uncorrected concentrations of the PCB homologs for duplicate CMA-A samples from a more extensive study are presented in Tables 29 and 30. Sample 2005 was spiked only with the internal standard so that any interferences corresponding to the 13C-labeled PCBs could be measured. Samples 2010 and 2020 are duplicate samples of CMA-A. The four surrogate compounds were added to approximately 0.1 g of each sample. The mixture was diluted to 1.0 ml and the internal standard added. Sample 2025Q is a sample that was submitted for PCB analysis by the MRI quality control department. This sample was weighed by QC personnel and the final preparation completed as described for the previous samples. The MRI QC coordinator calculated the final concentration for 2025Q from the extract concentration of each PCB homolog and weight of the CMA-A sample recorded in the QC laboratory record book. The surrogate-corrected values reported for samples 2010 through 2025Q are in good agreement with the total PCB concentration and homolog distribution reported in the CMA round robin (Pittaway and Horner, 1982). Tables 31 and 32 present the data from a standard addition experiment with the CMA-A sample matrix. The 11 PCB congener calibration standard was added to three separate aliquots of the CMA-A matrix to give spike levels ranging from approximately 20 to 100 | g of the monochlorobiphenyl and 50 to j 200 (Jg of decachlorobiphenyl. Samples 2030, 2040, and 2050 were prepared in the analytical laboratory. Sample 2060Q was prepared as a blind spike of the CMA-A matrix by MRI quality control personnel. The uncorrected amount found did not increase linearly with the spike level. In fact, at the highest spike level (Sample 2050) the amounts found for each homolog were less than the spike. No explanation is immediately available for this data trend, although the low recoveries of the 13C-octa- and tetrachlorobiphenyl surrogates indicated that the data are at best marginally valid. Tables 33 and 34 present data for CMA-A samples that were subjected to three different cleanup methods (concentrated H2S04, Florisil column chromatography, and saponification with alcoholic KOH). The data from the sulfuric acid cleanup was difficult to interpret because of interferences. As noted previously (Erickson and Stanley, 1982), the acid cleanup results in large losses of lower chlorinated PCB homologs. The poor recoveries of the surrogates shown in Table 33 are clearly outside of the QC criteria in Section 14.2.2 of Appendix B and indicate that the analyses are invalid. These results would not be reported as analyses for compliance with the proposed regulation. All of the blank samples (2001, 1080, 2100, and 2120) were analyzed along with the sample discussed above and found to contain no detectable PCBs. 48 �TABLE 28. UNCORRECTED AND CORRECTED PCS CONCENTRATIONS (pg/g) IN CMA-E SAMPLE (DILUTION PREPARATION) Congener no. 1 3 7 30 50 97 143 183 202 207 209 PCB homolog 110 Uncorrected 110 Corrected 1 1 2 3 4 5 6 7 8 9 10 1.2 1.8 10.5 0 0 0 0.02 0 0.05 0 0.06 1.5 2.4 13.8 0 0 0 0.02 0 0.03 0 0.04 13.4 17.7 b Total 211 212 214 a 76a 103/91C 151 1 4 10 Surrogate recovery (percent). b Not applicable. c Samples run twice on magnetic sector instrument for low and high masses. Congener no. 212 monitored in both runs. 49 �TABLE 29. UNCORRECTED PCB CONCENTRATION ((Jg/g) IN THE CMA-A SAMPLE MATRIX (INTERNAL STANDARD CALCULATION) CMA-A 2005 8/4/82 CMA-A 2010 8/4/82 CMA-A 2020 8/5/82 CMA-A 2025 8/5/82 Monochlorobiphenyl 26 23 37 40 Dichlorobiphenyl 35 28 41 48 Trichlorobiphenyl 17 14 46 50 Tetrachlorobiphenyl 20 31 33 36 Pentachlorobiphenyl 32 29 29 31 Hexachlorobiphenyl 29 23 21 22 Heptachlorobiphenyl 18 12 12 14 PCB homo log CGC/EIMS analysis date Octachlorobiphenyl 5.4 4.1 3.4 4.2 Nonachlorobiphenyl 2.6 2.2 2.0 3.5 Decachlorobiphenyl 12 10 Total PCB 197 176 9.7 11 234 260 Recovery ( ) of Surrogate Compounds % 13 NSa 64 84 89 13 NS 96 96 101 Ci 2 -octachlorobiphenyl NS 73 67 72 C12-decachlorobiphenyl NS 68 69 73 Ce-monochlorobiphenyl C12~tetrachlorobiphenyl 13 13 a NS = no surrogate added. b Final concentration determined from sample weight recorded by QC coordinator. c 302 Daltons used for quantitation. 50 �TABLE 30. CORRECTED PCB CONCENTRATION (|Jg/g) IN THE CMA-A SAMPLE MATRIX CMA-A 2010 8/4/82 CMA-A 2020 8/5/82 CMA-A, 2025Q 8/5/82 Monochlorobiphenyl 37 44 44 Dichlorobiphenyl 44 48 53 Trichlorobiphenyl 15 47 49 Tetrachlorobiphenyl 33 34 34 Pentachlorobiphenyl 30 30 31 Hexa chlo robipheny 1 24 21 22 Heptachlorobiphenyl 16 18 19 PCB homo log CGC/EIMS analysis date Octachlorobiphenyl 5.4 4.9 5.7 Nona chlo rob ipheny 1 3.1 3.0 4.8 Decachlorobiphenyl 15 14 16 Total PCB 223 264 280 a NS = no surrogates added. b Final concentration determined from sample weight recorded by QC coordinator. 51 �TABLE 31. PCB homolog CGC/EIMS analysis date UNCORRECTED PCB CONCENTRATION (|Jg/g) Of SPIKED CHA-A SAMPLES DETERMINED RY THE INTERNAL STANDARD QUANTITATION METHOD CMA-A 2030 Total sample Spike concentration level 8/5/82 CMA-A 2040 Total sample Spike level concentration 8/5/82 CMA-A 2050 Total sample Spike concentration level 8/6/82 CMA-A 2060Q Total sample Spike concentration level 8/6/82 Blind quantilat ion standard Total sample Spike concentration level 8/6/82 Monochlorobiphenyl 60 20 80 49 92 100 100 82 140 184 Dichlorobiphenyl 56 10 58 25 58 51 69 42 53 94 Trichlorobiphenyl 65 10 75 25 39 51 44 42 87 94 Tetrachlorobiphenyl 47 15 55 36 43 75 50 61 110 137 Pentachlorobiphenyl 48 17 58 42 64 86 73 70 140 157 Hexachlorobiphenyl 40 19 48 46 61 95 67 77 160 173 Heptachlorobiphenyl 40 25 58 62 87 130 87 100 340 234 Octachlorobiphenyl 46 45 82 110 100 230 110 180 560 414 Nonachlorobiphenyl 51 49 93 120 130 250 140 200 530 450 Decachlorobiphenyl 60 42 110 100 140 210 140 170 430 369 513 252 717 615 814 1,280 920 2,550 2,306 N5 Total PCB 1,020 Recovery ( ) of surrogate compounds % I3 C6-monochlorobiphenyl 89 79 76 93 88 I3 Ci2-tetrachlorobiphenylb 94 93 84 93 88 Cj2-octachlorobiphenyl 62 56 41 53 78 65 57 48 64 79 13 13 C12-decachlorobiphenyl a Concentration in ng/ml rather than |jg/g since this sample was prepared by dilution of stock solutions of standards by QC personnel, b 302 Daltons used for quantitation. �TABLE 32. CORRECTED PCB CONCENTRATION (pg/g) OF SPIKED CHA-A SAMPLES DETERMINED BY SURROGATE RECOVERY CORRECTION PCB homolog CGC/EIMS analysis date CMA-A 2030 Spike Total sample concentration cone. 8/5/82 CMA-A 2040 Total sample Spike concentration cone. 8/5/82 CHA-A 2050 Spike Total sample concentration cone. 8/6/82 CMA-A 2060 Spike Total sample cone. concentration 8/6/82 Blind quantitation standard 2070Q Spike Total sample concentration cone. 8/6/82 Monochlorobiphenyl 67 20 100 49 120 100 110 82 160 184 Dichlorobiphenyl 63 10 74 25 76 51 74 42 60 94 TrichJorobiphenyl 70 10 80 25 46 51 47 42 99 94 Tetrachlorobiphenyl 50 15 58 36 52 75 53 61 130 137 Pentachlorobipheriyl 51 17 63 42 77 86 78 70 160 157 Hexachlorobiphenyl 43 19 52 46 72 95 72 77 190 173 Heptachlorobiphenyl 64 25 100 62 210 130 160 100 430 234 Octachlorobiphenyl 74 45 150 110 250 230 210 180 720 414 Nonachlorobiphenyl 81 49 170 120 330 250 270 200 680 450 Decachlorobiphenyl 91 42 180 100 280 210 220 170 540 369 Total PCB 650 250 1,030 60 2 1,280 1,290 1,020 3,190 2,310 a 1,510 Concentration in ng/ml rather than pg/g since this sample was a blind quantitation sample. �TABLE 33. PCB CONCENTRATION ((Jg/g) OF CMA-A SAMPLES TREATED WITH DIFFERENT CLEANUP PROCEDURES (INTERNAL STANDARD QUANTITATION) PCB homolog CGC/EIMS analysis date CMA-A 2090 acid cleanup 8/9/82 Monochlorobiphenyl ND3 CMA-A 2110 Florisil cleanup 8/9/82 CMA-A 2130 alcoholic KOH cleanup 8/9/82 4.4 31 Dichlorobiphenyl 4.4 14 44 Trichlorobiphenyl 0.4 31 44 Tetrachlorobiphenyl 25 18 25 Pentachlorobiphenyl 19 17 20 Hexachlorobiphenyl 7.9 5.6 6.3 Heptachlorobiphenyl 5.9 2.2 3.8 Octachlorobiphenyl 2.4 6.0 2.6 Nonachlorobiphenyl 38 2.4 2.6 Decachlorobiphenyl 16 9.5 6.4 Total PCB 119 110 186 % Recovery ( ) of surrogate compounds 13 74 8 145 13 0 0 367 13 115 97 110 13 173 129 64 C6-monochlorobiphenyl Ci2~tetrachlorobiphenyl Ci2'°ctachlorobiphenyl C12~decachlorobiphenyl a ND = not detected. b 302 Daltons used for quantitation. 54 �TABLE 34. PCB CONCENTRATION (|Jg/g) OF CMA-A SAMPLES TREATED WITH VARIOUS CLEANUP PROCEDURES (SURROGATE COMPOUND CORRECTED) CMA-A 2090 acid cleanup 8/9/82 CMA-A 2110 Florisil cleanup 8/9/82 CMA-A 2130 alcoholic KOH cleanup 8/9/82 Monochlorobiphenyl NDa 28 11 Dichlorobiphenyl 30 86 15 PCB homolog CGC/EIMS analysis date Trichlorobiphenyl 0.3 (0.2)b 200 (16) 15 (20) 110 (9.3) 8.4 (11) (8.3) 110 (8.9) 6.8 (9.0) (3.5) 3.5 (2.2) 2.9 (2.9) Tetrachlorobiphenyl 17 (11) Pentachlorobiphenyl 13 Hexachlorobiphenyl 5.3 Heptachlorobiphenyl 2.6 1.2 1.8 Octachlorobiphenyl 1.1 3.1 1.2 1.2 1.2 3.1 4.2 Nona chl o r ob ipheny 1 Decachlorobiphenyl Total PCB 17 3.9 546 (159) 90 (78) 68 (77) a ND = not detected. b 13 C12-tetrachlorobiphenyl was not quantifiable due to interferences. The values reported were calculated using 1*Ce-inonochlorobiphenyl. Values in parenthesis were calculated using 13Ci2~°ctachlorobiphenyl. 55 �DCMA Pigment Samples-Eight DCMA pigment samples were analyzed following the preparation described in the experimental section (Table 15). The results are presented in Table 35. The diarylide yellow pigment (DCMA-1) was analyzed in duplicate and as a blind spike supplied by the MRI quality control department. This sample is reported to contain 3,3'-dichlorobiphenyl at levels of approximately 70 M8/g (Dry Colors Manufacturing Association, 1981). No analyte or surrogate PCBs were detected in the duplicate 1-g samples of the pigment and a known spike of the sample. The lack of detected PCBs indicates a loss of analytes in the sample preparation. The CGC/EIMS analysis of a sample of the yellow pigment spiked by MRI quality control personnel yielded an uncorrected concentration of 76 |Jg/g of 3,3'-dichlorobiphenyl based on the internal standard quantitation and a corrected concentration of 63 Mg/g, based on 120% recovery of the 13C6-4-monochlorobiphenyl surrogate. The level of the 3,3'-dichlorobiphenyl added by the QC personnel was reported to be 60 M8/8- Hence, the total dichlorobiphenyl concentration should have been approximatey 130 |Jg/g (70 |Jg/g endogenous plus 60 (Jg/g added). The phthalocyanine green pigment (DCMA-4) was also analyzed in duplicate following dissolution and fractionation with a Florisil column. This pigment reportedly contains only decachlorobiphenyl at approximately 40 |Jg/g based on the results of the DCMA round robin study (Dry Color Manufacturing Association, 1981). Our analysis of duplicate samples yielded uncorrected concentration levels of 24 and 27 [Jg/g of decachlorobiphenyl by the internal quantitation method. The corrected concentration for both samples was 13 (Jg/g with recovery of the 13Ce-decachlorobiphenyl surrogate at 190 and 210%. Phthalocyanine blue (DCMA-8) was also analyzed as a single sample. Pentachloro- and hexachlorobiphenyls were detected but the concentrations were below the quantitation limits for that particular day. The total PCB concentration of this pigment, as discussed in the results of the DCMA round robin (1981), is reported to be 90 |Jg/g. The DCMA pigment sample analyses did not produce valid results. These data suggest that further development or validation of extraction/cleanup procedure would be necessary to provide acceptable PCB analyses of these samples. All of the blank samples (2001, 2080, and 2100) analyzed along with the DCMA samples were found to contain no detectable PCBs. DISCUSSION The determination of PCBs is a complex problem. The inaccessability of standards for all 209 congeners has traditionally been circumvented by the use of commercial mixtures (e.g., Aroclors) as standards. Quantitation has often been addressed in terms of relating the analyte to an Aroclor standard to give a "total PCB" concentration. Determination of PCBs synthesized as by-products in commercial products or product waste presents three special problems: (a) the analyte does not generally resemble a commercial PCB mixture, so quantitation against Aroclor standards would be incorrect; (b) the matrix often contains high concentrations of other chlorinated organics which are not easily separated during a cleanup procedure and which interfere with the qualitative and quantitative analysis; and (c) the matrix is undefined and can include gases, liquids, or solids of any purity and complexity. 56 �TABLE 35. RECOVERY ( ) OF CARBON-13 LABELED SURROGATE COMPOUNDS FROM DIARYLIDE YELLOW % AND PHTHALOCYANINE BLUE AND GREEN PIGMENTS PCB surrogate DCMA-1 21403 DCMA-1 21503 DCMArl 2160b DCMA-1 2170QC 2175d 2180d DCMA-8 21906 DCMA-8 2200Q 13 ND8 ND ND 120 ND ND ND 12 l3 ND ND ND ND ND ND 94 52 13 ND ND ND 200 120 107 92 71 13 ND ND ND 250 190 210 150 77 C6-Monochlorobiphenyl C12-Tetrachlorobiphenyl C12-Octachlorobiphenyl C12-Decachlorobiphenyl a Samples 2140 and 2150 are duplicates prepared by the DCMA-B method. b Sample 2160 was spiked with 50 |jg/g of 3,3'-dichlorobiphenyl and prepared by the DCMA-B method. c Sample 2170Q was spiked by MRI quality control personnel with 3,3'-dichlorobiphenyl and was prepared by the DCMA-B method. d Samples 2175 and 2180 are duplicates prepared by the DCMA-B method, e Sample 2180 was prepared by the DCMA-A method. f Sample 2200Q was weighed by MRI quality control personnel. preparation by the DCMA-A method. g The four surrogate compounds were added but not detected. An unknown mass of sample was supplied for �In this situation, analytical methods require a different philosophy than the classic approach for a single analyte in a defined matrix where all steps, reagents, and apparatus are specified. The method proposed here leaves many of the analytical steps to the discretion of the analyst while ensuring the reliability of the results with a strong quality control program. Thus, an analyst familiar with general analytical techniques for a product, may readily adapt in-house extraction/cleanup procedures to incidental PCB analysis. Even when the recoveries are not optimized, the 13C-labeled surrogate recoveries will mimic those of the analyte PCBs. As long as the 13C recovery surrogates are thoroughly incorporated, their recoveries can be used to derive corrected analyte PCB concentrations. Several of the method validation analyses presented above, especially Tables 33 and 35, illustrate the importance of the recovery surrogates in QC. The techniques employed are common methods validated for PCB analysis by other laboratories without the 13C-surrogate data. Analyses of this type may have been used by a testing laboratory and erroneous results reported. The complexity of the matrix and the high probability of chlorinated organic interferents precluded the use of GC/ECD. The best available technique is GC/EIMS. During the validation work presented above, the anticipated difficulty of qualitatitve and quantitative data interpretation was confirmed. In addition to the inherent problems resulting from extrapolation from a standard to several analytes, interpretation of the complex peak clusters is a tedious, subjective, and error-prone process. The volume of data for one sample is staggering; for sample 2110, 286 peaks were identified and integrated in the PCB mass chromatograms as shown in Figures 6 through 16. Of these, 58 peaks met the qualitative criteria and were identified as PCBs. Clearly different analysts will obtain different results for those peaks which marginally fit the qualitative criteria. This very high data density relative to other common GC/MS analyses has a much higher potential for error and mistakes. In addition it should be noted that, for many of the samples analyzed in this study, the data interpretation is more time-consuming than the rest of the analytical process. The integration methods are also prone to error. Integration is always conducted interactively with the mass spectrometric data system, either manually or automatically. The selection of baseline criteria, background sensitivity, integration method (valley-to-valley, baseline-to-baseline, etc.), and retention window all affect automatic quantitation. The position of the cursor and integration method affect the manual quantitation results. The day-to-day instrumental variability with quadrupole systems also appears to adversely affect data quality, despite tight calibration specifications. The magnitude of this error soruce should be further documented. The above discussion presents some of our understanding of some of the major problems with analysis for by-product PCBs. Further work will be devoted to characterizing and reducing these problem areas. Even with forseeable improvements in the method, the data for by-product PCBs in many commercial product and product waste samples will exhibit low precision and accuracy. 58 �lie 08/09/82 16:20:60 SAMPLE: SAMPLE 12110 dlA-A FLOMSIL 1/IODIL ItAHCE; G 1.1759 LABEL: H 0. 4.9 OMAN: A 1ULIMJ A. 1.9 DATA: 49011109^5 91 CALI: IHDCAIJWWl 1)1 BASE: U 2fl. SCAIJS 1 TO 1750 3 128450^1. 169.0 Ln vo 1009 15:59 Figure 6. Reconstructed ion chromatogram for 1299 19:00 1409 22:10 1660 25:29 SCAil mm SIM analysis of the CMA-A sample No. 2110. �MASS CHBOIIATOCRAIIS 88/09/82 16:20:00 SAIIPLE: SAIIPLE I2MO CIIA-A FLOftlStL RAflCE: C I.I750 LAHEI.: H 0. 4.0 DATA: 198III09U5 HI CAM: MIDTAUWW H1 l/IODIL 1ULHU OVAII: A 0. 1.0 BASE: 020. 83 SCAIJS 703 TO 900 3 I80.0-, 700 11:95 850 M:27 Figure 7. SIM ion plots for monochlorobiphenyls (188 and 190 Daltons) and the1 monochlorobiphenyl surrogate (194 Daltons) in CMA-A sample No. 2110. 900 SCAJf 14:15 TlliH �MASS aiCOHATOGRAIIS 08/09/82 16:28:60 SANFLE: SAIIPLE 12118 QIA-A FLOBISIL RANGE: C 1.1758 LABEL: H «. 4.9 DATA: 1991II99V5 81 CAII: IIIDCAUWWI HI 1/IODIL 1UI.IHJ QUAII: A 9. 1.9 BASE: U 28. SCAIIS 768 TO 1199 3 222. 1198 SC'J! 17:25 Tim- Figure 8. SIM ion plots for dichlorobiphenyls No. 2110. (222 and 224 Daltons) in Cl'A-A san.nle �HASS CHROMA TOGRAIIS DATA: 4991II09V5 fll 08/69/82 16:28:98 CALI: IIIDCALIttOVI (14 SAMPLE: SAIIPLE 12lie QIA-A FLORISIL I/IODIL IULHU RANGE; G 1.1759 LABEL: H 9. 4.9 QUAII: A 9. 1.9 BASE: U 29. 3 199.8 SCAMS 859 TO 1150 torn 256 1156129. 256.977 ^ 9.5W _8Z5_ Ni 1093 89. 258 1384579. 258.877 ± 8.599 859 13:27 Figure 9. 990 14:15 SIM ion plots for 950 15:92 15:59 1959 1199 16:37 17:25 1150 SCAM 18:12 TKIE trichlorobiphenyls (256 and 258 Daltons) in CVA-A sample No. 2110. �71.3-1 NASS OffiOmTOGRAIIS DATA: 499IIM9V5 II •8/09/82 (6:20:99 HALI: IIIDCAIJWWI SI SAHTLE: SAIIPLE 12110 QIA-A FLOWS! L I/I00IL IULUU BAMCE: G 1.1759 LABEL: II 9. 4.9 OUAII: A 0. 1.9 BASE: U 29. 3 1222 SCAHS 1959 TO 13TO 217296. 299 . 109.9-, 292 . CTi OJ 39.7-1 137728. 298. 298.089 0.590 62.7-1 217344. 304. 304.091 0.509 1350 SCAN 21:22 TIME Figure 10. SIM ion plots for tetrachlorobiphenyls (290 and 292 Daltons), 3 , 3 ' , 4 , 4 ' - t e t r a c h l o r o biphenyl-dg (298 Daltons), and the 13 C 1 2-tetrachlorobiphenyl surrogate (304 Daltons) in O'A-A sample No. 2110. �MASS CUBOIIATOGRAMS DATA: 4901IW9V5 II1669 08/09/82 16:28:89 CALI: IIIDCAI.II09VI 11 SAMPLE: SAMPLE I2II0 CMA-A FLORISIL I/IOD1L 1ULIIU RANGE: G 1.1750 LABEL: II 9, 4.0 WAII: A 9. I.0 BASE: U 29. 3 SCAIB J2W TO f? 15897B. 12.57 326 326.097 * 0.500 i 81.3- 129280. 328 J 328.098 * 0.500 1200 19:06 1250 19:47 1390 20:35 Figure 11. SIM ion plots for pentachlorobiphenyls No. 2110. 1400 22:10 1450 22:57 SCAN 23:15 TIIIE (326 and 328 Daltons) in CMA-A sample �MASS dffiOHATOCRAItS 08/09/82 16:20:08 SAMPLE: SAMPLE 12110 QIA-A FLORISIL I/IOniL RANGE: G IHIAH: A 0. 1.1759 LADEL: II 0. 4.0 1ULIHJ 1.0. SCAMS 1250 TO I500 DATA: 198IIWOV5 81 CALI: IHDCALHOT"! M ;SE: 0 28. 3 15C784. 97.-In 308.108 * 9.599 CT> Ln 106. 154880. 362 302.108 0.S00 1450 22:57 Figure 12. SCAN 23:15 TIHE SIM ion plots of hexachlorobiphenyls ( 6 and 362 Daltons) in CFA-A sample No. 2110. 30 �MASS UIROIIATOGRAIIS DATA: 4991H99V5 II 08/09/82 16:29:89 CAM: HI DCAll »W I 111 SAMPLE: SAIIPLE 12119 C1IA A FLORISIL 1/I0DIL 1ULIHJ RANGE: G 1.1756 LABEL: II 9. 4.9 OWAH: A 9. 1.9 HASP: U 29. 3 99.9 SCAMS 1359 TO 45568. ^91.118 ± 9.599 56384. 396.118 ± 9.599 1359 21:22 Figure 13. 1459 22:57 1559 SCAN 21:32 TIME SIM ion plots of heptachlorobiphenyls (39* and 306 Daltons) in CMA-A sample No. 2110. �MASS dlBffllATOGRAIIS 08/W82 16:20:80 SAMPLE: SAMPLE 12110 QIA-A FLORISIL RAMGE: G 1.1750 LABEL: II 0. 4.0 79.3-1 DATA: 4901H09V5 »l CALIt HIDCAUI0WI HI I/16DIL 1ULIHJ CUAII: A 0. 1.0 BASE: 0 20. ISC SCAMS IfM TO 165!) 3 29728. 428 . 128 •11856. 430.129 0.500 42304. I00.9-! 442 * 1450 22:57 Figure 14. 1509 23:45 1550 24:32 1608 25:20 SIM ion plots of octachlorobiphenyls ( 4 2 8 and 430 Daltons) and the chlorobiphenyl surrogate ( 4 4 2 Daltons) in CMA-A sample No. 2llo. 112.132 0.5W IC50 SCAN :!fi:07 TIME ^ �MASS aiBOHATOGRAIIS DATA: 4961II99V5 II 68/09/82 16:28:60 CALI: IIIDCALIWWI lit SAMPLE: SAMPLE I2H6 CIIA-A FLOIIISIL I/10D1L 1ULIMJ RAMCE: G 1.17561 LABEL: II 6. 4.9 OUAII: A 6. 1.6 BASE: U 29. 3 STAIIS 1559 TO 89.5-1 24256. 1565 1631 164.139 ± 0.599 461 . 109.6- < 27164. oo 466 466.139 * 6.590 1569 24:42 Figure 15. 1580 25:01 1620 25: 39 1616 25:58 SCAN TIME SIM ion plots of nonachlorobiphenyl (464 and 466 Daltons) in CMA-A sample No. 2110, �MASS aiBWIATOGRAIIS 98/89/82 16:28:88 SAHTLEi SAIITLE 12118 QIA-A FLOW SI L BAHGE: G 1.1758 LABEL: N 9. 4.8 99.9-1 1669 DATA: 49eil»9V5 11669 CALI: IHDCALI»WI 01 I/1001L 1ULIHJ QUAH: A 9. 1.8 BASE: U 29. SCAIB 1658 TO 17B!) 3 79232. 4W. 149 * 8.599 498 . 88I92. 499.500 * e.598 43849. 5S9.588 8.588 1709 Figure 16. SIM ion plots of decachlorobiphenyl (498 and 500 Daltons) and the decachlorobiphenyl (510 Daltons) in QIA-A sample No. 2110. 13C i2~ SCAN TIME �SECTION 5 REFERENCES Dry Color Manufacturers Association. 1981. An analytical procedure for the determination of polychlorinated biphenyls in dry phthalocyanine blue, phthalocyanine green, and diarylide yellow pigments. 1117 North 19th Street, Arlington, VA 22209. Erickson MD, Stanley JS. 1982. Midwest Research Institute. Methods of analysis for incidentally generated PCBs literature review and preliminary recommendations. Draft interim report no. 1. Washington, DC: Office of Toxic Substances, U.S. Environmental Protection Agency. Contract 68-01-5915. Haile CL, Baladi E. 1977. Midwest Research Institute. Methods for determining the total polychlorinated biphenyl emissions from incineration and capacitor and transformer filling plants. Washington, DC: U.S. Environmental Protection Agency. Contract 68-02-1780. EPA 600/4-77-048. Pittaway AR, Horner TW. 1982. Heiden, Pittaway Associates. Statistical analysis of data from a round robin experiment on PCB samples. Washington, DC: Chemical Manufacturers Association report. Roth RW, Keys JR, Chien DHT, et al. 1982. Midwest Research Institute. Methods of analysis for incidentally generated PCBs--synthesis of 13C-PCB surrogates. Draft interim report no. 3. Office of Toxic Substances, U.S. Environmental Protection Agency. Contract 68-01-5915. Stanley JS, Erickson MD. 1982. Midwest Research Institute. Peer review and authors' replies to 'methods of analysis for incidentally generated PCBs-literature review and preliminary recommendations.1 Draft interim report no. 2. Washington, DC: Office of Toxic Substances, U.S. Environmental Protection Agency. Contract 68-01-5915. USEPA. 1979a (December 3). U.S. Environmental Protection Agency. neutrals, acids, and pesticides—method 605. 44 FR 69540. Base/ USEPA. 1979b (December 3). U.S. Environmental Protection Agency. Organochlorine pesticides and PCBs—method 608. 44 FR 69501. USEPA. 1982 (June 8). U.S. Environmental Protection Agency. Polychlorinated biphenyls (PCBs); manufacture, processing, distribution, and use in closed and controlled waste manufacturing processes. FR 74 24976. 70 �APPENDIX A SUPPLEMENTARY GC/EIMS DATA ON PCB CONGENERS A-l �The following data support the method validation section for gas chromatography/electron impact mass spectrometry (GC/EIMS) of polychlorinated biphenyls (PCB). Table A-l lists the average relative response factors (RRF) for the 77 commercially available PCB congeners determined as four replicates. Table A-2 presents results of the Student's t-test used to determine the significance of differences for average RRFs for PCB homologs measured on a single day versus multiple days. The data in Table A-2 indicate that only the average RRFs for the heptachlorobiphenyl homolog are significantly different. Table A-3 presents the results of the Student's t-test used to determine the significance of differences for the average RRFs for the PCB homologs determined with the quadrupole and magnetic sector mass spectrometers. All 77 PCB congeners were determined in a single day for each of the instrument studies. This comparison indicates that the average RRF values are significantly different, which was expected. However, the relative standard deviations are not significantly different, indicating that the selection of the calibration standards is appropriate. These conclusions are discussed more fully in the text. Table A-4 presents results of the Student's t-test used to determine significance of differences for the RRFs for the 11 congeners in Solution No. 1, which was analyzed daily. An example of the data generated for multiple analysis of Solution No. 1 is presented in Figures 1 to 23. This information includes a capillary GC/EIMS chromatogram of Solution No. 1, the mass spectra of each component in this solution, and a graphic illustration of the distribution of several measurements of each congener about the average response factor. It should be noted that the standard deviation and relative standard deviation presented in these plots are different from that reported in the text due to calculation of the standard deviation using N weighting rather than the correct N-l weighting. All other standard deviations reported in this document are based on the N-l weighting. The relative retention times of the 77 PCB congeners with respect to 3,3",4,4'-tetrachlorobiphenyl-de determined with the Finnigan 4023 quadrupole and the Varian MAT 311A mass spectrometers are presented in Table A-5. A relative retention time unit of 0.01 (10 sec) is required for resolution of two specific congeners based on the gas chromatography parameters used to generate these numbers. A-2 �TABLE A-l. RELATIVE RESPONSE FACTORS FOR COMMERCIALLY AVAILABLE PCB CONGENERS (QUADRUPOLE) Congener no. Degree of chlorination Average relative response factor Standard deviation Coefficient of variation (%) 1 2 3 1 1 1 4.073 2.951 2.969 0.118 0.056 0.028 2.905 1.894 0.956 4 5 7 8 9 10 11 12 14 15 2 2 2 2 2 2 2 2 2 2 1.232 1.959 2.008 2.049 2.148 1.880 3.073 1.929 2.083 1.909 0.008 0.035 0.027 0.023 0.061 0.031 0.073 0.036 0.098 0.089 0.646 1.803 1.366 1.134 2.846 1.658 2.363 1.877 4.702 4.686 18 21 24 26 28 29 30 31 33 3 3 3 3 3 3 3 3 3 1.104 1.586 1.051 1.714 1.587 2.195 1.526 1.706 1.688 0.012 0.018 0.033 0.013 0.028 0.048 0.067 0.024 0.031 1.089 1.110 3.105 0.731 1.733 2.188 4.418 1.409 1.863 40 44 47 49 50 52 53 54 61 65 66 69 70 72 75 77 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 0.597 0.712 1.062 0.831 0.957 0.732 0.750 0.958 0.975 1.086 1.139 1.058 1.091 0.980 1.185 1.095 0.013 0.007 0.059 0.019 0.025 0.011 0.008 0.013 0.069 0.022 0.068 0.012 0.050 0.048 0.061 0.050 2.152 0.946 5.591 2.245 2.574 1.504 1.006 1.344 7.094 1.994 5.966 1.110 4.548 4.870 5.113 4.595 1 (continued) A-3 �TABLE A-l (continued) Degree of chlorination Average relative response factor Standard deviation Coefficient of variation ( ) % 87 88 93 97 100 101 103 104 115 116 119 121 5 5 5 5 5 5 5 5 5 5 5 5 0.617 0.611 0.574 0.719 0.727 0.653 0.566 0.824 0.853 0.785 0.762 0.948 0.011 0.005 0.010 0.008 0.003 0.004 0.009 0.025 0.061 0.013 0.022 0.020 1.710 0.744 1.677 1.139 0.428 0.538 1.627 3.048 7.146 1.654 2.911 2.127 128 129 136 137 138 139 141 143 151 153 154 154 155 156 6 6 6 6 6 6 6 6 6 6 6 6 6 6 0.499 0.431 0.689 0.533 0.433 0.462 0.419 0.490 0.473 0.549 0.221 0.511 0.587 0.599 0.005 0.004 0.016 0.008 0.008 0.026 0.010 0.005 0.013 0.050 0.001 0.010 0.011 0.044 1.093 0.813 2.336 1.582 1.946 5.686 2.353 0.986 2.826 9.101 0.570 2.039 1.828 7.431 171 181 183 185 7 7 7 7 0.346 0.383 0.380 0.336 0.002 0.009 0.010 0.006 0.640 2.379 2.501 1.729 195 198 200 202 204 8 8 8 8 8 0.263 0.262 0.301 0.250 0.221 0.003 0.008 0.007 0.007 0.007 1.184 2.887 2.392 2.663 3.200 206 207 208 9 9 9 0.193 0.237 0.259 0.003 0.008 0.003 1.723 3.547 1.315 209 10 0.213 0.006 2.837 Congener no. Relative to 3,3',4,4'-tetrachlorobiphenyl-dg. All relative response factors were calculated as the average of four replicate measurements made on the same day. A-4 �TABLE A-2. STUDENT'S TWO-SIDED t-TEST TO DETERMINE SIGNIFICANT DIFFERENCES BETWEEN QUADRUPOLE RESPONSE FACTORS CALCULATED ON THE SAME DAY VERSUS MULTIPLE DAYS PCB homolog MonochloroDichloroTrichloroTetrachloroPentachloroHexachloroHeptachloroOctachloroNonachloroDecachloro- Number of isomers 3 10 9 16 12 13 4 6 3 1 Average RRF from replicate measurements 3.331 2.027 1.573 0.950 0.720 0.513 0.361 0.253 0.229 0.213 Standard deviation 0.643 0.447 0.341 0.175 0.120 0.078 0.024 0.030 0.034 0.006 Average RRF from single , measurement Standard deviation 2.739 2.048 1.592 0.946 0.725 0.500 0.308 0.224 0.188 0.179 0.254 0.322 0.289 0.189 0.127 0.096 0.025 0.039 0.030 c t-Statistic 1.478 -0.119 -0.131 0.0618 -0.1085 0.377 3.119 1.398 1.5,91 Significant at 95% level? No No No No No No Yes No 5 > a Four replicate measurements of the RRF were made for each isomer. For example, the three monochlorobiphenyl isomers were measured four times each. Hence, the average RRF and standard deviation were calculated from 12 distinct values. b A single measurement for each of the 77 PCB congeners was completed in a single day. Hence, the average RRF reported is the average of one measured RRF for each isomer within a homolog. For example, the average RRF and standard deviation reported for the monochlorobiphenyl was calculated from three distinct values. c Single measurement. d Cannot test significance of difference between single measurements. �TABLE A-3. COMPARISON OF THE AVERAGE RELATIVE RESPONSE FACTORS (RRF) DETERMINED WITH QUADRUPOLE (FINNIGAN 4023) AND MAGNETIC SECTOR (VARIAN MAT 311A) MASS SPECTROMETERS'" PCB homolog Number of isomers MonochloroDichloroTrichloroTetrachloroPentachloroHexachloroHeptachloroOctachloroNonachloroDecachloro- 3 10 9 16 12 13 4 6 3 1 Finnigan 4023 quadrupole MS Standard deviation RRF 2.739 2.038 1.592 0.946 0.725 0.500 0.308 0.224 0.188 0.179 0.250 0.32 0.29 0.19 0.13 0.10 0.025 0.04 0.93 Varian MAT 311A magnetic sector MS Standard RRF deviation 2.329 1.663 1.167 0.902 0.780 0.640 0.497 0.463 0.467 0.586 RRFs significantly different at the , 95% confidence level Variances significantly different at the 95% confidence level No Yes Yes No No Yes Yes Yes Yes e No No No No No No No No No 0.199 0.229 0.248 0.130 0.136 0.124 0.060 0.071 °;a05 a The RRF and standard deviation reported in this table for the quadrupole and magnetic sector mass spectrometers were determined as single measurements of all congeners in a single day with each instrument. b Student's two-sided t-test was used to determine significant differences of the RRFs. c An F-test was used to determine significant differences of the standard deviations, where F = (std dev!)2/(std dev2)2 with (n-1, n-1) degrees of freedom. d Single measurement. e Cannot test significance of difference between single measurements. �TABLE A-4. PCB STUDENT'S TWO-SIDED t-TEST TO DETERMINE SIGNIFICANT DIFFERENCES OF THE AVERAGE RELATIVE RESPONSE FACTOR (RRF) FOR SOLUTION NO. 1 FOR REPLICATE ANALYSIS ON A SINGLE DAY VERSUS SINGLE ANALYSES ON MULTIPLE DAYS Replicate analyses on single day no. RRF Standard deviation 1 11 29 47 121 136 181 195 207 209 4.073 3.073 2.195 1.062 0.948 0.689 0.383 0.263 0.237 0.213 0.118 0.073 0.048 0.059 0.020 0.016 0.009 0.003 0.008 0.006 congener Single analyses, on multiple days Standard RRF deviation 3.241 2.538 1.899 1.015 0.959 0 . 683 0.374 0.275 0.269 0.230 0.201 0.161 0.100 0.059 0.043 0.058 0.035 0.028 0.032 0.027 t-Statistic Significant differences of RRF at 95% confidence limit? 7.468 6.204 5.483 1.268 -0.479 0.186 0.662 -1.137 -2.479 -1.599 Yes Yes Yes No No No No No Yes No a The RRF and standard deviations were calculated from four replicate measurements completed in the same day. b The RRF and standard deviatons were calculated from seven single measurements from seven different days. �TABLE A-5. RELATIVE RETENTION TIMES (RRT) OF 77 COMMERCIALLY AVAILABLE PCB CONGENERS MEASURED VERSUS 3,3'4,4'-TETRACHLOROBIPHENYL-d6 DETERMINED WITH MAGNETIC SECTOR (VARIAN MAT 311A) AND QUADRUPOLE (FINNIGAN 4023) MASS SPECTROMETERS RRT RRT PCB congener no. 311A 4023 Monochloro1 2 3 0.403 0.481 0.474 0.425 0.490 0.499 Dichloro4 5 1 8 9 10 11 12 14 15 0.518 0.598 0.559 0.590 0.563 0.521 0.649 0.660 0.616 0.677 0.536 0.606 0.579 0.606 0.577 0.534 0.660 0.671 0.628 0.681 Trichloro18 21 24 26 28 29 30 31 33 0.665 0.762 0.685 0.729 0.745 0.719 0.641 0.741 0.760 0.678 0.767 0.694 0.738 0.753 0.728 0.653 0.752 0.769 Tetrachloro40 44 47 49 50 52 53 54 61 65 66 69 70 72 75 77 0.870 0.838 0.814 0.811 0.746 0.804 0.763 0.720 0.898 0.822 0.905 0.800 0.880 0.853 0.816 1.002 0.875 0.843 0.819 0.817 0.751 0.810 0.773 0.731 0.898 0.826 0.908 0.807 0.904 0.856 0.821 1.003 PCB congener no. 311A 4023 Pentachloro87 88 93 97 100 101 103 104 105 116 119 121 0.979 0.913 0.907 0.976 0.878 0.945 0.870 0.829 0.988 0.985 0.964 0.911 0.978 0.915 0.908 0.979 0.884 0.945 0.874 0.836 0.987 0.986 0.965 0.914 Hexachloro128 129 136 137 138 139 141 143 151 153 154 155 156 1.163 1.128 0.994 1.118 1.108 1.037 1.096 1.050 1.020 1.074 1.002 0.929 1.194 1.156 1.127 0.996 1.115 1.103 1.038 1.093 1.051 1.021 1.073 1.004 0.931 1.188 Heptachloro171 181 183 185 1.189 1.178 1.154 1.166 1.187 1.174 1.148 1.161 Octachloro194 195 198 200 202 204 1.355 1.326 1.275 1.203 1.194 1.209 1.351 1.317 1.265 1.199 1.188 1.203 (continued) A-8 �TABLE A-5 (continued) RRT RRT PCB congener no. 311A 4023 Nonachloro206 207 208 1.414 1.336 1.319 1.399 1.330 1.318 PCB congener no. 311A 4023 Decachloro209 1.453 1.440 A-9 �-Iflfl.O E1C DATA: r-<:91VQ2 1765 SCAIIS CS/01/82 8:33:00 CALI: C.ilOIUI 03 OUT OF SAMPLE: FCB MIXTURE—SOLUTION 01 1UL UIJ COUUS.t -16CO HIV (0-6 79EV .2IIA PCS--I1H 40-2II-335-C/ "OII-COl" R.MKE: C I.I2PO LABEL: t! 0, 4.0 OUAH-. A tt. I 0 J 0 BA<;| BASE: H 20. I TO I209 I TO I2CO 3 1. 328 599016. i-H C (U r^ c J3 •H 43 0) a. fX, vj O 1 : rH o t-i 0 --I rC <~M ^» U •H 43 O C 01 in 1 CN ^ 01 W.«20 -H 43 o 43 O -*" o 0. R59 CM u § PH 1 1 J^l 4J 0) to 1 X a 1 If If) U ^. <fr „ •^ 1 II en CM « •* m •* •• i CM i—i 43 CJ cd CO ^ u n CM * CM CNI | I CM I * || to CM CM 1 en - <r •\ •V ^ ro - ro en -; •s si- 1 ** - to H ^ **ll * ^ ^" O OJ M ** § :• in " to to •* <u cd 4J 43 cd C •£* 43 ft 0> PC 1 43 y cfl O CO - 43 0 M 0 rH 43 CJ 6-Octachloro : ,H -g 43 0 0.912 c ^ [-; a. (U 4= Ul ^ P« H-J tachlorobiph O i—* _r? t.1 C iphenyl-dg a) 2 t-i 0 C i orobiphenyl ^*. lorobiphenyl *rt 1 w ft (i 96 .orobiphenyl i-H t CM C °- CM ' 0.315 i 200 3:20 • i ISO 6-.10 . 1 COO je : eo , '" L "*"I~*~ *• -^ • i 1.102 "—* COO 13:20 _'•£' ItJ.0 1C: IB SCAII TIME Figure A-l. Fused silica capillary gas chromatogram of PCB Solution No. 1 analyzed with electron impact mass spectrometry. Experimental conditions for separation and analysis of the PCBs are presented in the experimental section of report. �UV.'. SVECi'tlUil GV20/32 16:42:00 * PATA: 498IE20W7 (Witt CAM: CALF.20UI 07 1:50 E: rest iimiiue — SOLUTIOII BI IUL iiu -J%8 L-lfV 10 6 70I:V .2IIA Bi!5 -1MI 1 10 -211- 525- 10/ "IJJI-COL J33IAKCED <S 15B 2H OT) ICO 9523:-. 50.0- 126 ^ T ti/n no Figure A-2. 1161 i*U ico 173 101! Electron impact ionization mass spectrum of 2-monochlorobiphenyl. �tt'I-CII'JRI U>:12:W * HATA: :: FOB IIIXTIII::;—soi.Minni ni »« mi IKO iaiv io-6 /ot-:v .M\ imr> j-.u (S I5K 2II OTI U17J DASJ; II/'E: CAM: CAM-JttH /:'j» >/ "n;i toi." C329C. 189.0-1 I I-1 NJ 59.0- 99 U~ IP -^-r-^M-. , T-p-^^-, ll/E (00 Figure A-3. 120 220 Electron impact ionization mass spectrum of 3,5-dichlorobiphenyl, �W5/20/I52 1C:'12:00 ^ I!:'H f.AI.1; r;ALK2ft»l IU SAIIPLH: PCB IIIXTIIRE—f.OLUYIOII 11 IIH. IIU COJ1DS.: -1550 HIV 10-6 /OCV .2IIA DBS-1511 110-211-325-10/ "iKlBillAilCIiD (S 15D 211 01) HAM: H/J-: KW niC: 271072. f 255 100.0 1928( II 16 59.0- U) • 10 93 *>-n ft Z W ,1 1?3 if 1 T .Vii^ l._ .iii Li 169 i IIIwll ll/n Figure A-4. \yt 110 ICO ll iiill 100 1 1. I?G »LiJ.. 280 KG I-L.^ i-l 22t} \l •l ?]S 240 2»iC» Electron impact ionization mass spectrum of 2,4,5-trichlorobiphenyl. �MASS M'ECIRHII 05/2t)/82 IS:4?:W + ?):r.1 (Ml: SA!ir»>;: rcc tiixTiim-—SIU.UTIMI n iw. MM BA'JK II/C: 2 '9 P.IIJ: 3&10J2. 0» C«!l»S.: -1550 HIV Mi> 7»JiV .2IIA 1)115-1511 IIR-2II-325--IO/ "iKJ-ljm.1 211 OT> 2 0 iee.0-i 39369 2- i2 1 e 1 • in r\ 8 I8 255 i: 3 '9 1 ii/n J Jilu^jll k 1eo 1 T ' lk-,,1 1 1. l vlll ^ lll^ l 1 50 l ?1 J,l. » ?96 H"--j |—"* 2Wt .,| I,2?i_T213_rJ r-*" 250 I ' 1 ' 1 ' XY3«3 Figure A-5. Electron impact ionization mass spectrum of 2,2',4,A'-tetrachloroblphenyl. �HAW srcciwiii DMA: 'icuil:2r;'f»/ 07:>0 (J7 85/29/82 I6:12:0fl ^ I V : 00 CAM: CAI.I-WI SAUPLE: riJtt HlXTUnC-—SOLUTIOJI ill IUL J I U UGHIS.: -I550 HIV IO-6 70I-V .2IIA 08'j-Ctll IIO-2II-325-IO/ "«;] OH." BASE Il/Et 2:i:} HIC: 12C3M. EKIIAHCED (s isn 211 OT» 59.9 --r^ 350 Figure A-6. Electron impact ionization mass spectrum of 3,3'4 ,4' -tetrachlorobiphenyl-d^. �IIAS-; «)AiA: I'micyv-)"/ nwi UAI.I: tMf.»»l 05/M/«? 10:12:00 + 11:01 SAiim-t res MiXTunti— r,oi.ufi(iii n sw. IIM ar,»S.: -15W) BIV 10-0 /tfl-Y .2IW IWi Till IIO-2II-323-IO/ (s isn 211 on r.n: 05 "»KJ COI." 3078 180.0 2-36 "7 59.9- 191 99 2 «J 163 I "*"* 19-1 2*> .MUl'^rJ it/'-: Figure A-7. |llL,_,,U J50 li5.lV2.ix 'J«0 Electron impact ionization mass spectrum of 2,3',4,5',6-pentachlorobiphenyl. �IIACS SH3MWUI 85/20/J52 16:12:00 ' 12:06 SAtimi: rcc MIXTURE—SOLUTION ni HJI. im I>ATA: CAM: CAI.K20WI II3 323501. COIffiS.; -I5!i0 EIIY 10-5 70EV .2!l-\ DB5 I'jll 110-211 325- It)/ "OSI-COI." DBIAUCl-n <S ISO 2\\ 811 100.0- 25661. 29fl 50.0- It'li 350 u Figure A-8. Electron impact ionization mass spectrum of 2,2',3,3',6,6'-hexachlorobiphenyl. �IIVJS SrtCTtnfll ItAfA: 1?iUE2T,V»7 IKKJI aV.W/R:> H>:'I2:Ofl ^ M:2I iMl : CAI.E2WI 01 SAllI'LE: r«.B IIIXUli::;— -SDUiriO'l 01 HH. 111! COi!D:i.: -IWU DIY 10 0 TDtV .2JIA llffi I'-ll 1 10-211-325- IO/ "flH-Cor BA';C ll/li: :>V3 niC: 3755J?. nr.i,\i:ci;u is 150 211 OT» IOH.O IGOC1. i? 120 50.0 1 i 1 |i 7 liy 6 i: 0 4 .1 ir 129 IGO 3 1 149 ll I [1 lijW^M K9 IK? I 1" Ti 2W 22» 249 1680 3 1 i i—' oo 50.0- 3'.9 1 289 ,11 JJ-r^L•^*-i-» I111,.. ??*!|,. . 274 >*Wvi 1 ' i il • » • . | i • ILL - i i 3CO | • 300 • •i 1 i ' [ • • * •r ISO Figure A-9. Electron impact ionization mass spectrum of 2,2',3,4,4',5,6-heptachlorobiphenyl. �HAY; si'tornini JIATA: 1!HMI-2Wli/ ()r)7l CA.I.I: CAI.E?C!M 0.1 95/20/82 lti:42:W + Ifi: II SAiirLE: njB uixiimi: — souinoii ai mi. iiu cniros.: -I559 ia iv so r, voi-v .?IIA m^-r.ii n')-2ll-32r.-l'-)/ EII)IAt:r;GD (S 1511 211 OD BASE II/E: I/O KM): 'JJOB11. "OII-COI." i; ') 100.0 I6C7. 50.0- 2 r, i '3 p«125 T 95 I5C lee 120 149 100 183.0-1 1 |R7 159 51 "T8 II I «S ' 209 229 ^ 269 21« IGT. 3! 0 29 1 50.8393 271 I I . ll/fi 209 Figure A-10. I I1 1 \\.,m 388 3ji 1,1.1,,, 320 I 1 •I, i | 310 368 I 1, • • i • | •i i . |i i • • | . i . . ! i i . . | • . ' 308 1«8 120 «0 Electron impact ionization mass spectrum of 2,2',3,3',4,4',5,6-octachlorobiphenyl. �I»ATA: 'IOOIE5AW7 0970 CMI: CAI.K2MM OJ 05/20/82 l«:12:0fl ^ lfi:l'.» SAiffLE: rcis mxiuiw- — soumnii n iw. RASE »/f: Ifti HIO: 123016. lt!J OKWS.t -IfBW DIV 10 li 70EV .2IIA IWK-IriH IIO-2II-325-IO/ "UH--COI." BOIMXED (S I5B 2H OT> 1 G 17661. 1 2 2 50.0- 111 1BO 10 fl 125 1 t 1* l ' f • 101 1 it i I-J-1-.-H J 10 o II •PI 2 1 1 T \L\ 11LjllLjll^llL I 20 N/E 183.9- 1 » IwJ*. iw» III ^?V IllllM^^-JJ llti*^,^ alii . 2^1 220 |i|iii _ 240 I 260 230 41-1 ji 3 2 50.0- 357 290 i l.it, ni . 3dO Figure A-ll. 1 .,332 320 31? 3W jjll. 3CO -,-r-i-i-r- 3(KI 4W> 120 13 1 - 4G8 I, Electron impact ionization mass spectrum of 2,2',3,3',4,4',5,6,6f-nonachlorobiphenyl. �liAG'J SI'ECTBUll r II-MA: '190i 9 >/.WC2 10:12:00 + 17:13 CAI.l: CAM SAiii'LK: ten iiixitirr- soi.urinn ui an. nu ClISiDS.: -1559 HIV 10-6 70EV .2IIA D»r»-I5ll 110-711-325-!0/ Eii!)Ata:nD (s i5n 211 OT> w'JW UIWJ3 '"I f)3 BAJiC H/C: 21 i '• Rir, : 439208. "DJl-COL" 2 1 100.0- 228I6. I7« 2 9 50.0- 107 160 96 II/E 'i 108 262 231 Jl|ljYjl|h. 159 1 2?7 . 3f« 259 - •J 356 50.0- 228I 3 14« 321 ii/n Figure A-12. J-lOJULff^-.J 350 L^^MJEL 1 1, 150 ... 5C3 Electron impact ionization mass spectrum of 2 , 2 ' , 3 , 3 ' , 4 , 4 ' , 5 , 5 ' , 6 , 6 ' - d e c a c h l o r o b i p h e n y l . �ra Ltn.-oi.ii mss: 200 <Rei-.amr:Oi.ii MASS.uiPtD-G TiiTi!Aanc!:iH!miHiYL. isosrei: 15210 J:tT:D-0 nnR.'.Oil.U!:i)-CIFi!BIY[.. ISOIII-i: 0210 r.t*)/«iiT./T.H;.,\:iT.i <AV : i.ocoi i. not ST.IJS-V.- (.1.000 * ST.DEV.0.009 i.e i ts3 N> 0.899 HATE 5/I1/82 Figure A-13. 5/21/02 6/ 3/02 Response factor plotted on a day-to-day basis for the internal standard, 3,3',4,4'-tetrachlorobiphenyl-d6, in Solution No. 1. �RESl'OllSIi L I B : 0 1 . 2 i IIASS: 1«8 (IW.COIIl'iOI. I. MASS: 2!l<!) OlP:2-liai!)ll1IU)l!0-Bin«3IYl..JSCHER fll EEF : I)-G TEHACIILORO-BiriltliYI.. ISOIIEI! 0210 tAKEA/atF.ALEAl/tAIIT./Rb'F.AIIT.) (AV: 'J.51-11 5.600 ... 0.431 ST.TiEV.- n.ni Sl.'il 4. oca 1 NJ u> 3.C80 BATH ^'1-1/02 5/2-1/32 Figure A-14. Relative response factor for 2-monochlorobiphenyl in Solution No. 1 calculated versus 3,3',4,4'-tetrachlorobiphenyl-d6 on a day-to-day basis. �B:Ol.^i MASS: 222 (KKI-.ailinOI. l> IIAC,S: 2!«5> aiP:3.3--nioii.onn-Binn3ivL. ISWIHR tin B1-F:D-C TKTKACmjmO-ltiniHIYL. 1SOIJER B2IO T./l:m:.AIIT. > (AV: 2.7.}2» .t.AW ST.niv.- t 0.206 X ST.DEY.II-..1G6 3 r n.naa S'tt =• l i . • 2.J18H :; 2.C8!l 2.VC3 i to *• X 2.683 X X X 2.5*3 :< 2.480 •H 2.393 •> oon 5/24/02 6/ 3/32 Figure A-15. Relative response factor for 3,3'-dichlorobiphenyl in Solution No. 1 calculated versus 3,3',4,4'-tetrachlorobiphenyl-dg on a day-to-day basis. �1.111:01.1i IttSS: 256 (RiiF-COllPiOI. li IHSS.- 2!)OI aip:2.i.5-TRiciiinr.o-niriiiiiiYj.. isotiER r.29 REFrH-O TL:inAaiI.Ol;0~BirilFJIVI.. ISOMER II2IO i:nA)/(AiiT./Ki;r.AiiT.) <AV : 2.0051 2.^(10 ST.DLV.0. IR3 £ ST.IEY.n. I38 SMI = i.e 2. x I NJ 1.999 1.899 1.709 DATE 5/11/82 5/2-1/82 6/ 3/B2 Figure A-16. Relative response factor for 2,4,5-trichlorobiphenyl in Solution No. 1 calculated versus 3,3',4,4'-tetrachlorobiphenyl-d6 on a day-to-day basis. �RESrOKSF 1.111:01.5. IUSS: 202 (MF. COUP: 01. I , IKSS: 29CI niP:2.2M.'r-TFiRAciii.ojH)-iiirii0iYL. isoum: iw REf:D-0 lEll:AaiLi?r.O-Diri!BM.. ISOIinn U2IO (AnP,\/nCI:.A!:i'A)/(AIIT./!iLl;.AIIT.) (AV: 1.032) 1.280 ST.PMV.0.058 Z Sf.DEY.5.596 1.9 >( 1.190 •• i ro * O" X X X X ): O.C33 DATK l'l/82 5/21/02 G/ 3/02 Figure A-17. Relative response factor for 2,2',4,4'-tetrachlorobiphenyl in Solution No. 1 calculated versus 3,3',4,4'-tetrachlorobiphenyl-c^ on a day-to-day basis. �r-rsroiiSK UB.