Document 4aK9dvvogapBbgJgZJBBnLRjG

l t :santo f RUM 0 Iname-location-phonc H, C. Godt, Jr. " 04B (4*2616) DATE ' .May 24, 1984 m u cct REFERENCE SPECIAL REPORT ON NON-INDUSTRIAL SOURCES OF -C-H-L-O-R-I-N-A-TE--D--D-EBENZ>O-P-DIOXINS (CDDs) - REPO.R.T-- NO..- MSL - 3506 TO 1 ;R. J. Barne-s* * Townley & Updike T. M. Bistiine - E2ND A. M. Fdrd > G3WG W. J. McCarville - G3WG P. M. Pleska - Bowles, McDavid, Graff & Love G. Roush - G2WG D. F. Snively - E2ND M. S. Weinberg - Weinberg Consulting Group, Inc. J. D. Wilson - G3WG Litigation Technical Support Archives (A. M. Ford) (2 copies) Technical Reports Library (R. G. Lee - R2C) (2 copies) 9 A month has elapsed since the subject report was' issued. In the meantime, Allan Ford, Jim Wilson and I have carefully perused the report looking for errors and misrepresentations of all kinds. Attached are the following corrected pages: Volume 1: .. % 2, 4, 5, 6, 7, 9, 13, 14, 15, 16, 18, 19, 21, 22, 23, 26, 27, 35, 37, 39, 40, 41, 43, 47, 4.9, 50, 51, 53, 54, 56, 58, 59, 60, 69, 70, 71, 73, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 92, 96, 97, 99, 103, 104, 109, 110, .112, 122, 125, 128. *. * Volume 2: . 15, 229, 230. Please replace the appropriate pages in your copy of the report. If you have any questions, please contact me. /mil Attachments ;v. C. H. C. Godt, Jr. COMF1 sue iect to PROTECTIVE ORDER. C2386:> ATTORNEY WORK PRODUCT ATTORNEY-CLIENT PRIVILEGE ^ Table I a Chlorinated Dibgnzo-p-dioxin Congeners by Homolog and Isomer M o n o c h l o r o (2 isomers) l-;2- Dichloro (10 isomers) 1,2-; 1 ,3-; 1,4-; 1,6-; 1 ,7-; 1,8-; 1,9-; 2 ,3-; 2,7-; 2,8-. Trichloro (14 isomers) 1,2,3-; 1,2,4-; 1,2,6-; 1,2,7-; 1,2,8-; 1,2,9-; 1,3,6-; 1,3,7-; 1,3,8-; 1,3,9-} 1,4,6-; 1,4,7-; 1,7,8-; 2,3,7-. Tetrachloro (22 isomers) 1,2,3,4-; 1,2,3,6-; 1,2,3,7-; 1,2,3,8-; 1,2,3,9-; 1,2,4,6-; 1,2,4,7-; 1,2,4,8-, 1,2,4,9-; 1,2,6,8-; 1,2,6,7-; 1,2,7,8-; 1,2,6,9-; 1,2,8,9-; 1,2,7,9-; 1,3,6,9-; 1,3,6,8-; 1*3,7,9-; 1,3,7,8-; 1,4,7,8-; 1,4,6,9-; 2,3,7,8-. Pentachloro (14 isomers) 1,2,3,4,6- J 1,2,3,4,7-; 1,2,3,6,7- ; 1,2,3,6,8-; 1,2,3,6,9- ; 1,2,3,7,8-; 1,2,3,7,9- ; 1,2,3,8,9-; 1,2,4,6,7- ; 1,2,4,6,8-; 1,2,4,6,9- ; 1,2,4,7,8-; 1,2,4,7,9- ; 1,2,4,8,9-. Hexachloro (10 isomers) 1,2,3,4,6,71,2,3,4,6,81,2,3,4,6,91,2,3,4,7,81,2,3,6,7,81,2,3,6,7,91,2,3,6,8,91,2,3,7,8,91,2,4,6,7,91,2,4,6,8,9- H e p t a c h l o r o (2 isomers) 1,2,3,4,6,7,8-, 1,2,3,4,.6,7,'3-. O c t a c h l o r o (1 i s o m e r ) 1,2,3,4,6,7,8,9-. Molecular Weights of Homologs Monochloro = Dichloro = Trichloro = Tetrachloro = Pentachloro = Hexachloro = Heptachloro = Octachloro = 218.64 253.08 287.52 321.96 356.40 390.84 425.28 459.72 a. Congener = a general term used to mean any isomer of any homolog; thus there are 75 PCDD congeners. b. Homolog (isomer cluster) = a group of PCDD isomers having a specified number of chlorine atoms; thus the tetrachloro homolog has 22 isomers. Isomer = a PCDD with a specific numerical arrangement of the . ` " N T ^ ^ c f t ^ l n e atoms within a homolog. 0^33*70 SU BJECT TO PROTECTIVE ORDER. 2 '9 One of the striking features of 2,3,7,8-TCDD's toxicity in' animals is that the lethal dose varies so much from one species to another. Table 4 illustrates this fact. Table 4 Variation of 2,3,7,8-TCDD Toxicity in Different Species Animal LDsoCug per kg body weight) Guinea pig Rat(male) Rat(female) Monkey Rabbit Mouse Dog Bullfrog Hamster 1 22 45 <70 115 114 >300 >500 5000 Source: Poland and Knutson cology & Toxicology, 1982 The guinea pig, the most sensitive animal tested, is at least 5000 times more sensitive than the hamster, the least sensitive animal yet tested. Rabbits, mice, and monkeys are somewhere in the middle - roughly 100 times less sensitive than guinea pigs, but 50 times more sensitive than hamsters. A. Structure-Biological Activity Relationship It is important to ,,note the structure-activity relationship among the CDDs. The unusual toxicity of the . chlorinated dibenzo-p-dioxin molecule is associated with the positioning of chlorine on the planar, tricyclic ring system. As early as 1973, Kende and Wade 111 analyzed the structure-activity relationship of about two dozen PCDD congeners. Recently, Poland 79,and Payne 117 noted certain other characteristics. The following presents the structure-activity relationship as now conceived: 654 1 . halogen atoms must occupy at least three, and for maximum potency, four of the lateral ring positions (positions 2,3,7 and 8 ) 2 . at least one ring position must be unsubstituted CONFIDENTIAL SUBJECT TO PROTECTIVE ORDER. * 023871 attorney w o r k product ATTORNEY-CUENT PRIVILEGE 3 . no amount of halogen substitution on only one ring leads to biological activity. 4 . halogenation in the peri positions (positions 1,4,6 and 9 ). reduces potency*, however, reduction in potency due to peri position halogenation is not nearly as significant as the removal of halogen from one or more of the lateral positions. 5 . the order of potency with regard to halogen substitution is bromine >chlorine >fluorine (potency of iodine analog apparently not determined). 6 . other functional groups [e.g., nitro (-NO2 )] render the dibenzo-p-dioxin ring toxic, but the potency is considerably less. 7. 2,3,7,8-TCDD and its approximate congeners carj be thought of as roughly fitting into a rectangle 3 x 10A (angstroms) with chlorine atoms in the four corners. It has been suggested that the extreme toxicity of 2,3,7,8-TCDD and the specific penta- and hexachloro isomers mentioned is related to the detoxification process of hydroxylation which takes place exclusively at the lateral positions. The presence of the chlorine atoms at these positions may prolong the in vivo life of the laterally substituted PCDDs, and thereby increase the toxicity of much smaller doses 117. Industrial processes that produce CDDs as unwanted by-products generally react specific materials to produce specific products. Thus, specific CDD congeners are found among the reaction products. This is not true of most of the non-industrial sources, which produce a much broader distribution of both toxic and relatively non-toxic CDDs. The non-industrial sources generally produce measureable quantities of 2,3,7,8-TCDD but it is not one of the major isomers present. The importance of the non-industrial sources of CDDs is related to the mass transport rate of the toxic CDD isomers and analogs (e.g., 2,3,7,8 -TCDF) rather than the proportionality of the individual components. Because of the broad distribution of the toxic CDDs present in the solid and gaseous effluents from combustion where 2 ,3 ,7 ,8 -TCDD represents a very small part of the total quantity of toxic PCDD isomers and PCDF analogs present, attempts have been made to estimate the "toxic equivalents" of the PCDDs and PCDFs emitted relative to 2 ,3 ,7 ',8 -TCDD. One such study is that of Olie, Lustenhouwer and Hutzinger, who calculated the "toxic equivalents" in the fly ash and flue gases from municipal incinerators in the Netherlands. 76 Since the toxicity of nine of the 19 probable > toxic PCDD isomers and only eight of the 39 probable toxic PCDFs have been studied, an assumption was made that the other probable toxic PCDD and PCDF isomers have a toxicity roughly 10% of that of CONFIDENTIAL SUBJECT TO PROTECTIVE ORDER. 