-OI.GI iwss: 326 (i!r.r.coiir oi.ii MASS: a\T-.2.y .1.5'.r.-rciiTAan.oi;o-mriiBiYi.. isomin : 11121 RCF:n-s iF.ii:Aaii.oi:o-iiii'iiaiYi.. isounK ir»io (Ar.t A/Cnp./.l:M>/(AIJT./r;nF./MIT.) (AV: O.'tIO) i. can ; -SU'liV.(1.952 Z ST.C-EV.T 477 Sl'il » 6. £39 i S3 -J DATE 5/11/82 5/21/82 6/ 3/82 Figure A-18. Relative response factor for 2,3',4,5',6-pentachlorobiphenyl in Solution No. 1 calculated versus 3,3',4,4'-tetrachlorobiphenyl-d6 on a day-to-day basis. �1.18:91.7. IIASS: 3C9 (l:EF.CO!H':fll. I. IIASS: diP^^'.s.s'.e.e'-iiiLVAaiLenn-BmiBiYJ.. isoiirc a i3t> BEF-.D-6 TElfiAaiI.OKO-|im!BIYI.. ISOIIKIt B2IO iiiiA^iAnT./iraF.AiiT.) <AV : o.ocs* O.J'O'.I ST.PEY.- O.OH £ ST.9EY.0.370 i.e X o.ynn * Is) 00 6.609 0.509 DATE 5/1V82 C/ 3/82 Figure A-19. Relative response factor for 2,2',3,',6,6'-hexachlorobiphenyl in Solution No. 1 calculated versus 3,3',4,4'-tetrachlorobiphenyl-d6 on a day-to-day basis. �RESPONSE i.in : oi.n, iwss: 391 (nnr.raiirrfli.ii MASS: air:2.2\3.i.i'.5.c-iiEi'T'.a!i.or.o-niniFiiYi.. isoiinn i lot EEF:«-c rETR/.ciu.ono-iiiMiBn'i.. iswim: 11210 : (AREA/niH .Ar,FA»/(AIIT./UtF.AIIT.) (AV: O.UO 0.377) ST.I'EV.0.827 x si.LEY.- tt.1?0 V.035 0. MO SMI - 1.0 x U.303 0.380 _• i isj VO I— 0.370 0.360 6.350 0.310 0.330 0.320 DATE 5/11/82 5/21/82 C/ 3/82 Figure A-20. Relative response factor for 2,2',3,4,4',5,6-heptachlorobiphenyl in Solution No. 1 calculated versus 3,3',4,4'-tetrachlorobiphenyl-d6 on a day-to-day basis. �r.Fsro;isr i.iitroi.o, IIASS 430 (nHP.cniir oi.i, IIASS REP:no Ti;u;.\aiLono-i;iriinj\'i.. isotinn U2io : : aiP:2,2'.3.3 t .1.4'.!;.6-Oi:TAOILORO-BIl l IK.ilV1.. ISWIFR 0 105 tlT. > (AV: 0.2/B> 1'. J/U sr.nr.v.s 0.021 ? Z Sf.DCV.7.972 '<; 0. ^-fi) sK 0. 203 - 1.0 •\c u_«. . .- , . _ _ i 1 0 O v *VM% 0 .270 , | 0.269 * X X 0.259 )C X n -MO — 5/2-1/82 Figure A-21. •• — • ' • II- »- — 1 C/ 3/02 Relative response factor for 2,2',3,3' ,4,4',5,6-octachlorobiphenyl in Solution No. 1 calculated versus 3,3' ,4,4'-tetrachlorobiphenyl-d^ on a day-to-day basis. �nnsronst LIB : OI.IOI HAGS: 101 (RnF.cniir oM air:2.2\3.3vi.i\5,G.<v-i;niiAciiLORiHiiriiniivi..: REF:iM> TErc/xiiioRo-DiFiintm.. isotmn 11210 (Am:4/RE»--.ARirA>/<AMT./r.FF.AHr.J <AV: 0.320 2301 0.257) ST.WiV.O.P29 5; sT.rtv.I I 239 SDil - 1.0 0.209 -v0.2J10 0.2/0 I u> 0.250 0.210 0.239 0.220 DATE Figure A-22. 5/14/82 5/21/82 6/ Relative response factor for 2,2',3,3',4,4',5,6,6'-nonachlorobiphenyl in Solution No.!1 calculated versus 3,3*,4,4'-tetrachlorobiphenyl-d6 on a day-to-day basis. �BESrOMSE L I B : O I . H i MASS: 498 (C£F.COIir : 0l. li MASS: 2991 air:2.2\3.r.4.1\5.5\G.6'-DECAaH.ORa-BIM!iaiYL. ISOIIHIi Q 209 BEF:D 6 TETIUdlLOBO-BlPliraril, ISOilEII (1210 i!EA>/(AIIT./i;EF.MIT.) <AV: 0.223) 0.270 ST.IlS-V.C.«22 Z Sf.OEV.9.8/7 n ?;T MM •= 1.8 r. • •• 0.210 I U> NJ 0.2JO 0.220 ) 0.210 I x 0.200 0. I'M DATIi 5/M/82 5/2-1/U2 C/ 3/32 Figure A-23. Relative response factor for 2,2' ,3,3',4,4',5,5',6,6'-decachlorobiphenyl in Solution No. 1 calculated versus 3,3',4,4'-tetrachlorobiphenyl-d6 on a day-to-day basis. �APPENDIX B ANALYTICAL METHOD: THE ANALYSIS OF BY-PRODUCT CHLORINATED BIPHENYLS IN COMMERCIAL PRODUCTS AND PRODUCT WASTES B-l �THE ANALYSIS OF BY-PRODUCT CHLORINATED BIPHENYLS IN COMMERCIAL PRODUCTS AND PRODUCT WASTES 1.0 Scope and Application 1.1 This is a gas chromatographic/electron impact mass spectrometric (GC/EIMS) method applicable to the determination of chlorinated biphenyls (PCBs) in commercial products and product wastes. The PCBs present may originate either as synthetic by-products or as contaminants derived from commercial PCB products (e.g., Aroclors). The PCBs may be present as single isomers or complex mixtures and may include all 209 congeners from monochlorobiphenyl through decachlorobiphenyl listed in Table 1. 1.2 The detection and quantitation limits are dependent upon the complexity of the sample matrix and the ability of the analyst to remove interferents and properly maintain the analytical system. The method accuracy and precision will be determined in future studies. 1.3 This method is restricted to use by or under the supervision of analysts experienced in the use of gas chromatography/mass spectrometry (GC/MS) and in the interpretation of gas chromatograms and mass spectra. Prior to sample analysis, each analyst must demonstrate the ability to generate acceptable results with this method by following the procedures described in Section 14.2. 1.4 The validity of the results depends on equivalent recovery of the analyte and 13C PCBs. If the *3C PCBs are not thoroughly incorporated in the matrix, the method is not applicable. 1.5 During the development and testing of this method, certain analytical parameters and equipment designs were found to affect the validity of the analytical results. Proper use of the method requires that such parameters or designs must be used as specified. These items are identified in the text by the word "must." Anyone wishing to deviate from the method in areas so identified must demonstrate that the deviation does not affect the validity of the data. Alternative test procedure approval must be obtained from the Agency. An experienced analyst may make modifications to parameters or equipment identified by the term "recommended." Each time such modifications are made to the method, the analyst must repeat the procedure in Section 14.2. In this case, formal approval is not required, but the documented data from Section 14.2 must be on file as part of the overall quality assurance program. B-2 �TABLE 1. NUMBERING OF PCB CONGENERS3 NO. Structure NO. Henoe(ilaroa<oh«ny1» 1 2 3 2 3 4 D<eh1orob1p(itny1t 4 5 6 7 3 9 10 11 12 13 14 15 }:]; 2)4 !:$' 2,6 3,3' 3,4 1:1' 4,4' Trlehlorablphtnyli 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 13 34 35 36 37 33 39 40 41 42 43 44 45 46 47 48 49 50 51 2.2'.3 2.2', 4 2.2', 5 2.2'. 6 2.3.3' 2.3,4 2.3.4' 2,3.5 2.3.6 2. 3', 4 2, 3'. 5 2, 3'. 6 2.4.4' NO. 52 53 54 5$ 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 105 106 107 108 109 110 2,2*. 5.5' 2,2',5.6* 2.2'.6.6' 2.3,3- .4 2.3,3'. 4' 2,3,3'. 5 2.3,3'.5' 2,3.3'. 6 2.3.4,42.3.4.5 2,3,4.6 2.3,4'. S 2, 3,4', 6 2,3.5,6 2,3' 2,3' 2.3' 2.3* 2.3' 2.3' 2.3' 2.3' in 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 4.4' 4,5 4,5' 4,6 4'. 5 4'. 6 5.5' S',8 2,4,4'. 5 2.4.4', 6 2'. 3, 4. 5 3,3'. 4. 4' 3,3',4,5 3,3'.4.5' 3,3'.5.5' 3,4.4'. 5 Structure 2,3.3' .4' 2.3.3' ,5 2.3.3' '.5 2.3.3' .5' 2.3.3' .6 2.3.3' ',6 2.3.3' ,5' 2.3,3' ,6 2.3.3' ',6 2.3,4.4 .5 2.3.4,4 ,6 2.3.4,5 6 2,3.4', .6 2.3'.*, 2.3'. 4. '.'6 2.3',4, ,5' 2.3". 4, 2'.3.3' 4J5 2'.3.4,4 '.5 2',3.4,. ,5' 2'.3.4.i .6' 3.3' .4.4 '.5 3.3'. 4.. .5' HemchU robtptienyls 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 15$ 156 157 158 159 160 PenUcM orcb< phtny 1 i 2.2'.3.3',4 2.2', 3. 3'. 5 2.2',3.3',6 2,2',3,4. 4' 2.2'. 3. 4. 5 2.2', 3, 4.5' 2,2'. 3,4.6 2,2'.3,4, 6' 2.2'.3.4',5 2,2'. 3,4'. 6 2.2',3,5,5' 2,2'. 3,5,6 2,2'.3.5,6' 2.2', 3, 5'. 6 2.2'. 3. 6.6' 2.2', 3'. 4.5 2,2'. 3' .4. 6 2.2', 4, 4' .5 2,2-.4.4'.6 2.2'. 4, 5,5' 2, 2' .4,5.6' 2.2',4.5',6 2, 2'. 4,6.6' 2.2', 4. 5 2.2' .4, 5' 2.2". 4, 6 2,2',4.6' NO. rtntacli oroMphenylj Tetneft 1 oreo 1 pinny 1 $ 82 83 84 2,4',S 85 2.4'.6 86 2'. 3. 4 87 2'. 3,5 88 3,3', 4 89 3.3',! 90 3,4,4' 91 3.4,5 92 3.4'.S 93 94 TttneM orob< ohnyl » 95 96 97 2.2'. 3.3' 2.2'. 3,4 98 2.2'.3,4' 99 2.2'. 3,5 100 2.2'.3.S' 101 2.2',3.6 102 2.2' .3,6' 103 2.2-.4.4 1 104 2,4.5 2.4.6 structure 2.2',3.. '.4,4' 2.2'. 3.. '.4.5 2.2'. 3. : '.4,5' 2.2'. 3.: '.4.6 2.2'. 3.: ',4.6' 2. 2'. 3,. '.5,5' 2.2' .3.. '.5,6 2.2* .3.: ',5, ' 2.2'. 3.: ',6, ' 2.2' .3,4 ,*'. 2.2'. 3. 4 .4'. ' 2,2' .3. 4 ,*', Z,2'.3,4 ,4' . ' 2.2'. 3.4 .5,5' 2,2'. 3, 4 .5.6 2.2'.3.4 .5,6' 2.2'. 3, 4 .5'. 6 2.2'. 3. 4 .6,6' 2.2', 3.4 '.5.5' 2,2',3.4 '.5,6 2,2'. 3, 4'.5.6' 2,2'. 3, 4 '.5- ,6 2.2'. 3.4 '.6.6' 2.2'.3.S .5' .6 2.2'. 3.S .6,6' 2,2' .4.4 '.5.5' 2,2'. 4.4 '.5.6' 2.2'. 4. 4 ',6,6' 2,3,3'. 4 .4'. 5 2.3, 3'. 4 ,4'.S' 2.3,3'. 4 .4'. 6 2.3,3'. 4 .5.5' 2.3.3'. 4 .5.6 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 2, 2'. 3, 3', 4. 4 ' , 5 2.2' .3. 3 ' . 4. 4 ' , 6 2, 2', 3. 3 ' , 4. 5. 5' 2, 2 ' , 3, 3 ' , 4. 5. 6 2,2'. 3. 3 ' , 4. 5 ' , 6 2, 2'. 3. 3 ' , 4.6. 6' 2, 2', 3. 3 ' , 5, 5 ' . 6 2. 2 ' . 3. 3 ' , 5, 6, 6' 2.2'. 3, 4, 4 ' , 5.5' 2. 2' .3, 4, 4'. 5, 6 2. 2' .3. 4. 4 ' , £ . 6 ' 2,2'.3,4,4',5',6 2, 2 ' , 3, 4, 4 ' . 6, 6' 2.2>.3,4,5.5'.6 2. 2'. 3, 4,5. 6, 6' 2,2', 3. 4 ' . 5. 5 ' . 6 2, 2', 3, 4 ' , 5,6, 6' 2.3, 3'. 4. 4', 5,5' 2, 3, 3', 4,4', 5, 6 2, 3, 3', 4, 4 ' , 5' ,6 2. 3, 3', 4, 5. 5 ' . 6 2.3,3' .4' .5.3'. 6 Ocucnlorotiionenyls 194 195 196 197 198 199 200 201 202 203 204 205 2,2', 3. 3'. .4', 5. 5' 2, 2'. 3, 3 ' , , 4 ' , 5, 6 2, 2 ' , 3, 3 ' . ,4', = , 5 ' 2, 2', 3, 3', ,4' ,5, 6' 2, 2', 3. 3 ' , S . S ' . S 2, 2', 3. 3 ' , ,5.6,6' 2.2', 3. 3 ' . .5'. 6. 6' 2, 2', 3, 3 ' , . S . 5 ' , 5 ' 2.2', 3, 3' .5. 5 ' , 6, 6' 2, 2'. 3, 4 . 4 ' , 5, 5 ' , 6 2.2* .3, 4, 4' ,5. 6.6' 2.3.3',4,4'f5,5',6 206 207 208 2.2'.3,3'.4.4',5,5',6 2, 2'. 3. 3 ' , 4, 4 ' . 5, 5, 6' 2. 2'. 3, 3'. 4, 5. 5 ' , 6, 6' NofueHloro61on«nyls Bteichlorebfomnyl Anil. Clw*., 302. 20-31 (I960). B-3 2.3,3'. 4, 5 ' , 6 2, 3, 3'. 4 ' . 5. 5' 2. 3.3'. 4'. 5, 6 2, 3. 3'. 4'. 5 ' , 6 2,3, 3'. 5, 5 ' . 6 2, 3, 4. 4 ' . 5.6 1 2.3',4,4'.5.5 2.3',4.4',5'.5 3. 3' .4. 4'. 5, 5' HtptieM orab< phtny 1 1 209 **dopt*d froa ttllicftrfttr, K. >nd Ztll, *., FrtstMus 2. Structure Htmehlorooiphtnylt Z.2',3.3'4,4',5,5',6.6' �2.0 Summary 2.1 The process or product must be sampled such that the specimen collected for analysis is representative of the whole. Statistically designed selection of the sampling position, time, or discrete product units should be employed. The sample must be preserved to prevent PCB loss prior to analysis. Customary inventory storage may be adequate for products. For intermediates, process samples, and other non-product specimens, storage at 4°C with optional preservation at low pH is recommended. 2.2 The sample is mechanically homogenized and subsampled if necessary. The sample is then spiked with four 13C PCB surrogates and the surrogates incorporated by further mechanical agitation. 2.3 The surrogate-spiked sample is extracted and cleaned up at the discretion of the analyst. Simple dilution or direct injection is permissible. Possible extraction techniques include liquidliquid partition, thermal desorption, and sorption onto resin columns followed by solvent desorption. Cleanup techniques may include liquid-liquid partition, sulfuric acid cleanup, saponification, adsorption chromatography, gel permeation chromatography, or a combination of cleanup techniques. The sample is diluted or concentrated to a final known volume for instrumental determination. 2.4 The PCB content of the sample extract is determined by capillary (preferred) or packed column gas chromatography/electron impact mass spectrometry (CGC/EIMS or PGC/EIMS) operated in the selected ion monitoring (SIM), full scan, or limited mass scan (IMS) mode. 2.5 PCBs are identified by comparison of their retention time and mass spectral intensity ratios to those in calibration standards. 2.6 PCBs are quantitated against the response factors for a mixture of 11 PCB congeners, using the response of the 13C surrogate to compensate for losses in workup and determination and instrument variability. 2.7 The PCBs identified by the SIM technique may be confirmed by full scan CGC/EIMS, retention on alternate GC columns, other mass spectrometric techniques, infrared spectrometry, or other techniques, provided that the sensitivity and selectivity of the technique are demonstrated to be comparable or superior to GC/EIMS. 2.8 The analysis time is dependent on the extent of workup employed. The time required for instrumental analysis of a single sample, excluding data reduction and reporting, is about 30 to 45 min. 2.9 Appropriate quality control (QC) procedures are included to assess the performance of the analyst and estimate the quality of the results . These QC procedures include the demonstration of laboratory capability: periodic analyst certification, the use of control B-4 �charts, and the analysis of blanks, replicates, and standard addition samples. A quality assurance (QA) plan must be developed for each laboratory. 2.10 While several options are available throughout this method, the recommended procedure to be followed is: 2.10.1 The sample is collected according to a scheme which permits extrapolation of the sample data to the whole product or product waste. 2.10.2 The sample is preserved to prevent any loss of PCBs or changes in matrix which may adversely affect recovery. 2.10.3 The sample is mechanically homogenized and subsampled if necessary. 2.10.4 The sample is spiked with four 13C PCB surrogates (4-chlorobiphenyl; 3,3',4,4'-tetrachlorobiphenyl; 2,2',3,3',5,5*,6,6"-octachlorobiphenyl; and decachlorobiphenyl). 2.10.5 Normally, the sample is extracted, although dilution may also be used. 2.10.6 The extract is cleaned up and concentrated to an appropriate volume. 2.10.7 An aliquot of the extract is analyzed by CGC/EIMS operated in the SIM mode. On-column injections onto a 15-m DB-5 capillary column, programmed (for toluene solutions) from 110° to 325°C at 10°/min after a 2-min hold is used. Helium at 45-cm/sec linear velocity is used as the carrier gas. 2.10.8 PCBs are identified by retention time and mass spectral intensities. 2.10.9 PCBs are quantitated against the response factors for a mixture of 11 PCB congeners. 2.10.10 The total PCBs are obtained by summing the amounts for each homolog found, and the concentration is reported as micrograms per gram. 3.0 Interferences 3.1 Method interferences may be caused by contaminants in solvents, reagents, glassware, and other sample processing hardware, leading to discrete artifacts and/or elevated baselines in the total ion current profiles. All of these materials must be routinely demonstrated to be free from interferences by the analysis of laboratory reagent blanks as described in Section 14.4. B-5 �3.1.1 3.1.2 3.2 4.0 Glassware must be scrupulously cleaned. All glassware is cleaned as soon as possible after use by rinsing with the last solvent used. This should be followed by detergent washing with hot water and rinses with tap water and reagent water. The glassware should then be drained dry and heated in a muffle furnace at 400°C for 15 to 30 min. Some thermally stable materials, such as PCBs, may not be eliminated by this treatment. Solvent rinses with acetone and pesticide quality hexane may be substituted for the muffle furnace heating. Volumetric ware should not be heated in a muffle furnace. After it is dry and cool, glassware should be sealed and stored in a clean environment to prevent any accumulation of dust or other contaminants. It is stored inverted or capped with aluminum foil. The use of high purity reagents and solvents helps to minimize interference problems. Purification of solvents by distillation in all-glass systems may be required. All solvent lots must be checked for purity prior to use. Matrix interferences may be caused by contaminants that are coextracted from the sample. The extent of matrix interferences will vary considerably from source to source, depending upon the nature and diversity of the sources of samples. Safety 4.1 The toxicity or carcinogenicity of each reagent used in this method has not been precisely defined; however, each chemical compound should be treated as a potential health hazard. From this viewpoint, exposure to these chemicals must be reduced to the lowest possible level by whatever means available. The laboratory is responsible for maintaining a current awareness file of OSHA regulations regarding the safe handling of the chemicals specified in this method. A reference file of material data handling sheets should also be made available to all personnel involved in the chemical analysis. 4.2 Polychlorinated biphenyls have been tentatively classified as known or suspected human or mammalian carcinogens. Primary standards of these toxic compounds should be prepared in a hood. Personnel must wear protective equipment, including gloves and safety glasses. Congeners highly substituted at the meta and para positions and unsubstituted at the ortho positions are reported to be the most toxic. Extreme caution should be taken when handling these compounds neat or in concentrated solutions. This class includes 3,3',4,4'-tetrachlorobiphenyl (both natural abundance and isotopically labeled). B-6 �4.3 4.4 5.0 Diethyl ether should be monitored regularly to determine the peroxide content. Under no circumstances should diethyl ether be used with a peroxide content in excess of 50 ppm, as an explosion could result. Peroxide test strips manufactured by EM Laboratories (available from Scientific Products Company, Cat. No. P1126-8 and other suppliers) are recommended for this test. Procedures for removal of peroxides from diethyl ether are included in the instructions supplied with the peroxide test kit. Waste disposal must be in accordance with RCRA and applicable state rules. Apparatus and Materials 5.1 Sampling containers - Amber glass bottles, 1-liter or other appropriate volume, fitted with screw caps lined with Teflon. Cleaned foil may be substituted for Teflon if the sample is not corrosive. If amber bottles are not available, samples should be protected from light using foil or a light-tight outer container. The bottle must be washed, rinsed with acetone or methylene chloride, and dried before use to minimize contamination. 5.2 Glassware - All specifications are suggestions only. Catalog numbers are included for illustration only. 5.2.1 Volumetric flasks - Assorted sizes. 5.2.2 Pipets - Assorted sizes, Mohr delivery. 5.2.3 Micro syringes - 10.0 Ml for packed column GC analysis, 1.0 pi for on-column GC analysis. 5.2.4 Chromatographic column - Chromaflex, 400 mm long x 19 mm ID (Kontes K-420540-9011 or equivalent). 5.2.5 Gel permeation chromatograph - GPC Autoprep 1002 (Analytical Bio Chemistry Laboratories, Inc.) or equivalent. 5.2.6 Kuderna-Danish Evaporative Concentrator Apparatus 5.2.6.1 Concentrator tube - 10 ml, graduated (Kontes K-570050-1025 or equivalent). Calibration must be checked. Ground glass stopper size (519/22 joint) is used to prevent evaporation of solvent. 5.2.6.2 Evaporative flask - 500 ml (Kontes K-57001-0500 or equivalent). Attached to concentrator tube with springs (Kontes K-662750-0012 or equivalent) . 5.2.6.3 Snyder column - Three ball macro (Kontes K-503000-0121 or equivalent). B-7 �5.3 Balance - Analytical, capable of accurately weighing 0.0001 g. 5.4 Gas chromatography/mass spectrometer system. 5.4.1 Gas chroraatograph - An analytical system complete with a temperature programmable gas chromatograph and all required accessories including syringes, analytical columns, and gases. The injection port must be designed for oncolumn injection when using capillary columns or packed columns. Other capillary injection techniques (split, splitless, "Grob," etc.) may be used provided the performance specifications stated in Section 7.1 are met. 5.4.2 Capillary GC column - A 12-20 m long x 0.25 mm ID fused silica column with a 0.25 (Jm thick DB-5 bonded silicone liquid phase (J&W Scientific) is recommended. Alternate liquid phases may include OV-101, SP-2100, Apiezon L, Dexsil 300, or other liquid phases which meet the performance specifications stated in Section 7.1. 5.4.3 Packed GC column - A 180 cm x 0.2 cm ID glass column packed with 3% SP-2250 on 100/120 mesh Supelcoport or equivalent is recommended. Other liquid phases which meet the performance specifications stated in Section 7.1 may be substituted. 5.4.4 Mass spectrometer - Must be capable of scanning from 150 to 550 daltons every 1.5 sec or less, collecting at least five spectra per chromatographic peak, utilizing a 70-eV (nominal) electron energy in the electron impact ionization mode and producing a mass spectrum which meets all the criteria in Table 2 when 50 ng of decafluorotriphenyl phosphine [DFTPP, bis(perfluorophenyl)phenyl phosphine] is injected through the GC inlet. Any GC-to-MS interface that gives acceptable calibration points at 10 ng per injection for each PCB isomer in the calibration standard and achieves all acceptable performance criteria (Section 10) may be used. Direct coupling of the fused silica column to the MS is recommended. Alternatively, GC-to-MS interfaces constructed of all glass or glasslined materials are recommended. Glass can be deactivated by silanizing with dichlorodimethylsilane. 5.4.5 A computer system that allows the continuous acquisition and storage on machine-readable media of all mass spectra obtained throughout the duration of the chromatographic program must be interfaced to the mass spectrometer. The data system must have the capability of integrating the abundances of the selected ions between specified limits and relating integrated abundances to concentrations using the calibration procedures described in this method. The computer must have software that allows B-8 �TABLE 2. DFTPP KEY IONS AND ION ABUNDANCE CRITERIA Mass Ion abundance criteria 197 198 199 Less than 1% of mass 198 100% relative abundance 5-9% of mass 198 275 10-30% of mass 198 365 Greater than 1% of mass 198 441 Present, but less than mass 443 442 Greater than 40% of mass 198 443 17-23% of mass 442 B-9 �searching any GC/MS data file for ions of a specific mass and plotting such ion abundances versus time or scan number to yield an extracted ion current profile (EICP). Software must also be available that allows integrating the abundance in any EICP between specified time or scan number limits. 6.0 Reagents 6.1 Solvents - All solvents must be pesticide residue analysis grade. New lots should be checked for purity by concentrating an aliquot by at least as much as is used in the procedure. 6.2 Calibration standard congeners - Standards of the PCB congeners listed in Table 3 are available from Ultra Scientific, Hope, Rhode Island; or Analabs, North Haven, Connecticut. 6.3 Calibration standard stock solutions - Primary dilutions of each of the individual PCBs listed in Table 3 are prepared by weighing approximately 1-10 mg of material within 1% precision. The PCB is then dissolved and diluted to 1.0 ml with hexane. The concentration is calculated in mg/ml. The primary dilutions are stored at 4°C in screw-cap vials with Teflon cap liners. The meniscus is marked on the vial wall to monitor solvent evaporation. Primary dilutions are stable indefinitely if the seals are maintained. The validity of primary and secondary dilutions must be monitored on a quarterly basis by analyzing four quality control check samples (see Section 14.2). 6.4 Working calibration standards - Working calibration standards are prepared that are similar in PCB composition and concentration to the samples by mixing and diluting the individual standard stock solutions. Example calibration solutions are shown in Table 3. The mixture is diluted to volume with pesticide residue analysis quality hexane. The concentration is calculated in ng/ml as the individual PCBs. Dilutions are stored at 4°C in narrow-mouth, screw-cap vials with Teflon cap liners. The meniscus is marked on the vial wall to monitor solvent evaporation. These secondary dilutions can be stored indefinitely if the seals are maintained. These solutions are designated "CSxxx," where the xxx is used to encode the nominal concentration in ng/ml. 6.5 Alternatively, certified stock solutions similar to those listed in Table 3 may be available from a supplier, in lieu of the procedure described in Section 6.4. 6.6 DFTPP standard - A 50-ng/|Jl solution of DFTPP is prepared in acetone or another appropriate solvent. 6.7 Surrogate standard stock solution - The four 13C-labeled PCBs listed in Table 4 may be available from a supplier as a certified solution. This solution may be used as received or diluted further. These solutions are designated "SSxxx," where the xxx is used to encode the nominal concentration in (Jg/ml. B-10 �TABLE 3. Homolog CONCENTRATIONS OF CONGENERS IN PCS CALIBRATION STANDARDS (ng/ml)a Congener no. CS100 CS1000 CS050 CS010 1 1 1,040 104 52 10 1 3 1,000 100 50 10 2 7 1,040 104 52 10 3 30 1,040 104 52 10 4 50 1,520 152 76 15 5 97 1,740 174 87 17 6 143 1,920 192 96 19 7 183 2,600 260 130 26 8 202 4,640 464 232 46 9 207 5,060 506 253 51 10 209 4,240 424 212 42 4 255 255 255 255 1 211 (RS) 104 104 104 104 4 212 (RS) 257 257 257 257 8 213 (RS) 407 407 407 407 10 a 210 (IS) 214 (RS) 502 502 502 502 Concentrations given as examples only. B-ll �TABLE 4. COMPOSITION OF SURROGATE SPIKING SOLUTION (SS100) CONTAINING 13 C-LABELED PCBsa Congener no. Compound Concentration ((jg/ml) 211 104 212 (13C12)3,3',4,4'-tetrachlorobiphenyl 257 213 (13C12)2,2(,3,3',5,5',6,6'-octachlorobiphenyl 395 214 a (l',2',3',4l,5',6l-13C6)4-chlorobiphenyl (13C12)decachlorobiphenyl 502 Concentrations given as examples only. B-12 �6.8 Internal standard solution - A solution of d6-3,3',4,4'-tetrachlorobiphenyl is prepared at a nominal concentration of 1-10 mg/ml in hexane. The solution is further diluted to give a working standard. 6.9 Solution stability - The calibration standard, surrogate, and DFTPP solutions should be checked frequently for stability. These solutions should be replaced after 6 months, or sooner if comparison with quality control check samples indicates compound degradation or concentration change. 6.10 Quality control check samples will be supplied by the Agency. 7.0 Calibration 7.1 The gas chromatograph must meet the minimum operating parameters shown in Tables 5 and 6, daily. If all criteria are not met, the analyst must adjust conditions and repeat the test until all criteria are met, 7.2 The mass spectrometer must meet the minimum operating parameters shown in Tables 2, 7, and 8, daily. If all criteria are not met, the analyst must retune the spectrometer and repeat the test until all conditions are met. 7.3 The PCB response factors (RF ) must be determined using Equation 7-1 for the analyte homologs? A x M. RF = -£ ~ Eq. 7-1 P A is x Mp where RF = response factor of a given PCB congener A = area of the characteristic ion for the PCB congener ™ peak M = mass of PCB congener injected (nanograms) A. = area of the characteristic ion for the internal standard peak M. = mass of internal standard injected (nanograms) IS Using the same conditions as for RF , the surrogate response factors (RF ) must be determined using Equation 7-2. s A x M. w *= t f ^ where A = area of the characteristic ion for the surrogate peak M s = mass of surrogate injected (nanograms) Other terms are the same as defined in Equation 7-1. B-13 �TABLE 5. OPERATING PARAMETERS FOR CAPILLARY COLUMN GAS CHROMATOGRAPHIC SYSTEM Recommended Parameter Tolerance Liquid phase Finnigan 9610 15 m x 0.255 mm ID Fused silica DB-5 (J&W) Liquid phase thickness 0.25 |Jm Carrier gas Helium Carrier gas velocity 45 cm/sec Injector Injector temperature On-column (J&W) c Optimum performance Other Optimum performance Injection volume 1.0 plc 70°C (2 min)d 70°-325°C at 10°C/min£ Other Other Other Transfer line temperature None 280°C Glass jet or othe p Optimum6 Tailing factor 0.7-1.5 0.4-3 Peak width 7-10 sec < 15 sec Gas chromatograph Column Other Other Other nonpolar or semipolar < 1 pro Hydrogen Optimum performance p Initial column temperature Column temperature program Separator a Substitutions permitted with any common apparatus or technique provided performance criteria are met. b Measured by injection of air or methane at 270°C oven temperature. c For on-column injection, manufacturer's instructions should be followed regarding injection technique. d With on-column injection, initial temperature equals boiling point of the solvent; in this instance, hexane. e C 12 Cl 1 o elutes at 270°C. Programming above this temperature ensures a clean column and lower background on subsequent runs. f Fused silica columns may be routed directly into the ion source to prevent separator discrimination and losses. g High enough to elute all PCBs, but not high enough to degrade the column if routed through the transfer line. h Tailing factor is width of front half of peak at 10% height divided by width of back half of peak at 10% height for single PCB congeners in solution CSxxx. i Peak width at 10% height for a single PCB congener is CSxxx. B-14 �TABLE 6. OPERATING PARAMETERS FOR PACKED COLUMN GAS CHRQMATOGRAPHY SYSTEM Tolerance Recommended Parameter Gas chromatograph Finnigan 9610 Other3 Column 180 cm x 0.2 cm ID glass Other Column packing 3% SP-2250 on 100/ 120 mesh Supelcoport Other nonpolar or semipolar Carrier gas Helium Hydrogen Carrier gas flow rate 30 ml/min Optimum performance Injector On-column Other Injector temperature 250°C Optimum Injection volume 1.0 pi S 5 |Jl Initial column temperature 150°C, 4 min Other Column temperature program 150°-260°C 3t 8°/min Other Separator Glass jet Other Transfer line temperature 280°C Optimum Tailing factor 0.7-1.5 0.4-3 Peak width 10-20 sec < 30 sec o a Substitutions permitted if performance criteria are met. b High enough to elute all PCBs. c Tailing factor is width of front half of peak at 10% height divided by width of back half of peak at 10% height for single PCB congeners in solution CSxxx. d Peak width at 10% height for a single PCB congener in CSxxx. B-15 �TABLE 7. OPERATING PARAMETERS FOR QUADRUPOLE MASS SPECTROMETER SYSTEM Parameter Recommended Tolerance Mass spectrometer Finnigan 4023 Other3 Data system Incos 2400 Other Scan range 95-550 Other Scan time 1 sec Otherb Resolution Unit Optimum performance Ion source temperature 280°C 200°-300°C Electron energy 70 eV Optimum performance Trap current 0.2 mA Optimum performance Multiplier voltage -1,600 V Optimum performance Preamplifier sensitivity 10"6 A/V Set for desired working range a Substitutions permitted if performance criteria are met. b Greater than five data points over a GC peak is a minimum. c Filaments should be shut off during solvent elution to improve instrument stability and prolong filament life, especially if no separator is used. B-16 �TABLE 8. OPERATING PARAMETERS FOR MAGNETIC SECTOR MASS SPECTROMETER SYSTEM Parameter Tolerance Recommended Mass spectrometer Finnigan MAT 31 1A Other3 Data system Incos 2400 Other Scan range 98-550 Other Scan mode Exponential Other Cycle time 1.2 sec Otherb Resolution 1,000 > 500 Ion source temperature 280°C 250°-300°C Electron energy 70 eV 70 eV Emission current 1-2 mA Optimum Filament current Optimum Optimum Multiplier -1,600 V Optimum a Substitutions permitted if performance criteria are met. b Greater than five data points over a GC peak is a minimum. c Filaments should be shut off during solvent elution to improve instrument stability and prolong filament life, especially if no separator is used. B-17 �If specific congeners are known to be present and if standards are available, selected RF values may be employed. For general samples, solutions CSxxx and SSxxx or a mixture (Tables 3 and 4), with a similar level of internal standard (de-3,31,4,4'-tetrachlorobiphenyl) added, may be used as the response factor solution. The PCB-surrogate pairs to be used in the RF calculation are listed in Table 9. Generally, only the primary ions of both the analyte and surrogate are used to determine the RF values. If alternate ions are to be used in the quantitation, the RF must be determined using that characteristic ion. The RF value must be determined in a manner to assure ±20% accuracy and precision. For instruments with good day-to-day precision, a running mean (RF) based on seven values determined once each day may be appropriate. Other options include, but are not limited to, triplicate determinations of a single concentration spaced throughout a day or determination of the RF at three different levels to establish a working curve. If replicate RF values differ by greater than ±10% RSD, the system performance should be monitored closely. If the RSD is greater than ±20%, the data set must be considered invalid and the RF redetermined before further analyses are done. 7.4 7.5 8.0 If the GC/EIMS system has not been demonstrated to yield a linear response or if the analyte concentrations are more than two orders of magnitude different from those in the RF solution, a calibration curve must be prepared. If the analyte and RF solution concentrations differ by more than one order of magnitude, a calibration curve should be prepared. A calibration curve should be established with triplicate determinations at three or more concentrations bracketing the analyte levels. The relative retention time (RRT) windows for the 10 homologs and surrogates must be determined. If all congeners are not available, a mixture of available congeners or an Aroclor mixture (e.g., 1016/1254/1260) may be used to estimate the windows. The windows must be set wider than observed if all isomers are not determined. Typical RRT windows for one column are listed in Table 10. The windows may differ substantially if other GC parameters are used. Sample Collection, Handling, and Preservation 8.1 Amber glass sample containers should have Teflon-lined screw caps. With noncorrosive samples, methylene chloride-washed aluminum foil liners may be substituted. The volume and configuration are determined by the amount of sample to be collected and its physical properties. For dry powders, other containers such as heavy-walled polyethylene bags may be appropriate. B-18 �TABLE 9. PAIRINGS OF ANALYTE, CALIBRATION, AND SURROGATE COMPOUNDS Analyte Congener no . 1 2,3 Calibration standard Compound 2-C12H9Cl 3- and 4-C12H9Cl 1 C "ID 1 £ OQ f* TT f> "I ^ 1 2 8 2 P IT PI /•A — Cl **U ~ o 1 P IIP! L ^ 2 6 ^ •*- 4 ^ oZ" 1Z / OO 1O T L.^2"5^'-'-5 128-169 170-193 194-205 206-208 209 C12H4C16 C12H3C17 C12H2C18 C12HC19 Ci2CliQ 4 ID" oy *-*l2**7'-'^-3 P TT O T Congener no. I 3 7 30 50 97 143 183 202 207 209 Compound 2 4 2,4 2,4 ,6 2,2 ',4,6 2,2 ',3', 4, 5 2,2 ',3,4,5 ,6' 2,2 ',3', 4, ,5', 6 4' 2»2 ',3,3',5, 5', 6, 6' 2,2 ',3,3',4, 4', 5, 6, 6' C12Clio w a Ballschmiter numbering system, see Table 1. Surrogate Congener no. Compound 211 211 211 212 212 212 212 213 213 213 214 13 C6-4 C6-4 13 C6-4 13 Cl2-3,3' ,4 ,4' 13 C12-3,3' ,4 ,4' 13 Cl2-3,3' ,4 ,4' 13 C12-3,3' ,4 ,4' 13 Cl2-2,2' ,3 ,3' , 5,5', 6, 6' 13 C12-2,2' ,3 ,3' ,5,5' ,6, 6' 13 C12-2,2' ,3 ,3' ,5, 5', 6, 6' 13 C12C110 13 �TABLE 10. PCB homo log RELATIVE RETENTION TIME (RRT) RANGES OF PCB HOMOLOGS VERSUS de-3,31.4,4'-TETRACHLOROBIPHENYL No. of isomers measured Observed range of RRTs3 Calibration solution Congener Observed no. RRT3 Projected range of RRTs 3 0.40-0.50 1 3 0.43 0.50 0.35-0.55 10 0.52-0.69 7 0.58 0.35-0.80 9 0.62-0.79 30 0.65 0.35-1.10 Tetrachloro 16 0.72-1.01 50 0.75 0.55-1.05 Pentachloro 12 0.82-1.08 97 0.98 0.80-1.10 Hexachloro 13 0.93-1.20 143 1.05 0.90-1.25 Heptachloro 4 1.09-1.30 183 1.15 1.05-1.35 Octachloro 6 1.19-1.36 202 1.19 1.10-1.50 Nonachloro 3 1.31-1.42 207 1.33 1.25-1.50 Decachloro 1 1.44-1.45 209 1.44 1.35-1.50 Monochloro Dichloro Trichloro a The RRTs of the 77 congeners and a mixture of Aroclor 1016/1254/1260 were measured versus 3,3',4,4'-tetrachlorobiphenyl-de (internal standard) using a 15-m J&W DB-5 fused silica column with a temperature program of 110°C for 2 min, then 10°C/min to 325°C, helium carrier at 45 cm/sec, and an oncolumn injector. A Finnigan 4023 Incos quadrupole mass spectrometer operating with a scan range of 95-550 daltons was used to detect each PCB congener. b The projected relative retention windows account for overlap of eluting homologs and take into consideration differences in operating systems and lack of all possible 209 PCB congeners. B-20 �8.2 Sample bottle preparation 8.2.1 8.2.2 Sample bottles are heated to 400°C for 15 to 20 min or rinsed with pesticide grade acetone or hexane and allowed to air dry. 8.2.3 8.3 All sample containers and caps should be washed in detergent solution, rinsed with tap water, and then with distilled water. The bottles and caps are allowed to drain dry in a contaminant-free area. Then the caps are rinsed with pesticide grade hexane and allowed to air dry. The clean bottles are stored inverted or sealed until use. Sample collection 8.3.1 8.3.2 Discrete product units - If the product is small enough that one or more discrete units would be used as the analytical sample, a statistically random sampling approach is recommended. 8.3.3 Liquids or free-flowing solids - If possible, the source is mixed thoroughly before collecting the sample. If mixing is impractical, the sample should be collected from a representative area of the source. If the liquid is flowing through an enclosed system, sampling through a valve should be randomly timed. 8.3.4 8.4 The primary consideration in sample collection is that the sample collected be representative of the whole. Therefore, sampling plans or protocols for each individual producer's situation will have to be developed. The recommendations presented here describe general situations. The number of replicates and sampling frequency also must be planned prior to sampling. Solids - Larger bulk solids which must be subsampled to get a reasonably sized analytical sample must be treated on a case-by-case basis. A representative sample should be obtained by designing a sampling location selection scheme such that all parts of the whole have a finite, known probability of inclusion. Based on such a scheme, the PCS content of the sample can be used to extrapolate to the content of the whole. Sample preservation - Product samples should be stored as the bulk or packaged product inventory would be stored, or in a cool, dry, dark area. Intermediates, process samples, or other non-product specimens should be stored at 4°C. If there is a possibility of microbial degradation, addition of HgSC^ during collection to a pH < 2 is recommended. A test strip is used to monitor pH. Storage times in excess of 4 weeks are not recommended. B-21 �If residual chlorine is present in the sample, it should be quenched with sodium thiosulfate. EPA Methods 330.4 and 330.5 may be used to measure the residual chlorine.1 Field test kits are available for this purpose. 9.0 Sample Preparation Since a wide variety of matrices may be subjected to analysis by this method, the extraction/cleanup procedure cannot be specified. This section describes general guidelines for subsampling, addition of 13C surrogates, dilution, extraction, cleanup, extract concentration, and other sample preparation procedures. 9.1 Sample homogenization and subsampling - The sample is homogenized by shaking, blending, shredding, crushing, or other appropriate mechanical technique. A representative subsample of 100 g or other known mass is then taken. The sample size is dependent upon the anticipated PCB levels and difficulty of the subsequent extraction/ cleanup steps. Note: The precision of the mass determination at this step will be reflected in the overall method precision. Therefore, an analytical balance must be used to assure that the weight is accurate to ±1% or better. 9.2 Surrogate addition - An appropriate volume of surrogate solution SSxxx is pipetted into the sample. The final concentration of the surrogates must be in the working range of the calibration and well above the matrix background. The surrogates are thoroughly incorporated by further mechanical agitation. For nonviscous liquids, shaking for 30 sec should be sufficient. For viscous liquids or free-flowing solids, 10-min tumbling is recommended. In cases where inadequate incorporation may be expected, such as solids, overnight equilibration with agitation is recommended. Note: The volume measurement of the spiking solution is critical to the overall method precision. The analyst must exercise caution that the volume is known to ±J% or better. Where necessary, calibration of the pipet is recommended. 9.3 Sample preparation (extraction/cleanup) - After addition of the surrogates, the sample is further treated at the discretion of the analyst, provided that the GC/EIMS response of the four surrogates meets the criteria listed in Section 7.0. The literature pertaining to these techniques has been reviewed.2 Several possible techniques are presented below for guidance only. The applicability of any of these techniques to a specific sample matrix must be determined by the precision and accuracy of the 13C PCB surrogate recoveries, as discussed in Section 14.2. B-22 �9.3.1 Extraction1 9.3.1.1 Dilution - In some cases, where the PCB concentration is high, a simple volumetric dilution with an appropriate solvent may be sufficient sample preparation. 9.3.1.2 Direct injection - If sample viscosity permits, direct injection with no dilution is permissible. 9.3.1.3 Liquid-liquid extraction - If the matrix is aqueous (or another solvent in which PCBs are only slightly soluble), a liquid-liquid partition may be effective. The solvent, number of extractions, solvent-to-sample ratio, and other parameters are chosen at the analyst's discretion. 9.3.1.4 Sorbent column extraction - PCBs may be isolated from free-flowing liquids onto sorbent columns. The selection of sorbent (XAD, Porapak, carbonpolyurethane foam, etc.) will depend on the nature of the matrix. The available methods have been reviewed.2 9.3.1.5 Thermal desorption - If the matrix is nonvolatile, thermal desorption of the PCBs onto a sorbent column, filter, or cold trap may be an effective extraction/cleanup method. 9.3.2 Cleanup - Several tested cleanup techniques are described below. All but the base cleanup (9.3.2.8) were previously validated for PCBs in transformer fluids.3 Depending upon the complexity of the sample, one or more of the techniques may be required to fractionate the PCBs from interferences. For most cleanups a concentrated (1-5 ml) extract should be used. 9.3.2.1 Acid cleanup 9.3.2.1.1 Place 5 ml of concentrated sulfuric acid into a 40-ml narrow-mouth screwcap bottle. Add the sample extract. Seal the bottle with a Teflon-lined screw cap and shake for 1 min. 9.3.2.1.2 Allow the phases to separate, transfer the sample (upper phase) with three rinses of 1-2 ml solvent to a clean container and concentrate to an appropriate volume. B-23 �9.3.2.1.3 9.3.2.1.4 9.3.2.2 Analyze as described in Section 10.0. If the sample is highly contaminated, a second or third acid cleanup may be employed. Florisil column cleanup 9.3.2.2.1 9.3.2.2.2 Place a 20-g charge of Florisil, activated overnight at 130°C, into a Chromaflex column. Settle the Florisil by tapping the column. Add about 1 cm of anhydrous sodium sulfate to the top of the Florisil. Pre-elute the column with 70-80 ml of hexane. Just before the exposure of the sodium sulfate layer to air, stop the flow. Discard the eluate. 9.3.2.2.3 Add the sample extract to the column. 9.3.2.2.4 Carefully wash down the inner wall of the column with 5 ml of hexane. 9.3.2.2.5 Add 200 ml of 6% ethyl ether/hexane and set the flow to about 5 ml/min. 9.3.2.2.6 Collect 200 ml of eluate in a KudernaDanish flask. All the PCBs should be in this fraction. Concentrate to an appropriate volume. 9.3.2.2.7 9.3.2.3 Variations among batches of Florisil (PR grade or equivalent) may affect the elution volume of the various PCBs. For this reason, the volume of solvent required to completely elute all PCBs must be verified by the analyst. The weight of Florisil can then be adjusted accordingly. Analyze the sample as described in Section 10.0. Alumina column cleanup 9.3.2.3.1 B-24 Adjust the activity of the alumina (Fisher A450 or equivalent) by heating to 200°C for 2 to 4 hr. When cool, add 3% water (wt:wt) and mix until uniform. Store in a tightly sealed bottle. Allow the deactivated alumina to equilibrate at least 1/2 hr before use. Reactivate weekly. �9.3.2.3.2 Variations between batches of alumina may affect the elution volume of the various PCBs. For this reason, the volume of solvent required to completely elute all of the PCBs must be verified by the analyst. The weight of alumina can then be adjusted accordingly. 9.3.2.3.3 Place a 50-g charge of alumina into a Chromaflex column. Settle the alumina by tapping. Add about 1 cm of anhydrous sodium sulfate. Preelute the column with 70-80 ml of hexane. Just before exposure of the sodium sulfate layer to air, stop the flow. Discard the eluate. 9.3.2.3.4 Add the sample extract to the column. 9.3.2.3.5 Carefully wash down the inner wall of the column with 5 ml of hexane. 9.3.2.3.6 Add 295 ml of hexane to the column. 9.3.2.3.7 Discard the first 50 ml. 9.3.2.3.8 Collect 250 ml Kuderna-Danish PCBs should be Concentrate to of the hexane in a flask. All of the in this fraction. an appropriate volume. 9.3.2.3.9 Analyze the sample as described in Section 10.0. 9.3.2.4 Silica gel column cleanup 9.3.2.4.1 Activate silica gel (Davison Grade 950 or equivalent) at 135°C overnight. 9.3.2.4.2 Variations between batches of silica gel may affect the elution volume of the various PCBs. For this reason, the volume of solvent required to completely elute all of the PCBs must be verified by the analyst. The weight of silica gel can then be adjusted accordingly. B-25 �9.3.2.4.3 Place a 25-g charge of activated silica gel into a Chromaflex column. Settle the silica gel by tapping the column. Add about 1 cm of anhydrous sodium sulfate to the top of the silica gel. 9.3.2.4.4 Pre-elute the column with 70-80 ml of hexane. Discard the eluate. Just before exposing the sodium sulfate layer to air, stop the flow. 9.3.2.4.5 Add the sample extract to the column. 9.3.2.4.6 Wash down the inner wall of the column with 5 ml of hexane. 9.3.2.4.7 Elute the PCBs with 195 ml of 10% diethyl ether in hexane (v:v). 9.3.2.4.8 Collect 200 ml Kuderna-Danish PCBs should be Concentrate to of the eluate in a flask. All of the in this fraction. an appropriate volume. 9.3.2.4.9 Analyze the sample as described in Section 10.0. 9.3.2.5 Gel permeation cleanup 9.3.2.5.1 Set up and calibrate the gel permeation chromatograph with an SX-3 column according to the Autoprep instruction manual. Use 15% methylene chloride in cyclohexane (v:v) as the mobile phase. 9.3.2.5.2 Inject 5.0 ml of the sample extract into the instrument. Collect the fraction containing the PCBs (see Autoprep operator's manual) in a Kuderna-Danish flask equipped with a 10-ml ampul. 9.3.2.5.3 Concentrate the PCB fraction to an appropriate volume. 9.3.2.5.4 Analyze the sample as described in Section 10.0. B-26 �9.3.2.6 Acetonitrile partition 9.3.2.6.1 Place the sample extract into a 125-ml separately funnel with enough hexane to bring the final volume to 15 ml. Extract the sample four times by shaking vigorously for 1 min with 30-ml portions of hexane-saturated acetonitrile. 9.3.2.6.2 Combine and transfer the acetonitrile phases to a 1-liter separatory funnel and add 650 ml of distilled water and 40 ml of saturated sodium chloride solution. Mix thoroughly for about 30 sec. Extract with two 100-ml portions of hexane by vigorously shaking about 15 sec. 9.3.2.6.3 Combine the hexane extracts in a 1-liter separatory funnel and wash with two 100-ml portions of distilled water. Discard the water layer and pour the hexane layer through an 8-10 cm anhydrous sodium sulfate column into a 500-ml Kuderna-Danish flask equipped with a 10-ml ampul. Rinse the separatory funnel and column with three 10-ml portions of hexane. 9.3.2.6.4 Concentrate the extracts to an appropriate volume. 9.3.2.6.5 Analyze as described in Section 10.0. 9.3.2.7 Florisil slurry cleanup 9.3.2.7.1 Place the sample extract into a 20-ml narrow-mouth screw-cap container. Add 0.25 g of Florisil (PR grade or equivalent). Seal with a Teflon-lined screw cap and shake for 1 min. 9.3.2.7.2 Allow the Florisil to settle; then decant the treated solution into a second container with rinsing. Concentrate the sample to an appropriate volume. Analyze as described in Section 10.0. B-27 �9.3.2.8 Base cleanup4 9.3.2.8.1 Quantitatively transfer the concentrated extract to a 125-ral extraction flask with the aid of several small portions of solvent. 9.3.2.8.2 Evaporate the extract just to dryness with a gentle stream of dry filtered nitrogen, and add 25 ml of 2.5% alcoholic KOH. 9.3.2.8.3 Add a boiling chip, put a water condenser in place, and allow the solution to reflux on a hot plate for 45 min. 9.8.2.8.4 After cooling, transfer the solution to a 250-ml separatory funnel with 25 ml of distilled water. 9.3.2.8.5 Rinse the extraction flask with 25 ml of hexane and add it to the separatory funnel. 9.3.2.8.6 Stopper the separatory funnel and shake vigorously for at least 1 min. Allow the layers to separate, and transfer the lower aqueous phase to a second separatory funnel. 9.3.2.8.7 Extract the saponification solution with a second 25-ml portion of hexane. After the layers have separated, add the first hexane extract to the second separatory funnel and transfer the aqueous alcohol layer to the original separatory funnel. 9.3.2.8.8 Repeat the extraction with a third 25-ml portion of hexane. Discard the saponification solution, and combine the hexane extracts. 9.3.2.8.9 Concentrate the hexane layer to an appropriate volume, and analyze as described in Section 10.0. B-28 �10.0 Gas Chromatographic/Electron Impact Mass Spectrometric Determination 10.1 Internal standard addition - An appropriate volume of the internal standard solution is pipetted into the sample. The final concentration of the internal standard must be in the working range of the calibration and well above the matrix background. The internal standard is thoroughly incorporated by mechanical agitation. Note: The volumetric measurement of the internal standard solution is critical to the overall method precision. The analyst must exercise caution that the volume is known to be ±1% or better. Where necessary, calibration of the pipet is recommended. 10.2 Tables 2, and 5 through 8 summarize the recommended operating conditions for analysis. Figure 1 presents an example of a chromatogram. 10.3 While the highest available chromatographic resolution is not a necessary objective of this protocol, good chromatographic performance is recommended. With the high resolution of CGC, the probability that the chromatographic peaks consist of single compounds is higher than with PGC. Thus, qualitative and quantitative data reduction should be more reliable. 10.4 After performance of the system has been certified for the day and all instrument conditions set according to Tables 2, and 5 through 8, inject an aliquot of the sample onto the GC column. If the response for any ion, including surrogates and internal standards, exceeds the working range of the system, dilute the sample and reanalyze. If the responses of surrogates, internal standards, or analytes are below the working range, recheck the system performance. If necessary, concentrate the sample and reanalyze . 10.5 Record all data on a digital storage device (magnetic disk, tape, etc.) for qualitative and quantitative data reduction as discussed below. 11.0 Qualitative Identification 11.1 Selected ion monitoring (SIM) or limited mass scan (LMS) data The identification of a compound as a given PCB homolog requires that two criteria be met: 11.1.1 (1) The peak must elute within the retention time window set for that homolog (Section 7.5); and (2) the ratio of two ions obtained by SIM (Table 11) or by LMS (Table 12) must match the natural ratio within ±20%. The analyst must search the higher mass windows, in particular M+70, to prevent misidentification of a PCB fragment ion cluster as the parent. B-29 �-a 4-1 x~* "nj C too 0 cn M QJ l-i QJ K C C/5 1101 100.0 0 TJ rH CO 1 rH 01 rH rH 1 rH rH K*~l rH r*. £*"! ^ C 0) O CO C C c !>•» QJ QJ OJ x; r^ C 0) C QJ XI P. XI P, XI -H rH P. •H Xl o XI O M >> C Ol rH rH xi . x;, pi p -H _fl -H _ /*-* •H -H O O X> XI /—^ O 0) O r-l V-l 4-1 rH CO 60 XI O O O rH H rC XI CJ O O O r-l 1-1 C C CO D 0 r-l 4-1 rH X o o cn S 2 — <r -<r 01 c oj p. H i rH ^r •H f>~> Xi r-iiG -> rH ^ rH > . Xi ro x; p. -H x: xi P* p. O •rt o o CO VJ S-i O rH 4-1 OJ (-I O 1 O rH ^O 4-1 CJ -H l-l H 1 ^D •* O> PL, | m <J P O rH X! O •H O | XI u o c o a x: CO - c CN CN co 117-1 OJ GI3 CN T T 1 I A V. ' 1 " I 1 rH ' £*-» C 0) XI o x ; p. ^-, XI C O rH >. >-l XI -H •H 0) 0 Xl XI 4-1 rC 0) CO rH O rH p>^ CO 0 - 0 rJ O o o O ZZ rH u i v£> v£> VD rH x: 1 -H XI 0 P 0 0 cfl )-i Ml 0 CO P, <-" C OJ 0) H XI 1 P. -H C P . 4-» XI 0 CO rH 4-1 m x; •• CO U CO X Ol SC 1 ^ r-1 cfl O CO D CO OJ Q QJ Q 0 a o ^ *> in v o. m Q) X: ^3•* ^r A pr] 1 ^O co •^ CO •s Ol •> ^ in »> --^ «> ^ ro m v£) *. n •- ^i" n ~^j" * r. CO »\ •CN r> Ol CO CO 01 *^ - l?.0f) oj ii™ A Ol i oi * Ol M78 l;ilc ^ Jl k ii -/20 , _ Off 027 ,'. r-l XI t-> O <T - O Ol <N 1 M 0 1 —1 XI O f-l Xi rH loViw P. m -H XI C QJ •H XI O )_! O 1 O rH 01 >^ 01 XI •-21 OJ *• XI - C rH C w O " C -* O M rH p, rH rJJp. , t 'l O I 112592. ^ O vo co vD ' ' Ki-.-t-j ' 12:111 ,!• >o-.e-:) J— ' v ~ •—~~ " _„ 1 2 J •.;.'.) Till,; Figure 1. Capillary gas chromatography/clectron impact ionization mass spectrometry (CGC/EIMS) chroniatogram or the calibration standard solution required for quantitation of PCBs by homolog. This chroniatogram includes PCBs representative of each homolog, three carbon-13 labeled surrogates, and the deuterated internal standard; The concentration of a J 1 components and the CGC/EIMS parameters are presented in Tables 3, 4, 5, and 7. �TABLE 11. CHARACTERISTIC SIM IONS FOR PCBs Ion (relative intensity) Secondary Tertiary Homolog Primary Ci2H9Cl 188 (100) 190 (33) - CiaHgCls 222 (100) 224 (66) 226 (11) C12H7Cl3 256 (100) 258 (99) 260 (33) Ci2H6Cl4 292 (100) 290 (76) 294 (49) Ci2H5Cl5 326 (100) 328 (66) 324 (61) C12H4Cl6 360 (100) 362 (82) 364 (36) CiaHsCl? 394 (100) 396 (98) 398 (54) CisH^Clg 430 (100) 432 (66) 428 (87) Ci2HCl9 464 (100) 466 (76) 462 (76) Ci2Clib 498 (100) 500 (87) 496 (68) Source: Rote, J. W., and W. J. Morris, "Use of Isotopic Abundance Ratios in Identification of Polychlorinated Biphenyls by Mass Spectrometry," J. Assoc. Offic. Anal. Chem.. 56(1), 188-199 (1973). B-31 �TABLE 12. LIMITED MASS SCANNING (LMS) RANGES FOR PCBs Compound Mass range (m/z) C.ACU 186-190 C12HgCl2 220-226 L 12nyC i-3 254-260 C ^2ngC J-3 288-294 C i2n5Cl5 322-328 Ci2H4Clg 356-364 C12H3C17 386-400 C12H2Clg 426-434 C12HC19 460-468 C12C110 494-504 C12D6C14 294-300 13 192-196 13 300-306 13 438-446 C612C6H9C1 C12H6C14 C12H2C18 13 506-516 C12C110 a Adapted from Tindall, G. W., and P. E. Wininger, "Gas Chromatography-Mass Spectrometry Method for Identifying and Determining Polychlorinated Biphenyls," J. Chromatogr., 196, 109-119 (1980). B-32 �11.1.2 If one or the other of these criteria is not met, interferences may have affected the results, and a reanalysis using full scan EIMS conditions is recommended. 11.2 Full scan data 11.2.1 The peak must elute within the retention time windows set for that homolog (as described in Section 7.5). 11.2.2 The unknown spectrum must match that of an authentic PCB. The intensity of the three largest ions in the molecular cluster (two largest for monochlorobiphenyls) must match the natural ratio within ±20%. Fragment clusters with proper intensity ratios must also be present. 11.2.3 Alternatively, a spectral search may be used to automatically reduce the data. The criteria for acceptable identification include a high index of similarity. For the Incos 2300, a fit of 750 or greater must be obtained. 11.3 Disputes in interpretation - Where there is reasonable doubt as to the identity of a peak as a PCB, the analyst must either identify the peak as a PCB or proceed to a confirmational analysis (see Section 13.0). 12.0 Quantitative Data Reduction 12.1 Once a chromatographic peak has been identified as a PCB, the compound is quantitated based either on the integrated abundance of the SIM data or EICP for the primary characteristic ion in Tables 11 and 12. If interferences are observed for the primary ion, use the secondary and then tertiary ion for quantitation. If interferences in the parent cluster prevent quantitation, an ion from a fragment cluster (e.g., M-70) may be used. Whichever ion is used, the RF must be determined using that ion. The same criteria should be applied to the surrogate compounds (Table 13). 12.2 Using the appropriate analyte-internal standard pair and response factor (RF ) as determined in Section 7.3, calculate the concentration of^each peak using Equation 12-1. A M. V Concentration (yg/g) = ^ • ~ • ^ • ^ Eq. 12-1 is p e i where A = area of the characteristic ion for the analyte PCB ^ peak A. = area of the characteristic ion for the internal standard peak RF = response factor of a given PCB congener B-33 �TABLE 13. CHARACTERISTIC IONS FOR 13 C-LABELED PCS SURROGATES Primary Ion (relative intensity) Secondary 13 194 (100) 196 (33) 13 304 (100) 306 ( 9 4) 302 (78) 13 442 (100) 444 (65) 440 (89) 13 510 (100) 512 (87) 514 (50) Specific compound C612C6H9C1 C12H6C14 C12H2C18 C12C110 B-34 Tertiary �M. = mass of internal standard injected (micrograms) IS M = mass of sample extracted (grams) V. = volume injected (microliters) V = volume of sample extract (microliters) 12.3 If a peak appears to contain non-PCB interferences, which cannot he circumvented by a secondary or tertiary ion, either: 12.3.1 12.3.2 Perform additional chemical cleanup (Section 9) and then reanalyze the sample; or 12.3.3 12.4 Reanalyze the sample on a different column which separates the PCB and interf erents ; Quantitate the entire peak as PCB. Calculate the recovery of the four 13C surrogates using the appropriate surrogate-internal standard pair and response factor (RF. ) as determined in Section 7.4 using Equation 12-2. A M. Recovery ( ) = ^ • - r • ^ • 100 % |Eq. 12-2 is s s where A S = area of the characteristic ion for the surrogate peak A. = area of the characteristic ion for the internal standard 18 peak RF = response factor for the surrogate compound with respect to the internal standard (Equation 7-2) M. = mass of internal standard injected (nanograms) 3.S MS - mass of surrogate, assuming 100% recovery (nanograms) 12.5 Correct the concentration of each peak using Equation 12-3. This is the final reportable concentration. Corrected concentration (pg/g) = 12.6 . 100 Eq. 12-3 Sum all of the peaks for each homolog, and then sum those to yield the total PCB concentration in the sample. Report all numbers in pg/g. The reporting form in Table 14 may be used. If an alternate reporting format (e.g., concentration per peak) is desired, a different report form may be used. The uncorrected concentrations, percent recovery, and corrected recovery are to be reported. 12.7 Round off all numbers reported to two significant figures. B-35 �TABLE 14. ANALYSIS REPORT INCIDENTAL PCBs IN COMMERCIAL PRODUCTS OR PRODUCT WASTES Sample No. Sample Matrix Sample Source Notebook No. or File Location Volume Extracted Extraction/Cleanup Int. Std. Procedure Mass Added (pg) (Circle one) 298 4-Cl(d6) Surrogates Mass Added (pg) (Circle one) 300 Ratio 194 196 100/33 4-C1 304 306 100/49 8-C1 442 444 100/65 10-C1 510 512 100/87 B-36 Intensity 100/49 1-C1 (continued) Ratio Intensity % Recovery �TABLE 14 (continued) Qualitative Analyte 1° 2° I l° T 1 2° Ratio Theoretical 1-C1 188 190 100/33 2-C1 222 224 100/66 3-C1 256 258 100/99 4-C1 292 290 100/76 5-C1 326 328 100/66 6-C1 360 362 100/82 7-C1 394 396 100/98 8-C1 430 432 100/66 9-C1 464 466 100/76 10-C1 498 500 Quantitative Uncorr. Corr. Ion Cone. Cone. OK? Used RF (|Jg/g) (Hg/g) 100/87 Total M8/8 Uncorr. Reported by: Internal Audit: Name Name EPA Audit: Name Signature/Date Signature/Date Signature/Date Organization Organization Organization B-37 M8/8 Corr. �13.0 Confirmation If there is reason to question the qualitative identification (Section 11.0), the analyst may choose to confirm that a peak is not a PCB. Any technique may be chosen provided that it is validated as having equivalent or superior selectivity and sensitivity to GC/EIMS. Some candidate techniques include alternate GC columns (with EIMS detection), GC/CIMS, GC/NCIMS, high resolution EIMS, and MS/MS techniques. Each laboratory must validate confirmation techniques to show equivalent or superior selectivity between PCBs and interferences and sensitivity (limit of quantitation, LOQ). If a peak is confirmed as being a non-PCB, it may be deleted from the calculation (Section 12). If a peak is confirmed as containing both PCB and non-PCB components, it must be quantitated according to Section 12.3. 14.0 Quality Control 14.1 Each laboratory that uses this method must operate a formal quality control (QC) program. The minimum requirements of this program consist of an initial demonstration of laboratory capability and the analysis of spiked samples as a continuing check on performance. The laboratory must maintain performance records to define the quality of data that are generated. After a date specified by the Agency, ongoing performance checks should be compared with established performance criteria to determine if the results of analyses are within accuracy and precision limits expected of the method. 14.2 The analysts must certify that the precision and accuracy of the analytical results are acceptable by: 14.2.1 14.2.2 14.3 The absolute precision of surrogate recovery, measured as the RSD of the integrated EIMS area (A ) for a set s • of samples^ must be ±10%. The mean recovery (R ) of at least four replicates of a QC check sample to be supplied by the Agency must meet Agency-specified accuracy and precision criteria. This forms the initial data base for establishing control limits (see Section 14.3 below). Control limits - The laboratory must establish control limits using the following equations: Upper control limit (UCL) = RC + 3 RSDc Upper warning limit (UWL) = R + 2 RSD Lower warning limit (LWL) = R - 2 RSD Lower control limit (LCL) = R - 3 RSD B-38 �These may be plotted on control charts. If an analysis of a check sample falls outside the warning limits, the analyst should be alerted that potential problems may need correction. If the results for a check sample fall outside the control limits, the laboratory must take corrective action and recertify the performance (Section 14.2) before proceeding with analyses. The warning and control limits should be continuously updated as more check sample replicates are added to the data base. 14.4 Before processing any samples, the analyst should demonstrate through the analysis of a reagent blank that all glassware and reagent interferences are under control. Each time a set of samples is analyzed or there is a change in reagents, a laboratory reagent blank should be processed as a safeguard against contamination. 14.5 Procedural QC - The various steps of the analytical procedure should have quality control measures. These include but are not limited to: 14.5.1 GC performance - See Section 7.1 for performance criteria. 14.5.2 MS performance - See Section 7.2 for performance criteria. 14.5.3 Qualitative identification - At least 10% of the PCB identifications, as well as any questionable results, should be confirmed by a second mass spectrometrist. 14.5.4 Quantitation - At least 10% of all manual calculations, including peak area calculations, must be checked. After changes in computer quantitation routines, the results should be manually checked. 14.6 A minimum of 10% of all samples, one sample per month or one sample per matrix type, whichever is greater, selected at random, must be run in triplicate to monitor the precision of the analysis. An RSD of ±30% or less must be achieved. If the precision is greater than ±30%, the analyst must be recertified (see Section 14.2). 14.7 A minimum of 10% of all samples, one sample per month or one sample per matrix type, whichever is greater, selected at random, must be analyzed by the standard addition technique. Two aliquots of the sample are analyzed, one "as is" and one spiked (surrogate spiking and equilibration techniques are described in Section 9.2) with a sufficient amount of Solution CSxxx to yield approximately 100 |Jg/g of each compound. The samples are analyzed together and the quantitative results calculated. The recovery of the spiked compounds (calculated by difference) must be 80-120%. If the sample is known to contain specific PCB isomers, these isomers may be substituted for solution CSxxx. If the concentrations of PCBs are known to be high or low, the amount added should be adjusted so that the spiking level is 1.5 to 4 times the measured PCB level in the unspiked sample. B-39 �14.8 Interlaboratory comparison - Interlaboratory comparison studies are planned. Participation requirements, level of performance, and the identity of the coordinating laboratory will be presented in later revisions. 14.9 It is recommended that the participating laboratory adopt additional QC practices for use with this method. The specific practices that are most productive depend upon the needs of the laboratory and the nature of the samples. Field duplicates or triplicates may be analyzed to monitor the precision of the sampling technique. Whenever possible, the laboratory should perform analysis of standard reference materials and participate in relevant performance evaluation studies. 15.0 Quality Assurance Each participating laboratory must develop a quality assurance plan according to EPA guidelines.5 The quality assurance plan must be submitted to the Agency for approval. 16.0 Method Performance The method performance is being evaluated. Limits of quantitation; average intralaboratory recoveries, precision, and accuracy; and interlaboratory recoveries, precision, and accuracy will be presented. 17.0 Documentation and Records Each laboratory is responsible for maintaining full records of the analysis. Laboratory notebooks should be used for handwritten records. GC/MS data must be archived on magnetic tape, disk, or a similar device. Hard copy printouts may be kept in addition if desired. QC records should be maintained separately from sample analysis records. The documentation must describe completely how the analysis was performed. Any variances from the protocol must be noted and fully described. Where the protocol lists options (e.g., sample cleanup), the option used and specifics (solvent volumes, digestion times, etc.) must be stated. B-40 �REFERENCES 1. "Methods 330.4 (Titrimetric, DPD-FAS) and 330.5 (Spectrophotometric, DPD) for Chlorine, Total Residual," Methods for Chemical Analysis of Water and Wastes, U.S. Environmental Protection Agency, Environmental Monitoring and Support Laboratory, Cincinnati, Ohio, March 1979, EPA 600-4/79-020. 2. Erickson, M. D., and J. S. Stanley, "Methods of Analysis for Incidentally Generated PCBs—Literature Review and Preliminary Recommendations," Interim Report No. 1, EPA Contract No. 68-01-5915, Task 51, 1982. 3. Bellar, T. A., and J. J. Lichtenberg, "The Determination of Polychlorinated Biphenyls in Transformer Fluid and Waste Oils," Prepared for U.S. Environmental Protection Agency, (1981) EPA-600/4-81-045. 4. American Society for Testing and Materials, "Standard Method for Analysis of Environmental Materials for Polychlorinated Biphenyls," pp. 877-885 in Annual Book of ASTM Standards, Part 40, Philadelphia, Pennsylvania (1980). ANSI/ASTM D 3304 - 77. 5. "Quality Assurance Program Plan for the Office of Toxic Substances," Office of Pesticides and Toxic Substances, U.S. Environmental Protection Agency, Washington, D.C., October 1980. B-41 �APPENDIX C ANALYTICAL METHOD: THE ANALYSIS OF BY-PRODUCTS CHLORINATED BIPHENYLS IN AIR C-l �THE ANALYSIS OF BY-PRODUCT CHLORINATED BIPHENYLS IN AIR 1.0 Scope and Application 1.1 This is a gas chromatographic/electron impact mass spectrometric (GC/EIMS) method applicable to the determination of chlorinated biphenyls (PCBs) in air emitted from commercial production through stacks, as fugitive emissions, or static (room, other containers, or outside) air. The PCBs present may originate either as synthetic by-products or as contaminants derived from commercial PCB products (e.g., Aroclors). The PCBs may be present as single isomers or complex mixtures and may include all 209 congeners from monochlorobiphenyl through decachlorobiphenyl listed in Table 1. 1.2 The detection and quantitation limits are dependent upon the volume of sample collected, the complexity of the sample matrix and the ability of the analyst to remove interferents and properly maintain the analytical system. The method accuracy and precision will be determined in future studies. 1.3 This method is restricted to use by or under the supervision of analysts experienced in the use of gas chromatography/mass spectrometry (GC/MS) and in the interpretation of gas chromatograms and mass spectra. Prior to sample analysis, each analyst must demonstrate the ability to generate acceptable results with this method by following the procedures described in Section 14.2. 1.4 The validity of the results depends on equivalent recovery of the analyte and 13C PCBs. If the *3C PCBs are not thoroughly incorporated in the matrix, the method is not applicable. 1.5 During the development and testing of this method, certain analytical parameters and equipment designs were found to affect the validity of the analytical results. Proper use of the method requires that such parameters or designs must be used as specified. These items are identified in the text by the word "must." Anyone wishing to deviate from the method in areas so identified must demonstrate that the deviation does not affect the validity of the data. Alternative test procedure approval must be obtained from the Agency. An experienced analyst may make modifications to parameters or equipment identified by the term "recommended." Each time such modifications are made to the method, the analyst must repeat the procedure in Section 14.2. In this case, formal approval is not required, but the documented data from Sectin 14.2 must be on file as part of the overall quality assurance program. C-2 �No. scnjcturt NO* NMoenlorooloAfWlf 1 2 1eh1oreb1pntnyli 4 5 6 7 8 9 10 11 12 13 14 15 .2' Is,'5 ,6 .3' ,4 3.4' 3.5 4,4' THc)i1erab1plnny1> 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 2,2' .3 2,2',4 2,2', 5 2.2'. 6 2,3,3' 2,3.4 2.3.4' 2.3,5 2,3,6 2. 3', 4 2,3', 5 2, 3', 6 2,4,4' 2.4.5 2.4,6 2.*' .5 2.4', 6 2'. 3,4 2'. 3.5 3.3'. 4 3,3'.5 3.4,4' 3.4.5 3.4', 5 SI 57 SI 59 10 61 62 13 M IS M 67 68 69 70 71 72 73 74 75 71 77 78 79 80 81 2,3' 2.3* 2.3' 2.3' 2,3' 2.3' 2.3' 2.3' 4,4' 4,5 4.5' 4.6 4', 5 4', 6 1 5.S 5',6 2.4.4-.S 2.4.4'.$ 2'. 3 4.5 3,3' 3,3* 3.3' 3.3' 82 83 84 as 2.3,3'. 4,4' 2,3,3', 4,5 2,3,3'. 4'. 5 2.3.3'. 4. 5' 2.3,3'. 4, 6 2.3,3', 4' .6 2.3.3' .5.5' 2.3.3'. 5.6 2.3.3'. 5'. 6 2.3,4, 4'. 5 2.3,4,4'. 6 2.3.4.5.6 2.3.4', 5,6 2.31. 4, 4' .5 2.3'. 4,4'. 6 2,3' ,4,5, 5' 2.3' ,4.5', 6 2'.3,3'.4,5 2 1 .3.4.4' ,5 2'. 3. 4.5. 5' 2'.3,4. 5.6' 3.3',4,4',5 3.3'. 4, 5,5' Htuehloreblphtnyla 4,4' 4,5 4.5' 5.5' 3.4.4' .5 2.2'. 3, 3'. 4 2.2'.3.3',5 2.2'. 3.3'. 6 2.2', 3,4,4' 2.2',3.4,5 2. 2' ,3.4.5' 2,2',3,4.6 2,2'. 3.4,6' 2,2',3,4',5 2,2'. 3, 4'. 6 2.2' ,3.5,5' 2,2'. 3, 5.6 2,2'. 3,5.6' 2.2'. 3. 5'. 8 2.2',3.6.6' 2.2'. 3'. 4. 5 2.2'.3'.4.8 2.2 1 .4.4'.S 2.2',4,4',6 2.2', 4,5.5' 2.2' .4.5.6' 2,2'. 4,5', 6 2.2'.4,«,6' NO. P«nt«cn lorottlphlnyll 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 2.2'.S.S' 2,2'. 5. 6' 2.2'.6.6' 2,3.3'. 4 2.3,3'. 4' 2,3,3', 5 2.3.3'. 5' 2.3,3'. 6 2.3.4.4' 2.3.4.5 2,3.1.6 2.3. 4' .5 2.3.4'. 6 2.3,5,6 Ptntlctll orettl phtny 1 1 88 87 88 39 90 91 92 93 94 95 Tttnrt 1 orob 1 phtfiy 1 » 98 2.2;,3 ' 97 98 99 2i2'l3 100 2,2'. 3 ' 101 2,2'. 3 102 103 2 < 2 ''4 ' 104 2> '.4 NUMBERING OF PCB CONGENERS3 MO. Structurt Tttnen 1 oreH etmiy 1 1 52 S3 54 B 2 3 TABLE 1. structure 128 129 130 137 132 133 134 135 131 137 138 139 140 141 142 143 144 14$ 141 147 148 149 150 151 152 153 154 155 151 157 1S8 159 160 2,2'.3.3'.4.4' 2.2'.3,3' .4.5 2.2',3.3',4.5' 2.2'.3.3'.4,« 2,2' .3.3' ,4.1' 2.2'.3,3'.5,5' 2,2'. 3,3'. 5,6 2.2* .3.3' ,5,1' 2,2' ,3,3'. 6,6' 2.2', 3,4, 4' .5 2,2'. 3, 4. 4 ' . £' 2,2' .3,4, 4 ' . 6 2.21, 3, 4, 4' .6' 2.2' .3, 4, 5, 5' 2,2'. 3.4. 5.6 2. 2' ,3, 4,5.6' 2,2'.3.4.5',6 2,2'.3.4, 6. 6' 2,2'. 3.4'. 5.5' 2.2*. 3,4- ,5,6 2,2' .3.4'. 5, 6' 2.2',3,4' .5', 6 2.2'. 3.4'. 6.6' 2,2',3.5,5',6 2.2', 3.5.6. 6' 2,2' ,4, 4 - , 5,5' 2,2', 4,4', 5.6' 2,2'. 4,4', 6.6' 2.3.3', 4.4- .5 2.3,3' .4, 4 ' . 5' 2.3,3', 4.4', 5 2.3,3' .4.5. 5' 2.3.3', 4.5,6 Il l 112 163 114 115 166 167 161 169 2.3,3' .4. 5'. 6 2.3. 3'. 4 ' . 5. S' 2,3.3' .4'. 5. 6 2,3. 3' .4'. 5'. 6 2.3.3'.5,5',6 2,3' .4,4'. 5,5' 2. 3'. 4. 4' ,5' .6 3.3',4.4'.5.5' Htptiehl orebf phtny 1 1 170 171 172 173 174 175 178 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 2,2'. 3. 3'. 4. 4 ' , 5 2.2' .3.3' ,4, 4 ' , 6 2, 2', 3. 3'. 4, 5. 5' 2, 2'. 3, 3'. 4, 5, 6 2, 2', 3, 3 ' . 4, 5, 6' 2,2'.3.3'.4,5',6 2,2', 3, 3 ' . 4' ,5. 6 2, 2', 3. 3', 5. 5 ' , 6 2. 2' .3, 3' ,5, 6, 6' 2,2',3,4,4',5,6 2, 2', 3, 4, 4 ' ,5,1' 2. 2' .3. 4, 4 ' . 5 ' . 5 2, 2', 3, 4, 4' ,6,6' 2,2', 3. 4,5, 5 ' . 6 2, 2', 3,4, 5, 6, 6' 2,2'.3,4',5,i l .S 2. 3, 3'. 4, 4 ' . 5,5' 2. 3, 3', 4. 4 ' . 5. 6 2.3, 3' .4. 4 ' , S ' , 5 2.3. 3'. 4,5. 5 ' . 1 2,3,3' .4'. 5,5'. 6 OetlcBlorottiphtnyli 194 195 191 197 198 199 200 201 202 203 204 205 2,2' ,3, 3' ,4.4', 5, 5' 2. 2 ' , 3. 3'. 4, 4 ' , 5, 6 2,2', 3. 3 ' , 4, 4 ' , S, 6' 2. 2', 3, 3', 4. 4 ' , 6, 6' 2. 2', 3. 3', 4, 5, 6, 6' 2,2'. 3,3' ,4. 5' .6, 6' 2.2'. 3. 3'. 4. 5, 5'. 6' 2.2' .3.3' ,5. 5 ' . 6 . 6 2.2'.3,4, 4'. 5, 5'. 6 2. 2' ,3.4. 4' .5, 6, 6' 2.3.3'. 4,4'. 5,5'. 6 Monichlorob<Pntnyli 206 207 208 2.2' .3, 3'. 4.4', 5.5', 6 2.2' .3,3* .4, 4' .5,6,1' 2.2'. 3, 3', 4, 5, 5'. 6, 6' OteieBlorottfohtnyl ' 209 •Adopt** fro* Stmcturt Htuchloroblphtnyls tollscMUr, X. ind Zcll, M., FrtMirius Z. Anal.CUM., 302. 20-31 (I960). C-3 2,2',3,3'4.4'.5,5'.6,5' �2.0 Summary 2.1 The air must be sampled such that the specimen collected for analysis is representative of the whole. Statistically designed selection of the sampling position (stack, flue, port, etc.) or time should be employed. Gaseous and particulate PCBs are withdrawn isokinetically from stacks, room air exhausts, process point exhausts, and other flowing gaseous streams using a sampling train.1 The PCBs are collected in the Florisil adsorbent tube and in the impingers in front of the adsorbent. PCBs are sampled from ambient air and other static gaseous sources onto a Florisil adsorbent tube. The sample must be preserved to prevent PCB loss prior to analysis. Storage at 4°C is recommended. 2.2 The Florisil adsorbent is extracted with hexane in a Soxhlet extractor, the aqueous condensate is extracted with hexane and the acetone/hexane impinger rinse is back-extracted with water. All three organic extracts are then combined. Optional cleanup techniques may include sulfuric acid cleanup and Florisil adsorption chromatography. The sample is concentrated to a final known volume for instrumental determination. 2.3 The PCB content of the sample extract is determined by capillary (preferred) or packed column gas chromatography/electron impact mass spectroraetry (CGC/EIMS or PGC/EIMS) operated in the selected ion monitoring (SIM), full scan, or limited mass scan (LMS) mode. 2.4 PCBs are identified by comparison of their retention time and mass spectral intensity ratios to those in calibration standards. 2.5 PCBs are quantitated against the response factors for a mixture of 11 PCB congeners using the internal standard technique. 2.6 The PCBs identified by the SIM technique may be confirmed by full scan CGC/EIMS, retention on alternate GC columns, other mass spectrometric techniques, infrared spectrometry, or other techniques, provided that the sensitivity and selectivity of the technique are demonstrated to be comparable or superior to GC/EIMS. 2.7 The analysis time is dependent on the extent of workup employed. The time required for instrumental analysis of a single sample excluding data reduction and reporting, is about 30 to 45 min. 2.8 Appropriate quality control (QC) procedures are included to assess the performance of the analyst and estimate the quality of the results. These QC procedures include the demonstration of laboratory capability: periodic analyst certification, the use of control charts, and the analysis of blanks, replicates, and standard addition samples. A quality assurance (QA) plan must be developed for each laboratory. C-4 �2.9 While several options are available throughout this method, the recommended procedure for stack gases to be followed is: 2.9.1 2.9.2 The sample is preserved at 4°C to prevent any loss of PCBs or changes in matrix which may adversely affect recovery. 2.9.3 The three sample fractions are extracted and combined. 2.9.4 The extract is cleaned up and concentrated to an appropriate volume. Internal standards are added. 2.9.5 An aliquot of the extract is analyzed by CGC/EIMS operated in the SIM mode. On-column injections onto a 15-m DB-5 capillary column, programmed (for toluene solutions) from 110° to 325°C at 10°/min after a 2 min hold is used. Helium at 45-cm/sec linear velocity is used as the carrier gas. 2.9.6 PCBs are identified by retention time and mass spectral intensities. 2.9.7 PCBs are quantitated against the response factors for a mixture of 11 PCB congeners. 2.9.8 3.0 The sample is collected using a modified Method 5 train1 according to a scheme which permits extrapolation of the sample data to the source being assessed. The total PCBs are obtained by summing the amounts for each homolog found, and the concentration is reported as micrograms per cubic meter. Interferences 3.1 Method interferences may be caused by contaminants, in sample collection media, solvents, reagents, glassware, and other sample processing hardware, leading to discrete artifacts and/or elevated baselines in the total ion current profiles. All of these materials must be routinely demonstrated to be free from interferences by the analysis of laboratory reagent blanks as described in Section 14.4. 3.1.1 Glassware must be scrupulously cleaned. All glassware is cleaned as soon as possible after use by rinsing with the last solvent used. This should be followed by detergent washing with hot water and rinses with tap water and reagent water. The glassware should then be drained dry and heated in a muffle furnace at 400°C for 15 to 30 min. Some thermally stable materials, such as PCBs, may not be eliminated by this treatment. Solvent rinses with acetone and pesticide quality hexane may be substituted C-5 �for the muffle furnace heating. Volumetric ware should not be heated in a muffle furnace. After it is dry and cool, glassware should be sealed and stored in a clean environment to prevent any accumulation of dust or other contaminants. It is stored inverted or capped with aluminum foil. 3.1.2 3.2 4.0 The use of high purity reagents and solvents helps to minimize interference problems. Purification of solvents by distillation in all-glass systems may be required. All solvent lots must be checked for purity prior to use. Matrix interferences may be caused by contaminants that are coextracted from the sorbent material or impingers. The extent of matrix interferences will vary considerably from source to source, depending upon the nature and diversity of the sources of samples. Safety 4.1 The toxicity or carcinogenicity of each reagent used in this method has not been precisely defined; however, each chemical compound should be treated as a potential health hazard. From this viewpoint, exposure to these chemicals must be reduced to the lowest possible level by whatever means available. The laboratory is responsible for maintaining a current awareness file of OSHA regulations regarding the safe handling of the chemical specified in this method. A reference file of material data handling sheets should also be made available to all personnel involved in the chemical analysis. 4.2 Polychlorinated biphenyls have been tentatively classified as known or suspected human or mammalian carcinogens. Primary standards of these toxic compounds should be prepared in a hood. Personnel must wear protective equipment, including gloves and safety glasses. Congeners highly substituted at the meta and para positions and unsubstituted at the ortho positions are reported to be the most toxic. Extrme caution should be taken when handling these compounds neat or in concentrated solution. The class includes 3,3',4'4'-tetrachlorobiphenyl (both natural abundance and isotopically labeled). 4.3 5.0 Waste disposal must be in accordance with RCRA and applicable state rules. Apparatus and Materials All specifications are suggestions only. are included for illustration only. C-6 Catalog numbers and suppliers �5.1 Stack sampling train1 - See Figure 1; a series of four impingers with a solid adsorbent trap between the third and fourth impingers. The train may be constructed by adaptation from a Method 5 train.2 Descriptions of the train components are contained in the following subsections. 5.1.1 Probe nozzle - Stainless steel (316) with sharp, tapered leading edge. The angle of taper shall be £ 30° and the taper shall be on the outside to preserve a constant internal diameter. The probe nozzle shall be of the buttonhook or elbow design, unless otherwise specified by the Agency. The wall thickness of the nozzle shall be less than or equal to that of 20 gauge tubing, i.e., 0.165 cm (0.065 in.) and the distance from the tip of the nozzle to the first bend or point of disturbance shall be at least two times the outside nozzle tubing. Other configurations and construction material may be used with approval from the Agency. 5.1.2 Probe liner - Borosilicate or quartz glass equipped with a connecting fitting that is capable of forming a leakfree, vacuum tight connection without sealing greases; such as Kontes Glass Company "0" ring spherical ground ball joints (model K-671300) or University Research Glassware SVL teflon screw fittings. A stainless steel (316) or water-cooled probe may be used for sampling high temperature gases with approval from the Agency. A probe heating system may be used to prevent moisture condensation in the probe. 5.1.3 Pitot tube - Type S, or equivalent, attached to probe to allow constant monitoring of the stack gas velocity. The face openings of the pitot tube and the probe nozzle shall be adjacent and parallel to each other but not necessarily on the same plane, during sampling. The free space between the nozzle and pitot tube shall be at least 1.9 cm (0.75 in.). The free space shall be set based on a 1.3 cm (0.5 in.) ID nozzle, which is the largest size nozzle used. The pitot tube must also meet the criteria specified in Method 22 and be calibrated according to the procedure in the calibration section of that method. 5.1.4 Differential pressure gauge - Inclined manometer capable of measuring velocity head to within 10% of the minimum measured value. Below a differential pressure of 1.3 mm (0.05 in.) water gauge, micromanometers with sensitivities of 0.013 mm (0.0005 in.) should be used. However, micromanometers are not easily adaptable to field conditions and are not easy to use with pulsating flow. Thus, other methods or devices acceptable to the Agency may be used when conditions warrant. C-7 �Thermometer Florisil Tube Probe (r^. Reverse-Type' Pitot Tube Manometer Tight /—TN Pump Control Box Figure 1. PCB sampling train for stack gases. C-8 Check Valve �5.1.5 Impingers - Four impingers with connecting fittings able to form leak-free, vacuum tight seals without sealant greases when connected together as shown in Figure 1. The first and second impingers are of the GreenburgSmith design. The final two impingers are of the Greenburg-Smith design modified by replacing the tip with a 1.3 cm (1/2 in.) ID glass tube extending to 1.3 cm (1/2 in.) from the bottom of the flask. One or two additional modified Greenburg-Smith impingers may be added to the train between the third impinger and the Florisil tube to accommodate additional water collection when sampling high moisture gases. Throughout the preparation, operation, and sample recovery from the train, these additional impingers should be treated exactly like the third impinger. 5.1.6 5.1.7 Metering system - Vacuum gauge, leak-free pump, thermometers capable of measuring temperature to within ±3°C (y 5°F), dry gas meter with 2% accuracy at the required sampling rate, and related equipment, or equivalent, as required to maintain an isokinetic sampling rate and to determine sample volume. When the metering system is used in conjunction with a pitot tube, the system shall enable checks of isokinetic rates. 