5 4 ATTORNEY WORK PRODUCT ATTORNEY-CLIENT PRIVILEGE 2,3,7,8-TCDD. From the concentrations of the isomer clusters of PCDDs and PCDFs with the same number of chlorine atoms (TCDDs, PeCDDs, HPDFs, etc.) determined in the fly ash and flue gas extracts and knowing that 2,3,7,8-TCDD is about 3.3% of the tetra isomers, the "toxic equivalents" relative to 2 ,3 ,7 ,8 -TCDD were calculated. Table 5 gives the results. Table 5 "Toxic Equivalents"(as 2,3,7,8-TCDD) in Emissions from Municipal Incinerators in the Netherlands Compounds TCDDs PeCDDs HCDDs HpCDDs TCDFs PeCDFs HCDFs HpCDFs To'tal Fly Ash_________ Total Toxic 2 ,3,7,8 Amt./Yr., Equiv./Yr. TCDD (kg) (kg) (kg) 5.6 15.2 36.2 45.6 10.4 18.7 27.5 18.8 0.3 0.8 2.5 2.3 0.2 0.9 2.1 0.9 0.2 178.0 10.0 ___________Flue Gas Total Amt./Yr. (kg) 0.8 3.4 6.2 4.9 2.3 3.8 7.4 4.1 Toxic 2,3,7,8 ' Equiv./Yr. TCDD (g) (*) 41 26 170 434 245 48 190 555 205 32.9 1888 Assumptions : 2,3,p7,8 -TCDD = 3.3% of total TCDDs. All "toxic" compounds have a toxicity of 10% compared to 2,3,7,8-TCDD. Isomers are equally distributed in each isomer group. Based on the above results, the authors estimate that the toxicity of the fly ash extracts is about 50 times greater than that expected from the 2,3,7,8-TCDD content. For the flue gas, the factor is about 80. It is very obvious that the penta, hexa, and hepta CDDs and CDFs contribute greatly to the toxicity of the emission extracts, and 2,3,7,8-TCDD probably contributes very Little. An article by Helder, 36 published in 1982, confirms this estimate. Fly ash extract was extremely toxic to rainbow trout yolk sac fry, which was attributable for the greater part to the other chorinated PCDD congeners and only a small part to 2,3,7,8-TCDD. B. Histopathology and Cause of Death The toxicological and histopathic response to 2,3,7,8-TCDD and its congeners include 79: "'N 1. Lethality - The ultimate target organ, that tissue whose functional disruption leads to death, is unknown. Animals show a slow wasting syndrome with a latent period before death. 023873 SUBJECT. TO PROTECTIVE ORDER. 6 %* if / ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE 2. Lymphoid involution - produces thymus, lymph node, and spleen atrophy accompanied by suppression of cellular immunity. 3. Embryotoxicity and teratogenesis - it is a potent embryotoxin and teratogen in mice, rats and chicken embryo. 4. Chloracne and hyperkeratosis - most common and characteristic sign of toxicity in humans. Also observed in rabbits, monkeys and hairless mice. 5. Liver damage - heptacellular necrosis observed in rats, mice and rabbits. Also, a disturbance in hepatic porphyrin synthesis (hepatic porphyria). 6 . Edematous syndrome - a characteristic lesion in chickens. Also seen in mice. 7. Bone marrow depression - reported in monkeys. 8 . Carcinogenesis 118 - demonstrated in rats treated with 0.1 ug/kg body weight of 2,3,7,8-TCDD for two years, but no adverse effects'at 0 . 0 0 1 ug/kg body weight. NIOSH and EPA water quality criteria currently list it as a carcinogen. Carcinogenesis in humans is indefinite. There is evidence that 2 %3,7,8-TCDD is slowly biologically attacked yielding metabolites appearing in the bile. There appears to be no reports of toxicity on any cell type culture. Most significant is the fact that 2,3,7,8-TCDD is an excellent enzyme inducer, and produces a 10 to 12 fold increase in hepatic aryl hydrocarbon hydroxylase (AHH) and 6 -arainolevulinic acid synthetase (ALAS) activity. This is very important because there is an excellent correlation between the potency of the PCDDs and their congeners to induce AHH and ALAS activity and their toxicity. A comparison of the enzyme inducing efficiency of various PCDDs is given in Table 6 ll7. Table 6 Induction of ALAS and AHH by PCDDs in Liver of Chick Embryos CUorodioni ALAS AHH Uwtakiied -- la THIJ.4 2J.7 Tan 100*4 I0.M WOO tau 100.4.7 Has I0OO.7J 10,4.4,7.4<40pa etn) Ocu 100.4.4.7A4 7 - -- - + -- - + ++ - -- -- -- + _ -- + + + + -- rn Kj 3871 4* ATTORNEY WORK PRODUCT ATTORNEY-CLIENT PRIVILEGE III. NON-INDUSTRIAL SOURCES Until about six years ago, most of the attention given to the sources of CDDs centered around their production as impurities in industrial processes.. This was brought about primarily because of certain accidents which occurred such as the "run-away" reaction at Monsanto Company's Nitro plant in 1949 in the production of sodium 2 ,4 ,5 -trichlorophenate (NaTCP), an intermediate in the production of 2,4,5-T, and a similar, more violent incident in Seveso, Italy in 1976. In 1977, Olie and co-workers 73 reported the finding of CDDs as trace components in the fly ash and flue gas from some municipal incinerators in the Netherlands. This was the first reported case of the observance of CDDs from a non-industrial source. Since then, numerous studies from industrial, academic and government laboratories all over the world have reported finding CDDs and the closely related chlorinated dibenzofurans (CDFs) in many different forms of combustion - some being adequately confirmed and others remaining controversial. To the best of our knowledge, the only non-industrial sources of CDDs reported to date involve combustion in one form or another. The possibility of CDDs being formed in biological environmental processes has been suggested 54 , but no reports of such findings have been documented. In this section, we will report the findings of the researchers regarding the various combustion sources where CDDs have been identified. The major combustion sources implicated will be discussed individually and chronologically. Laboratory studies of the combustion or thermal destruction of chemical compounds to determine the mechanism of formation of CDDs will be covered in the Chemistry section of this report. Laboratory studies directly related to non-industrial sources of CDDs will be covered under the specific source. A. Municipal Incinerators The first report citing CDDs from a non-industrial source was by Olie 73 and co-workers in 1977. They collected fly ash from the electrostatic precipitator (ESP), and particulates and liquid condensate from flue gas from three municipal incinerators in Arnheim, Amsterdam, and Alkmaar, Holland. The incinerators were modern, large and representative installations geographically distributed in different areas of Holland. Chemical waste, as a rule, was not incinerated. The emission samples were extracted with appropriate organic solvents. After removing most of the solvent, the extracts were examined by packed column gas chromatography/ mass spectrometry (GC/MS) without further purification. In the fly ash were found tetrachloro-, pentachloro-, hexachloro-, heptachlorodibenzo-p-dioxin isomers (TCDDs, SUBJECT TO PROTECTIVE ORDER. 9 023875 * ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE Table 11 PCDDs and PCDFs in Ash Samples of the Zurich- Hagenholtz Trash Incinerator (in ppb)* Ash 1 (furnace) 900C. Ash 2 (behind superheater) 480C. Ash 3 (in front of economizer) 480C. Ash 4 (behind electrofilter) 260C. OCDD HpCDDs HCDDs PeCDDs TCDDs 6 n.d. n.d. n.d. n.d. 18 n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. 120 60 30 8 2 OCDF HpCDFs HCDFs PeCDFs TCDFs n.d. n.d. n.d. n.d. n.d. 15 12 1 n.d. n.d. n.d. n.d. n.d. n.d. n.d. 10 40 30 4 1 Ju Samples of Oct. 5, 1977 n.