5.1.8 5.2 Solid adsorbent tube - Glass with connecting fittings able to form leak-free, vacuum tight seals without sealant greases (Figure 2). Exclusive of connectors, the tube has a 2.2 cm inner diameter, is at least 10 cm long, and has four deep indentations on the inlet end to aid in retaining the adsorbent. Ground glass caps (or equivalent) must be provided to seal the adsorbent-filled tube both prior to and following sampling. Barometer - Mercury, aneroid, or other barometers capable of measuring atmospheric pressure to within 2.5 mm Hg (0.1 in. Hg). In many cases, the barometric reading may be obtained from a nearby weather bureau station, in which case the station value shall be requested and an adjustment for elevation differences shall be applied at a rate of -2.5 mm Hg (0.1 in. Hg) per 30 mm (100 ft) elevation increase. Static air sampling train1 - The sampling train, see Figure 3, consists of a glass-lined probe, an adsorbent tube containing Florisil, and the appropriate valving and flow meter controls for isokinetic sampling as described in Section 5.1. The sampling apparatus in Figure 3 is the same as that in Figure 1 and Section 5.1, except that the Smith-Greenburg impingers and heated probe are not used. If condensation of significant quantities of moisture prior to the solid adsorbent is expected, Section 5.1 of the C-9 �} 28/12 10cm j28/12 Figure 2. Florisil adsorbent tube. C-10 �Probe (to sample from duct) •* Glass- lined Probe Florisil Glass Wool Check Valve Vacuum Line ! Manometer - Integrated | Flow Meter I Figure 3. Air Tight Pump PCB sampling train for static air. C-ll �method should be used. Since probes and adsorbent tubes are not cleaned up in the field, a sufficient number must be provided for sampling and allowance for breakage. 5.3 Sample recovery 5.3.1 5.3.2 Teflon FEP® wash bottle - Two, 500 ml, Nalgene No. 0023A59 or equivalent. 5.3.3 Sample storage containers - Glass bottles, 1 liter, with TFE®-lined screw caps. 5.3.4 Balance - Triple beam, Ohaus Model 7505 or equivalent. 5.3.5 Aluminum foil - Heavy duty. 5.3.6 5.4 Ground glass caps - To cap off adsorbent tube and the other sample exposed portions of the train. Metal can - To recover used silica gel. Analysis 5.4.1 Glass Soxhlet extractors - 40 mm ID complete with 45/50 S condenser, 24/40 S 250 ml round-bottom flask, heating mantle for 250 ml flask, and power transformer. 5.4.2 Teflon FEP wash bottle - Two, 500 ml, Nalgene No. 0023A59 or equivalent. 5.4.3 Separatery funnel - 1,000 ml with TFE® stopcock. 5.4.4 Kuderna-Danish concentrators - 500 ml. 5.4.5 Steam bath. 5.4.6 Separatory funnel - 50 ml with TFE® stopcock. 5.4.7 Volumetric flask - 25.0 ml, glass. 5.4.8 Volumetric flask - 5.0 ml, glass. 5.4.9 Culture tubes - 13 x 100 mm, glass with TFE®-lined screw caps. 5.4.10 Pipette - 5.0 ml glass. 5.4.11 Teflon®-glass syringe - 1 ml, Hamilton 1001 TLL or equivalent with Teflon® needle. 5.4.12 Syringe - 10 (Jl, Hamilton 701N or equivalent. C-12 �5.4.13 Disposable glass pipettes with bulbs - To aid transfer of the extracts. 5.4.14 Gas chromatography/mass spectrometer system. 5.4.14.1 Gas chromatograph - An analytical system complete with a temperature programmable gas chromatograph and all required accessories including syringes, analytical columns, and gases. The injection port must be designed for oncolumn injection when using capillary columns or packed columns. Other capillary injection techniques (split, splitless, "Grob," etc.) may be used provided the performance specifications stated in Section 7.1 are met. 5.4.14.2 Capillary GC column - A 12-20 m long x 0.25 mm ID fused silica column with a 0.25 pm thick DB-5 bonded silicone liquid phase (J&W Scientific) is recommended. Alternate liquid phases may include OV-101, SP-2100, Apiezon L, Dexsil 300, or other liquid phases which meet the performance specifications stated in Section 7.1. 5.4.14.3 Packed GC column - A 180 cm x 0.2 cm ID glass column packed with 3% SP-2250 on 100/120 mesh Supelcoport or equivalent is recommended. Other liquid phases which meet the performance specifications stated in Section 7.1 may be substituted. 5.4.14.4 Mass spectrometer - Must be capable of scanning from 150 to 550 daltons every 1.5 sec or less, collecting at least five spectra per chromatographic peak, utilizing a 70-eV (nominal) electron energy in the electron impact ionizaton mode and producing a mass spectrum which meets all the criteria in Table 2 when 50 ng of decafluorotriphenyl phosphine [DFTPP, bis(perfluorophenyDphenyl phosphine] is injected through the GC inlet. Any GC-to-MS interface that gives acceptable calibration points at 10 ng per injection for each PCB isomer in the calibration standard and achieves all acceptable performance criteria (Section 10) may be used. Direct coupling of the fused silica column to the MS is recommended. Alternatively, GC to MS interfaces constructed of all glass or glasslined materials are recommended. Glass can be deactivated by silanizing with dichlorodimethylsilane. C-13 �TABLE 2. DFTPP KEY IONS AND ION ABUNDANCE CRITERIA Mass Ion abundance criteria 197 198 199 Less than 1% of mass 198 100% relative abundance 5-9% of mass 198 275 10-30% of mass 198 365 Greater than 1% of mass 198 441 442 443 Present, but less than mass 443 Greater than 40% of mass 198 17-23% of mass 442 C-14 �5.4.14.5 A computer system that allows the continuous acquisition and storage on machine-readable media of all mass spectra obtained throughout the duration of the chromatographic program must be interfaced to the mass spectrometer. The data system must have the capability of integrating the abundances of the selected ions between specified limits and relating integrated abundances to concentrations using the calibration procedures described in this method. The computer must have software that allows searching any GC/MS data file for ions of a specific mass and plotting such ion abundances versus time or scan number to yield an extracted ion current profile (EICP). Software must also be available that allows integrating the abundance in any EICP between specified time or scan number limits. 6.0 Reagents 6.1 Sampling 6.1.1 6.1.2 Glass wool - Cleaned by thorough rinsing with hexane, dried in a 110°C oven, and stored in a hexane-washed glass jar with TFE®-lined screw cap. 6.1.3 Water - Deionized, then glass-distilled, and stored in hexane-rinsed glass containers with TFES-lined screw caps. 6.1.4 Silica gel - Indicating type, 6-16 mesh. If previously used, dry at 175°C for 2 hr. New silica gel may be used as received. 6.1.5 6.2 Florisil - Floridin Company, 30/60 mesh, Grade A. The Florisil is cleaned by 8 hr Soxhlet extraction with hexane and then by drying for 8 hr in an oven at 110°C and is activated by heating to 650°C for 2 hr (not to exceed 3 hr) in a muffle furnace. After allowing to cool to near 110°C transfer the clean, active Florisil to a clean, hexane-washed glass jar and seal with a TFE®-lined lid. The Florisil should be stored at 110°C until taken to the field for use. Florisil that has been stored more than 1 month must be reactivated before use. Crushed ice. Solvents - All solvents must be pesticide residue analysis grade. New lots should be checked for purity by concentrating an aliquot by at least as much as is used in the procedure. C-15 �6.3 Calibration standard congeners - Standards of the PCB congeners listed in Table 3 are available from Ultra Scientific, Hope, Rhode Island; or Analabs, North Haven, Connecticut. 6.4 Calibration standard stock solutions - Primary dilutions of each of the individual PCBs listed in Table 3 are prepared by weighing approximately 1-10 mg of material within 1% precision. The PCB is then dissolved and diluted to 1.0 ml with hexane. The concentration is calculated in mg/ml. The primary dilutions are stored at 4°C in screw-cap vials with Teflon cap liners. The meniscus is marked on the vial wall to monitor solvent evaporation. Primary dilutions are stable indefinitely if the seals are maintained. The validity of primary and secondary dilutions must be monitored on a quarterly basis by analyzing four quality control check samples (see Section 14.2). 6.5 Working calibration standards - Working calibration standards are prepared that are similar in PCB composition and concentration to the samples by mixing and diluting the individual standard stock solutions. Example calibration solutions are shown in Table 3. The mixture is diluted to volume with pesticide residue analysis quality hexane. The concentration is calculated in ng/ml as the individual PCBs. Dilutions are stored at 4°C in narrow-mouth, screw-cap vials with Teflon cap liners. The meniscus is marked on the vial wall to monitor solvent evaporation. These secondary dilutions can be stored indefinitely if the seals are maintained. These solutions are designated "CSxxx," where the xxx is used to encode the nominal concentration in ng/ml. 6.6 Alternatively, certified stock solutions similar to those listed in Table 3 may be available from a supplier, in lieu of the procedures described in Section 6.4. 6.7 DFTPP standard - A 50 ng/(Jl solution of DFTPP is prepared in acetone or another appropriate solvent. 6.8 Internal standard stock solution - The four 13C-labeled PCBs listed in Table 4 may be available from a supplier as a certified solution. This solution may be used as received or diluted further. 6.9 Solution stability - The calibration standard, surrogate and DFTPP solutions should be checked frequently for stability. These solutions should be replaced after 6 months, or sooner if comparison with quality control check samples indicates compound degradation or concentration change. 6.10 Quality control check samples will be supplied by the Agency. C-16 �TABLE 3. CONCENTRATIONS OF CONGENERS IN PCB CALIBRATION STANDARDS (ng/ml)a Homo log Congener no. CS1000 CS100 CS050 CS010 1 1 1,040 104 52 10 1 3 1,000 100 50 10 2 7 1,040 104 52 10 3 30 1,040 104 52 10 4 50 1,520 152 76 15 5 97 1,740 174 87 17 6 143 1,920 192 96 19 7 183 2,600 260 130 26 8 202 4,640 464 232 46 9 207 5,060 506 253 51 10 209 4,240 424 212 42 4 210 (IS) 255 255 255 255 1 211 (RS) 104 104 104 104 4 212 (RS) 257 257 257 257 8 213 (RS) 407 407 407 407 10 214 (RS) 502 502 502 502 a Concentrations given as examples only. C-17 �TABLE 4. COMPOSITION OF INTERNAL STANDARD SPIKING SOLUTION (SS100) CONTAINING 13C-LABELED PCBs3 Congener no. Compound Concentration (pg/ml) 211 (l',2l,3',4',5',6'-13C6)4-chlorobiphenyl 104 212 (13C12)3,3',4,4'-tetrachlorobiphenyl 257 213 (13C12)2,2',3,3',5,5',6,6'-octachlorobiphenyl 395 214 (13C12)decachlorobiphenyl 502 a Concentrations given as examples only. C-16 �7.0 Calibration Maintain a laboratory log of all calibrations. 7.1 Sampling train 7.1.1 Probe nozzle - Using a micrometer, the inside diameter of the nozzle is measured to the nearest 0.025 mm (0.001 in.). Three separate measurements are made using different diameters each time and obtain the average of the measurements. The difference between the high and low numbers must not exceed 0.1 mm (0.004 in.). When nozzles become nicked, dented, or corroded, they must be reshaped, sharpened, and recalibrated before use. Each nozzle must be permanently and uniquely identified. 7.1.2 Pitot tube - The pitot tube must be calibrated according to the procedure outlined in Method 2.2 7.1.3 Dry gas meter and orifice meter - Both meters must be calibrated according to the procedure outlined in APTD0581.3 When diaphragm pumps with bypass valves are used, proper metering system design is checked by calibrating the dry gas meter at an additional flow rate of 0.0057 m3/min (0.2 cfm) with the bypass valve fully opened and then with it fully closed. If there is more than ±2% difference in flow rates when compared to the fully closed position of the bypass valve, the system is not designed properly and must be corrected. 7.1.4 Probe heater calibration - The probe heating system must be calibrated according to the procedure contained in APTD-0581.3 7.1.5 Temperature gauges - Dial and liquid filled bulb thermometers are calibrated against mercury-in-glass thermometers, Thermocouples should be calibrated in constant temperature baths. 7.2 The gas chromatograph must meet the minimum operating parameters shown in Tables 5 and 6, daily. If all of the criteria are not met, the analyst must adjust conditions and repeat the test until all criteria are met. 7.3 The mass spectrometer must meet the minimum operating parameters shown in Tables 2, 7, and 8, daily. If all criteria are not met, the analyst must retune the spectrometer and repeat the test until all conditions are met. C-19 �TABLE 5. OPERATING PARAMETERS FOR CAPILLARY COLUMN GAS CHROMATOGRAPHIC SYSTEM Parameter Recommended Tolerance Liquid phase Finnigan 9610 15 m x 0.255 mm ID Fused silica DB-5 (J&W) Liquid phase thickness Carrier gas Carrier gas velocity Injector Injector temperature Injection volume Initial column temperature Column temperature program Separator Transfer line temperature 0.25 pm Helium 45 cm/sec r* On-column (J&W) c Optimum performance 1.0 (Jlc 70°C (2 min)d 70°-325°C at 10°C/mine None 280°C Other nonpolar or semipolar < 1 HID Hydrogen Optimum performance Other Optimum performance Other Other Other Glass jet or othe Optimum8 Tailing factorh 0.7-1.5 0.4-3 Peak width1 7-10 sec < 15 sec Gas chromatograph Column a Other Other Substitutions permitted with any common apparatus or technique provided performance criteria are met. b Measured by injection of air or methane at 270°C oven temperature. c For on-column injection, manufacturer's instructions should be followed regarding injection technique. d With on-column injection, initial temperature equals boiling point of the solvent; in this instance, hexane. e C^Clio elutes at 270°C. Programming above this temperature ensures a clean column and lower background on subsequent runs. f Fused silica columns may be routed directly into the ion source to prevent separator discrimination and losses. g High enough to elute all PCBs, but not high enough to degrade the column if routed through the transfer line. h Tailing factor is width of front half of peak at 10% height divided by width of back half of peak at 10% height for single PCB congeners in solution CSxxx. i Peak width at 10% height for a single PCB congener is CSxxx. C-20 �TABLE 6. OPERATING PARAMETERS FOR PACKED COLUMN GAS CHROMATOGRAPHY SYSTEM Gas chromatograph Column Tolerance Recommended Parameter Finnigan 9610 Other3 180 cm x 0.2 cm ID Other glass Column packing 3% SP-2250 on 100/ 120 mesh Supelcoport Other nonpolar or semipolar Carrier gas Helium Hydrogen Carrier gas flow rate 30 ml/min Optimum performance Injector On-column Other Injector temperature 250°C Optimum Injection volume 1.0 pi ^ 5 Ml Initial column temperature 150°C, 4 min Other Column temperature program 150°-260°C at 8°/min Other Separator Glass jet Other Transfer line temperature 280°C Optimum8 Tailing factor0 ,0.7-1.5 0.4-3 10-20 sec < 30 sec Peak widthd a Substitutions permitted if performance criteria are met. b High enough to elute all PCBs. c Tailing factor is width of front half of peak at 10% height divided by width of back half of peak at 10% height for single PCB congeners in solution CSxxx. d Peak width at 10% height for a single PCB congener is CSxxx. C-21 �TABLE 7. OPERATING PARAMETERS FOR QUADRUPOLE MASS SPECTROMETER Parameter Recommended SYSTEM Tolerance Mass spectrometer Finnigan 4023 Other3 Data system Incos 2400 Other Scan range 95-550 Other Scan time 1 sec Otherb Resolution Unit Optimum performance Ion source temperature 280°C 200°-300°C Electron energy 70 eV Optimum performance Trap current 0.2 mA Optimum performance Multiplier voltage -1,600 V Optimum performance Preamplifier sensitivity 10"6 A/V Set for desired working range a Substitutions permitted if performance criteria are met. b Greater than five data points over a GC peak is a minimum. c Filaments should be shut off during solvent elution to improve instrument stability and prolong filament life, especially if no separator is used. C-22 �TABLE 8. OPERATING PARAMETERS FOR MAGNETIC SECTOR MASS SPECTROMETER SYSTEM Parameter Recommended Tolerance Mass spectrometer Finnigan MAT 311A Other8 Data system Incos 2400 Other Scan range 98-550 Other Scan mode Exponential Other Cycle time 1.2 sec Other Resolution 1,000 > 500 Ion source temperature 280°C 250-300° Electron energy0 70 eV 70 eV Emission current 1-2 mA Optimum Filament current Optimum Optimum Multiplier -1,600 V Optimum a Substitutions permitted if performance criteria are met. b Greater than five data points over a GC peak is a minimum. c Filaments should be shut off during solvent elution to improve instrument stability and prolong filament life, especially if no separator is used. C-23 �7.4 The PCS response factors (RF ) must be determined using Equation 7-1 for the analyte homologs? A x M. RFp = A. x j -£ ^ Eq. 7-1 H r is Mp where RF = response factor of a given PCB isomer A = area of the characteristic ion for the PCB congener " peak M = mass of PCB congener injected (nanograms) A. = area of the characteristic ion for the internal standard peak M. = mass of internal standard injected (nanograras) 1S If specific congeners are known to be present and if standards are available, selected RF values may be employed. For general samples, solutions CSxxx and SSxxx or a mixture (Tables 3 and 4) may be used as the response factor solution. The PCB-surrogate pairs to be used in the RF calculation are listed in Table 9. Generally, only the primary ions of both the analyte and surrogate are used to determine the RF values. If alternate ions are to be used in the quantitation, the RF must be determined using that characteristic ion. The RF value must be determined in a manner to assure ±20% accuracy and precision. For instruments with good day-to-day precision, a running mean (RF) based on seven values determined once each day may be appropriate. Other options include, but are not limited to, triplicate determinations of a single concentration spaced throughout a day or determination of the RF at three different levels to establish a working curve. If replicate RF values differ by greater than ±10% RSD, the system performance should be monitored closely. If the RSD is greater than ±20%, the data set must be c6nsidered invalid and the RF redetermined before further analyses are done. 7.5 If the GC/EIMS system has not been demonstrated to yield a linear response or if the analyte concentrations are more than one order of magnitude different from those in the RF solution, a calibration curve must be prepared. If the analyte and RF solution concentrations differ by more than one order of magnitude, a calibration curve should be prepared. A calibration curve should be established with triplicate determinations at three or more concentrations bracketing the analyte levels. C-24 �TABLE 9. PAIRINGS OF ANALYTE. CALIBRATION, AND SURROGATE COMPOUNDS Analyte Congener3 no. Compound 1 2,3 4-15 16-39 40-81 82-127 128-169 170-193 194-205 206-208 o 209 I 2-Ci2HgCl 3- and 4-C12H9Cl Ci2HgCl2 C^HyCls C12H6C14 c 12H5Cls C12HsCl7 Ci2H2Clg Cj^HClg CiaCl10 Calibration standard Congener Compound no. 1 3 7 30 50 97 143 183 202 207 209 2 4 2,4 2,4 ,6 2,2 ',4 ,6 2,2 ',3 ',4,5 2,2 ',3 ,4,5 ,6' 2,2 ',3 ',4,4' ,5', 6 2,2 ',3 ,3',5,5', 6, 6' 3' 2,2 ',3 ,° » 4, 4' ,5,6, 6' C 12Cli 0 NJ a Ballschmiter numbering system, see Table 1. Surrogate Congener no. Compound 211 211 211 212 212 212 212 213 213 213 214 13 C6-4 c "3 e-4 i C6-4 13 C12-3,3' ,4,4' 13 Ci2-3,3' ,4,4' 13 C12-3,3' ,4,4' 13 Ci2-3,3' ,4,4' 13 C12-2,2' ,3,3' ,5,5', 6, 6' 13 q 01 C12-2,2' ,O,O ,5, 5', 6,6' 13 q 01 C12-2,2' ,J,J ,5,5', 6, 6' 13 Ci2Cll0 �7.6 8.0 The relative retention time (RRT) windows for the 10 homologs and surrogates must be determined. If all congeners are not available, a mixture of available congeners or an Aroclor mixture (e.g., 1016/1254/1260) may be used to estimate the windows. The windows must be set wider than observed if all isomers are not determined. Typical RRT windows for one column are listed in Table 10. The windows may differ substantially if other GC parameters are used. Sample Collection, Handling, and Preservation The sampling shall be conducted by competent personnel experienced with this test procedure and cognizant of the constraints of the anaytical techniques for PCBs, particularly contamination problems. 8.1 Stack sampling1 8.1.1 Pretest preparation - All train components shall be maintained and calibrated according to the procedure described in APTD-0581,3 unless otherwise specified herein. This should be done in the laboratory prior to sampling. 8.1.1.1 Cleaning glassware - All glass parts of the train upstream of and including the adsorbent tube and impingers, should be cleaned as described in Section 3.1.1. Special care should be devoted to the removal of residual silicone grease sealants on ground glass connections of used glassware. These grease residues should be removed by soaking several hours in a chromic acid cleaning solution prior to routine cleaning as described above. 8.1.1.2 Solid adsorbent tube - 7.5 g of Florisil activated within the last 30 days and still warm from storage in a 110°C oven, is weighed into the adsorbent tube (prerinsed with hexane) with a glass wool plug in the downstream end. A second glass wool plug is placed in the tube to hold the sorbent in the tube. Both ends of the tube are capped with ground glass caps. These caps should not be removed until the tube is fitted to the train immediately prior to sampling. 8.1.2 Preliminary determinations - The sampling site and the minimum number of sampling points are selected according to Method I2 or as specified by the Agency. The stack pressure, temperature, and the range of velocity heads are determined using Method 22 and moisture content using Approximation Method 42 or its alternatives for the purpose of making isokinetic sampling rate calculations. Estimates may be used. However, final results must be based on actual measurements made during the test. C-26 �TABLE 10. RELATIVE RETENTION TIME (RRT) RANGES OF PCB HOMOLOGS VERSUS d6-3,3'.4.4'-TETRACHLOROBIPHENYL PCB homolog Monochloro No. of isomers measured Observed range of RRTsa Congener no. Observed RRTa Projected range of RRTs 3 0.40-0.50 1 3 0.43 0.50 0.35-0.55 Dichloro 10 0.52-0.69 7 0.58 0.35-0.80 Trichloro 9 0.62-0.79 30 0.65 0.35-0.10 Tetrachloro 16 0.72-1.01 50 0.75 0.55-1.05 Pentachloro 12 0.82-1.08 97 0.98 0.80-1.10 Hexachloro 13 0.93-1.20 143 1.05 0.90-1.25 Heptachloro 4 1.09-1.30 183 1.15 1.05-1.35 Octachloro 6 1.19-1.36 202 1.19 1.10-1.50 Nonachloro 3 1.31-1.42 207 1.33 1.25-1.50 Decachloro 1 1.44-1.45 209 1.44 1.35-1.50 a The RRTs of the 77 congeners and a mixture of Aroclor 1016/1254/1260 were measured versus 3,3',4,4'-tetrachlorobiphenyl-de (internal standard) using a 15-m J&W DB-5 fused silica column with a temperature program of 110°C for 2 min, then 10°C/min to 325°C, helium carrier at 45 cm/sec, and an oncolumn injector. A Finnigan 4023 Incos quadrupole mass spectrometer operating with a scan range of 95-550 daltons was used to detect each PCB congener. b The projected relative retention windows account for overlap of eluting homologs and take into consideration differences in operating systems and lack of all possible 209 PCB congeners. C-27 �The molecular weight of the stack gases is determined using Method 3.2 A nozzle size is selected based on the maximum velocity head so that isokinetic sampling can be maintained at a rate less than 0.75 cfm. It is not necessary to change the nozzle size in order to maintain isokinetic sampling rates. During the run, the nozzle size must not be changed. A suitable probe length is selected such that all traverse points can be sampled. Sampling from opposite sides for large stacks may be considered to reduce the length of probes. A sampling time is selected appropriate for total method sensitivity and the PCB concentration anticipated. Sampling times should generally fall within a range of 2 to 4 hr. A buzzer-timer should be incorporated in the control box (see Figure 1) to alarm the operator to move the probe to the next sampling point. 8.1.3 Preparation of collection train - During preparation and assembly of the sampling train, all train openings must be covered until just prior to assembly or until sampling is about to begin. Immediately prior to assembly, all parts of the train upstream of the adsorbent tube are rinsed with hexane. The probe is marked with heat resistant tape or by some other method at points indicating the proper distance into the stack or duct for each sampling point. 200 ml of water is placed in each of the first two impingers, and the third impinger left empty. CAUTION: Sealant greases must not be used in assembling the train. If the preliminary moisture determination shows that the stack gases are saturated or supersaturated, one or two additional empty impingers should be added to the train between the third impinger and the Florisil tube. See Section 5.1.5. Approximately 200 to'300 g or more, if necessary, of silica gel is placed in the last impinger. Each impinger (stem included) is weighed and the weights recorded to the nearest 0.1 g on the impingers and on the data sheet. Unless otherwise specified by the Agency, a temperature probe is attached to the metal sheath of the sampling probe so that the sensor is at least 2.5 cm behind the nozzle and pitot tube and does not touch any metal. C-28 �The train is assembled as shown in Figure 1. Through all parts of this method use of sealant greases such as stopcock grease to seal ground glass joints must be avoided. Crushed ice is placed around the impingers. 8.1.4 Leak check procedure - After the sampling train has been assembled, the probe heating system(s) is turned on and set (if applicable) to reach a temperature sufficient to avoid condensation in the probe. Time is allowed for the temperature to stabilize. The train is leak checked at the sampling site by plugging the nozzle and pulling a 380 mm Hg (15 in. Hg) vacuum. A leakage rate in excess of 4% of the average sampling rate or 0.0057 m3/min (0.02 cfm) whichever is less, is unacceptable. The following leak check instruction for the sampling train described in APTD-05813 may be helpful. The pump is started with bypass valve fully open and coarse adjust valve completely closed. The coarse adjust valve is partially opened and the bypass valve slowly closed until 380 mm Hg (15 in. Hg) vacuum is reached. The direction of bypass valve must not be reversed. This will cause water to back up into the probe. If 380 mm Hg (15 in. Hg) is exceeded, either the leak check is conducted at this higher vacuum or the leak check is ended as described below and start over. When the leak check is completed, the plug is first slowly removed from the inlet to the probe and the vacuum pump is immediately turned off. This prevents the water in the impingers from being forced backward into the probe. Leak checks, shall be conducted as described above prior to each test run and at the completion of each test run. If leaks are found to be in excess of the acceptable rate, the test will be considered invalid. To reduce lost time due to leakage occurrences, it is recommended that leak checks be conducted between port changes. 8.1.5 Train operation - During the sampling run, an isokinetic sampling rate within 10%, or as specified by the Agency, of true isokinetic shall be maintained. During the run, the nozzle or any other part of the train in front of and including the Florisil tube must not be changed. For each run, the data required on the data sheets must be recorded. An example is shown in Figure 4. The dry gas meter readings are recorded at the beginning and end of each sampling time increment, when changes in flow rates are made, and when sampling is halted. Other data point readings are taken at least once at each sample point during each time increment and whenever significant C-29 �FIELD DATA PLANT. OATE_ SAMPLING LOCATION. SAMPLE TYPE RUN NUMBER OPERATOR PROBE LENGTH AND TYPE. NOZZLE ID. ASSUMED MOISTURE. "„ SAMPLE BOX NUMBER METER BOX NUMBER METER AH p C FACTOR PROBE HEATER SETTING HEATER BOX SETTING REFERENCE Ap_ AMBIENT TEMPERATURE BAROMETRIC PRESSURE . STATIC PRESSURE. (P$)_ FILTER NUMBER ($) SCHEMATIC OF TRAVERSE POINT LAYOUT READ AND RECORD ALL DATA EVERY, MINUTES TRAVERSE POINT NUMBER s ^X CLOCK hTIME LiNG r_ \Aoc K, TIMt.iim N^ ~~ GAS METER READING <Vml. It3 VELOCITY HEAD (APSI. in. H?0 —— __ ORIFICE PRESSURE DIFFERENTIAL (AHI. in. H20l DESIRED STACK TEMPERATURE |TSI.°F ACTUAL n i COMMENTS: Figure 4. EPAlDur) 2K Field data sheet, DRY GAS METER TEMPERATURE INLET (Tm mt."F OUTLET •Tm^.-'F PUMP VACUUM, in. H| SAMPLE BOX TEMPERATURE. °F IMPINGCR TEMPERATURL "F �changes (20% variation in velocity head readings) necessitate additional adjustments in flow rate. The portholes are cleaned prior to the test run to minimize change of sampling deposited material. To begin sampling, the nozzle cap is removed, the probe heater operational and temperature up, and the pitot tube and probe positions are verified (if applicable). The nozzle is positioned at the first traverse point with the tip pointing directly into the gas stream. The pump is started and the flow adjusted to isokinetic conditions. Nomographs are available for sampling trains using type S pitot tubes with 0.85 ± 0.02 coefficients (C ), and when sampling in air or a stack gas with equivalent density (molecular weight, M,, equal to 29 ± 4), which aid in the rapid adjustment of the isokinetic sampling rate without excessive computations. If C and M, are outside the above stated ranges, the nomograph cannot be used unless appropriate steps are taken to compensate for the deviations. When the stack is under significant negative pressure (height of impinger stem), the coarse adjust valve must be closed before inserting the probe into the stack to avoid water backing into the probe. If necessary, the pump may be turned on with the coarse valve closed. When the probe is in position, the openings around the probe and porthole must be blocked off to prevent unrepresentative dilution of the gas stream. The stack cross section is traversed, as required by Method I2 or as specified by the Agency. To minimize chance of extracting deposited material, the probe nozzle should not bump into the stack walls when sampling near the walls or when removing or inserting the probe through the portholes. During the test run, periodic adjustments are made to keep the probe temperature at the proper value. More ice and, if necessary, salt is added to the ice bath to maintain a temperature of less than 20°C (68°F) at the impinger/silica gel outlet, to avoid excessive moisture losses. Also, the level and zero of the manometer should be periodically checked. If the pressure drop across the train becomes high enough to make isokinetic sampling difficult to maintain, the test run should be terminated. Under no circumstances should the train be disassembled during the test run to determine and correct causes of excessive pressure drops. C-31 �At the end of the sample run, the pump is turned off, the probe and nozzle removed from the stack, and the final dry gas meter reading recorded. A leak check is performed, with acceptability of the test run based on the same criteria as in Section 8.1.4. The percent isokinetic is calculated (see calculation section) to determine whether another test run should be made. If there is difficulty in maintaining isokinetic rates due to source conditions, the Agency should be consulted for possible variance on the isokinetic rates. 8.1.6 8.2 Blank train - For each series of test runs, a blank train is set up in a manner identical to that described above, but with the nozzle capped with aluminum foil and the exit end of the last impinger capped with a ground glass cap. The train is allowed to remain assembled for a period equivalent to one test run. The blank sample is recovered as described in Section 8.3. Static air sampling3 - The sampling procedure for static air is identical to that described in Section 8.1 with the following exceptions: (a) impingers and a heatable probe are not required prior to the adsorbent tube; and (b) the PCB concentrations may dictate a longer or shorter sampling time. The selection of sampling time and rate should be based on the approximate levels of PCB residues expected in the sample. The sampling rate should not exceed 14 liter/rain and may typically fall in the range of 5 to 10 liter/rain. Sampling times should be more than 20 min but should not exceed 4 hr. 8.3 Sample recovery - Proper cleanup procedure begins as soon as the probe is removed from the stack at the end of the sampling period. When the probe can be safely handled, all external particulate matter near the tip of the probe nozzle is wiped off. The probe is removed from the train and both ends closed off with aluminum foil. The inlet to the train is capped off with a ground glass cap. The probe and impinger assembly are transfered to the cleanup area. This area should be clean and protected from the wind so that the chances of contaminating or losing the sample will be minimized. The train is inspected prior to and during disassembly and any abnormal conditions noted. The samples are treated as follows: 8.3.1 Adsorbent tube - The Florisil tube is removed from the train and capped with ground glass caps. 8.3.2 Sample Container No. 1 - The first three impingers are removed. The outside of each impinger is wiped off to remove excessive water and other debris. The impingers C-32 �are weighed (stem included), and the weight recorded on a data sheet. The contents are poured directly into Container No. 1. 8.3.3 8.3.4 8.4 Sample Container No. 2 - Each of the first three impingers are rinsed sequentially with 30-ml acetone and then with 30-ml hexane, and the rinses put into Container No. 2. Material deposited in the probe is quantitatively recovered using 100-ml acetone and then 100-ml hexane and these rinses added to Container No. 2. Silica gel container - The last impinger is removed, and the outside wiped to remove excessive water and other debris. It is weighed (stem included), and the weight recorded on the data sheet. The contents are transferred to the used silica gel can. Sample preservation - Samples should be stored in the dark at 4°C. Storage times in excess of 4 weeks are not recommended. 9.0 Sample Preparation1 9.1 Extraction 9.1.1 Adsorbent tube - The entire contents of the adsorbent tube are expelled directly onto a glass wool plug in the sample holder of a Soxhlet extractor. Although no extraction thimble is required, a glass thimble with a coarsefritted bottom may be used. The tube is rinsed with 5-ml acetone and then with 15-ml hexane and these rinses put into the extractor. The extraction apparatus is assembled and the adsorbent extracted with 170-ml hexane for at least 4 hr. The extractor should cycle 10 to 14 times per hour. After allowing the extraction apparatus to cool to ambient temperature, the extract is transferred into a KudernaDanish evaporator. The extract is evaporated to about 5 ml on a steam bath and the evaporator allowed to cool to ambient temperature before disassembly. The extract is transferred to a 50-ml separatory funnel and the funnel set aside. 9.1.2 Sample Container No. 1 - The aqueous sample is transferred to a 1,000-ml separatory funnel. The container is rinsed with 20-ml acetone and then with two 20-ml portions of hexane, adding the rinses to the separatory funnel. The sample is extracted with three 100 ml portions of hexane and the sequential extracts transferred to a Kuderna-Danish evaporator. C-33 �The extract is concentrated to about 5 ml and allowed to cool to ambient temperature before disassembly. The extract is filtered through a micro column of anhydrous sodium sulfate into a 50-ml separatory funnel containing the corresponding Florisil extract from Section 9.1.1. The micro column is prepared by placing a small plug of glass wool in the bottom of the large portion of a disposable pipette and then adding anhydrous sodium sulfate until the tube is about half full. 9.1.3 Sample Container No. 2 - The organic solution is transferred into a 1,000-ml separatory funnel. The container is rinsed with two 20 ml portions of hexane and the rinses added to the separatory funnel. The sample is washed with three 100 ml portions of water. The aqueous layer is discarded and the organic layer transferred to a KudernaDanish evaporator. The extract is concentrated to about 5 ml and allowed to cool to ambient temperature before disassembly. The extract is filtered through a micro column of anhydrous sodium sulfate into the 50-ml separatory funnel containing the corresponding Florisil and impinger extracts (Section 9.1.2). 9.2 Cleanup - Two tested cleanup techniques are described below.4 Depending upon the complexity of the sample, one or both of the techniques may be required to fractionate the PCBs from interferences. If the sample extract is colored, the Florisil column cleanup may be indicated. 9.2.1 Acid cleanup 9.2.1.1 Add 5 ml of concentrated sulfuric acid to the separatory funnel containing the sample extract and shake for 1 min. 9.2.1.2 Allow the phases to separate, transfer the sample (upper phase) with three 1 to 2 ml solvent rinses to Kuderna-Danish evaporator and concentrate to an appropriate volume. 9.2.1.3 Analyze as described in Section 10.0. 9.2.1.4 If the sample is highly contaminated, a second or third acid cleanup may be employed. 9.2.2 Florisil column cleanup 9.2.2.1 Variations among batches of Florisil may affect the elution volume of the various PCBs. For this reason, the volume of solvent required to C-34 �completely elute all of the PCBs must be verified by the analyst. The weight of Florisil can then be adjusted accordingly. 9.2.2.2 Place a 20-g charge of Florisil, activated overnight at 130°C, into a Chroraaflex column. Settle the Florisil by tapping the column. Add about 1 cm of anhydrous sodium sulfate to the top of the Florisil. Pre-elute the column with 70-80 ml of hexane. Just before the exposure of the sodium sulfate layer to air, stop the flow. Discard the eluate. 9.2.2.3 Add the sample extract to the column. Add 225 ml of hexane to the column. Carefully wash down the inner wall of the column with a small amount of the hexane prior to adding the total volume. Discard the first 25 ml. 9.2.2.4 Collect 200 ml of hexane eluate in a KudernaDanish flask. All of the PCBs should be in this fraction. Concentrate to an appropriate volume. 9.2.2.5 Analyze the sample as described in Section 10.0. 10.0 Gas Chromatographic/Electron Impact Mass Spectrometric Determination 10.1 Internal standard addition - Pipet an appropriate volume of internal standard solution SSxxx into the sample. The final concentration of the internal standards must be in the working range of the calibration and well above the matrix background. The internal standards are thoroughly mixed by mechanical agitation. Note: The volume measurement of the spiking solution is critical to the overall method precision. The analyst must exercise caution that the volume is known ±1% or better. Where necessary, calibration of the pipet is recommended. Note: This same solution is used as a surrogate standard solution in the protocols for products/product waste and for water. In this protocol, the 13C-labeled PCBs are spiked after extraction, so are used as internal standards. Alternately, another internal standard solution such as the d63,3',4,4'-tetrachlorobiphenyl used in the product/product waste and water protocols may be used, if acceptable RF precision and accuracy are shown across the homolog range. 10.2 Tables 2, and 5 through 8 summarize the recommended operating conditions for analysis. Figure 5 presents an example of a chromatogram. C-35 �W O) ,H CO CO 00 0) 4-1 l-i 3 C ffl 4-1 c o K 10U.O vO ^O T3 U rH CO 1 rH rH C £*^ C QJ C QJ J^ p., fi Q) J2 p. -H £X ^Q -H -H XI XI ^-s O O Q) M l-l 4J O O O O i H C x; CJ co rJ 1-1 C 3 M 4-1 rH >> QJ QJ 1 o o en a I I ,c ^ )S *—' <f rH >-. r^> C <f H r-f <J •H J>"i *• _O £j ^3" c QJ x: XI -H rH jj") -H XI O IM C QJ P>i p. -H It 1C rH CO rH >% C o) x: P. ca l-i 0) -H l-i a co rH 1-1 >. C QJ 4-1 QJ H x: 1 p. 4-1 O |-i 0 tH QJ H 1 vO f. 0 cO 4J O rH X: O •H Q 1 ^J i X CJ -H M H | 04" 7 "** 01 P. ,--, P . x: 0 ca XI C 0 M O O CO -H o 1 vD 0 vO ^O v£) LO rH 0 X! O - 4-> rH 1-1 O l-i O XI 0 o 1-1 QJ XI rH 1 •H cfl M O x: 1-1 U 3 cfl CO CO a CJ —' QJ Q cfl - <r M - o rH CO X O) ffl rrj 1 vO - oo p. <J *. LO _ QJ rn x: r, LO f. 4-1 O •v ~^r ^ CO LO ** • t < CO CO n *, •- -<r ^ CN ^ CN] «s CM r. 0) Q r, CN ^ " J r, CO *x CO ' , 1208 o f. I3 C ' r. CN CN i ui ||7ll ™" ^ '' 7 | rH 1 2 f. r C7 CM" 1 r T II a o x : 4-1 XI m -<r ^-^ CJ O <f •s QJ PL, | LO - <3~ Ovl 04 fQ -H , n -H — >sO c - ^J" **3" , ; vo 0-1 CIO QJ x: P, rH M C &^ c rH 1 rH p, CO O XI P. •H XI O o ^i o rH x: o o 10: (W o X: •H XI «2_.Mh.. 1 c-.rj -H X3 rO o P. -~ p. n rH XI x; C X! o ,x C QJ rH o I 2 o l-l o rH x: rH i—i x: - 6 5 QJ QJ - CX rH CJ CO C JZ O O HC i t O O o o co rH rH (30 x: x: o 112592 CN rH J>^ 1 rH 1 rH rH >•>,>> ' CO 02? r 1M0 11:20 04 ™ I H (I »;« * "t lono IB: 1D I * 1I ,,,„ i uv: j ->0: CO ^_ 1 - V r-.'f' ^ 23:?'J soA'i Till.: Figure 5. Capillary gas chromatography/electron impact ionization mass spectrometry (CGC/EIMS) chromatogram or the calibration standard solution required for quantitation of PCBs by homolog. This chromatogram includes PCBs representative of each liomolog, three carbon-13 labeled surrogates, and the deuterated internal standard. The concentration of all components and the CGC/EIMS parameters are presented in Tallies 3, 4, 5, and 7. �10.3 While the highest available chromatographic resolution is not a necessary objective of this protocol, good chromatographic performance is recommended. With the high resolution of CGC, the probability that the chromatographic peaks consist of single compounds is higher than with PGC. Thus, qualitative and quantitative data reduction should be more reliable. 10.4 After performance of the system has been certified for the day and all instrument conditions set according to Tables 2, and 5 through 8, inject an aliquot of the sample onto the GC column. If the response for any ion, including surrogates and internal standard, exceeds the working range of the system, dilute the sample and reanalyze. If the responses of surrogates, internal standard, or analytes are below the working range, recheck the system performance. If necessary, concentrate the sample and reanalyze. 10.5 Record all data on a digital storage device (magnetic disk, tape, etc.) for qualitative and quantitative data reduction as discussed below. 11.0 Qualitative Identification 11.1 Selected ion monitoring (SIM) or limited mass scan (IMS) data The identification of a compound as a given PCS homolog requires that two criteria be met: 11.1.1 (1) The peak must elute within the retention time window set for that homolog (Section 7.6); and (2) the ratio of two ions obtained by SIM (Table 11) or by IMS (Table 12) must match the natural ratio within ±20%. The analyst must search the higher mass windows, in particular M+70, to prevent misidentification of a PCB fragment ion cluster as the parent. 11.1.2 If one or the other of these criteria is not met, interferences may have affected the results and a reanalysis using full scan EIMS conditions is recommended. 11.2 Full scan data 11.2.1 The peak must elute within the retention time windows set for that homolog (as described in Section 7.6). 11.2.2 The unknown spectrum must match that of an authentic PCB. The intensity of the three largest ions in the molecular cluster (two largest for monochlorobiphenyls) must match the natural ratio within ±20%. Frequent clusters with proper intensity ratios must also be present. 11.2.3 Alternatively, a spectral search may be used to automatically reduce the data. The criteria for acceptable C-37 �TABLE 11. CHARACTERISTIC SIM IONS FOR PCBs Ion (relative intensity) Secondary Tertiary Homolog Primary ^12^9^1 188 (100) 190 (33) C^HgC^ 222 (100) 224 (66) 226 (11) Ci2H7Cl3 256 (100) 258 (99) 260 (33) C12H6C14 292 (100) 290 (76) 294 (49) C12H5C15 326 (100) 328 (66) 324 (61) £12^4^16 360 (100) 362 (82) 364 (36) C12H3C17 394 (100) 396 (98) 398 (54) Ci2H2Cl8 430 (100) 432 (66) 428 (87) Cj^HClg 464 (100) 466 (76) 462 (76) C 498 (100) 500 (87) 496 (68) 12 C llO - Source: Rote, J. W., and W. J. Morris, "Use of Isotopic Abundance Ratios in Identification of Polychlorinated Biphenyls by Mass Spectrometry," J. Assoc. Offic. Anal. Chem., 56(1), 188-199 (1973). C-38 �TABLE 12. LIMITED MASS SCANNING (LMS) RANGES FOR PCBs Compound Mass range (m/z) CX2H.CU 186-190 C12H8C12 220-226 C12H7C13 254-260 C12H6C13 288-294 Ci2H5Cl5 322-328 Ci2H4Cl6 356-364 C12H3C17 386-400 C12H2C18 426-434 C12HC19 460-468 CisCl^ 494-504 C12D6C14 294-300 13 192-196 13 300-306 C612C6H9C1 C12H6C14 13 438-446 -C12C110 506-516 C12H2C18 a Adapted from Tindall, G. W., and P. E. Wininger, "Gas Chromatography-Mass Spectrometry Method for Identifying and Determining Polychlorinated Biphenyls," J. Chromatogr.t 196, 109-119 (1980). C-39 �identification include a high index of similarity. For the Incos 2300, a fit of 750 or greater must be obtained. 11.3 Disputes in interpretation - Where there is reasonable doubt as to the identity of a peak as a PCB, the analyst must either identify the peak as a PCB or proceed to a confirmational analysis (see Section 13.0). 12.0 Quantitative Data Reduction 12.1 Once a chromatographic peak has been identified as a PCB, the compound is quantitated based either on the integrated abundance of the SIM data or EICP for the primary characteristic ion in Tables 11 and 12. If interferences are observed for the primary ion, use the secondary and then tertiary ion for quantitation. If interferences in the parent cluster prevent quantitation, an ion from a fragment cluster (e.g., M-70) may be used. Whichever ion is used, the RF must be determined using that ion. The same criteria should be applied to the internal standard compounds (Table 13). 12.2 Using the appropriate response factor (RF ) as determined in Section 7.3, calculate the mass of each PCB peak (M ) using Equation P 12-1. A , M = -E • 'M p A.is RF p is Eq. 12-1 H * where A = area of the characteristic ion for the analyte PCB " peak A. = area of the characteristic ion for the internal standard peak RF = response factor of a given PCB congener M. S = mass of internal standard injected (micrograms) ~L 12.3 If a peak appears to contain non-PCB interferences which cannot be circumvented by a secondary or tertiary ion, either: 12.3.1 12.3.2 Perform additional chemical cleanup (Section 9) and then reanalyze the sample; or 12.3.3 12.4 Reanalyze the sample on a different column which separates the PCB and interferents; Quantitate the entire peak as PCB. Sum all of the peaks for each homolog and then sum those to yield the total PCB mass, MT, in the sample. If a concentration-perpeak or concentration-per-homolog reporting format is desired, carry each value through the calculations in an appropriate manner. C-40 �TABLE 13. CHARACTERISTIC IONS FOR Specific compound C612C6H9C1 ^ Primary 13 C-LABELED PCS SURROGATES Ion (relative intensity) Secondary Tertiary 13 194 (100) 196 (33) 13 304 (100) 306 (49) 302 (78) 13 442 (100) 444 (65) 440 (89) 510 (100) 512 (87) 514 (50) C12H6C14 C12H2C18 13 C12C110 C-41 �12.5 Calculation of air sample volume1 12.5.1 Nomenclature M = Mass of PCB represented by a chromatographic peak " micrograms M~ = Total mass of PCBs in sample, micrograms C = Concentration of PCBs in air, micrograms per cubic meter, corrected to standard conditions of 20°C, 760 mm Hg (68°F, 29.92 in. Hg) on dry basis A = Cross-sectional area of nozzle, square meter (square n feet) B = Water vapor in the gas stream, proportion by volume I = Percent of isokinetic sampling MWw = Molecular weight of water, 18 g/g-mole (18 lb/ Ib-mole) P, = Barometric pressure at the sampling site, mm Hg oar /. (in. IT \ Hg) Ps = Absolute stack gas pressure, mm Hg (in. Hg) Pstd, = Standard absolute pressure, 760 mm Hg (29.92 in Hg) R = Ideal gas constant, 0.06236 mm Hg-m3/K-g-raole (21.83 in. Hg-ft3/°R-lb-mole) T = Absolute average dry gas meter temperature °K (°R) TS = Absolute average stack gas temperature °K (°R) = Standard absolute temperature, 293°K (528°R) V, = Total volume of liquid collected in impingers and silica gel, milliliters. Volume of water collected equals the weight increase in grams times 1 ml/g Vm = Volume of gas sample as measured by dry gas meter, dcm (dcf) V , ,x = Volume of gas sample measured by the dry gas meter corrected to standard conditions, dscra (dscf) C-42 �,N = Volume of water vapor in the gas sample corrected to standard conditions, son (scf) V = Total volume of sample, railliliter V = Stack gas velocity, calculated by EPA Method 2, s m/sec (ft/sec) AH = Average pressure differential across the orifice meter, mm H20 (in. H20) Pw = Density of water, 1 g/ml (0.00220 Ib/ral) 6 = Total sampling time, minutes 13.6 = Specific gravity of mercury 60 = Seconds per minute 100 = Conversion to percent 12.5.2 Average dry gas meter temperature and average orifice pressure drop - See data sheet (Figure 4). 12.5.3 Dry gas volume - Correct the sample volume measured by the dry gas meter to standard conditions [20°C, 760 mm Hg (68°F, 29.92 in. Hg)] by using Equation 12-2. ., Vstd) - T „ *std V m - P +M *bar 13.6 _ „„ = P + AH bar 1376" Eq. 12-2 V where K = 0.3855°K/mm Hg for metric units = 17.65 °R/in. Hg for English units 12.5.4 Volume of water vapor P w RTstd Vw(std) = Ic MW P 5,^ = K Ic , , ,, V, ajT- .V, v ' w std where K = 0.00134 m3/ml for metric units = 0.0472 ft3/ml for English units 12.5.5 Eq. ^ 12-3 Moisture content = -J^Iltd)E 1. 2 4 V +V m(std) w(std) If the liquid droplets are present in the gas stream, assume the stream to be saturated and use a psychrometric chart to obtain an approximation of the moisture percentage. B ws C-43 �12.6 Concentration of PCBs in stack gas - Determine the concentration of PCBs in the air according to Equation 12-5 and report in micrograms per cubic meter using Table 14. If an alternate reporting format (e.g., concentration per peak) is desired, a different report form may be used. M C a = K y—-- Eq. 12-5 m(std) where K = 35.31 ft3/m3 12.7 Isokinetic variation 12.7.1 Calculations from raw data. [K V, + (V /T ) (P, ) + AH/13.6)] 1nn Ts l 100 Ic m m bar 10 , T r i = -60s e v P A- Eq- 12' s n 3 where K = 0.00346 mm Hg-m /ml-°K for metric units = 0.00267 in. Hg-ft3/ml-°R for English units 12.7.2 Calculations from intermediate values T V f «. ,, P . . 100 T s m(std) std 1 " T , , s 6n s 60 (1-B J A P std V ws _ ,7 ' /"/ tq ,, _s Vm(std) _ T Ps Vs An 0 (1-B ws) where K = 4.323 for metric units = 0.0944 for English units 12.7.3 Acceptable results - The following range sets the limit on acceptable isokinetic sampling results: If 90% < I < 110%, the results are acceptable. If the results are low in comparison to the standards and I is beyond the acceptable range, the Agency may opt to accept the results. 12.8 Round off all numbers reported to two significant figures. 13.0 Confirmation If there is reason to question the qualitative identification (Section 11.0), the analyst may choose to confirm that a peak is not a PCB. Any technique may be chosen provided that it is validated as having equivalent or superior selectivity and sensitivity to GC/EIMS. Some candidate techniques include alternate GC columns (with EIMS detection), GC/CIMS, GC/NCIMS, high resolution EIMS, and MS/MS techniques. Each laboratory C-44 �TABLE 14. ANALYSIS REPORT INCIDENTAL PCBs IN AIR Sample No. Sample Matrix Sample Source Notebook No. or File Location m3 Volume Collected [V , fcdJ Mass of Internal Stanaara Injected, M. is pg Qualitative Analyte 1° 2° I l° T 2° Ratio Theoretical IS 298 246 100/76 1-C1 188 190 100/33 2-C1 222 224 100/66 3-C1 256 258 100/99 4-C1 292 290 100/76 5-C1 326 328 Quantitative Ion Mass OK? Used RF M (pg) P 1.000 100/66 i 6-C1 360 362 100/82 7-C1 394 396 100/98 8-C1 430 432 100/66 9-C1 464 466 100/76 10-C1 498 500 100/87 Total (MT) Concentration (C.) M£ 3 Mg/m Reported by: Internal Audit: Name Name EPA Audit: Name Signature/Date Signature/Date Signature/Date Organization Organization Organization C-45 �must validate confirmation techniques to show equivalent or superior selectivity between PCBs and interferences and sensitivity (limit of quantitation, LOQ). If a peak is confirmed as being a non-PCB, it may be deleted from the calculation (Section 12). If a peak is confirmed as containing both PCB and non-PCB components, it must be quantitated according to Section 12.3. 14.0 Quality Control 14.1 Each laboratory that uses this method must operate a formal quality control (QC) program. The minimum requirements of this program consist of an initial demonstration of laboratory capability and the analysis of spiked samples as a continuing check on performance. The laboratory must maintain performance records to define the quality of data that are generated. After a date specified by the Agency, ongoing performance checks should be compared with established performance criteria to determine if the results of analyses are within accuracy and precision limits expected of the method. 14.2 The analysts must certify that the precision and accuracy of the analytical results are acceptable by: 14.2.1 14.2.2 14.3 The absolute precision of surrogate recovery, measured as the RSD of the integrated EIMS area (A ) for a set of samples, must be ±10%. The mean recovery (R ) of at least four replicates of a QC check sample to be supplied by the Agency must meet Agency-specified accuracy and precision criteria. This forms the initial data base for establishing control limits (see Section 14.3 below). Control limits - The laboratory must establish control limits using the following equations: Upper control limit (UCL) = R + 3 RSD Upper warning limit (UWL) = R + 2 RSD Lower warning limit (LWL) = RC - 2 RSDc Lower control limit (LCL) = R - 3 RSD These may be plotted on control charts. If an analysis of a check sample falls outside the warning limits, the analyst should be alerted that potential problems may need correction. If the results for a check sample fall outside the control limits, the laboratory must take corrective action and recertify the performance C-46 �(Section 14.2) before proceeding with analyses. The warning and control limits should be continuously updated as more check sample replicates are added to the data base. 14.4 Before processing any samples, the analyst should demonstrate through the analysis of a reagent blank that all glassware and reagent interferences are under control. Each time a set of samples is analyzed or there is a change in reagents, a laboratory reagent blank should be processed as a safeguard against contamination. 14.5 Procedural QC - The various steps of the analytical procedure should have quality control measures. These include but are not limited to: 14.5.1 GC performance - See Section 7.1 for performance criteria. 14.5.2 MS performance - See Section 7.2 for performance criteria. 14.5.3 Qualitative identification - At least 10% of the PCB identifications, as well as any questionable results, should be confirmed by a second mass spectrometrist. 14.5.4 Quantitation - At least 10% of all manual calculations, including peak area calculation, must be checked. After changes in computer quantitation routes, the results should be manually checked. 14.6 A minimum of 10% of all samples, one sample per month or one sample per matrix type, whichever is greater, must be selected at random, sampled, and analyzed in triplicate to monitor the precision of the analysis. An RSD of ±30% or less must be achieved. If the precision is greater than ±30%, the analyst must be recertified (see Section 14.2). 14.7 A minimum of 10% of all samples, one sample per month or one sample per matrix type, whichever is greater, selected at random, must be analyzed by the standard addition technique. Two aliquots of the sample are analyzed, one "as is" and one spiked with a sufficient amount of solution CSxxx to yield approximately 100 pg/ sample of each compound. The spiking compounds are thoroughly incorporated by mechanical agitation. For the liquid impinger contents, shaking for 30 sec should be sufficient. For the Florisil, 10 min tumbling is recommended. For filters where inadequate incorporation may be expected, overnight equilibration with agitation is recommended. Note: The volume measurement of the spiking solution is critical to the overall method precision. The analyst must exercise caution that the volume is known to ±1% or better. Where necessary, calibration of the pipet is recommended. C-47 �The samples are analyzed together and the quantitative results calculated. The recovery of the spiked compounds (calculated by difference) must be 80-120%. If the sample is known to contain specific PCB isomers, these isomers may be substituted for solution CSxxx. If the concentrations of PCBs are known to be high, the amount added should be adjusted so that the spiking level is 1.5 to 4 times the measured PCB level in the unspiked sample. 14.8 Sampling efficiency - The efficiency of PCB collection during sampling should be monitored. This may be achieved by adding a known amount of the 13C surrogate spiking solution (Section 6.4) sufficient to give an analytical signal well above background to the first impinger prior to sampling. The recovery of the four compounds should be > 14.9 Interlaboratory comparison - Interlaboratory comparison studies are planned. Participation requirements, level of performance, and the identity of the coordinating laboratory will be presented in later revisions. 14.10 It is recommended that the participating laboratory adopt additional QC practices for use with this method. The specific practices that are most productive depend upon the needs of the laboratory and the nature of the samples. Field duplicates or triplicates may be analyzed to monitor the precision of the sampling technique. Whenever possible, the laboratory should perform analysis of standard reference materials and participate in relevant performance evaluation studies. 15.0 Quality Assurance Each participating laboratory must develop a quality assurance plan according to EPA guidelines.5 The quality assurance plan must be submitted to the Agency for approval. 16.0 Method Performance The method performance is being evaluated. Limits of quantitation; average intralaboratory recoveries, precision, and accuracy; and interlaboratory recoveries, precision, and accuracy will be presented. 17.0 Documentation and Records Each laboratory is responsible for maintaining full records of the analysis. Laboratory notebooks should be used for handwritten records. GC/MS data must be archived on magnetic tape, disk, or a similar device. Hard copy printouts may be kept in addition if desired. QC records should be maintained separately from sample analysis records. C-48 �The documentation roust describe completely how the analysis was performed. Any variances from the protocol must be noted and fully described. Where the protocol lists options (e.g., sample cleanup), the option used and specifies (solvent volumes, digestion times, etc.) must be stated. C-49 �REFERENCES 1. Haile, C. L., and E. Baladi, "Methods for Determining the Polychlorinated Biphenyl Emissions from Incineration and Capacitor and Transformer Filling Plants," U.S. Environmental Protection Agency, (1977) EPA-600/4-73-048. 2. U.S. Environmental Protection Agency, Federal Register, 42(160), Thursday, August 18, 1977. 3. Martin, R. M., "Construction Details of Isokinetic Source Sampling Equipment," Environmental Protection Agency, Air Pollution Control Office Publication No. APTD-0581. 4. Bellar, T. A., and J. J. Lichtenberg, "The Determination of Polychlorinated Biphenyls in Transformer Fluid and Waste Oils," Prepared for U.S. Environmental Protection Agency, (1981) EPA-600/4-81-045. 5. Quality Assurance Program Plan for the Office of Toxic Substances, Office of Pesticides and Toxic Substances, U.S. Environmental Protection Agency, Washington, D.C., October 1980. C-50 �APPENDIX D ANALYTICAL METHOD; THE ANALYSIS OF BY-PRODUCT CHLORINATED BIPHENYLS IN INDUSTRIAL WASTEWATER D-l �THE ANALYSIS OF BY-PRODUCT CHLORINATED BIPHENYLS IN INDUSTRIAL WASTEWATER i.0 Scope and Application 1.1 This is a gas chromatographic/electron impact mass spectrometric (GC/EIMS) method applicable to the determination of chlorinated biphenyls (PCBs) in industrial wastewater. The PCBs present may originate either as synthetic by-products or as contaminants derived from commercial PCB products (e.g., Aroclors). The PCBs may be present as single isomers or complex mixtures and may include all 209 congeners from monochlorobiphenyl through decachlorobiphenyl listed in Table 1. 1.2 The detection and quantitation limits are dependent upon the volume of sample extracted the complexity of the sample matrix and the ability of the analyst to remove interferents and properly maintain the analytical system. The method accuracy and precision will be determined in future studies. 1.3 This method is restricted to use by or under the supervision of analysts experienced in the use of gas chromatography/mass spectrometry (GC/MS) and in the interpretation of gas chromatograms and mass spectra. Prior to sample analysis, each analyst must demonstrate the ability to generate acceptable results with this method by following the procedures described in Section 14.2. 1.4 The validity of the results depends on equivalent recovery of the analyte and 13C PCBs. If the *3C PCBs are not thoroughly incorporated in the matrix, the method is not applicable. 1.5 During the development and testing of this method, certain analytical parameters and equipment designs were found to affect the validity of the analytical results. Proper use of the method requires that such parameters or designs must be used as specified. These items are identified in the text by the word "must." Anyone wishing to deviate from the method in areas so identified must demonstrate that the deviation does not affect the validity of the data. Alternative test procedure approval must be obtained from the Agency. An experienced analyst may make modifications to parameters or equipment identified by the term "recommended." Each time such modifications are made to the method, the analyst must repeat the procedure in Section 14.2. In this case, formal approval is not required, but the documented data from Section 14.2 must be on file as part of the overall quality assurance program. D-2 �TABLE 1. No. Structure NO. Menoeftloraoiphenylt 2 52 3 4 54 55 56 S7 IS » 60 61 (2 63 64 65 66 67 68 01chlorob<BHttiy1i a 9 10 n 12 13 14 IS 2.2' 2.3 2.3' 2.4 2.4' 2.5 2.6 3,3' 3,4 3.4' 3|5 4.4' Trlchloraolphtnyli 16 17 18 19 20 21 22 23 24 2S 26 27 28 29 JO 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 SO SI 2. 2'. 3 2.2'. 4 2. 2'. 5 2, 2', 6 2.3.3' 2,3.4 2.3.4' 2,3.5 2,3.6 2,3', 4 2, 3', 5 2,3', 6 2.4,4' 2.4,5 2.4,6 2, 4'. 5 2. 4' ,5 2', 3,4 2', 3,5 3,3', 4 3.3'.5 3,4,4' 3.4.5 3.4'. 5 NO. 69 70 71 72 73 74 75 76 77 78 79 80 81 10$ 2.3,3 ,4,4' 1 1 1 C 2 .J.J ,4,9 2,3,3 .4'. 5 2.3.3 ,4.5' 2,3.3 ,4.6 2,3.3 .4* ,6 2.3,3 ,5,5' 2,3,3 .5.6 2.3,3 .5', 6 2,3.4 4'.5 2.3,4 4' ,6 2.3,4 5,6 2.3,4', 5, 6 2,3'.4.4-.5 2.3'. 4. 4' .6 2.3'. 4, 5. 5' 2.3". 4,1 6 5', 21 .3. 3 . 4.5 2'. 3.4. 4' .5 21 .3.4.5. 52'.3.4.5,6' 3. 3'. 4. 4' .5 3,3'. 4. 5.5' 170 171 172 173 174 175 176 177 178 179 180 Htmeliloroblptitnylt 2.2'. «,S' 9* « *' 2 •* •*»• 2,2'. 5,62.3. 3'. 4 2.3.3'. 4' 2.3.3' .5 2.3.3-. S' 2.3,J'.< 2.3.4.4' 2.3.4.5 2,3.4,6 2.3. 4' .5 2,3.4-. 6 2.3,5.6 2.3', 4.4' 2.3', 4,5 2.3', 4.5" 2.3'. 4.6 2.3-.4'.5 2.3-,4-.6 .3'.S.5' ,3-.5',6 .4.4'. 5 ,4. 4', 6 '.3 4.5 .3' 4,4.3' 4,5 .3' 4,5' .3' 5.5.4.4-.S 2.2-.3.3'.4 2.2'. 3. 3', 5 2.2'.3.3',6 2. 2'. 3.4,4' 2,2' .3.4.5 2.2- .3, 4, 5' 2.2'. 3, 4,6 2,2'. 3.4,6"' 2.2'. 3, 4'. 5 2.2'. 3. 4'. 6 2.2', 3,5.5' 2.2'. 3, 5.6 2.2-.3.S.6' 2.2'. 3,5', 6 2.2-,3,6.6' 2.2-,3'.4.5 2.2'.3',4,6 2.2'. 4,4', 5 2.21,4.4'.6 2.2'. 4, 5.5' 2,2'. 4.5.5' 2.2-,4.S',6 2,2'. 4.6. S' 10 . 182 P«nt»ctil orebt phtny 1» Irtit IUO 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 Pmticftl arotl pHtny 1 » 82 S3 84 85 86 87 88 39 90 91 92 93 94 T«tricMorob<Bh«nyl» 95 96 97 2.2', 3.3' 2.2'. 3,4 98 2.2'. 3,499 2.2'.3,$ 100 2.2'.3.5' 101 2.2'. 3.6 10Z 2.2'.3,6' 103 2.2'. 4,4104 2,2' ,4, 5 2,2'. 4.J' 2 2 1 4 61 2.2 .4.6 structure Tttnetil arott Ph«nv1 s 1 4 5 6 7 NUMBERING OF PCB CONGENERS3 structure 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 2,2'.3.3-,4.42,2'. 3,3' .4. 2.2'.3.3',4. 2.2' .3,3'. 4. .2', 3.3'. 4, ,2'. 3, 3', 5, .2'.3,3',S. .2'.3.3'.5, ,2'. 3,3' 6, ' ,2'.3,4,4- ( 1 ,2'. 3,4. 4'. .2'. 3. 4.4' , 1 2.2". 3,4.4'. 2.2' .3,4, 5, 5 2,2'. 3,4, 5,6 2.2',3,4.5,6 2.2'. 3. 4,5'. 2.2'. 3,4, 6. 6 2,2' ,3,4'. 5. 2.2' ,3, 4', S. 2.2'.3,4'.5, 2.2', 3,4'. 5' 6 2.2', 3, 4' .5, 2.2-.3,5.5'. 2.2', 3.5.6.6' 2.2' ,4.4'. 5.5' 2.21, 4,4-. 5. S2.2',4.4'.M' 2,3,3'. 4, 4'. 5 2.3. 3' .4,4' .5' 2.3.3', 4, 4'. 6 2.3.3'. 4,5, 5' 2,3,3'. 4,i,« HtueMorobiphtnyl* 161 1X9 IK 163 164 165 166 167 168 169 tallieMttr, 181 183 184 185 186 187 188 189 190 191 192 193 2.3.3'. 4'. 5.6 2,3, 3', 4'. 5 ' , 6 2;3;3'.S.S',6 2.3.4. 4- .5,6 2,3' .4.4'. 5. 5' 2.3'. 4, 4', £'.6 3,3'.4.4'.5,5 1 2,2',3,3',4,4',5 2.2',3,3'.4,4',6 2.2', 3, 3 ' , 4, 5, 5' 2, 2'. 3, 3', 4,5,6 2. 2'. 3, 3', 4, 5,6' 2.2' .3,3' .4. 5'. 6 2,2'. 3.3', 4.6. 6' 2. 2 1 . 3,3' ,4' ,5,6 2. 2', 3,3'. 5, 5'. 6 2,2',3,3',5,6,6' 2.2-.3.4,4'.5.5' 2. 2', 3, 4, 4', 5. 6 2. 2'. 3, 4, 4', 5, 6' 2,2'. 3. 4, 4'. 5'. 6 2.2',3.4,4',j,6 1 2.2' .3. 4,5, 5' ,6 2,2' .3,4,5, 6, 6' 2,2',3,4 1 ,5.5 1 .6 2,2'. 3,4', 5. 6, 6' 2. 3, 3', 4, 4', 5,5' 2,3.3', 4. 4 ' , 5, 6 2. 3. 3' .4. 4' .5'. 6 2, 3, 3'. 4, 5,5' 6 2.3, 3'. 4' .5,5' ,6 Octiehl orobi phtny 1 s 194 195 196 197 198 199 200 201 202 203 204 205 2,2'. 3. 3', 4. 4 ' , 5,5' 2,2'.3,3'.4,4'.5,6 2,2'.3,3'.4.4',5,5' 2, 2', 3. 3'. 4, 4 ' , 6, 6' 2.2'.3.3',4,5.5'.S 2,2'. 3, 3' .4. 5,6, 6'1 2.2'.3.3 1 ,4.S'.6.6 2.2'. 3, 3'. 4. 5, 5 ' , 6' 2.2' .3. 3' ,5, 5'. 6.6' 2,2', 3, 4. 4'. 5, 5 ' . 6 2.2' .3,4,4' .5, 6, 6' 2,3.3' ,4.4'.5.5'.6 NomehlarobfohtnyM 206 207 208 2,2-.3,3'.4.4',5,5 < .S 2.2'.3.3'.4.4'.5,6,6' 2,2',3.3 I ,4,5,5'.6,6' 0«eicMorot>(pn«ityl X. ind Z«11. H., Frwnlui I. Anil. ChM., 302. 20-31 (1980). D-3 2.3. 3', 4, 5', 6 H»otieH1orattfphtnyl » 209 •Adapted 1rm structure 2.2'.3,3'4,4',$,5',6.6' �2.0 Summary 2.1 The wastewater must be sampled such that the specimen collected for analysis is representative of the whole. Statistically designed selection of the sampling position (valve, port, outfall, etc.) or time should be employed. The sample must be preserved to prevent PCB loss prior to analysis. Storage at 4°C with optional preservation at low pH is recommended. 2.2 The sample is mechanically homogenized and subsampled if necessary. The sample is then spiked with four 13C PCB surrogates and the surrogates incorporated by further mechanical agitation. 2.3 The surrogate-spiked sample is extracted and cleaned up at the discretion of the analyst. Possible extraction techniques include liquid-liquid partition and sorption onto resin columns followed by solvent elution. Cleanup techniques may include liquid-liquid partition, sulfuric acid cleanup, saponification, adsorption chromatography, gel permeation chromatography or a combination of cleanup techniques. The sample is diluted or concentrated to a final known volume for instrumental determination. The EPA Method 6081 and 6252 extraction and cleanup procedures may be used. 2.4 The PCB content of the sample extract is determined by capillary (preferred) or packed column gas chromatography/electron impact mass spectrometry (CGC/EIMS or PGC/EIMS) operated in the selected ion monitoring (SIM), full scan, or limited mass scan (IMS) mode. 2.5 PCBs are identified by comparison of their retention time and mass spectral intensity ratios to those in calibration standards. 2.6 PCBs are quantitated against the response factors for a mixture of 11 PCB congeners, using the response of the 13C surrogate to compensate for losses in workup and instrument variability. 2.7 The PCBs identified by the SIM technique may be confirmed by full scan CGC/EIMS, retention on alternate GC columns, other mass spectrometric techniques, infrared spectrometry, or other techniques, provided that the sensitivity and selectivity of the technique is demonstrated to be comparable or superior to GC/EIMS. 2.8 The analysis time is dependent on the extent of workup employed. The time required for instrumental analysis, excluding data reduction and reporting, is about 30 to 45 min. 2.9 Appropriate quality control (QC) procedures are included to assess the performance of the analyst and estimate the quality of the results. These QC procedures include the demonstration of laboratory capability: periodic analyst certification, the use of control charts, and the analysis of blanks, replicates, and standard addition samples. A quality assurance (QA) plan must be developed for each laboratory. D-4 �2.10 While several options are available throughout this method, the recommended procedure to be followed is: 2.10.1 The sample is collected according to a scheme which permits extrapolation of the sample data to the body or containers of water being sampled. 2.10.2 The sample is preserved at low pH and at 4°C to prevent any loss of PCBs or changes in matrix which may adversely affect recovery. 2.10.3 The sample is-mechanically homogenized and subsampled if necessary. 2.10.4 The sample is spiked with four 13C-PCB surrogates (4-chlorobiphenyl; 3,3',4,4'-tetrachlorobiphenyl; 2,2',3,3',5,5',6,6'-octachlorobiphenyl; and decachlorobiphenyl). 2.10.5 The sample is extracted. 2.10.6 The extract is cleaned up and concentrated to an appropriate volume. 2.10.7 An aliquot of the extract is analyzed by CGC/EIMS operated in the SIM mode. On-column injections onto a 15-m DB-5 capillary column, programmed (for toluene solutions) from 110° to 325°C at 10°/min after a 2 min hold is used. Helium at 45-cm/sec linear velocity is used as the carrier gas. 2.10.8 PCBs are identified by retention time and mass spectral intensities. 2.10.9 PCBs are quantitated against the response factors for a mixture of 11 PCB congeners. 2.10.10 The total PCBs are obtained by summing the amounts for each homolog found and the concentration is reported as micrograms per liter. 3.0 Interferences 3.1 Method interferences may be caused by contaminants in solvents, reagents, glassware, and other sample processing hardware, leading to discrete artifacts and/or elevated baselines in the total ion current profiles. All of these materials must be routinely demonstrated to be free from interferences by the analysis of laboratory reagent blanks as described in Section 14.4. D-5 �3.1.1 Glassware must be scrupulously cleaned. All glassware is cleaned as soon as possible after use by rinsing with the last solvent used. This should be followed by detergent washing with hot water and rinses with tap water and reagent water. The glassware should then be drained dry and heated in a muffle furnace at 400°C for 15 to 30 min. Some thermally stable materials, such as PCBs, may not be eliminated by this treatment. Solvent rinses with acetone and pesticide quality hexane may be substituted for the muffle furnace heating. Volumetric ware should not be heated in a muffle furnace. After it is dry and cool, glassware should be sealed and stored in a clean environment to prevent any accumulation of dust or other contaminants. It is stored inverted or capped with aluminum foil. 3.1.2 The use of high purity reagents and solvents helps to minimize interference problems. Purification of solvents by distillation in all-glass systems may be required. All solvent lots must be checked for purity prior to use. 3.2 Matrix interferences may be caused by contaminants that are coextracted from the sample. The extent of matrix interferences will vary considerably from source to source, depending upon the nature and diversity of the sources of samples. 4.1 The toxicity or carcinogenicity of each reagent used in this method has not been precisely defined; however, each chemical compound should be treated as a potential health hazard. From this viewpoint, exposure to these chemicals must be reduced to the lowest possible level by whatever means available. The laboratory is responsible for maintaining a current awareness file of OSHA regulations regarding the safe handling of the chemicals specified in this method. A reference file of material data handling sheets should also be made available to all personnel involved in the chemical analysis. 4.2 Polychlorinated biphenyls have been tentatively classified as known or suspected human or mammalian carcinogens. Primary standards of these toxic compounds should be prepared in a hood. Personnel must wear protective equipment, including gloves and safety glasses. 4.0 Congeners highly substituted at the meta and para positions and unsubstituted at the ortho positions are reported to be the most toxic. Extreme caution should be taken when handling these compounds neat or in concentration solution. The class includes 3,3',4,4'-tetrachlorobiphenyl (both natural abundance and isotopically labeled). D-6 �4.3 4.4 5.0 Diethyl ether should be monitored regularly to determine the peroxide content. Under no circumstances should diethyl ether be used with a peroxide content in excess of 50 ppm as an explosion could result. Peroxide test strips manufactured by EM Laboratories (available from Scientific Products Company, Cat. No. P1126-8 and other suppliers) are recommended for this test. Procedures for removal of peroxides from diethyl ether are included in the instructions supplied with the peroxide test kit. Waste disposal must be in accordance with RCRA and applicable state rules. Apparatus and Materials 5.1 Sampling containers - Amber glass bottles, 1-liter or other appropriate volume, fitted with screw caps lined with Teflon. Cleaned foil may be substituted for Teflon if the sample is not corrosive. If amber bottles are not available, samples should be protected from light using foil or a light-tight outer container. The bottle must be washed, rinsed with acetone or methylene chloride, and dried before use to minimize contamination. 5.2 Glassware - All specifications are suggestions only. Catalog numbers are included for illustration only. 5.2.1 Volumetric flasks - Assorted sizes. 5.2.2 Pipets - Assorted sizes, Mohr delivery. 5.2.3 Micro syringes - 10.0 pi for packed column GC analysis, 1.0 |Jl for on-column CGC analysis. 5.2.4 Chromatographic column - Chromaflex, 400 mm long x 19 mm ID (Kontes K-420540-9011 or equivalent). 5.2.5 Gel permeation chromatograph - GPC Autoprep 1002 (Analytical Bio Chemistry Laboratories, Inc.) or equivalent. 5.2.6 Kuderna-Danish Evaporative Concentrator Apparatus 5.2.6.1 Concentrator tube - 10 ml, graduated (Kontes K-570050-1025 or equivalent). Calibration must be checked. Ground glass stopper size (S19/22 joint) is used to prevent evaporation of solvent. 5.2.6.2 Evaporative flask - 500 ml (Kontes K-57001-0500 or equivalent). Attach to concentrator tube with springs (Kontes K-662750-0012 or equivalent). 5.2.6.3 Snyder column - Three ball macro (Kontes K5030000121 or equivalent). D-7 �5.3 Balance - Analytical, capable of accurately weighing 0.0001 g. 5.4 Gas chromatography/mass spectrometer system. 5.4.1 Gas chromatograph - An analytical system complete with a temperature programmable gas chromatograph and all required accessories including syringes, analytical columns, and gases. The injection port must be designed for oncolumn injection when using capillary columns or packed columns. Other capillary injection techniques (split, splitless, "Grob," etc.) may be used provided the performance specifications stated in Section 7.1 are met. 5.4.2 Capillary GC column - A 12-20 m long x 0.25 mm ID fused silica column with a 0.25 |Jm thick DB-5 bonded silicone liquid phase (J&W Scientific) is recommended. Alternate liquid phases may include OV-101, SP-2100, Apiezon L, Dexsil 300, or other liquid phases which meet the performance specifications stated in Section 7.1. 5.4.3 Packed GC column - A 180 cm x 0.2 cm ID glass column packed with 3% SP-2250 on 100/120 mesh Supelcoport or equivalent is recommended. Other liquid phases which meet the performance specifications stated in Section 7.1 may be substituted. 5.4.4 Mass spectrometer - Must be capable of scanning from 150 to 550 Daltons every 1.5 sec or less, collecting at least five spectra per chromatographic peak, utilizing a 70-eV (nominal) electron energy in the electron impact ionization mode and producing a mass spectrum which meets all the criteria in Table 2 when 50 ng of decafluorotriphenyl phosphine [DFTPP, bis(perfluorophenyl)phenyl phosphine] is injected through the GC inlet. Any GC-to-MS interface that gives acceptable calibration points at 10 ng per injection for each PCB isomer in the calibration standard and achieves all acceptable performance criteria (Section 10) may be used. Direct coupling of the fused silica column to the MS is recommended. Alternatively, GC-toMS interfaces constructed of all glass or glass-lined materials are recommended. Glass can be deactivated by silanizing with dichlorodimethylsilane. 5.4.5 A computer system that allows the continuous acquisition and storage on machine-readable media of all mass spectra obtained throughout the duration of the chromatographic program must be interfaced to the mass spectrometer. The data system must have the capability of integrating the abundances of the selected ions between specified limits and relating integrated abundances to concentrations using the calibration procedures described in this method. The computer must have software that allows D-8 �TABLE 2. DFTPP KEY IONS AND ION ABUNDANCE CRITERIA Mass Ion abundance criteria 197 198 199 Less than 1% of mass 198 100% relative abundance 5-9% of mass 198 275 10-30% of mass 198 365 Greater than 1% of mass 198 441 442 443 Present, but less than mass 443 Greater than 40% of mass 198 17-23% of mass 442 D-9 �searching any GC/MS data file for ions of a specific mass and plotting such ion abundances versus time or scan number to yield an extracted ion current profile (EICP). Software must also be available that allows integrating the abundance in any EICP between specified time or scan number limits. 6.0 Reagents 6.1 Solvents - All solvents must be pesticide residue analysis grade. New lots should be checked for purity by concentrating an aliquot by at least as much as is used in the procedure. 6.2 Stock standard solutions - Standards of the PCB congeners listed in Table 3 are available from Ultra Scientific, Hope, Rhode Island; or Analabs, North Haven, Connecticut. 6.3 Calibration standard stock solutions - Primary dilutions of each of the individual PCBs listed in Table 3 are prepared by weighing approximately 1-10 mg of material within 1% precision. The PCB is then dissolved and diluted to 1.0 ml with hexane. Calculate the concentration in mg/ml. The primary dilutions are stored at 4°C in screw-cap vials with Teflon cap liners. The meniscus is marked on the vial wall to monitor solvent evaporation. Primary dilutions are stable indefinitely if the seals are maintained. The validity of primary and secondary dilutions must be monitored on a quarterly basis by analyzing four quality control check samples (see Section 14.2). 6.4 Working calibration standards - Working calibration standards are prepared that are similar in PCB composition and concentration to the samples by mixing and diluting the individual standard stock solutions. Example calibration solutions are shown in Table 3. The mixture is diluted to volume with pesticide residue analysis quality hexane. The concentration is calculated in ng/ml as the individual PCBs. Dilutions are stored at 4°C in narrow-mouth, screw-cap vials with Teflon cap liners. The meniscus is marked on the vial wall to monitor solvent evaporation. These secondary dilutions can be stored indefinitely if the seals are maintained. These solutions are designated "CSxxx," where the xxx is used to encode the nominal concentration in ng/ml. 6.5 Alternatively, certified stock solutions similar to those listed in Table 3 may be available from a supplier, in lieu of the procedures described in Section 6.4. 6.6 DFTPP standard - A 50-ng/pl solution of DFTPP is prepared in acetone or another appropriate solvent. 6.7 Surrogate standard stock solution - The four 13C-labeled PCBs listed in Table 4 may be available from a supplier as a certified solution. This solution may be used as received or diluted further. These solutions are designated "SSxxx," where the xxx is used to encode the nominal concentration in ng/ml. D-10 �TABLE 3. CONCENTRATIONS OF CONGENERS IN PCS CALIBRATION STANDARDS (ng/ml)a Homolog Congener no. CS1000 CS100 CS050 CS010 1 1 1,040 104 52 10 1 3 1,000 100 50 10 2 7 1,040 104 52 10 3 30 1,040 104 52 10 4 50 1,520 152 76 15 5 97 1,740 174 87 17 6 143 1,920 192 96 19 7 183 2,600 260 130 26 8 202 4,640 464 232 46 9 207 5,060 506 253 51 10 209 4,240 424 212 42 4 255 255 255 255 1 211 (RS) 104 104 104 104 4 212 (RS) 257 257 257 257 8 213 (RS) 407 407 407 407 10 a 210 (IS) 214 (RS) 502 502 502 502 Concentrations given as examples only. D-ll �TABLE 4. COMPOSITION OF SURROGATE SPIKING SOLUTION (SS100) CONTAINING 13C-LABELED PCBs3 Congener no. Compound Concentration (Hg/ml) 211 104 212 (13C12)3,3' ,4,4'-tetrachlorobiphenyl 257 213 (13C12)2,2' ,3,3' ,5,5' ,6,6'-octachlorobiphenyl 395 214 a (I1 ,2' ,3' ,4' ,5' ,6'-13C6)4-chlorobiphenyl (13C12)decachlorobiphenyl 502 Concentrations given as examples only. D-12 �6.8 6.9 Solution stability - The calibration standard, surrogate and DFTPP solutions should be checked frequently for stability. These solutions should be replaced after 6 months, or sooner if comparison with quality control check samples indicates compound degradation or concentration change. 