<d. = not detectable. Detection limits: 0.2 (TCDDs/TCDFs) to 1 ppb (OCDD/OCDF). From Table 11, the largest concentration of PCDDs and PCDFs is in the fly ash from behind the elctrofilter, the last attainable spot before the chimney. It is also the coolest sample point. Ash samples 1 to 3, taken at considerably higher temperatures, contained only the most chlorinated (least volatile) PCDDs and PCDFs. Considerable quantities of chlorobenzenes (PCBzs), chlorinate biphenyls (PCBs) and chlorinated naphthalenes (PCNs) were also found. The authors indicated that they did not know if the PCDDs were present in the waste and passed through the incinerator unchanged or were formed in the incinerator from chlorinated aromatic precursors. They suggested that the latter was most likely. Another publication in 1978 by the authors together with C. Rappe- 80 gives the identification of many of the PCDD isomers found in the Zurich-Hagenholtz incinerator ash by sharpening up the GC/MS analysis by using the SIM technique with standards. Although the data was not quantitative, 12 of the possible 22 TCDD Isomers were present. The main isomers were 1.3.6.8 - and 1,3,7,9-TCDD. A small amount of 2,3,7,8-TCDD was identified. Up to 10 of the possible 14 PeCDD isomers were found with 1,2,3,7,8 - and 1.2.4.7.8 - PeCDD being identified. Up to 8 of the possible 10 HCDD isomers were present, the main ones being 1 ,2 ,3 ,4,6,8 -, 1,2,3,6 ,8 ,9-, 1 ,2,3,6,7,8- and 1,2,3,7,8,9-HCDD. Both possible HpCDD isomers (1,2,3,4,6 ,8,9- and 1.2.3.4.6.7.8 - HpCDD) and OCDD were also present. Most important was the fact that micropyrolysis of a mixture of the potassium salts of 2,4,6-tri, 2,3,4,6-tetra and pentachlorophenol gave the same pattern of main isomers identified in the fly ash from the Zurich-Hagenholtz incinerator as well as from an industrial waste incinerator. This was the first concrete evidence that chlorophenols could be precursors of PCDDs in combustion. In the section, we will elaborate on this point. IA J SU B JE C T TO PROTECTIVE ORDER. 13 C23S76 *3 ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE In November, 1978, Dow issued a report by its Chlorinated Dioxin Task Force to the Michigan Dept, of Natural Resources, which indicated that PCDDs are found in widely differing incineration processes including municipal incinerators. This report first mentioned their "Trace Chemistries of Fire" hypothesis, which has led to much controversy. We do not have a copy of this original report, but the technical data appeared in an article by Bumb, et al, in Science in 1980 13. Dow's "Trace Chemistries of Fire" hypothesis says "Chlorinated dioxins appear to be ubiquitous. Their ubiquity is due to the existence of natural phenomena, trace chemistries of fire, which consist of numerous chemical reactions occurring during combustion of concentrations as low as 10" 10 percent." Dow supports their case for the "Trace Chemistries of Fire" hypothesis by the data presented in Table 12, which shows that PCDDs are emitted from various combustion sources, as well as being found in soil, dust, and wastewater sludge (Milorganite) samples. Table 12 Chlorinated Dioxins in the Environment Sample Mum. ber of sampies Other isomers Apparent dioxin content. ng/g (ppb)* TCDD 2,3,7,8- Total HCDD HTCDD OCDD Soil Rural (Gaylord, Michigan) Urban (Lansing and East Lansing, Michigan) Major metropolitan (Chicago. Illinois) Dow Chemical (Midland, Michigan) Dust Dow Chemical laboratory Midland, Michigan Metropolitan (Detroit, Michigan) Metropolitan (St. Louis, Missouri) Metropolitan (Chicago, Illinois) Wastewater treatment sludge Commercial sludge fertilizer (Milwaukee, Wisconsin) Incinerators, powerhouse Dow powerhouse Dow rotary incinerator stack (present normal operation with supplemental fuel)$ Dow stationary tar burner stack (normal operation with supplemental fuel) U.S. municipal incinerator (electrostatic precipitator) (Nashville. Tennessee) European municipal incinerators Mufflers Diesel truck muffler Auto muffler Other sources Home fireplace soot Home electrostatic precipitator Cigarette smoke Charcoal-broiled steak J 5 Ss 0.8-18 6 0.5-2 2 4 1 0.16 2 1 0.29 0.3-100 0.7-3 0.12 0.02 N.D.t N.D. 0.005-0.03 1-120 * M 0.03-0.04 (0.02) N.D.-0.03 0.3 0.04 (0.04) N.D. 0.03-1.2 0.03-0.3 7-280 9-35 0.2-0.4 N.D.-0.3 2 N.D.-0.3 0.31 2 N.D.-0.05 0.03-2 0.1-3 70-3,200 N.D.-0.2 0.05-2 0.4-22 490-20,000 140-1,200 2-4 0.3-4 34 0.6-3 650-7,500 20-30 0.1-4 210 3-8 30 180 1 38(20) 5 N.D. (10) 38 (20) N.D. (2) 5 N.D. N.D. N.D. 1 7.3 0.4 7.7 2-20 2 0.02 0.003 0.023 4 N.D.-0.004 N.D.-0.004 N.D.-0.008 2 N.D.-0.3 1 0.4 (0.4) 2 N.D. 4 N.D. N.D.-0.I 0.6 N.D. N.D. N.D.-0.4 1.0 N.D. N.D. 2 4 24 1-5 4-100 9-950 1-20 27-160 190-440 14 28 30 30-200 60-130 4CM20 0.020 N.D. 0.100 0.26 0.003-0.01 0.02-0.07 0.2-3 0.7-16 34 430 0.004-0.008 0.009 N.D. N.D. 0.9-25 1.300 0 02-0 05 0 0) (0.03) miJhi|>lmnpl were Analyzed from similar tourcei, the range of observed value* i* shown. tN .D. indicale* that ih* signal observed was less ihan 2.5 1.U m ju oi gencraMy observed wereTCDD, 0.001 10O.OI ppb; HCDD, 0.01 to 0.09 ppb; H,CDD. 0.003 to 0.0) ppb: and OCDD. 0 01 io 0 0) ppb. Limili or m u tim i mu within these ranges are shows parenthetically after the value reported for signals between 2.J and Li)times noise. tD iia without supple mental fuel no included here because this practice has bee* eliminated. /* .'-a q SU B JE C T TO PROTECTIVE ORDER. 14 023877 ATTORNEY W ORK PRODUCT ATTORNEY-CUENT PRIVILEGE Note that Dow examined the fly ash from a municipal incinerator in Nashville, Tenn. and found PCDDs. Also approximate data for European municipal incinerators are shown for comparison. Notice that 2,3,7,8-TCDD is only a small part of the TCDDs present in the fly ash from the Nashville incinerator. The analytical methodology including sampling was reviewed by a group of scientists, who found it to represent state-of-the-art. However, in October, 1979, A. Hay 35 reported that C. Rappe was severely critical of the Dow study after reading the original November, 1978 report. He questioned the methodology, particularly the chromatographic separation techniques used by Dow to identify PCDD isomers. He said Dow's techniques were poor (see Figure 1), leading to an over estimation of the amount of 2,3,7,8-TCDD present in the fly ash Figure 1 from municipal incinerators. Rappe also indicated that the ubiquitous nature of PCDDs was far from proven. At that time, Rappe considered chlorinated phenols and diphenyl ethers to be the main precursors to PCDDs in municipal incinerator emissions, based on laboratory combustion experiments on these two classes of compounds giving PCDD isomers in similar proportions to those found in commercial incinerators. He said he reported these findings to Dow, who apparently ignored them because there are no analyses for these apparent precursors in the Dow report. ,*?*> T k T T H T T T T * ^ m TV T7 AW SU B JE C T TO. PROTECTIVE ORDER 15 C23878 fr ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE In 1982, an article by Crnmmett 23 reviews much of the data presented by Bumb, et al, 13 and elaborates on the analytical methodology. The analytical procedures employed several liquid chromatographic (LC) cleanup processes and various column GC/low and high resolution MS techniques. Crummett reemphasizes the apparent ubiquitous nature of PCDDs and cites Mahle and Whiting's study 67 on the pyrolysis of essentially chlorinated dioxin-free coal in the presence of sodium chloride, hydrogen chloride and chlorine giving rise to PCDDs. He says these studies support the "Trace Chemistries of Fire" hypothesis and the fact that PCDDs appear to be ubiquitous is reasonable. Table 13 C hlorinated O io n n s Fro * Thermal O atdition and Chlorination of Cl N ite riil He ted C l, A ir COfrl * A ir Coil * A ir * MiCl Coil * A ir * MCI COil A ir t c l , Aooirent C h lo rin ite d D l o i in i, nq/q (ppb) TCDQ HCDO H,CDD OCDD tiD(n.Q7) K0(0 .2 ) NO(0 .3 ) N0{0.5) ND(0.Q8) N0(0.1) 0 .6 (0.3 ) 1 .3 (0 .7 ) N 0{0.07) NO(0 .2 ) 0.5 (0.2 ) 2 .7 (0 .6 ) 1 .4 (0 .0 0 6 ) a- 26 64 1.2 29. 91. 290. In 1979, Eiceman, Clement, and Karasek 20 analyzed fly ash samples from municipal incinerators in Canada, Japan and the Netherlands. The Canadian samples were from two city incinerators, one located in a heavily industrialized area and the other located in a more domestic region of the city. The Netherlands sample was from 0. Hutzinger at the Univ. of Amsterdam. The five samples were extracted with benzene, and the concentrated extracts examined by GC/MS using the SIM technique. PCDDs and PCDFs (Cl4 through Cl8 isomers) were present in all samples. This suggested to the researchers that PCDDs and PCDFs are formed universally during incineration although the combustible materials and conditions may vary. In 1981 and 1982, these researchers extended their studies of PCDDs from Canadian municipal incinerators 19j 21 . Eight samples of fly ash were taken weekly from a single municipal incinerator in Ontario, Canada. Table 14 shows the results using GC/MS analysis with SIM technique. The values ranged from 0.4ng/g for OCDD in sample 7 to 53ng/g for HCDDs in sample 3. The largest change in concentration of a chlorine number isomer cluster occurred in the HCDDs where samples 3 and 7 differed by 51ng/g. PeCDDs and HCDDs were present in the highest concentrations. PAHs were also measured. D N F ID E N IIM SUBJECT TO PROTECTIVE ORDER. 