6.10 7 .0 Internal standard solution - A solution of de-3,3" ,4,4" -tetrachlorobiphenyl is prepared at a nominal concentration of 1-10 mg/ml in hexane. The solution is further diluted to give a working standard. Quality control check samples will be supplied by the Agency. Calibration 7.1 The gas chroma tograph must meet the minimum operating parameters shown in Tables 5 and 6, daily. If all of the criteria are not met, the analyst must adjust conditions and repeat the test until all criteria are met. 7.2 The mass spectrometer must meet the minimum operating parameters shown in Tables 2, 7, and 8, daily. If all criteria are not met, the analyst must retune the spectrometer and repeat the test until all conditions are met. The PCB response factor (RF ) must be determined using Equat 7-1 for the analyte homologi. where is p RF = response factor of a given PCB isomer A = area of the characteristic ion for the PCB congener P peak M = mass of PCB congener injected (nanograms) A. = area of the characteristic ion for the internal standard peak M. = mass of internal standard injected (nanograms) IS Using the same conditions as for RF , the surrogate response factors (RF ) must be determined usSng Equation 7-2. A x M. IS S where A S = area of the characteristic ion for the surrogate peak MS = mass of surrogate injected (nanograms) Other items are the same as defined in Equation 7-1. D-13 �TABLE 5. OPERATING PARAMETERS FOR CAPILLARY COLUMN GAS CHROMATOGRAPHIC SYSTEM Recommended Parameter Tolerance Gas chromatograph Finnigan 9610 Other Column 15 m x 0.255 mm ID Fused silica Other Liquid phase DB-5 Other nonpolar or semipolar Liquid phase thickness 0.25 urn < 1 |Jm Carrier gas Helium Hydrogen Carrier gas velocity 45 cm/sec Optimum performance Injector On-column (J&W) (J&W) Injector temperature Optimum performance Injection volume Other c 1.0 plc Optimum performance Other d Initial column temperature 70°C (2 min) Column temperature program 70°-325°C at 10°C/min£ Separator None Glass jet or other Transfer line temperature 280°C Optimum** Tailing factor 0.7-1.5 0.4-3 Peak width 7-10 sec < 15 sec Other Other a Substitutions permitted with any common apparatus or technique provided performance criteria are met. b Measured by injection of air or methane at 270°C oven temperature. c For on-column injection, manufacturer's instructions should be followed regarding injection technique. d With on-column injection, initial temperature equals boiling point of the solvent; in this instance, hexane. e C^Clio elutes at 270°C. Programming above this temperature ensures a clean column and lower background on subsequent runs. f Fused silica columns may be routed directly into the ion source to prevent separator discrimination and losses. g High enough to elute all PCBs, but not high enough to degrade the column if routed through the transfer line. h Tailing factor is width of front half of peak at 10% height divided by width of back half of peak at 10% height for single PCB congeners in solution CSxxx. i Peak width at 10% height for a single PCB congener is CSxxx. D-14 �TABLE 6. OPERATING PARAMETERS FOR PACKED COLUMN GAS CHROMATOGRAPHY SYSTEM Tolerances Recommended Parameter Gas chromatograph Finnigan 9610 Other3 Column 180 cm x 0.2 cm ID glass Other Column packing 3% SP-2250 on 100/ 120 mesh Supelcoport Other nonpolar or semipolar Carrier gas Helium Hydrogen Carrier gas flow rate 30 ml/min Optimum performance Injector On-column Injector temperature 250°C Optimum Injection volume 1.0 pi g 5 pi Initial column temperature 150°C, 4 min Other Column temperature program 150°C-260° at 8°/min Other Separator Glass jet Other Transfer line temperature 280°C Optimum3 Tailing factor 0.7-1.5 0.4-3 Peak widthd 10-20 sec < 30 sec p a Substitutions permitted if performance criteria are met. b High enough to elute all PCBs. c Tailing factor is width of front half of peak at 10% height divided by width of back half of peak at 10% height for single PCS congeners in solution CSxxx. d Peak width at 10% height for a single PCB congener in CSxxx. D-15 �TABLE 7. OPERATING PARAMETERS FOR QUADRUPOLE MASS SPECTROMETER SYSTEM Parameter Recommended Tolerance Mass spectrometer Finnigan 4023 Other3 Data system Incos 2400 Other Scan range 95-550 Other Scan time 1 sec Otherb Resolution Unit Optimum performance Ion source temperature 280°C 200°-300°C Electron energy 70 eV Optimum performance Trap current 0.2 mA Optimum performance Multiplier voltage -1,600 V Optimum performance Preamplifier sensitivity 10"6 A/V Set for desired working range a Substitutions permitted if performance criteria are met. b Greater than five data points over a GC peak is a minimum. c Filaments should be shut off during solvent elution to improve instrument stability and prolong filament life, especially if no separator is used. D-16 �TABLE 8. OPERATING PARAMETERS FOR MAGNETIC SECTOR MASS SPECTROMETER SYSTEM Parameter Tolerance Recommended Mass spectrometer Finnigan MAT 311A Other3 Data system Incos 2400 Other Scan range 98-550 Other Scan mode Exponential Other Cycle time 1.2 sec Otherb Resolution 1,000 > 500 Ion source temperature 280°C 250°-300°C Electron energy 70 eV 70 eV Emission current 1-2 mA Optimum Filament current Optimum Optimum Multiplier -1,600 V Optimum a Substitutions permitted if performance criteria are met. b Greater than five data points over a GC peak is a minimum. c Filaments should be shut off 'during solvent elution to improve instrument stability and prolong filament life, especially if no separator is used. D-17 �If specific congeners are known to be present and if standards are available, selected RF values may be employed. For general samples, solutions CSxxx and SSxxx or a mixture (Tables 3 and 4), with a similar level of internal standard (dg-3,3',4,4*-tetrachlorobiphenyl) added, may be used as the response factor solution. The PCB-surrogate pairs to be used in the RF calculation are listed in Table 9. Generally, only the primary ions of both the analyte and surrogate are used to determine the RF values. If alternate ions are to be used in the quantitation, the RF must be determined using that characteristic ion. The RF value must be determined in a manner to assure ±20% accuracy and precision. For instruments with good day-to-day precision, a running mean (RF) based on seven values determined once each day may be appropriate. Other options include, but are not limited to, triplicate determinations of a single concentration spaced throughout a day or determination of the RF at three different levels to establish a working curve. If replicate RF values differ by greater than ±10% RSD, the system performance should be monitored closely. If the RSD is greater than ±20%, the data set must be considered invalid and the RF redetermined before further analyses are done. 7.4 7.5 8.0 If the GC/EIMS system has not been demonstrated to yield a linear response or if the analyte concentrations are more than two orders of magnitude different from those in the RF solution, a calibration curve must be prepared. If the analyte and RF solution concentrations differ by more than one order of magnitude, a calibration curve should be prepared. A calibration curve should be established with triplicate determinations at three or more concentrations bracketing the analyte levels. The relative retention time (RRT) windows for the 10 homologs and surrogates must be determined. If all congeners are not available, a mixture of available congeners or an Aroclor mixture (e.g., 1016/1254/1260) may be used to estimate the windows. The windows must be set wider than observed if all isomers are not determined. Typical RRT windows for one column are listed in Table 10. The windows may differ substantially if other GC parameters are used. Sample Collection, Handling, and Preservation 8.1 Amber glass sample containers should have Teflon-lined screw caps. With noncorrosive samples, methylene chloride-washed aluminum foil liners may be substituted. The volume is determined by the amount of sample to be collected but will usually be 1 liter or 1 qt. The sample size is dependent on the anticipated PCB levels and difficulty of the subsequent extraction/cleanup steps. D-18 �TABLE 9. Analyte Congener no . 1 2,3 PAIRINGS OF ANALYTE, CALIBRATION, AND SURROGATE COMPOUNDS Compound 2-C12H9Cl 3- and 4-C12H9Cl 1 3 /•* TT r>~t Li2ngLl2 -f ' i £. on 1O~ jy r* TJ r*i L. 12^7 *•*-!- 3 An_fti HU O 1 p l2^6*->-'-4 u n L. 82-127 128-169 C12H5C15 C12H4C16 1"7A1OO 1/0-19J r* ur*! Li2h.3Ll7 194-205 206-208 209 C12H2C18 C12HC19 C12C110 Compound Congener no. 2 4 2,4 2,4, 6 2,2' ,4 ,6 2,2' ,3 ',4,5 2,2' ,3 ,4,5 ,6' 2,2' ,3 ',4,4', 5', 6 2,2' ,3 ,3',5, 5', 6, 6' J 2,2' ,3 ,3'» 4, 4', 5, 6, 6' /"* f L12Lli 0 211 211 211 212 212 212 212 213 213 213 214 Congener no. 4— 1 r 1j O 30 50 97 143 183 202 207 209 v£> a Surrogate Calibration standard Ballschmiter numbering system, see Table 1. Compound 13 C6-4 C6-4 13 C6-4 13 Ci2-3,3' ,4 ,4' 13 Ci2-3,3' ,4 ,4' 13 C12-3,3' ,4 ,4' 13 Ci2-3,3' ,4 ,4' 13 ,6,6' C12-2,2' ,3 ,3', 5, 5' 13 C12-2,2' ,3 ,3' , J V J 3 5 , ,6,6' 13 ,6,6' C12-2,2' ,3 ,3', 5, 5' 13 13 c12ci10 �TABLE 10. PCB homolog RELATIVE RETENTION TIME (RRT) RANGES OF PCB HOMOLOGS VERSUS d6-3,3',4.4'-TETRACHLOROBIPHENYL No. of isomers measured Observed range of RRTs3 Calibration solution Congener Observed RRT3 no. Projected range of RRTs 3 0.40-0.50 1 3 0.43 0.50 0.35-0.55 10 0.52-0.69 7 0.58 0.35-0.80 9 0.62-0.79 30 0.65 0.35-1.10 Tetrachloro 16 0.72-1.01 50 0.75 0.55-1.05 Pentachloro 12 0.82-1.08 97 0.98 0.80-1.10 Hexachloro 13 0.93-1.20 143 1.05 0.90-1.25 Heptachloro 4 1.09-1.30 183 1.15 1.05-1.35 Octachloro 6 1.19-1.36 202 1.19 1.10-1.50 Nonachloro 3 1.31-1.42 207 1.33 1.25-1.50 Decachloro 1 1.44-1.45 209 1.44 1.35-1.50 Monochloro Dichloro Trichloro a The RRTs of the 77 congeners and a mixture of Aroclor 1016/1254/1260 were measured versus 3,3',4,4'-tetrachlorobiphenyl-d6 (internal standard) using a 15-m J&W DB-5 fused silica column with a temperature program of 110°C for 2 min, then 10°C/min to 325°C, helium carrier at 45 cm/sec, and an oncolumn injector. A Finnigan 4023 Incos quadrupole mass spectrometer operating with a scan range of 95-550 daltons was used to detect each PCB congener. b The projected relative retention windows account for overlap of eluting homologs and take into consideration differences in operating systems and lack of all possible 209 PCB congeners. D-20 �8.2 Sample bottle preparation 8.2.1 8.2.2 Sample bottles are heated to 400°C for 15 to 20 min or rinsed with pesticide grade acetone or hexane and allowed to air dry. 8.2.3 8.3 All sample bottles and caps should be washed in detergent solution, rinsed with tap water and then with distilled water. The bottles and caps are allowed to drain dry in a contaminant-free area. Then the caps are rinsed with pesticide grade hexane and allow to air dry. The clean bottles are stored inverted or sealed until use. Sample collection 8.3.1 8.3.2 If possible, mix the source thoroughly before collecting the sample. If mixing is impractical, the sample should be collected from a representative area of the source. If the liquid is flowing through an enclosed system, sampling through a valve should be randomly timed. 8.3.3 8.4 The primary consideration in sample collection is that the sample collected be representative of the whole. Therefore, sampling plans or protocols for each individual producer's situation will have to be developed. The recommendations presented here describe general situations. The number of replicates and sampling frequency also must be planned prior to sampling. Fill the bottle with water, add preservative (Section 8.4), cap tightly, and shake well. To prevent the caps from working loose during storage tape the caps on with a water-insoluble tape. Sample preservation - Samples should be stored at 4°C. Since there is a possibility of microbial degradation, addition of H2S04 during collection to a pH < 2 is recommended. A test strip is used to monitor the pH. Storage times in excess of 4 weeks are not recommended. If residual chlorine is present in the sample, it should be quenched with sodium thiosulfate. EPA Methods 330.4 and 330.5 may be used to measure the residual chlorine.3 Field test kits are available for this purpose. 9.0 Sample Preparation 9.1 Sample homogenization and subsampling - The sample is homogenized by shaking, blending, or other appropriate mechanical technique, if necessary. If the density of the sample is not between 0.9 D-21 �and 1.1, the density should be determined and reported. Consideration should be given to treating the sample as a product waste (see separate protocol). Note: 9.2 Surrogate addition - An appropriate volume of surrogate solution SSxxx is pipetted into the sample. The final concentration of the surrogates must be in the working range of the calibration and well above the matrix background. Note: 9.3 The precision of the mass determination at this step will be reflected in the overall method precision. Therefore, an analytical balance must be used to assure that the weight is accurate to ±1% or better. The volume measurement of cal to the overall method exercise caution that the better. Where necessary, recommended. the spiking solution is critiprecision. The analyst must volume is known to ±1% or calibration of the pipet is Sample preparation (extraction/cleanup) - After addition of the surrogates, the sample is further treated at the discretion of the analyst, provided that the GC/EIMS response of the four surrogates meets the criteria listed in Section 7.0. The literature pertaining to these techniques has been reviewed.4 Several possible techniques are presented below for guidance only. The applicability of any of these techniques to a specific sample matrix must be determined by the precision and accuracy of the *3C PCB surrogate recoveries, as discussed in Section 14.2. 9.3.1 Extraction - The entire sample must be transferred to the extraction vessel with PCB-free water washing, if necessary, to transfer all solids. The container is then rinsed with the extraction solvent to recovery any PCBs adhering to' the container wall. The solvent rinses are combined with the extracts from below. Measure the sample volume to the nearest 0.5%. 9.3.1.1 Liquid-liquid extraction - The solvent, number of extractions, solvent-to-sample ratio, and other parameters are chosen at the analyst's discretion. A suggested extraction from water is presented in EPA Methods 60S1 and 625.2 9.3.1.2 Sorbent column extraction - PCBs may be isolated from water onto sorbent columns, although these techniques are not as widely used or thoroughly validated as liquid-liquid extraction. The selection of sorbent (XAD, Porapak, carbonpolyurethane foam, etc.) will depend on the nature of the matrix. The available methods have been reviewed.4 D-22 �9.3.2 Cleanup - Several tested cleanup techniques are described below. All but the base cleanup (9.3.2.8) were previously validated for PCBs in transformer fluids.5 Depending upon the complexity of the sample, one or more of the techniques may be required to fractionate the PCBs from interferences. For most cleanups a concentrated (1-5 ml) extract should be used. 9.3.2.1 Acid cleanup 9.3.2.1.1 Place 5 ml of concentrated sulfuric acid into a 40-ml narrow-mouth screwcap bottle. Add the sample extract. Seal the bottle with a Teflon-lined screw cap and shake for 1 min. 9.3.2.1.2 Allow the phases to separate, transfer the sample (upper phase) with three rinses of 1-2 ml solvent to a clean container and concentrate to an appropriate volume. 9.3.2.1.3 Analyze as described in Section 10.0. 9.3.2.1.4 If the sample is highly contaminated, a second or third acid cleanup may be employed. 9.3.2.2 Florisil column cleanup 9.3.2.2.1 Variations among batches of Florisil (PR grade or equivalent) may affect the elution volume of the various PCBs. For this reason, the volume of solvent required to completely elute all of the PCBs must be verified by the analyst. The weight of Florisil can then be adjusted accordingly. 9.3.2.2.2 Place a 20-g charge of Florisil, activated overnight at 130°C, into a Chromaflex column. Settle the Florisil by tapping the column. Add about 1 cm of anhydrous sodium sulfate to the top of the Florisil. Pre-elute the column with 70-80 ml of hexane. Just before the exposure of the sodium sulfate layer to air, stop the flow. Discard the eluate. D-23 �9.3.2.2.3 Add the sample extract to the column. 9.3.2.2.4 Carefully wash down the inner wall of the column with 5 ml of the hexane. 9.3.2.2.5 Add 220 ml of hexane to the column. 9.3.2.2.6 Discard the first 25 ml. 9.3.2.2.7 Collect 200 ml Kuderna-Danish PCBs should be Concentrate to of hexane eluate in a flask. All of the in this fraction. an appropriate volume. 9.3.2.2.8 Analyze the sample as described in Section 10.0. 9.3.2.3 Alumina column cleanup 9.3.2.3.1 Adjust the activity of the alumina (Fisher A540 or equivalent) by heating to 200°C for 2 to 4 hr. When cool, add 3% water (wt:wt) and mix until uniform. Store in a tightly sealed bottle. Allow the deactivated alumina to equilibrate at least 1/2 hr before use. Reactivate weekly. 9.3.2.3.2 Variations between batches of alumina may affect the elution volume of the various PCBs. For this reason, the volume of solvent required to completely elute all of the PCBs must be verified by the analyst. The weight of alumina can then be adjusted accordingly. 9.3.2.3.3 Place a 50-g charge of alumina into a Chromaflex column. Settle the alumina by tapping. Add about 1 cm of anhydrous sodium sulfate. Pre-elute the column with 70-80 ml of hexane. Just before exposure of the sodium sulfate layer to air, stop the flow. Discard the eluate. 9.3.2.3.4 Add the sample extract to the column. 9.3.2.3.5 D-24 Carefully wash down the inner wall of the column with 5 ml volume of hexane. �9.3.2.3.6 Add 295 ml of hexane to the column. 9.3.2.3.7 Discard the first 50 ml. 9.3.2.3.8 Collect 250 ml Kuderna-Danish PCBs should be Concentrate to of the hexane in a flask. All of the in this fraction. an appropriate volume. 9.3.2.3.9 Analyze the sample as described in Section 10.0. 9.3.2.4 Silica gel column cleanup 9.3.2.4.1 Activate silica gel (Davison grade 950 or equivalent) at 135°C overnight. 9.3.2.4.2 Variations between batches of silica gel may affect the elution volume of the various PCBs. For this reason, the volume of solvent required to completely elute all of the PCBs must be verified by the analyst. The weight of silica gel can then be adjusted accordingly. 9.3.2.4.3 Place a 25-g charge of activated silica gel into a Chromaflex column. Settle the silica gel by tapping the column. Add about 1 cm of anhydrous sodium sulfate to the top of the silica gel. 9.3.2.4.4 Pre-elute the column with 70-80 ml of hexane. Discard the eluate. Just before exposing the sodium sulfate layer to air, stop the flow. 9.3.2.4.5 Add the sample extract to the column. 9.3.2.4.6 Wash down the inner wall of the column with 5 ml of hexane. 9.3.2.4.7 Elute the PCBs with 195 ml of 10% diethyl ether in hexane (v:v). 9.3.2.4.8 Collect 200 ml Kuderna-Danish PCBs should be Concentrate to D-25 of the eluate in a flask. All of the in this fraction. an appropriate volume. �9.3.2.4.9 9.3.2.5 Analyze the sample according to Section 10.0. Gel permeation cleanup 9.3.2.5.1 Set up and calibrate the gel permeation chromatograph with an SX-3 column according to the Autoprep instruction manual. Use 15% methylene chloride in cyclohexane (v:v) as the mobile phase. 9.3.2.5.2 Inject 5.0 ml of the sample extract into the instrument. Collect the fraction containing the PCBs (see Autoprep operator's manual) in a Kuderna-Danish flask equipped with a 10-ml ampul. 9.3.2.5.3 Concentrate the PCB fraction to an appropriate volume. 9.3.2.5.4 Analyze as described in Section 10.0. 9.3.2.6 Acetonitrile partition 9.3.2.6.1 Place the sample extract into a 125-ml separatory funnel with enough hexane to bring the final volume to 15 ml. Extract the sample four times by shaking vigorously for 1 min with 30-ml portions of hexane-saturated acetonitrile. 9.3.2.6.2 Combine and transfer the acetonitrile phases to a 1-liter separatory funnel and add 650 ml of distilled water and 40 ml of saturated sodium chloride solution. Mix thoroughly for about 30 sec. Extract with two 100-ml portions of hexane by vigorously shaking about 15 sec. 9.3.2.6.3 Combine the hexane extracts in a 1-liter separatory funnel and wash with two 100-ml portions of distilled water. Discard the water layer and pour the hexane layer through a 8-10 cm anhydrous sodium sulfate column into a 500-ml Kuderna-Danish flask equipped with a 10-ml ampul. Rinse the separatory funnel and column with three 10-ml portions of hexane. D-26 �9.3.2.6.4 Concentrate the extracts to an appropriate volume. 9.3.2.6.5 Analyze as described in Section 10.0. 9.3.2.7 Florisil slurry cleanup 9.3.2.7.1 Place the sample extract into a 20-ml narrow-mouth screw-cap container. Add 0.25 g of Florisil (PR grade or equivalent). Seal with a Teflon-lined screw cap and shake for 1 min. 9.3.2.7.2 Allow the Florisil to settle; then decant the treated solution into a second container with rinsing. Concentrate the sample to an appropriate volume. Analyze as described in Section 10.0. 9.3.2.8 Base cleanup6 9.3.2.8.1 Quantitatively transfer the concentrated extract to a 125-ml extraction flask with the aid of several small portions of solvent. 9.3.2.8.2 Evaporate the extract just to dryness with a gentle stream of dry filtered nitrogen, and add 25 ml of 2.5% alcoholic KOH. 9.3.2.8.3 Add a boiling chip, put a water condenser in place, and allow the solution to reflux on a hot plate for 45 min. 9.8.2.8.4 After cooling, transfer the solution to a 250-ml separatory funnel with 25 ml of distilled water. 9.3.2.8.5 Rinse the extraction flask with 25 ml of hexane and add it to the separatory funnel. 9.3.2.8.6 D-27 Stopper the separatory funnel and shake vigorously for at least 1 min. Allow the layers to separate and transfer the lower aqueous phase to a second separatory funnel. �9.3.2.8.7 Extract the saponification solution with a second 25-ml portion of hexane. After the layers have separated, add the first hexane extract to the second separatory funnel and transfer the aqueous alcohol layer to the original separatory funnel. 9.3.2.8.8 Repeat the extraction with a third 25-ml portion of hexane. Discard the saponification solution, and combine the hexane extracts. 9.3.2.8.9 10.0 Concentrate the hexane layer to an appropriate volume and analyze according to Section 10.0. Gas Chrotnatographic/Electron Impact Mass Spectrometric Determination 10.1 Internal standard addition - An appropriate volume of the internal standard solution is pipetted into the sample. The final concentration of the internal standard must be in the working range of the calibration and well above the matrix background. The internal standard is thoroughly incorporated by mechanical agitation. Note: The volumetric measurement of the internal standard solution is critical to the overall method precision. The analyst must exercise caution that the volume is known to be ±1% or better. Where necessary, calibration of the pipet is recommended. 10.2 Tables 2, and 5 through 8 summarize the recommended operating conditions for analysis. Figure 1 presents an example of a chromatogram. 10.3 While the highest available chromatographic resolution is not a necessary objective of this protocol, good chromatographic performance is recommended. With the high resolution of CGC, the probability that the chromatographic peaks consist of single compounds is higher than with PGC. Thus, qualitative and quantitative data reduction should be more reliable. 10.4 After performance of the system has been certified for the day and all instrument conditions set according to Tables 2, and 5 through 8, inject an aliquot of the sample onto the GC column. If the response for any ion, including surrogates and internal standards, exceeds the working range of the system, dilute the sample and reanalyze. If the responses of surrogates, analyte, or internal standard are below the working range, recheck the system performance. If necessary, concentrate the sample and reanalyze. D-28 �tH CO cd C 0) 4J fi C Ol 0 Xi X* '""•*• i—4 f~! CJ x; x: o CJ CJ O r-l C cd M r-l O C 3 o o to rH 0 01 X! p. •H p. _r\ O M 0 •H Xi o M o rH Xi CJ O o a •rt ;s CN 1 "^ o (~) CN c <r XI O 0) *• Xi P. CO t-l o tH x: o rH iH ho rH -» ^» •* •nH G/ 5 VO 01 H 1 •- ^ C 01 X! P. <r -* 0* 01 r; 4J iH a s^ 1 1 a i C 01 X! P. •H 523 575 1 JL. GS3 10:0!) X! 11 - rH >. 0 x; o rH r^) 4-1 C 01 01 H XI 1 P. -H ^3 » 0 r-l " O " O ^ •H XI rH 01 pr* | Xi ~ o 4-1 01 0 o H 1 vo r-l O rH X! CJ •rH r-l H t Cd vO CO * if) 4-1 K C •s CO -3" 01 CN 1 CNI *. ~3~ CN - CN vO -H Xi O M M 0 rH Xi o t) <T X! P. P> •H X I rH v^- <U r*, P. -H XI 0 CO r-l C C Ol ~* .d rH O £>-. C ) ^ C 01 o x ; iH Xi CJ cfl cd O H2592. CN rH 0 CO iH 1 rH 5*"\ Xi CO O Vi ^0 p. ^ P . -H •H Q) XI 4-1 O cfl M 00 O O Xi 0 O a 1 rH .C rH Xi l-l r-l O 1 •" vo ** O CO CJ 0 3 CO 0) r! vo iO CJ - P. -H XI O M O 4J CJ vO vO rH O ^~- Ol O Ol Q cfl 4j in p. 01 *n ^ _" -* -d~ vO CO CO *v •• m *. «\ co co •• • X V CN ~3~ " CN »v CN *i CN -* co n ^ I /!v/U CNJ iiao I31G •t CN 1 "^ CN 07-1 CO -" 7-W V i—i rH cO r4 613 iii" Ji r-H(V,*< O O 0) M M 4-1 O O cfl tH rH 00 me. G 0) Xi P. •H XI 0 01 P. P. •H -H - M 1 rH C 01 Xi s>* rH ^ CO T3 1 tH rH o VO 0 CO tH 1 ^W -' vO r-l 3 V 198.0- oo 0 CN" RL C'59 13:20 [I 936 llfrl 1 fl | i JLJL_ IOSO 16:10 1178 l | |m J[ .L i l'J{-9 20: CD ' 1 I1';9 23:20 SCA'l Tllh: Figure 1. Capillary gas chromatography/electron impact ionization mass spectrometry (CGC/EIMS) chromatogram or the calibration standard solution required for quantitation of PCBs by homolog. This chromatogram includes PCBs representative of each homolog, three carbon-13 labeled surrogates, and the deuterated internal standard. The concentration of all components and the CGC/EIMS parameters are presented in Tables 3, 4, 5, and 7. �10.5 11.0 Record all data on a digital storage device (magnetic disk, tape, etc.) for qualitative and quantitative data reduction as discussed below. Qualitative Identification 11.1 Selected ion monitoring (SIM) or limited mass scan (IMS) data The identification of a compound as a given PCB homolog requires that two criteria be met: 11.1.1 11.1.2 11.2 (1) The peak must elute within the retention time window set for that homolog (Section 7.5); and (2) the ratio of two ions obtained by SIM (Table 11) or by LMS (Table 12) must match the natural ratio within ±20%. The analyst must search the higher mass windows, in particular M+70, to prevent misidentification of a PCB fragment ion cluster as the parent. If one or the other of these criteria is not met, interferences may have affected the results and a reanalysis using full scan EIMS conditions is recommended. Full scan data 11.2.1 11.2.2 12.0 The unknown spectrum must match that of an authentic PCB. The intensity of the three largest ions in the molecular cluster (two largest for monochlorobiphenyls) must match the natural ratio within ±20%. Fragment clusters with proper intensity ratios must also be present. 11.2.3 11.3 The peak must elute within the retention time windows set for that homolog (as described in Section 7.5). Alternatively, a spectral search may be used to automatically reduce the data. The criteria for acceptable identification include a high index of similarity. For the Incos 2300, a fit of 750 or greater must be obtained. Disputes in interpretation - Where there is reasonable doubt as to the identity of a peak as a PCB, the analyst must either identify the peak as a PCB or proceed to a confirmational analysis (see Section 13.0). Quantitative Data Reduction 12.1 Once a chromatographic peak has been identified as a PCB, the compound is quantitated based either on the integrated abundance of the SIM data or EICP for the primary characteristic ion in Tables 11 and 12. If interferences are observed for the primary ion, D-30 �TABLE 11. CHARACTERISTIC SIM IONS FOR PCBs Ion (relative intensity) Tertiary Secondary Homolog Primary Cj^HgCl 188 (100) 190 (33) - Ci2HgCl2 222 (100) 224 (66) 226 (11) C^HrCls 256 (100) 258 (99) 260 (33) Ci2HeCl4 292 (100) 290 (76) 294 (49) C12H5C15 326 (100) 328 (66) 324 (61) c i2H4Cle 360 (100) 362 (82) 364 (36) C12H3C17 394 (100) 396 (98) 398 (54) Ci^HfcClg 430 (100) 432 (66) 428 (87) Ci2HCl9 464 (100) 466 (76) 462 (76) CiaCljo 498 (100) 500 (87) 496 (68) Source: Rote, J. W., and W. J. Morris, "Use of Isotopic Abundance Ratios in Identification of Polychlorinated Biphenyls by Mass Spectrometry," J. Assoc. Offic. Anal. Chem., 56(1), 188-199 (1973). D-31 �TABLE 12. LIMITED MASS SCANNING (LMS) RANGES FOR PCBs Compound Mass range (m/z) Ci2H9Cli 186-190 Ci2H8Cl2 220-226 Cl2H?Cl3 254-260 Ci2H6Cl3 288-294 C12H5C15 322-328 C12H4C16 356-364 C12H3C17 386-400 Ci2H2Clg 426-434 Cl2HClg 460-468 Ci2Clio 494-504 C12D6C14 294-300 13 192-196 C612CeH9Cl 13 300-306 13 C12H2C18 438-446 13 C12Clio 506-516 C12H6C14 a Adapted from Tindall, G. W., and P. E. Wininger, "Gas Chromatography-Mass Spectrometry Method for Identifying and Determining Polychlorinated Biphenyls," J. Chromatogr., 196, 109-119 (1980). D-32 �use the secondary and then tertiary ion for quantitation. If interferences in the parent cluster prevent quantitation, an ion from a fragment cluster (e.g., M-70) may be used. Whichever ion is used, the RF must be determined using that ion. The same criteria should be applied to the surrogate compounds (Table 13). 12.2 Using the appropriate analyte-internal standard pair and response factor (RF ) as determined in Section 7.3, calculate the concentration ofpeach peak using Equation 12-1. A M. V Eq 12-1 Concentration (Mg/g) = ^ ' RF ' M^ ' \T ' is p e i where A = area of the characteristic ion for the analyte PCB peak A. = area of the characteristic ion for the internal standard peak RF = response factor of a given PCB congener M. = mass of internal standard injected (micrograms) 1S M = mass of sample extracted (grams) V. = volume injected (microliters) V 12.3 = volume of sample extract (microliters) If a peak appears to contain non-PCB interferences which cannot be circumvented by a secondary or tertiary ion, either: 12.3.1 12.3.2 Perform additional chemical cleanup (Section 9) and then reanalyze the sample; or 12.3.3 12.4 Reanalyze the sample on a different column which separates the PCB and interferents; Quantitate the entire peak as PCB. Calculate the recovery of the four 13C surrogates using the appropriate surrogate-internal standard pair and response factor (RF. & as determined in Section 7.4 using Equation 12-2. ) 1 A M. Recovery ( ) = j*- - ^- • ^ • 100 % Eq. 12-2 is s s where A S = area of the characteristic ion for the surrogate peak A. = area . the characteristic ion for the internal standard of IS peak D-33 �TABLE 13. CHARACTERISTIC IONS FOR 13C-LABELED PCS SURROGATES Primary Ion (relative intensity) Secondary 13 194 (100) 196 (33) l3 304 (100) 306 (49) 302 (78) 13 C12H2C18 442 (100) 444 (65) 440 (89) 13 Ci2Clio 510 (100) 512 (87) 514 (50) Specific compound C612C6H9C1 Ci2H6Cl4 D-34 Tertiary �RF = response factor for the surrogate compound with respect to the internal standard (Equation 7-2) M. = mass of internal standard injected (nanograms) is M s = mass of surrogate, assuming 100% recovery (nanograms) 12.5 Correct the concentration of each peak using Equation 12-3. is the final reportable concentration. Corrected concentration (pg/g) = Concentration ug/g .10Q ^re>i */ 12.6 Recovery ( ) % £ This _3 ^ 12 Sum all of the peaks for each homolog, and then sum those to yield the total PCB concentration in the sample. Report all numbers in |Jg/g. The reporting form in Table 14 may be used. If an alternate reporting format (e.g., concentration per peak) is desired, a different report form may be used. The uncorrected concentrations, percent recovery, and corrected recovery are to be reported. 12.7 Round off all numbers reported to two significant figures. 13.0 Confirmation If there is reason to question the qualitative identification (Section 11.0), the analyst may choose to confirm that a peak is not a PCB. Any technique may be chosen provided that it is validated as having equivalent or superior selectivity and sensitivity to GC/EIMS. Some candidate techniques include alternate GC columns (with EIMS detection), GC/CIMS, GC/NCIMS, high resolution EIMS, and MS/MS techniques. Each laboratory must validate confirmation techniques to show equivalent or superior selectivity between PCBs and interferences and sensitivity (limit of quantitation, LOQ). If a peak is confirmed as being a non-PCB, it may be deleted from the calculation (Section 12). If a peak is confirmed as containing both PCB and non-PCB components, it must be quantitated according to Section 12,3. 14.0 Quality Control 14.1 Each laboratory that uses this method must operate a formal quality control (QC) program. The minimum requirements of this program consist of an initial demonstration of laboratory capability and the analysis of spiked samples as a continuing check on performance. The laboratory must maintain performance records to define the quality of data that are generated. After a date specified by the Agency, ongoing performance checks should be compared with established performance criteria to determine if the results of analyses are within accuracy and precision limits expected of the method. D-35 �TABLE 14. ANALYSIS REPORT INCIDENTAL PCBs IN WASTEWATER Sample No. Sample Matrix Sample Source Notebook No. or File Location Volume Extracted Extraction/Cleanup Procedure Int. Std. liter Mass Added (|Jg) (Circle one) 4-Cl(d6) Surrogates 298 Mass Added ()Jg) 300 Intensity 100/49 (Circle one) Ratio 1-C1 194 196 100/33 4-C1 304 306 100/49 8-C1 442 444 100/65 10-C1 510 512 100/87 (continued) D-36 Ratio Intensity % Recovery �TABLE 14 (continued) Qualitative I Analyte 1° 2° l° 1-C1 188 190 100/33 2-C1 222 224 100/66 3-C1 256 258 100/99 4-C1 292 290 100/76 5-C1 326 328 100/66 6-C1 360 362 100/82 7-C1 394 396 100/98 8-C1 430 432 100/66 9-C1 464 466 100/76 10-C1 498 500 Quantitative Uncorr Corr Ion Cone. Cone OK? Used RF (pg/4) (M8/4) 100/87 2° Ratio Theoretical Total MS/4 Uncorr. Reported by: Internal Audit: Name Name EPA Audit: Name Signature/Date Signature/Date Signature/Date Organization Organization Organization D-37 M8/4 Corr. �14.2 The analysts must certify that the precision and accuracy of the analytical results are acceptable by: 14.2.1 14.2.2 14.3 The absolute precision of surrogate recovery, measured as the RSD of the integrated EIMS area (Ag) for a set of samples, must be ±10%. The mean recovery (R ) of at least four replicates of a QC check sample to be supplied by the Agency must meet Agency-specified accuracy and precision criteria. This forms the initial data base for establishing control limits (see Section 14.3 below). Control limits - The laboratory must establish control limits using the following equations: Upper control limit (UCL) = RL. + 3 RSD V_ Upper warning limit (UWL) = R + 2 RSD Lower warning limit (LWL) = R - 2 RSD Lower control limit (LCL) = R - 3 RSD These may be plotted on control charts. If an analysis of a check sample falls outside the warning limits, the analyst should be alerted that potential problems may need correction. If the results for a check sample fall outside the control limits, the laboratory must take corrective action and recertify the performance (Section 14.2) before proceeding with analyses. The warning and control limits should be continuously updated as more check sample replicates are added to the data base. 14.4 Before processing any samples, the analyst should demonstrate through the analysis of a reagent blank that all glassware and reagent interferences are under control. Each time a set of samples is analyzed or there is a change in reagents, a laboratory reagent blank should be processed as a safeguard against contamination. 14.5 Procedural QC - The various steps of the analytical procedure should have quality control measures. These include but are not limited to: 14.5.1 GC performance - See Section 7.1 for performance criteria. 14.5.2 MS performance - See Section 7.2 for performance criteria. D-38 �14.5.3 Qualitative identification - At least 10% of the PCB identifications, as well as any questionable results, should be confirmed by a second mass spectrometrist. 14.5.4 Quantitation - At least 10% of all manual calculations, including peak area calculations, must be checked. After changes in computer quantitation routines, the results should be manually checked. 14.6 A minimum of 10% of all samples, one sample per month or one sample per matrix type, whichever is greater, selected at random, must be run in triplicate to monitor the precision of the analysis. An RSD of ±30% or less must be achieved. If the precision is greater than ±30%, the analyst must be recertified (see Section 14.2). 14.7 A minimum of 10% of all samples, one sample per month or one sample per matrix type, whichever is greater, selected at random, must be analyzed by the standard addition technique. Two aliquots of the sample are analyzed, one "as is" and one spiked (surrogate spiking and equilibration techniques are described in Section 9.2) with a sufficient amount of Solution CSxxx to yield approximately 100 (Jg/liter of each compound). The samples are analyzed together and the quantitative results calculated. The recovery of the spiked compounds (calculated by difference) must be 80-120%. If the sample is known to contain specific PCB isomers, these isomers may be substituted for solution CSxxx. If the concentrations of PCBs are known to be high or low, the amount added should be adjusted so that the spiking level is 1.5 to 4 times the measured PCB level in the unspiked sample. 14.8 Interlaboratory comparison - Interlaboratory comparison studies are planned. Participation requirements, level of performance, and the identity of the coordinating laboratory will be presented in later revisions. 14.9 It is recommended that the participating laboratory adopt additional QC practices for use with this method. The specific practices that are most productive depend upon the needs of the laboratory and the nature of the samples. Field duplicates or triplicates may be analyzed to monitor the precision of the sampling technique. Whenever possible, the laboratory should perform analysis of standard reference materials and participate in relevant performance evaluation studies. 15.0 Quality Assurance Each participating laboratory must develop a quality assurance plan according to EPA guidelines.7 The quality assurance plan must be submitted to the Agency for approval. D-39 �16.0 Method Performance The method performance is being evaluated. Limits of quantitation; average intralaboratory recoveries, precision, and accuracy; and interlaboratory recoveries, precision, and accuracy will be presented. 17.0 Documentation and Records Each laboratory is responsible for maintaining full records of the analysis. Laboratory notebooks should be used for handwritten records. GC/MS data must be archived on magnetic tape, disk, or a similar device. Hard copy printouts may be kept in addition if desired. QC records should be maintained separately from sample analysis records. The documentation must describe completely how the analysis was performed. Any variances from the protocol must be noted and fully described. Where the protocol lists options (e.g., sample cleanup), the option used and specifics (solvent volumes, digestion times, etc.) must be stated. D-40 �REFERENCES 1. Environmental Protection Agency, Organochlorine Pesticides and PCBs— Method 608," Fed. Reg., 44, 69501-69509 (December 3, 1979). 2. Environmental Protection Agency, "Base/Neutrals, Acids, and Pesticides— Method 625," Fed. Reg., 44, 69540-69552 (December 3, 1979), and subsequent revisions. 3. "Methods 330.4 (Titrimetric, DPD-FAS) and 330.5 (Spectrophotometric, DPD) for Chlorine, Total Residual," Methods for Chemical Analysis of Water and Wastes, U.S. Environmental Protection Agency, Environmental Monitoring and Support Laboratory, Cincinnati, Ohio, March 1979, EPA 600-4/79-020. 4. Erickson, M. D., and J. S. Stanley, "Methods of Analysis for Incidentally Generated PCBs Literature Review and Preliminary Recommendations," Interim Report No. 1, EPA Contract No. 68-01-5915, Task 51, 1982. 5. Bellar, T. A., and J. J. Lichtenberg, "The Determination of Polychlorinated Biphenyls in Transformer Fluid and Waste Oils," Prepared for U.S. Environmental Protection Agency (1981). EPA-600/4-81-045. 6. American Society for Testing and Materials, "Standard Method for Analysis of Environmental Materials for Polychlorinated Biphenyls," pp. 877-885, in Annual Book of ASTM Standards, Part 40, Philadelphia, Pennsylvania (1980). ANSI/ASTM D 3304-77. 7. Quality Assurance Program Plan for the Office of Toxic Substances, Office of Pesticides and Toxic Substances, U.S. Environmental Protection Agency, Washington, D.C., October 1980. D-41 �TECHNICAL REPORT DATA (Please read Instructions on the reverse before completing) 3. RECIPIENT'S ACCESSION NO. 2. 1. REPORT NO. EPA-560/5-82-006 4. TITLE AND SUBTITLE 5. REPORT DATE Analytical Methods for By-Product PCBs—Initial Validation and Interim Protocols 6. PERFORMING ORGANIZATION CODE October 11, 1982 7. AUTHORIS) Mitchell D. Erickson, John S. Stanley, Gil Radolovich, Kay Turman, Karin Bauer, Jon Onstot, Donna Rose, and Margaret Wickham 8. PERFORMING ORGANIZATION REPORT NO. 9. PERFORMING ORGANIZATION NAME AND ADDRESS 10. P R O G R A M E L E M E N T NO. Midwest Research Institute 425 Volker Boulevard Kansas City, MO 64110 11. CONTRACT/GRANT NO. 12. SPONSORING AGENCY NAME AND ADDRESS 13. TYPE OF REPORT AND PERIOD COVERED MRI Project No. 4901-A51 EPA 68-01-5915, Task 51 U.S. Environmental Protection Agency Office of Toxic Substances, Field Studies Branch TS-798 Washington, DC 20460 Interim 4, 4/24-8/31/82 14. SPONSORING AGENCY CODE 15. SUPPLEMENTARY NOTES The task manager is David P. Redford; the project officer is Frederick W. Kutz. 16. A B S T R A C T This document presents proposed analytical methods for analysis of by-product PCBs in commercial products, product waste streams, wastewaters, and air. The analytical method for commercial products and product waste streams consist of a flexible approach for extraction and cleanup of particular matrices. The 13c-labeled PCB surrogates are added as part of a strong quality assurance program to determine levels of recovery. The wastewater method is based on EPA Methods 608 and 625 with revisions to include use of the 13c-labeled PCB surrogates. The air method is a revision of a proposed EPA method for the collection and analysis of PCBs in air and flue gas emissions. Capillary or packed column gas chromatography/electron impact ionization mass spectrometry is proposed as the primary instrumental method. Response factors and retention times of 77 PCB congeners relative to tetrachlorobiphenyl-d6 are presented in addition to statistical analysis to project validity of the data and extrapolation of relative response factors to all 209 possible congeners. Preliminary studies using the ISClabeled surrogates to validate specific cleanup procedures and to analyze several commercial products and product wastes indicate that the proposed analytical methods are both feasible and practical. KEY WORDS AND DOCUMENT ANALYSIS 17. DESCRIPTORS Polychlorinated biphenyls PCBs Incidentally generated Analytical protocols Air Wastewater Commercial products b.IDENTIFIERS/OPEN ENDED TERMS Commercial waste Capillary column Electron impact EIMS Response factors Relative retenti Surrogates 18. DISTRIBUTION STATEMENT EPA Form 2220-1 (R»v. 4-77) Unclassified PREVIOUS EDITION is OBSOLETE COSATI Field/Group relative response factor n times 19. SECURITY CLASS (This Report) Unclassified 20. SECURITY CLASS (This page) Unlimited c. streams gas chromatography onization mass spectromet 21. NO. OF PAGES 243 22. PRICE �United States Environmental Protection Agency Office of Toxic Substances Washington DC 20460 � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Alvin L. Young Collection on Agent Orange Description An account of the resource <p style="margin-top: -1em; line-height: 1.2em;">The Alvin L. Young Collection on Agent Orange comprises 120 linear feet and spans the late 1800s to 2005; however, the bulk of the coverage is from the 1960s to the 1980s and there are many undated items. The collection was donated to Special Collections of the National Agricultural Library in 1985 by Dr. Alvin L. Young (1942- ). Dr. Young developed the collection as he conducted extensive research on the military defoliant Agent Orange. The collection is in good condition and includes letters, memoranda, books, reports, press releases, journal and newspaper clippings, field logs and notebooks, newsletters, maps, booklets and pamphlets, photographs, memorabilia, and audiotapes of an interview with Dr. Young.</p> <p>For more about this collection, <a href="/exhibits/speccoll/exhibits/show/alvin-l--young-collection-on-a">view the Agent Orange Exhibit.