16 C23e 9 * ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE Table 15 Representative Quantitative Results of Analyses of PCDDs and PCDFs(ppb) in Solid Emissions from 25 Municipal Incinerators Isomer Groups TCDDs PeCDDS HCDDs HpCDDs OCDD TCDFs PeCDFs HCDFs HpCDFs OCDF unpublisheda Solid Emissions____ fly asha part, matter (electr. prec.) (stack) S 64 no 31 182 488 80 326 1200 190 288 902 266 106 110 13 111 223 42 196 510 109 361 670 89 177 407 20 18 26 10C 800 1370 1370 , 310 460 960 1600 1130 140 a. low, medium and high values chosen from 80 analyses (over 25 installations studied) In 1980 and 1981 16 83 A. Cavallaro studied the emissions from various sources including municipal incinerators in the Lombardy Region of Italy. The slag, fly ash and liquid condensate samples were extracted and the extracts purified by layered column liquid chromatography and analyzed for PCDDs and PCDFs by HRGC/LRMS with El or MNCI source and SIM technique with sensitivity in the pg (ppt) range. One of the objects of these studies was to determine if PCDDs and PCDFs were present in the emissions of all types of incinerators fed with different type wastes. Every examined sample from the different incinerators exhibited a similar profile concerning the presence of PCDDs and PCDFs, a variable number of isomers from TCDDs to OCDD and from TrCDFs to OCDF being detected, with many specific isomers identified. Generally the hexa, hepta, and octa CDDs and CDFs predominate over the penta and tetrachloro isomers. This has significance because degradation proceeds by dechlorination rather than ring cleavage, thus the higher chlorinated species (higher than tetra) degrade through the tetrachloro isomers. Table 16 gives the analytical results from 6 municipal solid waste incinerators. Notice that in nearly all cases the vapor phase (condensate) contains considerably larger amounts of PCDDs and PCDFs than does the dusts (solid phase). Many researchers have only looked at fly ash because sampling is easier. SUBJECT TO PROTECTIVE ORDER. 18 C2388U  ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE Table 16 Analyses of Emissions from Six Municipal Incinerators in Lombardy Region (Milan) of Italy ation n. 1 n .2 n3 n .4 n .5 t D. 6 T y p e of sample Dusts from d^gg traps Emitted dusts ng/Nm^U Emitted condensate ng/Nfn' U Dusts from dust traps Emitted dusts ne/Nm^U Emitted condensate ng/Nm' U Dusts from duspraps Emitted dusts ng/Nm^U Emitted condensate ng/Nm' u Dusts from dust traps ppta Emitted dusts ng/Nm-^U Emitted condensate ng/Nm' u Dusts from dust traps Emitted dusts'ng/NmJU Emitted condensate ng/Nm *U Dusts from dust traps PPG. Emitted dusts ng/NmJU Emitted condensate ng/Nm1U r 4-co d l.l 19,6 0,25 172,2 17,0 N .n. 0,037 19,0 46,4 10,9 60,0 0.7 0,34 9.6 N.D. N . . 19,0 3 CDD 2,7 4CDD >1.5 iCDD 1.03 27.9 176,2 159.6 1.7 172,3 294,0 12015,0 8.9 575.0 107.0 26620,0 828,0 0,92 t ,8 3.1 0.3 6.7 0.2 40.0 6542.0 124.0 65.4 2496.0 7,9 2.8 0.54 3.2 33.0 1390.0 167.0 0,05 o,o:> 0,007 2,4 196.0 9.9 21.0 328.0 46.0 N.D. N.D. 0,0012 0,010 0,28 N.D. n .o 480;0 6.0 ICDD 8.0 63.9 295.0 7312.0 1179.0 1.5 1.7 776.0 841.5 39.0 2703.0 0.1 173,0 244.0 5,86 .0.51 71.0 * * not investigated N,D. not determined 4CDI` aCDF 2,2 59.3 0.46 75.0 15.8 2863.0 108.6 4390.0 0.8 2.57 33 0.08 429.0 1010,0 61.7 3.7 255,0 0,06 I8 I4 .Q 1760.0 1,18 75.3 0.0015 3.2 305.0 89.0 N.D. N.l). 1.93 N.D. 27.0 24.0 From 1980 to 1982, Liberti, Brocco and co-workers u >58*59,60,61,62 also investigated the emissions from municipal incinerators in Italy. The emissions included bottom ash (from furnace), fly ash (from ESP), sludge (fly ash collected by water spraying the fumes), particulates (from stack fumes) and organic vapors (condensed from stack fumes). The extracts of the samples were cleaned up by use of multilayered column liquid chromatography and analysis was by capillary HRGC/MS with El source and MID technique. In one of their publications 58 , they analysed the starting refuse to be incinerated and found no PCDDs or PCDFs. This is one of the few reports which clearly shows that the PCDDs and PCDFs are formed "by pyrolytic reactions of precursors". These investigators support Dow's "Trace Chemistries of Fire" hypothesis, and seem to concur that PCDDs are ubiquitous. They say "Our investigations show that the incineration processes of urban wastes definitely yield a "A number of PCDDs and PCDFs through a variety of possible reactions, the concentration being low definitely but much larger than those usually found in the environment and arising from other sources." CONFIDENT*** SUBJECT TO PROTECTIVE ORDER. 19 C23S8t ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE Table 19 62 gives more specific analyses of emissions of an urban Italian incinerator. The dry wastes are a sample of the wastes which were incinerated. Notice that no PCDDs were found; however, PCBs and chlorophenols were found. Also the higher the PCBs in the emission sample, the higher were the PCDDs in the same sample. Likewise, the highly chlorinated PCDDs are present in the largest concentration with OCDD being the most. Table 19 Concentration of Some Chloroorganic Compounds in the Emissions of an Urban Italian Incinerator Peak No. Compounds Dry wastes (nq/g) pcbtot 200 13 2,3,7,9- and/or 1,2,3,4-tetra-CDD - 16 not id. - 17 penta-CDD - 20 not id. - 21 1,2,4,6,7,9-hexa-CDD - 22 not id. - 23 hexa-CDD - 24 1,2,3,6,7,8-hexa-CDO - 26 not id. - 27 hepta-CDD - 28 hepta-CDO - 32 octa-COO * Ash (ng/g) 550 trac. 2 4 4 2 6 4 4 15 8 8 35 Fly ash (ng/g) 1,100 trac. 9 6 -7 7 24 15 7 210 25 30 50 Fumo part, matter (ng/g) 7,800 40 115 110 160 150 400 250 140 - 400 500 900 In another publication by Liberti, Brocco and co-workers 60 , various emissions from Italian urban waste incinerators classified as follows were analysed: A. incinerators -from highly populated areas where wastes are incinerated without any treatment. B. incinerators where waste is mainly agricultural products which are burned without any treatment. C. incinerators where waste passes through a recycling process where paper, plastics and vegetable material are removed. D._________________ incinerators where the input material is the residue of wastes which went through an homogenization and __ oxidation process (the unfermented residue being CONFIDENT!8, ' 21 SUBJECT TO PROTECTIVE ORDER. 49 ATTORNEY WORK PRODUCT ATTORNEY-CLIENT PRIVILEGE The various emission samples were extracted, and the extracts purified by passage through silica gel and alumina chromatographic columns. Analyses were performed by packed or capillary column GC/MS with El source and SIM technique. Table 20 gives the analytical Table 20 Concentration (ppb) of PCDDs and PCDFs in Emissions from Italian Municipal Incinerators Burning Wastes of Different Types Encinerator A A A 5 B C D D Emission analysed n. Cl-4 fly ash ash Particulated in fumes fly ash sludge sludge Particulated in fumes sludge . 20 16 n.e n .e n.e n.e n.e n.e PCDD PCDF S 6 ? 8 n. Cl-7 100 160 230 490 20 33 90 340 80 130 290 510 n ,e n.e 30 40 n.e n.e 10 15 n. e n.e 6 12 n.e n.e n.e n.e 55 1l 70 25 110 n.e n.e. n.e n.e n.e a 50 50 60 n.e n.e n .c n.e n.e n . e . nec evaluable results. These researchers propose the following simple mechanism of formation of PCDDs and PCDFs: Precursor A + (Organic matrix) Phenols Polyphenols Materials with phenolic structure Precursor B -----(Clg or HC1 donor) PVC Other chlorinated plastics Chlorinated pesticides > PCDDs + PCDFs Examples of Precursor A are tannins widely diffused in vegetables, flowers and fruits which upon heating yield phenols and polyhydroxyphenols. This simple hypothesis explains the analytical results observed from the various types of incinerators. Type B incinerators give lower concentrations of PCDDs and PCDFs than Type A incinerators. Vegetable matter will have only small amounts of precursor B, thus only small amounts of chlorinated compounds will appear in the emissions. For type C and D incinerators, most A precursors are removed and combustion of the material mainly containing Precursor B gives rather small amounts of chlorinated organics. Another report 11 relates the amount of HC1 present in the incinerator stack fumes to the concentration of PCDDs and PCDFs found in the emissions. It would seem that the higher the concentration of HC1 present in the fumes the higher the concentration of chlorinated organics. Table 21 shows exactly this relationship. cgnf: SUBJECT TO PROTECTIVE ORDER. 22 023833 44 ATTORNEY WORK PRODUCT ATTORNEY-CLIENT PRIVILEGE Table 21 Cencentrations of PCDDs, PCDFs and HC1 in Emissions from Municipal Incinerators Burning Various Type Materials erators Ke-CDD He-CDF Hp-CDD 0CDF OCOD HCl -CDF (ppbi (pnb) (ppb) (p?bi (ppb) (ppm) 9C 63 U S 65 152 10-200 *2 39 30 42 31 118 10-200 i *; *3 3r 35 90 10-200 0 ---- 6 -- ' 12 0- 10 C ---- 5 --` 5 0-10 A * ine. wither any treatment B * ine. after reeve ling C * ine. after compost production s In December, 1980, A. D. Little, Inc. issued a report for the American Society of Mechanical Engineers on the MState-of-the-Art of Dioxin from Combustion Sources" 66. This was done because of the great concern by the solid waste resource recovery industry regarding the presence of 2,3,7,8-TCDD in particulates. This report does a respectable job of summarizing the published literature on the status of knowledge regarding all facets related to PCDD emissions - identifying all documented combustion sources; the chemistry of PCDDs; the sampling, extraction,clean-up of extracts, and analysis for PCDDs; and toxicological considerations; as well as conclusions and recommendations. Naturally, this was most helpful in assuring that we covered the literature thoroughly up to June, 1980. In 1981, Redford of the EPA and co-workers 95 collected gaseous, aqueous, and solid samples of the emissions from a municipal incinerator burning 100% raw refuse and one burning 85% coal and 15% refuse. Table 22 gives the results of analyses of the purified extracts of the flue gas by capillary HRGC/HRMS in the SIM mode. No PCDDs or PCDFs were found in the extracts of fly or bottom ash or aqueous emissions. Also no PCDDs or PCDFs were found in any of the samples from the coal/refuse derived fuel (RDF) facility. SUBJECT TO PROTECTIVE ORDER. 23 C23884 ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE protocol requiring much handling and analyses by two different laboratories, which obviously did not agree. This suggests that the extraction procedure or analytical methods leave something to be desired. The authors hint at a possible inefficient extraction procedure. Also in 1982, Gizzi, et al, 32 analysed the stack emissions from a modern municipal incinerator in Como, Italy. Samples were taken over a nine month period on two consecutive days each month. Sampling method, extractions, and clean-up of the extracts were state-of-the-art. Analyses were performed by packed column GC/MS in the SIM mode, the data being comparative but not strictly quantitative. Tables 25 and 26 show the concentrations of PCDDs and PCDFs in the emissions. Table 25 Concentrations of PCDDs in Emissions from a Municipal Incinerator in Como, Italy (ng/Nm3) Sample Tetra Penca Hexj Htpta G^ta Total 1 2 3 4 5 6 7 6 9 10, * 11 12 13 14 15 lh 17 He anC 17 ( u . 9)a 32 19 683 (16. 3) 1127 (10.7) 31 <1 1 .U 12 ( 7. 9) 10 ( 6 ..7) 134 (15..9) 23 (12..4) 43 ( 4..1 ) 7 ( 9,.6) 12 (13. n 15 (1 2 ..2) 11 (13..1 ) 19 ( 7..0) 13 ( 7..6) 128.4 n .d .b n ,d. 30 (14. 0) 701 (16. 6) 20S9 (19. 8) 37 (13. 2) n .d. n .d. 230 (27. 2) 52 (27. 9) 112 (1 0.8) 13 (18. 3) 20 (2 1 .7) 23 (1 0.7) 19 (22.6) 35 (1 2 .,9) 23 (13. 5) b 18 (15 .6) 97 (37 -5) 89 (41 6) 1 3*91 (33 .2) 3805 (36 -1 ) 66 (23 -5) 42 (27 .6) 26 (17 .4) 290 (34 .4) 69 <37 -1 ) 161 (15 .5) 19 (26 -S) 24 (26 -1 ) 31 (25 .2) 15 (17 -9) 46 u t . .9) 33 (19 3) 366 15 (13 .2) 65 (25 .0) 45 (21 .0) 1045 (25 .0) 2884 (27 .4) 73 (26 .1 ) 37 (24 -4) 43 (28 -9) 125 (14 .6) 19 U O .2) 351 (33 .8) 13 (15 .3} 14 (15 .2) 21 <i7 .1 ) 16 (19 .0) 63 (23 U 36 (22 .2) 28.3 64 (56. 1 ) 97 (37. 5) 50 (23. 4) 365 ( 8 .7) 6 31 ( 6 .0 ) 73 (26. 1) 61 (40. l) 70 (47. 0) 65 ( 7. 7) 23 (1 2 .4) 371 (35. b) 15 (2b. 6) 2 C 3 . 9 ) 3 1 (26. f) 23 (27. 4) U-7 (40..1 ) ^ (37.,4) 1 2 5 . Sr 114 259 214 4185 10536 280 152 149 844 186 1038 71 92 123 64 272 1 71 1104 1 The numbers i(1 brackets give the pe rcen tape Cif the individual c. 1n cse s .1 f is;T e r > . 1 n.d. - Scnsitlivicy limi ts not delr e m i nate bL`c .i1iie no anal yci.cal rd w:is avail ah lo. For c The tlic sari- reason mean was calc averag ulated e as values signin g foar v P5CDDS alue eq hive not uivalent Seen Calculated. to one half the detection linit to samples 2 and 3. rH'I i SUBJECT TO PROTECTIVE ORDER. 26 -o VJ Ad 388o 4 ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE Table 26 Concentrations of PCDFs in Emissions from a Municipal Incinerator in Como, Italy (ng/Nm3) Sarplc 1 2 3 4 5 6 7 S 9 10 11 12 13 14 13 16 17 i'< an Tirtra PeiUi He** H*pta Oct* 9* (11 . n b 130 (15 .4) 1S3 (17 .7) 171 (20 .0) 1046 (26 -2) 620 0 5 5) 2846 (20 -9) 2261 (23 .0) 56 (10 .0) 74 0 3 .3) 25 ( 7 .8) 54 0 6 .9) 17 (10 .6) 17 (10 .6) 288 (31 .6} 214 (23 .5) 53 (25 -5) 6 (22 -1) 102 (21 -3) 103 (21 .5) 30 (23 .6) 27 (21 .3) 24 (21 1) 55 (24 .0) 25 (21 9) 57 (24 .9) 36 (21 .7) 52 O l 3) 4 ( 2 0 .9) 93 (26 .3) 46 (19 . 2 ) 60 (25 - 0 ) 309 250. 3 204 (24 .2) 196 (23..1) 890 (22..3) 2928 (29..8) 177 (31 .7) 73 (22,8) 25 (15,.6) 164 (18, 0) 3? (18,.8) 101 (21..0) 22 (17, 3) 23 C O . 2) 37 (16 .2) 25 (15,.1) 80 <22 . 6 ) 41 (17 . 1 ) 314.2 24 5 (29. 0) 171 (20.3) 2 76 (32. 2) 60 ( 7.0) 692 (17. 3) 74 9 (18.7) 1414 (14. 4) ` 382 ( 3.9) 141 (25. 3) 110 (19.7) 75 (23. 4) 47 (29. 4) 93 (29.1) 54 (33.8) 178 (19. S) 68 ( 7.4) 36 (17. 3) 34 (16.3) 100 (20. 8) 74 (15.4) 29 (22. 8) 19 (15.0) 25 (21. 9) 1 7 (14.9) 41 (17. 9) 39 (17.0) 31 (18. 7) 22 0 3 . 2 ) 64 (18.,1) 43 (12.1) 48 (20. 0) 45 (18.7) 215.1 123.8 Total 844 857 3997 9831 558 320 160 912 208 480 127 114 229 166 54 240 1212. * Simple 1 was not analyzed because PCDF standards were not available at that time. The numbers in brackets give the percentage of the individual classes of isomers. It is worth emphasizing that considering the stack flow 33,000 Nm3/hr.) on the days of samplings U and 5, about 2 g each of PCDDs and PCDFs were released during 8 hours each day. On these days, the incinerator operating temperature was the lowest (~ 500C.), probably because of the high moisture content of the waste, which contained a lot of vegetables, and because of rainy weather. In 1983, examination of the emissions from municipal incinerators continued strongly, but not much significantly new knowledge was gained. Chiu, et al, 17 analysed emissions from two Canadian municipal incinerators (a large, modern heat recovery incinerator and a small, modular controlled-air one) along with coal-fired power and heating plants. The sample preparations and clean-up procedures are well established and analysis was performed by HRGC/LRMS using negative oxygen chemical Ionization in the MID mode. Table 27 gives the analytical results, which contain no surprises. CONFIDENTIAL SUBJECT TO PROTECTIVE ORDcR. 