</a></p> Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Box The box containing the original item. 186 Folder The folder containing the original item. 5385 Series The series number of the original item. Series VIII Subseries I Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Creator An entity primarily responsible for making the resource Erickson, Mitchell D. John S. Stanley Kay Turman Gil Radolovich Karin Bauer Jon Onstot Donna Rose Margaret Wickham Description An account of the resource <strong>Corporate Author: </strong>Midwest Research Institute Date A point or period of time associated with an event in the lifecycle of the resource October 11 1982 Title A name given to the resource Analytical Methods for By-Product PCBs - Preliminary Validation and Interim Methods ao_seriesVIII https://www.nal.usda.gov/exhibits/speccoll/files/original/ac4ce6a56fe2889670fbb8a6fecd2a5a.pdf 07cce4ca40a0a0c75364b6e4db961875 PDF Text Text ItemD Number °5495 D (Jot Scanned Author Corporate Author The President's Commission on Industrial Competitiven Report/Article Tltlfl Appendix C: Balancing Environmental, Health, and Safety Regulation with the Needs of Research, Development, and Technological Innovation, A Special report on the Need for Further Regulatory Review Journal/Book Title Year Month/Day Color Number of Images December? D ° About 1/4 of page 279 is missing. Friday, March 15, 2002 Page 5495 of 5571 �APPENDIX C BALANCING ENVIRONMENTAL, HEALTH. AMD SAFETY REGULATION WITH MB MEEDS OP RESEARCH, DEVELOPMENT. AND TECHNOLOGICAL INNOVATION A SPECIAL REPORT ON THE NEED FOR FURTHER REGULATORY REVIEW The President's Commission on Industrial Competitiveness Committee on Research, Development, and Manufacturing December 7, 1984 �INTRODUCTION The need to regulate new technologies and products is reflected in the explosive growth in Federal health, safety, and environmental regulations over the last three decades.1 Regulation is certainly legitimate and necessary to protect public interests. With the extensive growth of regulation, however, every major industry now faces thousands of social regulations, many of which unnecessarily impede the competitiveness of American firms, especially in high-technology sectors of the economy. It is therefore worryisome that regulation now threatens to slow the emergence of entirely new industries. Clearly, societal benefits arise from governmental regulation, but in recent years a large number of well-documented economic and policy studies point to the fact that in all too many cases the costs to society of so much regulation, if imprudently conceived and implemented, could far exceed the apparent benefits. Many of these studies also demonstrate that regulation can have a powerful negative impact on technological innovation. For example, the Advisory Committee on Industrial Innovation, U.S. Department of Commerce, found: There is no doubt in impacted industries that regulation has a serious negative effect on industrial innovation and on productivity and contributes to inflation.2 And a National Academy of Sciences study reports that: . . . the connection between regulation and innovation is found in the costs entailed in meeting regulatory requirements, as these reduce the availability of R&D funds for innovative new products, the capital available for new plants to manufacture such products, or the competitiveness of the products in the U.S. world markets.3 Industrial innovation is crucial to the continued vitality of the U.S. economy and our ability to compete in world markets. The United States must therefore safeguard its comparative advantage in high technology if our Nation is to remain a world leader. In 1980, high-technology trade contributed a large surplus to the U.S. balance of trade—approaching $40 billion.4 Nine out of the ten fastest growing industries are technology-intensive. Moreover, real output growth for high-technology industries is more than double that for all businesses in the United States.5 Poorly conceived regulation, however, poses a potential threat to our continued leadership. �Regulation is a significant factor, but not the only factor, affecting innovation and product development. Economic factors, such as market size, and other factors also have important roles. The negative impact of regulation on product research, development, and technological innovation is particularly apparent for the high technology industries discussed in this paper (see exhibit 1). They illustrate how industrial innovation is adversely affected by health, environmental, and safety regulation. The impact is felt not only in these industries, but throughout our economy. Increased business uncertainty, delays in research and development, decline in product approvals, slower productivity growth, and the diversion of capital resources put U.S. industry at a competitive disadvantage. Thus, infeasible regulations not only erode our technological lead but result in fewer jobs for U.S. workers, and delay the availability of new products for consumers. Ultimately, the issue is not whether regulations should exist. Their benefits are clear and the public supports the concept of regulation. The public has the right to be informed on the level of risk relating to new products. Workers have the right to a safe work place or at least to an informed opinion as to the associated risk. The Government's role in securing their rights through the regulatory process is not in issue. However, 1 L there are many unnecessary or unintended regulatory constraints on inno- | vation, research, development, and product approval which should be eliminated. The problem of excessive regulation has developed primarily from regulatory mandates which do not call for a balance between competing national public policy objectives. Specifically, issues of international competitiveness and technological innovation have been neglected in the regulatory process. As a result, the mature regulatory process is excessively devoted to a single purpose and often regulates substances and conditions that it should not. Also the regulator should be dependent on good technological advice. This advice, however, often can constitute nothing more than uninformed opinion. If a reliable source of advice is not available, or if the regulatory mandate does not require or permit an appropriate balancing of interests, the inevitable result is overregulation. Concern about overregulation is not new. A national consensus has emerged that inept Federal regulation in too many instances has become more of a hindrance to progress than a solution to the problems it addresses.6 In fact, the last four Administrations have reviewed the problem and taken actions to ease specific economic regulations. The pace of new regulations has slowed. Little has been achieved, however, in the review of existing health, safety, and environmental regulations as they affect the innovation of new products. It is, therefore, critically important that the new Administration place this issue high on its agenda in 1985. The following discussion examines the impact of health, safety, and environmental regulations on product innovation, research, development, and product approval. The findings and recommendations are limited to the impact of regulation on these narrow but highly important areas. They also do not address the issue of regulatory codes that other countries use to achieve similar objectives. Since many regulations adversely impact innovation and competitiveness, the challenge is to identify those that are unnecessary. A regulation-by-regulation analysis is beyond the scope of this paper, but the examples presented here indicate that such a review is 280 �needed. Regulations should be applied with skill and reasonableness. The I major recommendation in this paper calls for an expansion of the role of the | Office of Science and Technology Policy in the regulatory process to take actions in the regulatory process in balancing the need to enhance industrial innovation and competitiveness, while continuing to protect public health, safety, and the environment. THE IMPACT OF HEALTH, SAFETY, AND ENVIRONMENTAL REGULATION ON SELECTED IMPORTANT INDUSTRIES The relationship between regulation and innovation is complex and varies from industry to industry. A summary of current U.S. regulatory policy in selected industries highlights the damage done to technological innovation and competitiveness (see also exhibit 1). Chemicals, pesticides, and pharmaceuticals have been under mature and intense regulation for many years. The ever-growing regulatory barriers faced by these industries are indicative of problems shared by many U.S. businesses. They also reveal what the future may hold for two other highly innovative industries—medical devices and biotechnology—if Government regulation follows the same policies and 1 practices of overregulation and attempts to protect against all possible | risks as in the past. All point to the need for a more balanced regulatory process, which also considers the need to enhance research and technological innovation. REGULATION OF BIOTECHNOLOGY Biotechnology is one of today's most promising new industries, with the market for genetically engineered products expected to leap from less than $100 million this year to tens of billions of dollars annually during the 1990's.^ The United States has taken the lead in the worldwide race to develop new genetically engineered products. This technology promises new advances ranging from disease-resistant plants to microbes that clean up oil spills. Because only one genetically engineered product—human insulin—has been developed to the point where it is sold on the market, the regulation of biotechnology is still in its incipiency. The specter of overregulation, however, is threatening to stifle innovation and affect strategic decisions by U.S. firms. Therefore, regulation could have a major impact on the ability of the United States to compete in this infant industry. For the past 8 years, most Federal Government oversight of genetic engineering has been exercised by the Recombinant DNA Advisory Committee (RAC) of the National institutes of Health (NIH). RAC's purpose is to review proposals for publicly funded research and construct guidelines for the safe handling of recombinant DNA material in the laboratory. It is believed that private companies have also voluntarily complied with NIH guidelines and both public and private researchers have made requests to the RAC to release genetically altered organisms into the environment for field testing. To date, two regulatory issues threaten the United States' preeminence in this field: (1) export restrictions that could severely limit the ability of American firms to market products abroad, and (2) the turf fight among several Federal regulatory agencies for control over biotechnology. 281 �EXHIBIT 1 Impact of Regulation on Research, Development, and Innovation in Selected Important Industries .Mucky Biotechnology Msdleal Devices Value* (to millions) toss than $100* Principal Regulations/Regulator NIH— research regulations EPA —proposing regulation* FDA —drug approval regulations USOA Antitrust/Export Laws SS.OOO regulations UUC__naft M*Mil»MnM Chemical EPA—TSCA $16.000 Pharmaceutical $142,000 EPA-FIFRA $22,000 FOA—drug approval process Export Laws Impact HurisdMional squabble over which agency wM regulate research and commercialization -inability to export products licensed abroad but not in U.S —uncertain effect of antitrust law* on joint RAO EHect • • Hiding product development, marketing and commercialization •— ttv •*( of kX6*Qft oofiapMNion In roMfch And oomimfci nN* zatton —increasing degree of regulation as FDA continue* to issue regulation* under Medical Device —when fuNy implemented, impacl on innovation la likely to be severe because economiM ol —increasing impacl on use as HUG implement* DflGs —export restrictions —70-90% reduction in rate of new product innovation since TSCA' —SS.000-12.000 cos) ol Ming a premanufacture notification'' —over $20 million cost in developing a new pesticide* —14-22 year delay from discovery to full production* —$70 million in R4D costs per marketed new drug1 —since 1962 doubling of R»D investment required for a new drug" —greatly increased time from discovery to marketing these specialty product* —rising R»0 costs —greater competition from abroad —small manufacturers hurl —rising RAO costs —declining rate of new product introductions -rising RU> costs -declining effective patent Me ol new drug —U.S. lags behind other countries in approving new drugs for -"drug lag" Value ol industry shtoments tor 1961 (in miMons of dollars). 1981 Annual Survey of Manufacturers. Bureau of Census RepnMents estimated sale*. "Biotech: WM the United States Loce Its Edge?". Dun'* Btainta Month, August 1984 OTA. rechnofeoica/ fcnevafon and Httlth, Safety, antf QiWronmenfa/ AeguMftyi. Vtll-53 (1981) Davies, "The Effects of Federal Regulation on Chemical Industry Innovation". 46 Law and Contemporary ftobfems SI (Summer 1983). Chemical Manufacturers Association. "Preserving Innovation Under the Toxic Substances Control Act," p. 18 (January 20.1982) flucfcefenau* v. Monsanro, 52 U.S.L.W. 4886.4889 (June 26.1964) OTA. COmmercW Stortcnnotooyr An Mematfonar Anaysfe 361 (1984) Statman. CompeMkm in (he miarmaceuticaf Industry: Tht Declining ProftfaMily of drug Innovation (American Enterprise Institute 1983) 282 �Reacting to congressional concern that deliberate release of modified organisms could pose risks to the environment, Federal agencies have begun to formulate a comprehensive plan to regulate biotechnology. ..under the recent direction of the White House Office of Science and Technology Policy, agencies, including Environmental Protection Agency, NIH, U.S. Department of Agriculture, Food and Drug Administration, and Occupational Safety and Health Administration, have met to clarify the regulatory path that a company would follow to meet Federal health and safety requirements. EPA is also moving to regulate biotechnology by changing the review procedures for pesticides and toxic chemicals. Mew regulations would no longer exempt living microorganisms from the current regulations for field tests of small quantities of chemicals used solely for R&D.& These proposed rules have the potential for becoming inflexible and bureaucratized, as discussed in the later sections on chemicals and pesticides. Improperly conceived and administered, they would have a high probability of delaying or altogether extinguishing the research efforts needed to maintain biotechnology innovation. Lengthy court battles by opponents of genetic engineering over the interpretation of regulations could also stifle innovation. This spring, with the scientific community's overwhelming support for the experiment, scientists planned to field-test a modified bacterium that could save farmers as much as $1.5 billion a year by protecting crops from frost damage.9 Environmental activists, however, obtained an injunction to stop the scientific experiments already approved by the NIH. Now results from this key experiment will be delayed at least a year. In order to maintain U.S. competitiveness in biotechnology, Federal agencies must work to develop a coherent interagency strategy for clear and expeditious regulation based on rigorous scientific data. Agencies must avoid a patchwork of conflicting, overlapping, and unnecessary rules. Awkwardly conceived and poorly managed regulation would have a particularly damaging effect on smaller biotechnology firms, which are leaders in innovation but lack the financial resources to contend with governmental regulation. Unilateral regulation could lead to the removal of research and testing facilities to countries where oversight is less stringent and damage the fragile lead the United States has over its competitors in the global biotechnology race. REGULATION OF MEDICAL DEVICES The medical device industry is affected primarily by the 1976 Medical Device Amendments, which extended FDA's regulatory control over a large spectrum of medical products, and by the Department of Health and Human Services (HHS) new regulations for Diagnostic Related Groups (DRG) aimed at containing health care costs. The device industry is currently facing a number of new FDA regulatory pressures. Many regulations required to implement the 1976 Medical Device Amendments are still pending. With increased pressure coming from Congress, consumer groups, and Federal agencies such as the General Accounting Office, several major regulations are expected to be implemented in the next year. Industry observers fear that medical device innovation and technology will 283 �be affected by excessively cumbersome regulations that will attempt to control the industry along lines similar to FDA regulation of Pharmaceuticals (discussed below). Furthermore, a study from the National Academy of Sciences concludes: If the FDA brings a "safety imperative* regulatory philosophy to bear on this sector similar to that which it has exhibited in Pharmaceuticals, the costs in foregone innovation are likely to be quite high indeed. This is particularly so because innovation in many medical device fields (such as heart pacemakers) has not been characterized by large economies of scale. Several major new products have emanated from small firms. Such firms would be least able to finance or bear the costs and risks of an expensive, lengthy, and uncertain premarket regulatory approval process.1° The 1976 Medical Device Amendments also impose new requirements regarding the export of unapproved devices. These controls pose problems for medical innovation and technological progess in this industry. Delays for obtaining export permits act as a disincentive to trade. The emphasis on health care cost containment has led to the recent enactment of new DRG regulations. Whereas the DRG's have a useful purpose in promoting competition in hospital services, they are also forming a potential barrier to the development and dissemination of new medical devices. Cost constraints tend to create disincentives for innovation in virtually all medical technologies, .regardless of their cost. This difficulty is illustrated by examining the prospective payment system under the Department of Health and Human Services. Prospective payment is an attempt to force cost-conscious management upon hospitals by capping overall reimbursement. Based on the illness diagnosed, hospitals will be paid a fixed rate per patient admitted. In contrast to the "fee for service" system of payment, prospective payment gives hospital administrators a vested interest in controlling all costs. If costs are higher than the specified reimbursement, the hospitals will incur a "loss." Under prospective payment, health care technology will have to compete for hospital resources with salaries, malpractice insurance, energy, cleaning supplies, and the like. Moreover, some of these costs, such as salaries and wages, are responsive to vocal constituencies well entrenched in the hospital. In contrast, innovative technology, being new, is unlikely to have many effective champions to press its case. Furthermore, medical technology represents a very small part of total hospital costs. In 1981 drugs constituted only 2.6 percent of these costs, and surgical and medical instruments and supplies were 2.1 percent. By contrast, wages and salaries were the largest item at 56.6 percent and employee benefits were next with 8.2 percent.11 The high-quality, leading-edge technology of U.S. medical devices should permit continuation of innovation and export growth, unless overregulation in the United States restricts opportunities for expansion. Recent regulatory initiatives for the industry, however, seem likely to have a negative impact on future medical innovation, as they have had on chemical and 284 �pharmaceutical innovation. Unless the DRG regulations are changed, industry observers think that the development of many useful new technologies will be slowed or stopped. Therefore, allowances for medical innovation should be made within the reimbursement system to recognize technology's pivotal importance in the future health care system of this country. Innovation in health care is of vital importance not only for our technological prowess but for the standard of living and quality of life as well. REGULATION OF CHEMICALS The chemical industry is affected by an array of Federal regulations, based on the Clean Air and Clean Water Acts, the Worker Safety Act, Resource Conservation and Recovery Act, and perhaps most important, the Toxic Substances Control Act (TSCA). This act requires manufacturers of all new chemical substances (not already regulated as drugs or pesticides) to give notification to the Environmental Protection Agency 90 days in advance of their first manufacture. Prior to marketing, EPA can then require manufacturers to test any substances deemed to have potentially unreasonable risks to health or the environment, or for which significant human or environmental exposure may occur. In recent years at least four studies—ADL,12 ICF,13 Regulatory Research Service for CMA,14 and CSMA15—have concluded that the review mechanism and testing requirements under TSCA have an adverse impact on innovation in the chemical industry. The costs imposed by these acts—including business uncertainties about regulatory actions, required paperwork, and additional testing—have had adverse effects on the chemical industry, particularly on innovation and the introduction of new products. Prior to the passage of TSCA and the introduction of complex screening requirements, chemical companies were developing and introducing between 1,000 and 2,200 new chemicals annually. By 1981, however, new chemical introductions, as measured by premanufacture notifications, had plummeted to only 627, a decline of as much as 72 percent.16 The Office of Technology Assessment has also reported that the Toxic Substances Control Act "has already reduced the rate of new product introductions by 70 to 90 percent."^ Industry studies report that R&D activity has become less innovative, with over 10 percent of R&D budgets directed at environmental and health activity.IB Emphasis is being placed on peripheral improvements in product lines, to the detriment of developing new ones. In the chemical industry, as in others, technological progress is usually incremental, consisting of small improvements over time—improvements now being made less frequently. Classes of chemicals are being disregarded totally if their molecular structure presents the possibility of incurring rigorous regulatory examination.19 The decrease in R&D discourages the development of new chemicals with more favorable risk benefits than those currently on the market. Increased costs, which manifest themselves in many ways, put innovative products at a severe disadvantage when competing with existing products. Where these trends are caused by unnecessary regulation, they pose potentially long-term problems and strong disincentives to the innovation process. Such regulation contributes to a reduction in the rate of new chemical innovation and to rising research and development costs. The impact of regulation on the innovative stature of this industry cannot be ignored if it is to remain competitive in global markets. 285 �A more flexible regulatory apparatus should be developed in the United States to balance risks against innovation and product development. The British regulatory approach, which rests on a foundation of trust, cooperation, and dialogue between industry, academia, and Government, merits examination. Environmental regulations in Britain are formulated through mandatory consultation among government officials, outside consultants, and industry. The outgrowth of this concept termed "cooperative regulation" is a less adversarial, more realistic atmosphere in which many factors-including innovation and competition—are taken into account.20 its effectiveness and feasibility, however, have not been evaluated. REGULATION OF PESTICIDES The Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) requires that, prior to marketing, all pesticides must be registered with the EPA along with test data demonstrating their safety and efficacy. The review mechanism and testing requirements under FIFRA increase the direct cost of marketing new products and impose significant time delays on new product development. The average research and development cost for each new pesticide registered under FIFRA is high.21 In a June 1984 decision, the Supreme Court of the United States accepted as "findings of fact" the following regulatory environment in which American companies try to develop new pesticides to protect the world's food supply. The process includes expenditures of $5 million to $15 million annually during the "development process," which "may take between 14 and 22 years." The success rate is infinitesimal: "For every manufacturing-use pesticide the average company finally markets, it will have screened and tested 20,000 others." A major cost in this process flows directly from regulation. The Supreme Court noted that one manufacturer "had incurred costs in excess of $23.6 million in developing the health, safety, and environmental data submitted by it under FIFRA."22 The increase in direct costs may result in a company's decision not to market a new product or develop any new products. The impact of this Act is demonstrated by an analysis of EPA data on new pesticides introduced before and after the full impact of FIFRA, which shows a marked decline in new product development. In 1978-79, for example, there were only nine new pesticides registered with the EPA against 58 in 1975 and 1976.23 Of the $450 million total pesticide industry R&D expenditures in 1981, 67 percent was devoted to development of new products, 25 percent to product expansion, and 8 percent to reregistration and product defense.24 The failure of pesticide producers to develop and market "narrow spectrum" pesticides also illustrates the consequences of the cost barrier to the development of new chemical products. Specialized pesticides have long been favored over "broad spectrum" products that kill not only a particular pest but many other organisms in the same terrain. But the high cost of obtaining a license under FIFRA makes such selective products commercially unattractive.25 As of 1978, more than 125 biological chemicals had been discovered that were not submitted to the EPA because of the high cost of regulation.26 286 �The impact of direct regulatory costs on product innovation is roughly proportional to the percentage such costs bear to the total investment necessary to commercialize a product. For a large firm deciding, whether to market a major new product, the direct regulatory costs are unlikely to influence the decision significantly. On the other hand, for a firm deciding whether to make a single small-volume batch of a new chemical, at a cost of $10,000 to $30,000, almost any regulatory costs may influence the decision to proceed.27 For major products involving large capital investments, time delays are a more important regulatory impact than direct costs. The National Agricultural Chemicals Association (NACA) estimates that in 1981 the average time consumed from submission of a registration application for a new pesticide chemical to granting of a conditional registration was 24 months. On average, more than 7 years elapsed between initial discovery and conditional registration. A 2-year delay in registering the pesticide would reduce the cumulative net income28 from the product over its total commercial life by more than 50 percent. In 1982, the EPA announced a more flexible policy toward pesticides that is designed to reduce some of the paperwork, data requirements, and delay involved in registration.29 It remains to be seen how much these relatively minor changes in FIFRA will improve the conditions for innovation in the pesticide industry. EPA must still work toward a more flexible policy that does not impede the development of new products which deliver improved social benefit in terms of less crop damage and a safer environment. REGULATION OF PHARMACEUTICALS Innovation in the pharmaceutical industry also suffers from excessive regulation. The detail and pervasiveness of FDA regulation of drug safety and effectiveness has been called "almost unique, both in comparison with U.S. regulation of other industries, and with foreign regulation of pharmaceutical markets."30 The innovative R&D process from discovery of a new drug to FDA approval for marketing takes a decade or more. As a result, the cost of discovering and developing a new drug 31 the stage where it receives to FDA approval averaged $70 million in 1980. The current Pharmaceutical Manufacturers Association (PMA) estimate is $84 million. Part of this cost is for basic research, but a significant portion reflects the extensive clinical testing required by the FDA regulations. It is not surprising then that a number of studies have concluded that FDA regulation has been one of the more important factors underlying the adverse trends in pharmaceutical innovation.32 A 1979 National Research Council Study concludes that "As regulatory control of this industry has become more stringent since 1962, a number of adverse trends in pharmaceutical innovation have become increasingly apparent."33 These adverse trends include • Increased costs of R&D and lower yields on drug R&D investments; • Declining rates of new drug introductions on the market as reflected by new product innovations; 287 �• 'A decline in the number of firms introducing innovative new drugs (New Chemical Entities) because, in the view of economists and the industry, the costs of regulation have discouraged innovations by smaller firms; • Declining growth rates for domestic R&D and shifts in R&D abroad; and • Earlier availability of new drugs abroad than in the United States. Last year a study from the National Academy of Engineering (NAE) found that "the data compiled...indicated a clear relative deterioration in the foundation of pharmaceutical competitive position...."34 Two aspects of the decline in the pharmaceutical industry's competitive position were atypical of the general U.S. industrial experiences (1) the proportion of world drug production located within U.S. boundaries has dropped precipitously; and (2) the steady decline in the American share of world pharmaceutical R&D efforts is markedly more severe than comparable changes in world R&D shares for other U.S. industries. The NAE study found FDA regulations to be a significant factor contributing to this adverse trend. It also found a relatively more favorable environment abroad for pharmaceutical research. Export controls effectively prevent American manufacturers from producing and exporting drugs approved for sale in other countries unless they are also approved for use in the United States. They must either wait to market their products abroad, license them to other companies, or build manufacturing facilities in other countries. A recent report by the Office of Technology Assessment concludes that export restrictions have a major adverse effect on American competitiveness, resulting in the transfer of technology and the loss of jobs." Skyrocketing costs and regulatory delays in the approval of new drugs have had a number of negative effects on the industry. There has been a precipitous decline in the number of new drugs approved by the FDA over the last 20 years. The long R&D process required to meet FDA regulations and delays in the FDA approval process substantially reduce the effective length of time during which a drug is protected by patent. The average effective patent term for the new drugs is less than 10 years, little more than half the statutory 17 years.36 Recent enactment of patent legislation to restore up to 5 years of the patent time lost to meet FDA regulations will ameliorate part of the problem.37 The most effective approach to the problems of regulatory cost and delay in the introduction of new, innovative drugs, however, is to streamline the entire FDA regulatory process with the aim of reducing barriers to innovation, commercialization of new products, and international competitiveness of this most important research-intensive industry. An indepth study of the FDA process was conducted by a Congressional Commission on the Federal Drug Approval Process, which in 1982 made extensive recommendations to speed up and improve the process through changes in FDA regulations, operations, and management practices without lowering standards for drug safety and efficacy.3^ Core recommendations regarding FDA regulations included the followingi 288 �• To test drugs more expeditiously preliminary requirements for early clinical research in humans should be simplified in a manner consistent with human safety. • The FDA Commissioner should clarify statutory and regulatory standards with regard to the evidence which must be submitted to establish the effectiveness of a new drug. Two or, when appropriate, one well-designed and controlled study would be sufficient to document such effectiveness. • New Drug Application (NDA) submissions and review should be streamlined, with summary presentations of data replacing individual case report forms. • Less restrictive interpretation of conflict of interest statutes should be put into place and experts outside the FDA should be given a more significant role in the Agency's new drug investigation and approval processes. • Mechanisms should be put in place to ensure equitable resolution of disputes regarding IND's (investigative New Drugs) and NDA's, with the aim of improving interaction between the FDA and industry. For several years, the FDA has recognized the need to reform its procedures and initiated a rewrite of the IND and NDA regulations.39 This process has stalled and should now be reviewed in light of the needs of innovation, research, and development. LESSONS LEARNED Regulation adversely affects, technological innovation in a variety of ways: it increases costs and business uncertainty, causes delays in research and product approval, slows productivity growth, and diverts research efforts that otherwise could be put to more beneficial uses. The cumulative effects, however, are more severe than the effects on individual companies. Regulation puts our Nation at a competitive disadvantage, adversely affects our quality of life when it costs American workers their jobs, and denies consumers the benefit of safer and better products. Where excessive, regulations should be modified or eliminated. Evidence clearly supports the proposition that health, safety, and environmental regulation of new products and processes by the Government .is a negative factor in the continued vitality of technological innovation. Chemicals, pesticides, and Pharmaceuticals are examples where the impact is particularly severe. The medical devices and biotechnology industries appear to be on the threshold of new and stifling regulatory control. We have the opportunity to minimize the harm to these new high-technology industries and help alleviate problems for older industries if policymakers and regulators heed the lessons learned in the years past from excessive health, safety, and environmental regulation. The problems that regulation poses for innovation are summarized below. They are clearly reflected in the chemical, drug, and pesticide industries, 289 �but studies indicate that they exist in varying degrees in a number of other industries. 1. Lengthened Product Development Cycles. The time span from concept to implementation has increased substantially for regulated products and processes , thereby denying the public rapid access to environmental, health, and safety benefits of innovative developments. These delays increase project costs, lower the efficiency and hence the availability of technical personnel, increase the risk of investment recovery because of large-entry costs and less foreseeable market conditions, and reduce the period of investment recovery by effectively reducing the period of patent protection. 2. Delays in Product Introduction. There is growing evidence that the United States is falling behind the world in the availability of innovative new products. This is of particular concern with respect to beneficial drugs and pesticides, it is a threat for biotechnology and medical devices. Safety should delay product availability only when delays are clearly necessary and the risk/cost/benefit decision is a balanced evaluation made with the people affected. 3. Decline in the Number of New Products. Uncertainties and delays have caused management to become overly cautious in regard to innovation and risk taking, even if the process is likely to succeed. 4. Increased Costs of R&D. Complex procedures for market entry and regulatory compliance produce excessive costs without apparent commensurate benefit to the public. They divert capital and human resources away from innovative R&D, making the process less efficient. Such costs, and the regulation of prices, affect the free market mechanisms and add uncertainty to the innovation process. 5. R&D and Manufacturing is Moving Abroad. The ever-increasing costs of R&D and product development are compelling many companies to manufacture products abroad. The highly regulated, costly business environment in the United States and the increasing trend toward manufacturing abroad also translate into new products and processes being denied to U.S. consumers. For many products not approved for sale in the United States, export laws and regulations give the manufacturer no choice but to manufacture the product abroad. Differences in international goals and regulations should be considered. 6. Excessive Assimilation and Reporting Requirements. Separate industrial groups have often been needed just to read the mass of regulations and to prepare the multiplicity of near-duplicate records. A separate bureaucracy has been created in industry just to cope with regulatory paperwork without any measurable benefit, adding costs that reduce resources for innovation. 7. Inadequate Provisions for Trade Secret Information. Regulatory agencies frequently are neither prepared nor required to provide adequate security for proprietary information submitted by industry in response to permit requirements, compliance reports, and other documentation. In addition, some agencies are unable to acknowledge the right of industry to place a confidentiality claim on the development of innovative new 290 �manufacturing processes and products. While there may be social needs to disseminate information for the public good, it should not be at the expense of the owners whose investment created it. When this occurs, further innovation decisions are stifled. 8. Effects on Capital. increased costs, lack of protection, and the trend to conduct R&D and manufacturing abroad contribute to the erosion of U.S.-based, innovation-oriented capital formation. Industry in the United States is hampered through the diversion of capital resources to meet everincreasing regulatory requirements. Large numbers of'U.S. firms rely more on existing technologies and less on new processes for product development. Investment in capital equipment also affects the long-term comparative advantage of a country. One report has stated that "to the extent the U.S. undertakes less real investment...than its major competitors, then the longrun international competition of U.S. industry will be reduced."40 In summary, excessive social regulation diverts capital from the construction of new plants and postpones the societal benefits of new technology and products. Ever-increasing regulatory costs drive innovative new firms from the marketplace, creating powerful disincentives to promising enterprise burdened by high startup costs tied to meeting regulatory standards. WHY THE PROBLEM HAS NOT BEEN SOLVED Regulations formulated without concern for technological innovation or U.S. competitiveness will have an adverse impact throughout our society. Regulators must, therefore, be sensitive both to those who press for more stringent regulations and to those who support innovation. They should also be required to review regulations as knowledge is gained. Regulation, by its very nature, is an adversarial process in a largely political arena. Unless the legislative mandates are clear, the proinnovative voice will not be heard and more stringent regulations than are necessary, based solely on a premise of greater safety, will result. Regulators, of course, must be sensitive to valid safety considerations. However, they should not contribute to the problem through either a concern for an extreme safety imperative,4*- or from an "inability...to make straightforward scientifically grounded risk-benefit determinations."42 What they clearly need is a better mechanism for sifting through conflicting claims and for distinguishing hypothetical fears from scientifically credible evidence. In addition, rto process exists for a systematic and continuous review of regulations. Such a process is necessary because even the most skillfully drawn regulatory requirements become obsolete over time. Regulatory bureaucracies usually ignore or minimize these problems. Furthermore, revising or eliminating old rules as a result of new information takes time and resources that agencies believe could better be used to promulgate additional rules. Therefore, the Federal regulatory agencies must be required to assess the effects of their rules on a continuous basis and to revise or eliminate them where excessive and/or outdated. 291 �present Administration has tried to correct some of these problems, but a permanent procedure for review of regulations is needed, with appropriate outside input by industry, scientists, and other experts. A review of existing regulations and initiation of new ones should be made by regulators, Congress, and panels of experts with a view toward protecting innovation and technological progress and ultimately our ability to enhance U.S. industrial competitiveness. PROPOSALS FOR THE NEXT ADMINISTRATION A key goal for the new Administration should be to put review of health, safety, and environmental regulations as they impact the innovation, production, or use of new products on the national agenda in 1985. The overall objective should be to implement improved processes for balancing values and pressures in both the review of current regulations and the promulgation of new regulations. We believe that a number of changes can be made to improve Government regulatory policies to lessen the negative effects of current and future regulations on innovation. The President can take one key action without sacrificing the essential goals of health, safety, and environmental regulations: The President should expand the role of the Office of Science and Technology Policy (and transfer this role to the new Department of Science and Technology when created) to require it to take actions in the regulatory process to balance the needs of science and technology with concerns about health, safety, and the environment. New responsibilities of this office would include • Assessing the impact that regulations promulgated by Government . agencies will have on technological innovation, research, development, product approval, and hence, industrial competitiveness of affected industries; • Facilitating and initiating actions to encourage the review and promulgation of regulations that can be supported by rigorous scientific evidence; and • Designating an ombudsman to seek and encourage the participation of qualified outside scientists and technical experts in the process of regulatory review and promulgation. In implementing this process, several guiding principles should be followed in reforming regulations to enhance innovation and competitiveness: • No health, safety, or environmental regulation of new products should seek or purport to eliminate every possible risk. Risks below de minimis levels should be left unregulated, except for possible labeling requirements. • Industry self-regulation should be given an opportunity to develop in new areas as the first alternative to Government regulation. 