27 C238815 B. Industrial Waste Incinerators ATTORNEY WORK PRODUCT ATTORNEY-CLIENT PRIVILEGE The first reported findings of PCDDs in an industrial waste incinerator was by Buser and co-workers in 1978 80*81. They analysed the fly ash from an industrial heating facility burning used industrial oils near Aarau, Switzerland. The sampling of the fly ash, its extraction and clean-up of the extract were state-of-the-art. Analysis was by capillary HRGC/MS with El source and SIM technique-certainly state-of-the-art. Table 36 gives the analytical results on the fly ash. This is one of the few reports that shows DCDDs and TrCDDs formed by combustion. Chlorobenzenes and chlorophenols were also identified in the fly ash. The amounts of PCDDs and PCDFs were higher in the fly ash from this industrial waste incinerator than in a municipal incinerator. Table 36 PCDDs and PCDFs in Fly Ash from a Switzerland Industrial Waste Oil Incinerator (ppb) DCDD TrCDD TCDD PeCDD HCDD HpCDD OCDD 10 20 100 160 180 130 40 TCDF PeCDF HCDF HpCDF OCDD 100 100 70 50 n.d. n.d. = not detectable The Dow reports of 1978, 1980 and 1982 on the "Trace Chemistries of Fire" 13i23 previously mentioned confirm the findings of Buser and co-workers. Dow analysed the particulates from a stationary tar burner and a rotary kiln incinerator at their Midland, Michigan plant. Their analytical methodology was state-of-the-art as previously discussed. Table 37 gives the detailed results of their analysis for PCDDs 66 [see Table 12 (p.14) for a comparison of analytical results with other combustion sources]. A very dramatic feature of the data in Table 37 is the reduction in total PCDD levels when clean supplementary fuel, rather than tarry waste, is burned in the secondary combustion chamber of the rotary kiln incinerator. The supplemenatary fuel causes a higher temperature in the secondary combustion chamber, thus PCDDs are more completely destroyed or cannot form to as great an extent. In 1981, Hryhorczuk and co-workers 38 examined scrapings from two of three stacks and from all three furnaces of a wire reclamation incinerator in the U.S. midwest. The samples were analysed for TCDDs and TCDFs by Gross of the University of Nebraska using state-of-the-art GC/HRMS techniques. Table 38 gives the results on two of the ash samples. The authors indicated that these compounds could have entered the incinerator intact in the fuel or charge, or could have formed from precursors in the wire insulation. CONFIDENTIAL SUBJECT TO PROTECTIVE ORDER. 35 023887 01 ATTORNEY W O RK PRODUCT ATTORNEY-CLIENT PRIVILEGE Table 38 TCDDs and TCDFs in Ash from a Wire Reclamation Incinerator (ppt) Sample Furnace No. 2 Stack No. 2 Total TCDDs 58 410 Total TCDFs 730 11,600 In 1982, Barnes,et al, 7 reported that the U.S. EPA, in 1979, carefully examined the emissions of two commercial incinerators used to burn PCB wastes. In each case, detectable levels of PCDDs and PCDFs were found. Also in 1982, Rappe,et al, 94 analysed a series of samples taken during a test study of the thermal decomposition of PCBs in a rotary cement kiln in Norway. The temperature of the oven was at 1450C. No PCDDs or PCDFs (tri-to octachloro isomers) could be found in any of the samples. This suggests that this might be a safe way to destroy PCB containing wastes. In 1983, Tiernan and co-workers 101 investigated the combustion products from incineration of chlorinated chemical wastes. Table 39 shows the analytical results along with the concentration of CDDs/CDFs in the original PCP wastes incinerated. " Table 39 Relative Concentrations of CDDs/CDFs in Chemical Wastes and Combustion Products Thereof M0F CfW:riAr.r (TGCS 4 .S . pentachlorophenol (PCP) Ifl OIL CODs CDFs PCP SLUDGE FROM WOOD ASH FROM treatment comoustion of PLANT PCP WASTES CDOs "CDFs CODs CDFs FLUE GASES FROM INCINERATION OF m ixtures of chlorinateq CHEMICAL WASTES CODs CDFs Tetn Pentd- H(WH9tt- 0.0G28 .C083 0.14 0.064 l.Q 0.0099 0.076 1.0 0.063 0.19 _ ,_ 0.0013 0.023 0.0055 0.062 0.31 1.0 1.0 0.25 0.018 0.16 0.41 0.59 1.0 0.88 1.0 0.63 0.75 0.25 0.80 0.80 0.20 0.40 1.0 0.17 1.0 0.83 0.40 0.33 V Normalized to chlorinated c la ss detected in largest concentration. These data suggest that some degradation of the higher CDDs/CDFs present in the wastes occurred during combustion with an increase in the relative concentrations of the lower CDDs/CDFs (particularly the Cl4 and CI5 isomers). The distribution of CDD/CDF homologs (isomer clusters) in the mass spectrum is distinctly different from those ' observed in the flue gases from municipal incinerators, but the major TCDD isomers were the same (1 ,3 ,6 ,8 -, 1,3,7,9-, and 1,3,7,8 -TCDD). Finally, at the National ACS Meeting in Washington, D. C. last September (1983), Gross, et al , 122 investigated the incineration of chemical wastes. They measured the TCDD and TCDF content of the C O N FIO E N T lM ; 6starting chemical wastes and in the emissions on incineration. The s 40 and 41. Except for one emission SU BJEC T TO PROTECTIVE ORDER. 37 023888 a4 ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE C. Oil/Coal-fired Powerhouses and Utility Facilities The first report of PCDDs detected in oil/coal-fired powerhouses or utility facilities was by Dow in 1978 66 in their original "Trace Chemistries of' Fire" paper. The study was elaborated on in publications of 1980 13 and 1982 23. Dow analysed their Midland, Michigan powerhouse operating under normal conditions with fuel oil and coal. The results are reported in Table 12 (p.14). With regard to HCDDs through OCDD, the concentration of these PCDDs in the emissions from the powerhouse is the lowest of the incinerator type sources; however, the concentration of TCDDs appears to be the highest compared to the., others. This seems unusual. In 1980, Kimble and Gross 39 analysed fly ash from the stack of a 750 MW power plant burning low sulfur, high ash (10%) coal with 50 ppm chlorine content. Quantitative analysis using packed column LRGC/HRMS with SIM mode showed no detectable TCDDs at a detection limit of 0.6 ppt. They did not analyse the fly ash for other CDD homologs. This detection limit is lower by a factor of at least 30,000 than that determined by Dow In their study. Kimble and Gross say that one explanation for the large difference in TCDD concentration may be the nature of the fuel sources. For this reason, the authors cautioned that the finding of significant amounts of PCDDs in emissions from municipal and chemical waste incinerators should not be used to infer that these compounds are also significant products of fossil-fueled power plants. Thus their results contradict Dow's findings. Also in 1980, Lustenhouwer, Olie and' Hutzinger 84 showed that the extraction procedure used by Kimble and Gross gives poor recoveries for PCDD. They also analysed fly ash from a coal-fired powerhouse with methodology suitable for the detection of ppb levels of chlorinated compounds. PCDDs were not detected but they did find chlorobenzenes in the fly ash. The authors did not comment whether oxygen was in excess or not. In 1981, Junk and Richard 46 analysed particulates and gaseous emissions from a utility facility producing 35 MW of electrical power from steam produced by co-burning coal and refuse derived fuel (RDF). Using HRGC/MS with MID technique with detection limits of 0.1 ppb for particulate samples and 0.5 ppt for vapor, no TCDDs were detected in the emissions. The influent coal and RDF also showed no detectable TCDDs. They explain no detectable TCDDs in the emissions by (1 ) the high operating temperature of 1200C. due to the coal supplement and the type of boiler (2 ) sufficiently long resident time (>1.3 sec.) (3) adequate O2 (4) small pieces of RDF produced by shredding. These four conditions have not existed for most incineration processes where TCDDs have been found. They say the one exception may be the negative TCDD results by Kimble and Gross with a presumed higher operating temperature. Prior to this study, very little data on the operating conditions of the various municipal incinerators and power plants where the emissions were found to contain PCDDs had been published. SU B JE C T TO PROTECTIVE ORDER. 