292 �• When necessary to implement new regulations, it should be done using a more flexible and cooperative approach based on trust and dialogue between industry, academe, Government, workers, and consumers. Highlevel commissions consisting of representatives of these groups should be established to review present regulations. • Where Federal safety regulation of new products is necessary, general performance standards should be used wherever possible, rather than specific design standards. • American producers should be permitted to export a product approved for use abroad even if the product has not yet been approved for use in the United States. • Protection should be afforded to trade secrets, patents, and the results of R&D, especially when firms must submit confidential data to the Government in order to obtain regulatory approvals. • The overlapping and often conflicting role of Federal regulatory agencies should be reviewed as to how the regulatory process might be streamlined, especially with regard to emerging high technology industries. Implementation of these new procedures will be a significant step toward achieving a better balance of regulatory requirements that are less harmful to innovation and industrial competitiveness. 293 � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Alvin L. Young Collection on Agent Orange Description An account of the resource <p style="margin-top: -1em; line-height: 1.2em;">The Alvin L. Young Collection on Agent Orange comprises 120 linear feet and spans the late 1800s to 2005; however, the bulk of the coverage is from the 1960s to the 1980s and there are many undated items. The collection was donated to Special Collections of the National Agricultural Library in 1985 by Dr. Alvin L. Young (1942- ). Dr. Young developed the collection as he conducted extensive research on the military defoliant Agent Orange. The collection is in good condition and includes letters, memoranda, books, reports, press releases, journal and newspaper clippings, field logs and notebooks, newsletters, maps, booklets and pamphlets, photographs, memorabilia, and audiotapes of an interview with Dr. Young.</p> <p>For more about this collection, <a href="/exhibits/speccoll/exhibits/show/alvin-l--young-collection-on-a">view the Agent Orange Exhibit.</a></p> Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Box The box containing the original item. 195 Folder The folder containing the original item. 5495 Series The series number of the original item. Series VIII Subseries I Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Description An account of the resource <strong>Corporate Author: </strong>The President's Commission on Industrial Competitiveness, Committee on Research, Development, and Manufacturing Date A point or period of time associated with an event in the lifecycle of the resource December 7 1984 Title A name given to the resource Appendix C: Balancing Environmental, Health, and Safety Regulation with the Needs of Research, Development, and Technological Innovation, A Special report on the Need for Further Regulatory Review ao_seriesVIII https://www.nal.usda.gov/exhibits/speccoll/files/original/7523a52c990eaa17682a7154a4ae5b08.pdf fe299585b42e2c8449746eb0b4b9f0f2 PDF Text Text Item D Number 05732 n JOT scanned Author Corporate Author Report/Article Title Appendix Four: Review of Epidemiologic Data on Humans Exposed to Dioxin-Contaminated Substances Journal/Book Title Year 000 ° Month/Day Color Number of Images DOSCrlptOn Notes G ° Alvin L Youn 9filed ttiese documents together with others under the label, "Agent Orange Exposure Project." Tuesday, March 26, 2002 Page 5732 of 5743 �APPENDIX FOUR REVIEW OF EPIDEMIOLOGIC DATA ON HUMANS EXPOSED TO DIOXIN-CONTAMINATED SUBSTANCES Agent Orange is composed of equal parts of esters of two phenoxy herbicides, 2,4-D and 2,4,5-T. During the production of the 2,4,5-T there was unintended generation of small amounts of a contaminant, 2,3,7,8-TCDD. The production was a two step process of making 2,4,5-trichlorophenol and then using this substance to make 2,4,5-T. The interest in the CDC study is exposure to Agent Orange contaminated with 2,3,7,8-TCDD in amounts up to about 50 ppm, with an average level of contamination of 2 ppm. In evaluating studies of exposed populations reported in the literature to determine whether they have relevance to exposures of the veterans to Agent Orange, the following exposure situations are of interest. The published reports were examined to learn whether the literature contains data to permit judgments about how much exposure to Agent Orange would be necessary to cause harmful medical effects after an individual has been exposed. 1. Chemical workers who made dioxin-contaminated 2,4,5-trichlorophenol and 2,4,5-T and/or who were exposed following industrial accidents. 2. Herbicide sprayers who sprayed 2,4,5-T in forests, fields, and rights of way and foresters exposed to pentachlorophenol. 3. Citizens exposed in the contamination of a large area in Seveso, Italy following an industrial explosion. 4. Citizens of Missouri, U.S.A. following exposure to soil contaminated with waste oils containing 2,3,7,8-TCDD. 5. Three British laboratory scientists who suffered health effects after they had synthesized 2,3,7,8-TCDD. 6. Instances of application of substances to humans which produce chloracne. 1. Chemical workers: Chemical workers who made the substances contaminated with 2,3,7,8 TCDD are generally considered to have had much heavier exposures than would have been experienced by most veterans because of the daily opportunity for exposure and because some workers worked for many years. These substances include 2,4,5trichlorophenol and 2,4,5-T. Severe medical disorders of the peripheral nervous system, liver and skin occurred following some industrial explosions, and some of the disorders have persisted for many years. It is generally assumed that the workers experienced heavy exposure, but there are no published data providing detailed assessments of the exposures. Since all explosions occurred in trichlorophenol reactors, the specific substances to which the workers were exposed were the reactants of the 2,4,5-trichlorophenol process, including the contaminating 2,3,7,8-TCDD. The actual amounts of �2,3,7,8-TCDD present are not known and would have been dependent on the particular stage and conditions under which the explosion occurred. In recent years epidemiologic medical and mortality studies have been conducted of chemical workers exposed during the industrial explosions and also during daily job duties. The major limitations of the studies have been small size and limited information about exposures of the individuals in the study. The results have suggested that the medical problems experienced following the explosions do persist in some workers. Unfortunately, no data are present to address the question whether persons with low levels of exposure are at increased risk of medical problems. Several current studies improve upon the earlier design limitations of small size and inadequate exposure assessment. The National Institute for Occupational Safety and Health (NIOSH) has gathered detailed exposure information for 7,000 U.S. chemical workers which is being applied in a large mortality study and two large medical studies of chemical workers. 2. Herbicide Sprayers and Pentachlorophenol Workers: The definition of "exposure" is unclear in studies of herbicide sprayers. Sprayers use numerous types of herbicides and, generally, the particular types and amounts sprayed by each individual are not known. A number of case control studies have evaluated the possible association of soft tissue sarcoma, lymphoma, nasal and colon cancer with exposures to phenoxy herbicides and chlorophenols by interviewing subjects regarding prior exposures. These studies defined exposed sprayers as those who worked more than 1 day. In these situations an individual was considered "exposed" even if the phenoxy herbicide, such as 2,4-D or MCPA, contained no 2,3,7,8-TCDD. Additionally, no distinction in exposure was made for individuals working with pentachlorophenol, which might not contain 2,3,7,8-TCDD, but could contain substantial amounts of more highly chlorinated and less toxic isomers of dioxin. 3 & 4 Citizens of Seveso and Missouri: Studies of citizens of Seveso have had major design problems and have included no measurements of levels of exposure, making it impossible to assess a relationship between medical problems and levels of exposure to the spewed contents of the trichlorophenol reactor. Cases of chloracne did occur following the Seveso explosion, especially among children. Recent studies of citizens in Missouri, U.S.A., who were exposed to soil contaminated with 2,3,7,8-TCDD in waste oils have noted no cases of chloracne, but have found indications of possible immune effects. 4. British laboratory workers: Three British laboratory workers who synthesized 2,3,7,8-TCDD experienced medical problems similar to the chemical workers exposed in industrial accidents, including chloracne and neurologic problems. However, there is no information on the levels of their exposures. �5. Application of chloracnegens to human skin: No published studies have examined the relationship between level of exposure and the appearance of chloracne in humans. NIOSH may be able to contribute information on this question when the evaluation of hundreds of medical records of chemical workers has been completed and the results Interpreted in light of individual exposures. Consequently, anecdotal situations of application of chloracnegens to humans are of Interest. At best these are very rough estimates because of the variability encountered among individuals. In the mid-19601s, sixty volunteer persons were treated on the forearm or mid-back region with between 0.2 and 8 ug dioxin and the application repeated two weeks later. No one developed chloracne, yielding the conclusion that humans can tolerate exposure to 16 ug dioxin without developing chloracne. (The study design was based on prior animal studies which showed that rabbits developed mild chloracne from application of 0.5 ug dioxin inside the rabbit ear. Application of 1-2 ug caused a more pronounced effect, and 4-8 ug, a severe effect). Subsequently, the researcher applied 7,500 ug in one square inch to the back area of ten volunteers, of whom 8 developed chloracne which lasted 4-7 months. No other medical information was described. Therefore, limited information suggests that the human threshhold for chloracne lies between 16 and 7,500 ug of dioxin applied in a small area of the back. Conclusions: Knowledge of the actual exposure experienced by study participants is the weakest characteristic of all published studies of human exposure to dioxincontaminated substances. Several current but not yet completed studies have good exposure estimates. The published studies do not provide definitions of exposure which are useful in evaluating how much exposure to Agent Orange would be necessary to cause harmful health outcomes for the veterans. Bibliography The information presented here can be explored in greater detail through the use of the following publications, which review and cite other valuable references. (1) Tucker, R., Young, A., and Gray, A., (eds.). Human and Environmental Risk of Chlorinated Dioxins. Plenum Pres, New York, 1983. (2) IARC (1978) Long term hazards of polychlorinated dibenzodioxins and polychlorinated dibenzofurans. IARC Technical Report No. 78/001. International Agency for Research on Cancer, Lyon. (3) Fingerhut, M., Sweeney, M.H., Halperin, W.E. The epidemiology of populations exposed to dioxin. American Chemical Society Annual Meeting, New York City, April, 1986. (4) Hay, A. The Chemical Scythe, Lessons £f_ 2,4,5-T and Dioxin. Plenum Press, New York, 1982. � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Alvin L. Young Collection on Agent Orange Description An account of the resource <p style="margin-top: -1em; line-height: 1.2em;">The Alvin L. Young Collection on Agent Orange comprises 120 linear feet and spans the late 1800s to 2005; however, the bulk of the coverage is from the 1960s to the 1980s and there are many undated items. The collection was donated to Special Collections of the National Agricultural Library in 1985 by Dr. Alvin L. Young (1942- ). Dr. Young developed the collection as he conducted extensive research on the military defoliant Agent Orange. The collection is in good condition and includes letters, memoranda, books, reports, press releases, journal and newspaper clippings, field logs and notebooks, newsletters, maps, booklets and pamphlets, photographs, memorabilia, and audiotapes of an interview with Dr. Young.</p> <p>For more about this collection, <a href="/exhibits/speccoll/exhibits/show/alvin-l--young-collection-on-a">view the Agent Orange Exhibit.</a></p> Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Box The box containing the original item. 202 Folder The folder containing the original item. 5732 Series The series number of the original item. Series VIII Subseries II Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Appendix Four: Review of Epidemiologic Data on Humans Exposed to Dioxin-Contaminated Substances ao_seriesVIII https://www.nal.usda.gov/exhibits/speccoll/files/original/748d2fc951fc3655f501225de1b4810b.pdf ff6ec2615992b3315a8485e3c315f550 PDF Text Text item D Number °5733 n (jot Scanned Author Corporate Author Report/Article Title Appendix Six: Utilization of Biological Samples to Assess Exposure to Agent Orange Journal/Book Title Year Month/Day Color Number of hiauos DBSCrlptOn NOtOS Alvin L Youn 9filed these documents together with others under the label, "Agent Orange Exposure Project." Tuesday, March 26, 2002 Page 5733 of 5743 �APPENDIX SIX Ultilization of Biological Samples to Assess Exposure to Agent Orange Recent advancements in the analytic sensitivity of laboratory instruments have made it possible to analyze very low concentrations of 2,3,7,8-TCDD in samples of human fat ( ) The results of several independent efforts (2-4) indicate 1. that there is a background average level of 2,3,7,8-TCDD in human fat of approximately 7 parts paseaillion (ppt) (range 0-20 ppt). One study analyzed fat samples from volunteer Vietnam veterans ( ) The 4. results indicated that two veterans classified by the Veterans Administration as "heavily exposed" to Agent Orange had fat levels of 2,3,7,8-TCDD of 35 and 99 ppt. The remaining 10 veterans who were classified as "lightly exposed" and "possibly exposed" had levels between 3 and 13 ppt. Four veterans who had no service in Vietnam had levels between 4 and 8 ppt. The results of this study indicate that it may be possible to distinguish high exposure to Agent Orange by analysis of fat samples. The results also indicate that veterans classified as "lightly exposed" to Agent Orange have only background levels of 2,3,7,8-TCDD in their fat, the same levels as are found in the U.S. population in general. Analysis of fat is a difficult method for several reasons. A surgical or suction procedure is necessary to obtain 20 grams of fat (about the size of an egg) and the cost is about $1,000 per sample. Efforts are underway currently to analyze a large volume of serum (200 ml) to detect low levels of 2,3,7,8-TCDD. Data are also being sought which would describe the distribution of 2,3,7,8-TCDD between adipose tissue and serurt in the human body. Success with the serum method would provide a method to recognize levels of exposure which were high enough to raise levels of 2,3,7,8-TCDD above background levels in the population. The recent advances in laboratory analytic techniques could be used to ascertain whether veterans in the various exposure categories of the CDC Agent Orange study have levels of 2,3,7,8-TCDD above the background levels in the population. For example, a sample of veterans currently meeting criteria for the CDC Agent Orange study category of "high likelihood of exposure" and a sample of veterans from the non-exposed category could be asked to provide fat (or possibly serum) specimens for analysis. An evaluation of the results should provide insight into the adequacy of the military records to select truly exposed and truly unexposed individuals. Additionally, the results should indicate whether the levels of 2,3,7,8-TCDD are significantly different from the levels in the general U.S. population. Analysis of fat (or serum) from other populations could also provide valuable insights. Several studies are currently underway in which analysis of fat is being conducted on Vietnam veterans, chemical workers, and persons with residential and recreational exposures to 2,3,7,8-TCDD. Analysis of fat (or serum) could also be conducted on selected individuals in the CDC Vietnam Experience study who have known high or low levels of exposure. Samples of fat already collected from Ranch Hand participants during elective surgery could be analyzed and compared to the levels of exposure experienced by the Individuals. �1. Patterson DG, et al. High resolution gas chromatography/high-resolution mass spectrophotometric analysis of human adipose tissue for 2,3,7,8-TCDD. Anal. Chem 1986; 58:705-716. 2. Graham M, Hileman FD, Kirk D, et al.: Background human exposure to 2,3,7,8-TCDD. Fourth International Symposium on Chlorinated Dioxins and Related Compounds, 1984; Ottawa, Canada; October 16-18. 3. Graham M, Hileman FD, Wendlong J, Wilson JD. Chlorocarbons in adipose tissue samples. Fifth International Symposium on Chlorinated Dioxins and Related Components, 1985. Bayreuth FRG, September 16-19.2508R 4. Gross ML, Lay JO, Lyon PA, et al.: 2,3,7,8-tetrachlorodibenzo-p-dioxin levels in adipose tissue of Vietnam veterans. Environ Res 1984; 33:261-268. � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Alvin L. Young Collection on Agent Orange Description An account of the resource <p style="margin-top: -1em; line-height: 1.2em;">The Alvin L. Young Collection on Agent Orange comprises 120 linear feet and spans the late 1800s to 2005; however, the bulk of the coverage is from the 1960s to the 1980s and there are many undated items. The collection was donated to Special Collections of the National Agricultural Library in 1985 by Dr. Alvin L. Young (1942- ). Dr. Young developed the collection as he conducted extensive research on the military defoliant Agent Orange. The collection is in good condition and includes letters, memoranda, books, reports, press releases, journal and newspaper clippings, field logs and notebooks, newsletters, maps, booklets and pamphlets, photographs, memorabilia, and audiotapes of an interview with Dr. Young.</p> <p>For more about this collection, <a href="/exhibits/speccoll/exhibits/show/alvin-l--young-collection-on-a">view the Agent Orange Exhibit.</a></p> Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Box The box containing the original item. 202 Folder The folder containing the original item. 5733 Series The series number of the original item. Series VIII Subseries II Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Appendix Six: Utilization of Biological Samples to Assess Exposure to Agent Orange ao_seriesVIII https://www.nal.usda.gov/exhibits/speccoll/files/original/b1fa6dbaa3acd8cc03ace14a021e44e1.pdf 46c74166a455a35e3917814e4139aa35 PDF Text Text Item D Numb en 05269 D NotScannBfl Author CorpOratB Author United States Environmental Protection Agency (EPA) Roport/Artido TltlB Appendix to Rebuttal Comments of Respondent Environmental Protection Agency in re: Emergency Suspension Order for 2,4,5-T and Silvex, FIFRA Docket Nos. 409-410 Journal/Book Title Year 1979 Month/Day A ril Color Number ofimages P D ° Also includes cover letter from L. Mark Wine of Kirkland and Ellis. Tuesday, March 05,2002 Page 5269 of 5363 �KIRKLAND&. ELLIS Washington Office Area Code 202 857-5000 To Call Writer Direct 202857- 5024 1776 K Street, N.W. Washington, D.C. 20006 April 5, 1979 chlcago ^ Area Code 312 861-2000 Telex 25-4361 200 E. Randolph Drive Chicago,III.60601 Enclosed is a copy of EPA's rebuttal comments which greatly expands its discussion of the data and sets forth the points it intends to prove at the hearing. We must respond as fully as possible to all of these claims both in preparation of Dow's testimony and in crossexamination of EPA witnesses. Sincerely yours/ L. Mark Wine LMW:bac Enclosure �U.S. ENVIRONMENTAL PROTECTION AGENCY in Fe: Emergency Suspension Order for 2.4.5-T and Silvex ) ) FIFFA Docket Nos. 409-410 APPENDIX TO REBUTTAL COMMENTS OF RESPONDENT ENVIRONMENTAL PROTECTION AGENCY Attached herwith are copies of the exhibits cited in "Rebuttal Comments of Respondent Environmental Protection Agency", dated April 4, 1979. Copies of this Appendix have been filed with the Hearing Clerk, the Hearing Panel and Counsel for Dow Chemical Corp. Any active party desiring a copy of the Appendix should contact the Office of General Counsel directly (202-755-2680). �APPENDIX I. Introduction - II. Exhibit 1, cited at p. 3. "USDA to Suspend Use of 2.4.5-T in Forest Management". USDA 1979. Animal Data Exhibit 1, cited at p. 14. "Reproductive Dysfunction. . ."« Barsotti, 1977. - Exhibit 2, cited at p. 14, "Hormonal Alternatives . . ." . Earsotti, 1979. Exhibit 3. cited at p. 15. "Environment and Birth Defects". 1973. - Exhibit 4. cited a p. 14. "TCDD Toxicity in Various Animal Models". Spencer 1979. - Exhibit 5. cited pp. 15-16. "The non-teratogenicity . . ." . Dougherty. 1975. Exhibit 6. cited at p. 16. "Abnormalities of Intrauterine Development . . ". 1971. Wilson. * III. - Exhibit 7. cited at p. 20. "Response to Rebuttal Comments. . ." . Albert. 1979. Alsea Exhibit 1. cited a p. 44. "Status Report". EPA. Feb. 8 . 1979. Exhibit 2, cited at p. 44. "Congential Abnormalities" , Infante. Exhibit 3. cited at p. 45. "Proceedings. Conference on Women in the Workplace. Infante 1976. - Exhibit 4. cited a p. 45. "Significant Factors . . .". Bailar. 1967. - Exhibit 5. cited p. 45. "The Effects of Maternally Inhaled Vinyl Chloride . . .". John. 1977. �-2- IV. Exposure - Exhibit 1. cited p. 66. letter. McGuire to Costle. 11-13-78. - Exhibit la. cited p. 69. Burkett affidavit. 4-3-79. - Exhibit 2. cited p. 68. Clary affadivit. 4-3-79. - Exhibit 3. cited p. 69. Cromwell affidavit. 3-14-74 - Exhibit 4. cited p. 77. "General Review of Environmental Movement of 2.4.5-T and Silvex". - Exhibit 5. cited p. 78. Exposure Report. "Ditch Contamination . - Exhibit 6. cited p. 79. "Soil Application of 2.4.5-T". - Exhibit 7. cited p. 79. "Railroad Rows". - Exhibit 8. cited p. 79. "Testimony of Hal Weber". 3-14-78. Exhibit 9. cited p. 79. "Testimony of Kathleen O'Hagan". 3-14-78. Exhibit 9a» cited p. 79a. "Chlorodioxins in Pesticides. Soils and Plants". Helling. 1973. - Exhibit 11. cited p. 79a. "Studies on Bioaccumulation . . .". Matsumura. 1973. - Exhibit 12. cited p. 79a. "Fate of 2,3.7.8 Tetrachlordibenzo-p-Dioxin. . . "..Ward & Matsumra. 1978. - V. Exhibit 10. cited p. 79a« "Persistence and Metabolism of Chlorodioxins in Soils". Kearney. 1972. Exhibit 13. cited p. 79a, "Conquering the Monster". Crosby. 1977. Benefits - Exhibit 1. cited at p. 83. "Memorandum of Understanding . . .". _- Exhibit-2. cited at p. 87. "USDA to Suspend Use of 2.4.5-T in Forest Management". 3-1-79. �U.S. ENVIRONMENTAL PROTECTION AGENCY ,in Re: Emergency Suspension Order for 2,4,5-T and Silvex ) ) FIFRA Docket Nos. 409-410 ' ERRATA SHEET FOR REBUTTAL COMMENTS OF RESPONDENT ENVIRONMENTAL PROTECTION AGENCY The following corrections should be incorporated into the text of the "Rebuttal Comments of Respondent Environmental Protection Agency," filed April 4, 1979. i 1) p. iii, §B., "chemical residue" not "chemicals residue" 2) p. iii, §(1), "causal" not "causual" 3) p. If line 4, "pages," not "pages" 4) p. 3, line 1, "Dow's" not "Dow s" 5) p. 6, line 2, "manufacturers" not "manufactures" 6) p. 6, line 6, "510 F2d. 1292," not "§10 2d. 1292," 7) p. 6, line 23, "variety" not "varity" 8) p. 7, line 6, "another study," not "other study," 9) p. 7, line 18, "hypotheses" not "nypothesis" 10) p. 8, line 3, delete "to" 11} p. 9, line 14, "510" not "§510" 12) p. 10, line 14, "at 10," not "at 13) p. 10, line 15, "at 10." not "at ," �-2- 14) p. 10, line 18, "teratogenic," not "terotogenic," .15) p. 11, l i n e 8, "pesticides" not "pesticide" ,16) p. 12, heading B . , l i n e 1, "Fetotoxic," not "Fetotoxic" 17) p. 13, l i n e 2, "has" not " h a d " 18) p. 13, **/, delete "a" in last line — NOEL had" 19) p. 14, line 22, "three" not "thre" 20) p. 17, line 19,."rather" not "rathr" 21) p. 21, line 10, "studies." not "studies"" 22) p. 22, line 1, delete second "effects" 23) p. 24, line 3, "ug/kg.," not "ug/kg." 24) p. 24, fn */ "at 19-20," not "at 25) p. 26, third indented paragraph; insert the following should read "that no ," after "ingesting": "TCDD was clearly affected at dose levels" 26) p. 29, line 15, delete second "in the" 27) p. 38, line 4, "hospitalized" not "hositalized" 28) p. 38, line 11, "risk" not "isk" 29) p. 39, 1'ine 5, "study area" not "study are" 30) p. 43, line 11, add "is" after "conclusion" 31) p. 43, line 12, "an" not "on" 32) p. 44, l i n e 17, "medical" not "radeical" 33) p . ' 4 6 , line 7, " f i s h i n g , " not " f i s h i n g " 34) p. 47, line 20, "plausibility." not "pluasibility." 35) p. 48, line 5, "(See, text, supra, at 9.)," not "(See Section , infra.)," �-3- 36) p. 49, line 14 "no more than 1-10 ppt in1- 37) p. 51, line 1, "respond" not "rspond" ,38) p. 51, line 2, "whether" not "nether" 39) p. 52, line 3, "may be" not "maybe" 40) p. 52, line 6, "cross-correlation a n a l y s i s . " not "cross-correlationanalys is." 41) p. 52, line 7, " r e g u l a t o r y " not " r e g u l a t i o n " 42) p. 52, S d ) , f i r s t l i n e , ""causal"" not " " c a u s u a l " " 43) p. 53, l i n e 9, " t h o u g h t f u l " not " t h g o u t h t f u l " 44) p. 53, line 15, "response" not "reponse" 45) p. 53, line 20, "waste of time and f u n d s . " not "waste time a n d f u n d . " 46) p. 53, line 24, "pursued" not "pursue" 47) p. 54, l i n e 20, "charges." not " c h a n g e s . " 48) p. 55, line 16, "uncler-reports" not "under r e p o r t s " 49) p. 56, l i n e 1, delete " t h e r e " 50) p. 57, line 3, "hypothesis" not "hpothesis" 51) .p. 58, line 3, "on" not "or" 52) p. 60, line 4, "questionable" not "questionabel" 53) p. 62, fn. **/, line 3, "correction" not "ccorrection" 54) p. 62, line 15, "Dow" not "Bow" 55) p. 63, 2nd indented paragraph, line 4, "ways" not "wasy" 56) p. 63, 2nd indented paragraph, line 7, "at 50-52;" not "at 50-51;" 57) ,p. 63,_ljLne_ 4, "proceeded" not "provided" �-4- 58) p. 64, line 2, "dealt" not "dealth" .59) p. 64, line 4, "they" not "it" ,60) p. 64, line 11, "unpersuasive," not "unpresuasive," 61) p. 64, line 21, "chemicals" not "chemicasls" 62) p. 65, line 8, "application." not "affication." 63) p. 65, line 16, "and habitation," not "hibitation," 64) p. 65, line 18, "conformance" not "conforming" 65) p. 65, line 19, "on" not "in" 66) p. 65, line 22, "unknowingly" not "unknowningly" 67) p. 66, line, "diffusion" not "difffusion" 68) p. 66, line 1, "through" not "though" 69) p. 66, line 2, delete "and" 70) p. 66, line 5, "also" not "aslo" 71) p. 67, line 5, "descriptions" not "discriptions" 72) p. 73, fn. */, line 4, "follow" not "follw" 73) p. 74, fn. */, line 6, place "generally" after "is" 74) p. 74, line 2, "*/" not "**/" 75) p. 74, line 14, "during" not "dueing" 76) p. 74, line 15, "same" not "ame" 77) p. 75, line 7, delete ", treated" 78) p. 76, line 13, "quantification" not "qualification" 79) p. 76, line 15, "from a qualitative" not "from qualitative" 80) p. 77, line 20, "from" not "for" 81) p. 78, line 11, "site" not "iste" �-5- 82) p. 80, line 19, " o f " not "a" 83) p. 81, line 2, " i n d i c a t e " not " i n d i c a t a e " 84) p. 82, line 16 "costs" not "cost" 85) p. 84, line 11, "exercised" not "exercized" 86) p. 84, line 13, "exercise" not "exercize" 87) p. 85, line 13, delete comma after "partially" 88) p. 87, line 10, insert ", Dow asserts" after "treatment" 89) p. 87, line 13, insert ", Dow claims" after "feasible" �INSERT AFTER P. 75 -75aDow contends that environmental monitoring has failed to reveal any significant TCDD residues and that studies fail to demonstrate the presence of TCDD as a residue in the American food supply. They contend that this is because the short half- lives of 2,4,5-T, silvex and TCDD prevent a significant contamination of environmental media. Doubt is cast upon these claims by the evidence. Monitoring studies done by EPA indicate that 2,4,5-T and/or silvex are present in human urine, ambient air and surface waters. Recent preliminary results from a national urine survey show trace amounts of 2,4,5-T in 3 of 1085 samples, quantifiable silvex residues in 4 samples, and trace amounts of silvex in an additional 13 samples. During the period May, 1976 - August, 1978, 2,4,5-T was detected in 6 of 1350 whole water samples; silvex was detected twice. Ambient air monitoring of agricultural areas from 1970-1971 3 3 shows residues of 2,4,5-T ranging from 0.5 ng/m to 36 ng/m . The U.S. Forest Service has done some limited monitoring studies in connection with herbicide spray projects on National Forest Lands in the Pacific Northwest. Analyses were only done for 2,4,5-T since Silvex is generally not used by the Forest Service. Some of those analyses were positive for 2,4,5-T. The Bureau of Land-Management analyzed a limited number of Oregon water samples for silvex only, since 2,4,5-T is not usually used by the Bureau. Silvex was detected within 72 hours after treatment. Data^collected by the U.S. Geological Survey for 1968 - 1971 showed that 2,4,5-T was the most commonly detected herbicide in western streams. �INSERT AFTER P. 79 -79a3. The environmental persistence of TCDD allows for exposure after the actual use-time. While 2,4,5-T and silvex are not environmentally persistent chemicals, TCDD is capable of great persistence. Helling et al. (1973) concluded that this persistence is not surprising since it is an insoluble, non-polar, chlorinated molecule, devoid of biologically labile functional groups. The half-life in soil has been estimated to be about one year in two different soils (Kearney et al., 1972). Of 100 strains of micro- organisms which have the capability to degrade other persistent pesticides, only five showed some ability to degrade TCDD (Matsumura and Benezet, 1973). Matsumura (1978) found that TCDD was stable in lake sediments and had a half-life of about 600 days. TCDD usually first appears in the environment at a very low concentration spraying. in a thin film on foliage or soil, as a result of The extent of its persistence depends upon the current environmental circumstances. Under certain conditions, including principally the presence of a hydrogen-donating solvent and adequate sunlight intensity, photodechlorination of TCDD to lower chlorinated dioxins can take place. Dow contends that TCDD has a short half-life in the environment when "on vegetation in the presence of a hydrogen donor and that photochemical degradation also occurs in soil." at 37. Dow Comments However, these conditions are clearly not always available. As noted by Crosby (1977), breakdown would be expected to be slower in the shade even if efficient hydrogen donors wore available. Much of the forest area in the Northwest is subject to long periods �CERTIFICATE OF SERVICE I hereby certify that copies of the foregoing "ERRATA Sheet for Rebuttal Comments of Respondent Environmental Protection Agency" and "Appendix to Rebuttal Comment of Respondent of Environmental Protection Agency" were hand-delivered or mailed first class postage prepaid, on April 9, 1979, to the following persons: Timothy Atkeson Steptoe & Johnson Chevron Chemical Company 1250 Connecticut Avenue, N.W, Washington, D.C. 20036 Shrikant V. Kulkarni, Ph. D. Manager, Pesticide Labeling Vertac Technical Center P.O. Box 941 West Memphis, AR 72301 Robert L. Ackerly Sellers, Conner & Cuneo Counsel for Lebanon Chemical Corp, 1625 K Street, N.W. Graham Purcell Doub, Purcell, Muntzing & Hansen Chartered 1775 Pennsylvania Avenue, N.W. Washington, D.C. 20006 Counsel for Riverdale Chemical Company Frank Miller & Sons Tobacco States Chemical Co. PBI Gordon Corp Pueblo Chemical & Supply Co. Plat'te Chemical Co. Roger A. Shores, President Bartels and Shores Chemical Co 1400 St. Louis Avenue Kansas City, HO 64101 Washington, D.C. 20006 Edward W. Warren, et al Kirkland & Ellis Counsel for Dow Chemical Company 1776 K Street, N.W. Washington, D.C. 20006 Gene R. Currie Technical Services Manager Imperial Inc. Box 423 West 6th Avenue Shenandoah, IA 51601 Larry R. Hodges AMCHEM PRODUCTS, Inc. Ambler, Pennsylvania 19002 Anthony P. Brown Pillsbury, Madison & Sutro Chevron Chemical Company P.O. Box 7880 San Francisco, CA 94120 Gene R. Currie Floyd E. Grabiel, II General Counsel Universal Cooperatives, Inc 3001 Metro Drive Minneapolis, MN 55420 Technical Services Manager MFA Oil Company Box 423 West 6th Avenue Shenandoah, IA 51601 �- 2Bonide Chemical Company 2 wurz Yorkviile, New York 13495 Jacqueline Warren Counsel for the Environmental Defense Fund, Inc. 1525 18th Street, W.W. Washington, D.C. 20036 Graham Purcell Doub, Purcell, Muntzing & Hansen Attorneys at Law Counsel for Platte Chemical Company 1775 Pennsylvania Avenue, N.W. Washington, D.C. 20006 Melvin R. Wilcox, III Roberts, Harbour, Smith, Harris French & Ritter Counsel for T.O. Bell/dba/Forage Unlimited 404 North Green Street Longview, Texas 75601 Frank B. Stewart, Vice President The Chas. H. Lilly Company 7237 N.E. Killingsworth Portland, Oregon 97218 John J. Balardo Corporate Counsel 2727 Walker, N.W. Grand Rapids, MI 49504 0. A. Wolcott, Manager Planning & Technical Services Farmers Union Central Exchange, Inc. Box 43089 St. Paul, MM 55164 John R. Diem, Vice President Southern Agricultural Insecticides,- Inc P.O. Box 218 Palmetto, PL 335'61 J. R. McCloud, Manager Specialty Products Smith-Douglass Division of Borden Chemical, Borden, Inc. P.O. Box 419 Norfolk, VA 23501 Richard deC. Hinds, Esq. Cleary, Gottlieb, Steen & Hamilton Counsel for USS Agri-Chemical Division of USS Corporation 1250 Connecticut Avenue, N.W. Washington, D.C. 20036 Bernard H. Lorant Counsel for Black Leaf Products Co. Lorant and Lorant, PC P.O. Box 668 Aldo Blasio, President Farmingdale Garden Labs, Inc. 136 Verdi Street Farmingdale, New York 11735 Highland Park, II 60035 John E. Soltes, General Manager WEGRO, Division of Old Fort Industries, Inc. Grand Rapids, Ohio 43522 Fernando Erazo, President Heritage House Product Corporation 1025 Northern Boulevard Roslyn, New York 11576 �- 3- - Ronald A. Meier, Manager Lawn Fertilizer Division The Andersons P.O. tiox 119 Mauraee, Ohio 43537 John R. Wittpenn, President Rockland Chemical Company, Inc. P.O. Box 809 West Caldwell, New Jersey 07006 Robert w. Cummings Assistant Vice President J. & L. Adikes, Inc. 182-12 93rd Avenue Jamaica, New York 11423 Jon D. Loft, President Lofts Pedigreed Seed, Inc. P.O. Box 146 Bound Brook, New Jersey 08805 JU_1 Kevin Lee April 9, 1979 � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Alvin L. Young Collection on Agent Orange Description An account of the resource <p style="margin-top: -1em; line-height: 1.2em;">The Alvin L. Young Collection on Agent Orange comprises 120 linear feet and spans the late 1800s to 2005; however, the bulk of the coverage is from the 1960s to the 1980s and there are many undated items. The collection was donated to Special Collections of the National Agricultural Library in 1985 by Dr. Alvin L. Young (1942- ). Dr. Young developed the collection as he conducted extensive research on the military defoliant Agent Orange. The collection is in good condition and includes letters, memoranda, books, reports, press releases, journal and newspaper clippings, field logs and notebooks, newsletters, maps, booklets and pamphlets, photographs, memorabilia, and audiotapes of an interview with Dr. Young.</p> <p>For more about this collection, <a href="/exhibits/speccoll/exhibits/show/alvin-l--young-collection-on-a">view the Agent Orange Exhibit.</a></p> Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Box The box containing the original item. 180 Folder The folder containing the original item. 5269 Series The series number of the original item. Series VIII Subseries I Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Description An account of the resource <strong>Corporate Author: </strong>United States Environmental Protection Agency (EPA) Date A point or period of time associated with an event in the lifecycle of the resource 1979-04-01 Title A name given to the resource Appendix to Rebuttal Comments of Respondent Environmental Protection Agency in re: Emergency Suspension Order for 2,4,5-T and Silvex, FIFRA Docket Nos. 409-410 ao_seriesVIII https://www.nal.usda.gov/exhibits/speccoll/files/original/6d2183b4cd00fc6734611fe1012f4998.pdf fd12328fce3937c2e0ab9f2c6ccb2914 PDF Text Text Rom D Number 0573 ° D ^t Scanned Author Corporate Author RlHJOPt/APtlclB Tltlfl Appendix V: Toxicity Data and Exposure Estimates Journal/Book Title Year 000 ° Month/Day Color n Number of Images ° Alvin L Youn gfiled these documents together with others under the label, "Agent Orange Exposure Project." Tuesday, March 26, 2002 Page 5730 of 5743 �APPENDIX-3JT TOXICITY DATA AND EXPOSURE ESTIMATES Concern has been expressed about the toxicity of the herbicides used in Vietnam, including the contaminant 2,3,7,8-TCDD. Therefore, the files of the Office of Pesticide Programs in the U.S. Environmental Protection Agency (EPA) were consulted to determine the toxicity .(both carcinogenicity and non-carcinogenicity) for the compounds in question: 2,4,5-T, 2,4-D, picloram, and cacodylic acid, as well as 2,3,7,8-TCDD. Attachment 1 summarizes these data. The "ADI" is an estimate of the level of exposure which could be received daily for a lifetime with little likelihood of deleterious effects to exposed humans. In addition, the Science Panel investigated the potential for toxicologically significant exposure under a variety of scenarios. The Bricker paper, found elsewhere in the appendix, presents much valuable data on exposure conditions in Vietnam and estimated exposures. In a separate, focused effort Kang summarized the exposure potential for "wet sprays" via Ranch Hand, as estimated by Flanders (CDC), Gough (in a recently published book), and Kingsley and Stevens (in a previously published article) (Attachment 2). Attachment 3 summarizes exposure estimates for a number of exposure scenarios. Finally, attachment 4 is an extract of a detailed EPA exposure assessment on the use of 2,4,5-T in various applications in the U.S. � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Alvin L. Young Collection on Agent Orange Description An account of the resource <p style="margin-top: -1em; line-height: 1.2em;">The Alvin L. Young Collection on Agent Orange comprises 120 linear feet and spans the late 1800s to 2005; however, the bulk of the coverage is from the 1960s to the 1980s and there are many undated items. The collection was donated to Special Collections of the National Agricultural Library in 1985 by Dr. Alvin L. Young (1942- ). Dr. Young developed the collection as he conducted extensive research on the military defoliant Agent Orange. The collection is in good condition and includes letters, memoranda, books, reports, press releases, journal and newspaper clippings, field logs and notebooks, newsletters, maps, booklets and pamphlets, photographs, memorabilia, and audiotapes of an interview with Dr. Young.</p> <p>For more about this collection, <a href="/exhibits/speccoll/exhibits/show/alvin-l--young-collection-on-a">view the Agent Orange Exhibit.</a></p> Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Box The box containing the original item. 202 Folder The folder containing the original item. 5730 Series The series number of the original item. Series VIII Subseries II Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Appendix V: Toxicity Data and Exposure Estimates ao_seriesVIII Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Alvin L. Young Collection on Agent Orange Description An account of the resource <p style="margin-top: -1em; line-height: 1.2em;">The Alvin L. Young Collection on Agent Orange comprises 120 linear feet and spans the late 1800s to 2005; however, the bulk of the coverage is from the 1960s to the 1980s and there are many undated items. The collection was donated to Special Collections of the National Agricultural Library in 1985 by Dr. Alvin L. Young (1942- ). Dr. Young developed the collection as he conducted extensive research on the military defoliant Agent Orange. The collection is in good condition and includes letters, memoranda, books, reports, press releases, journal and newspaper clippings, field logs and notebooks, newsletters, maps, booklets and pamphlets, photographs, memorabilia, and audiotapes of an interview with Dr. Young.</p> <p>For more about this collection, <a href="/exhibits/speccoll/exhibits/show/alvin-l--young-collection-on-a">view the Agent Orange Exhibit.</a></p> Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Box The box containing the original item. 202 Folder The folder containing the original item. 5707 Series The series number of the original item. Series VIII Subseries II Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Creator An entity primarily responsible for making the resource Gough, M. Source A related resource from which the described resource is derived Dioxin, Agent Orange - The Facts Date A point or period of time associated with an event in the lifecycle of the resource 1986 Title A name given to the resource Appendix: Calculation of the Amount of Dioxin Exposure of a Person Standing under a Ranch Hand Spray Mission ao_seriesVIII