39 * attorney w o rk product ATTORNEY-CLIENT PRIVILEGE l a 1981, R e d fo rd , e t a l , 95 exam ined th e g a s e o u s , a q u eo u s, and solid emissions of a utility facility burning 85% coal and 15% RDF. No PCDDs or PCDFs were detected using capillary GC/HRMS with SIM technique at a detection limit of 10 ppt. This finding is thus in agreement with those of Junk and Richard, Kimble and'Gross, and Lustenhouwer. In 1982, Krishnan and Hellwig 55 surveyed the literature regarding trace emissions from various oil/coal-fired type boilers. Using a set of assumptions, national emissions of the trace pollutants were estimated. A "dioxins" emission factor was only reported for coal-fired utility boilers (0.01 ppt/Joule of energy input). This translated to an estimated U.S. national emission in 1978 of --220 lbs. of "dioxins"/yr. This is considerably less than all other estimated trace emissions, such as metals or organics which are known or suspected carcinogens [ As , Be, Cd, Hg, formaldehyde and benzo(a)pyrene}. This survey indicates that coal combustion emits significant quantities of a greater number of trace pollutants than does oil combustion. In 1983, Ahlberg,et al, 1 analysed the particulates from an oil-fired boiler fired with heavy fuel oil (ash, 0.06%; sulfur, 2%) and a pulverized coal-fired boiler fired with Polish coal (ash, 0.13%; sulfur, 0.8%) for 2,3,7,8-TCDD and TCDF. In both cases, neither compound was detected with detection limits at 75-140 pg/g dry fuel for 2,3,7,8-TCDD and 12-22 pg/g dry fuel for 2,3,7,8 -TCDF. State-of-the-art procedures for sampling, extraction and cleanup of the extractants were used. Analysis was by GC/HPMS with SIM technique. More organics were found in the oil soot than in the coal fly-ash. Also, in 1983, Chiu, et al, analysed fly ash from a large, modern coal-fired thermal generating station and from a 30 year old coal-fired boiler at a central heating plant. Sampling, extraction, clean-up and analysis were state-of-the-art. Analysis was performed by HRGC/LRMS using negative oxygen chemical ionization source with MID technique. Table 42 gives the analytical results. Table 42 PCDDs and PCDFs in Coal-fired Power Plants (ppb) Fly ash Source Modern power plant 30 yr.old central heating plant CI4 CDD CDF 18 nd nd ____ CI5 CDD CDF 6 32 nd nd ____Cl6 CDD CDF 6 18 nd nd ____CI7 CDD CDF 5 18 nd nd ____ Clg CDD CDF '' nd nd nd = not detected CONFIDENTIAL SUBJECT TO. PROTECTIVE ORDER. 40 C2389G 4* ATTORNEY w o r k p r o d u c t ATTORNEY-CLIENT PRIVILEGE Regarding the modern power plant, the authors indicated there was not enough data to conclude whether it does or does not represent a potential source of PCDDs/PCDFs. The data reported above is from only one fly ash sample. Other samples showed only non-measureable traces of these compounds. Also in 1983, Eklund and Stromberg 28 analysed the emissions from a coal-fired fluidized bed combustor and a coal-fired automatically regulated, on/off boiler for greenhouse heating as well as a fluidized bed municipal incinerator. The authors indicate that PCDDs and PCDFs were found in flue gas samples from the municipal incinerator but makes no mention of their presence or absence in the emissions from the coal-fired power units. Analysis was performed by GC/MS but no quantitative data are given. The data and the article, per se, are judged poor. Finally, as mentioned earlier, Haile and co-workers33 analysed the emissions from the Ames, Iowa Municipal Coal Boiler Power Plant. Many compounds which could be precursors to PCDDs and PCDFs, such as PCBzs, PoCPs, and PCBs, were quantified but no PCDDs or PCDFs were found at a detection limit of 0.1-0.25 ng/dscra. Related to the above studies, Mahle and Whiting 67 reported some laboratory studies dealing with the pyrolysis of bituminous coal at 600C. in the presence of NaCl, HC1 and CI2 . Table 43 gives the results. This investigation clearly shows that combustion of coal under certain conditions can yield significant quantities of PCDDs. Regardless of Mahle and Whiting's findings, it appears that, in general, oil- and coal-fired power plants and utility facilities are unlikely sources of significant PCDD and PCDF emissions. D . Wood Burning Fireplace/Furnaces Here again, the first report of PCDDs detected in wood burning fireplaces and furnaces was by Dow in 1978 66 in their original "Trace Chemistries of Fire" paper. Elaborations of this study were published in 1980 13 and 1982 23. Dow analysed the carbonaceous material on the walls of a metal firebox of a 25 year old fireplace, whose flue had not been cleaned in 10 years. Burch, oak, willow, mountain ash and applewood untreated with preservatives (such as pentachlorophenol) had been burned along with some paper to start fires. Also the residue on firebrick in a 12 year old fireplace, whose chimney had never been cleaned, was sampled. Hardwood, mostly oak, untreated with preservatives, was used as fuel. The results are shown in Table 12 (p.14). The levels of TCDDs and HCDDs are quite small while HpCDDs and OCDD are present in more significant quantities. The A. D. Little report 66 suggests that the findings in the wood burning fireplaces and other commonplace sampLes seem tenuous when judged by the CONFIDENTIAL SU BJECT TO PROTECTIVE ORDER. a C23st ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE standards of the rest of the Dow report. The history of the sampled fireplace walls and mufflers is not well documented. The reported observations may reflect circumstantial, rather than causal, relations between the apparent PCDDs and the commonplace combustion process. Also, for the auto mufflers and the charcoal grilled steaks, the identity of the PCDD species reported was not confirmed by GC/MS to be present at levels exceeding two times the detection limit. In 1982 71 and 1983 70, Nestrick and co-workers reported on a study ,to determine if PCDDs were produced during the combustion of wood. They reasoned that if PCDDs can be detected from its combustion today, PCDDs have been present in the environment from at least as far back as the appearance of trees. Residential wood combustion (RWC) was selected as the source of particulate emissions, while uncontrolled open-burning situations (i.e., forest fires) or open-burning experiments were ruled out for various reasons. All combustion particulate samples were obtained from domestic heating systems. Preliminary evaluation samples were obtained from the flues of an exclusive oil heating system and an exclusive wood heating system located in Cape Cod, MA as well as a primary wood heating system located ~ 15 miles from Midland, Michigan. For the actual study, six individual samples of chimney particulates were collected from each of three rural regions in the U.S. No particulate samples were obtained from systems where any type of treated or manufactured wood was burned or where the wood or residence had ever been exposed to materials like 2,4-D, 2,4,5-T or pentachlorophenol. The Eastern regional samples were taken from residences near Farmington, NH, an agricultural community. The Central regional samples came from near Grand Marais, MN (a tourist and logging area), at an old farm site near Duluth, MN, and near Escanaba, MI in the upper peninsula. The Western regional samples came from a rural area near Creswell, OR. State-of-the-art sampling, extraction, clean-up and analytical methodology were employed. Reverse phase HPLC was employed in the clean-up procedure to assure removal of residual chemically similar interferences to PCDDs. Both HRGC/LRMS and HRGC/HRMS in the SIM and MPM (multiple peak monitoring) mode with several standards were used for isomer specific PCDD analysis. Table 44 gives the results of the preliminary evaluation sampling. Table 44 PCDDs in Combustion Particulates from Preliminary Experiments(ppt) flut pip# particulates,Cape Cod, MA RWC unit particulates, Midland, M Ifl exclusive oil burner exclusive chimney e wood burner base flue pipe back flue pipe front firebox mh total TCDDs total HCDDs total H,CDDa OCDD total CDDs ND6 (2.4) ND (4.2) 2S 130(20) 155 260 340 420 260 1200 950 3400 8100 12000 24450 7.5 17 89 250 563.5 2.9 16 81 242 341.9 1.6 ND (2.0) 30 120 151.6 " R tfrr to text for sample descriptions. * Not detected, refer to footnote d in Table46 e Isomer-specific neulb preeentod in Table 45 - CONFIDENTIAL SU BJECT TO PROTECTIVE ORDER. 43 A ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE The following comments and conclusions were given: (1) For all regional samples in which one or more TCDD isomer was observed, the % 2,3,7,8-TCDD relative to total TCDD was 0-16% with an average of 5%. This average composition for 2,3,7,8-TCDD is representative of rural regions and is similar to the 3.3% 2 ,3 ,7 ,8 -TCDD observed for the Midland RWC chimney particulates where locally obtained wood was from an industrialized area associated with production of chlorinated compounds. (2) PCDD formation appears to be related to the natural chlorine content of the wood fuel rather than environmental sources. In a published study by TRW 133, eleven different varieties of wood were quoted to contain chlorine concentrations of 14-84 ug/g. (3) Typical residential wood combustion units located in rural areas are a source of PCDD emissions to the environment. The observation of a relatively flat distribution pattern for TCDD isomer concentration suggests that their formation in wood fueled RWC units may be a random statistical process where each of the 22 TCDD isomers is equally favored. (4) Although these results suggest that wood combustion may be a potential significant contributor to environmental PCDD background levels, additional research is necessary to confirm "The trace chemistries of fire" hypothesis and to better characterize the magnitude of this source. (5) Because it has been estimated that particulate emissions from forest fires in the U.S. alone range from 0.5 x 10 and 54 x 106 tons/year, the authors find it difficult to believe that PCDDs are not ubiquitous in the environment. Also PCDFs were found to be major components of these flue particulates. The apparent isomer distribution for TCDFs is similar to that of TCDDs, in that a large number of isomers are present at reasonably similar levels. The fact that unchlorinated dibenzofuran, was identified whereas unchlorinated dibenzo-p-dioxin was not, may indicate that PCDDs and PCDFs are produced by different mechanisms during wood combustion. It also may indicate that the unchlorinated dibenzofuran ring system is more thermally stable than unchlorinated dibenzo-p-dioxin. PCDD concentrations increase with increasing degree of chlorination whereas PCDF concentrations increase with decreasing degree of chlorination. Also in 1982, Sraithjet al, 98 analysed a sample of common wood-burning fireplace soot from Clifton Park, New York for 2,3,7,8-TCDD and TCDF. Extraction and clean-up procedures were state-of-the-art. Analysis was performed using RP-HPLC and capillary column GC/HRMS with ion monitoring technique. The analytical results indicated 2.7 ppb 2,3,7,8-TCDD and 7.2 ppb 2,3,7,8 -TCDF. The results seem rather tenuous compared to the other data given in the report. CONFIDENTIAL SUBJECT TO PROTECTIVE ORDER. 47 attorney w o rk product a t t o r n e y -c l ie n t p r iv ile g e E. Diesel/Auto Exhausts This potential source has had very limited attention, but could have very significant repercussions. The Dow articles 13 23 suggest that the quantities of PCDDs emitted are very small Qsee Table 12, (p.14)] Mufflers from gasoline-powered automobiles were collected at muffler shops in Pontiac and Detroit, Michigan (more than 100 miles from Midland), as they were removed from cars. The car owners voluntarily gave the mufflers and information about mileage and gasoline type used. The make, model, and mileage of each car were noted. The mufflers were transported to the Dow Analytical Laboratories sealed in plastic bags. Mufflers from Diesel-powered trucks were collected at Auburn and Saginaw, Michigan. They were sealed and transported as above. At the Dow Analytical Laboratories, the inside chambers of the mufflers were scraped and residue was collected from these chambers and analyzed for chlorinated dioxins by both GC with electron capture detection and GC/MS after liquid chromatographic separation. The only other article dealing with PCDDs in Diesel/auto exhaust emissions was published by Cavallaro, et al, 62 in 1980. They mention that only OCDD was found in an auto exhaust sample. No quantitative data was given. The extraction, clean-up and analytical procedures appear to be state-of-the-art. Analysis was performed using capillary HRGC/MS with a MNCI (methane negative ion chemical ionization) source. With these "hints" of possible PCDD emissions from truck/auto exhausts, one wonders why more publications have not appeared on this subject. These two articles suggest that the quantities of PCDDs emitted are very small (if at all) and, it could be, that researchers in this field feel that this is not a significant emission source. Because of its potential significance, more studies in this area are warranted. F. Miscellaneous Sources The search of the literature revealed several other possible non-industrial sources of PCDDs, all of them being combustion processes. These include cigarette smoke, charcoal grill fumes, and accidential fires, particularly those involving PCBs. The only references to the possible presence of PCDDs in cigarette smoke and charcoal grill fumes are the Dow reports l3>23. The analytical results are shown in Table 12 (p.14). In cigarette smoke, trace quantities of hexa-, hepta- and octaCDDs near the detection limits were found. Smoke generated by the simulated smoking of cigarettes (an arbitrarily selected national brand without filters) was analyzed for trace levels of chlorinated dioxins. The smoke was collected on special collection matrices consisting of 100/120-mesh silica (coated in situ with 20 percent hexadecane by weight) contained in a disposable glass transfer pipette. The cigarette was attached to the pipette by cleaned latex rubber tubing and puffed by applying gentle vacuum with a rubber pipette bulb. CONFIDENTIAL SU B JEC T TO PROTECTIVE ORDER. i ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE Approximately 20 to 30 puffs nominally 2 to 3 seconds in duration were collected from each cigarette. An entire package of 20 cigarettes was combusted during each experiment. Combustion experiments were conducted in two different urban locations. Before use, the inlet end of each collection tube was spiked with 2.0 nanograms of 13C-labeled 2,3,7,8-TCDD, which served as an internal standard for TCDD quantitation after residue cleanup. Each collection tube was extracted with hexane, followed by a multistep cleanup, including reverse phase LC, to remove interferences. The LC eluate was subjected to both low resolution Gc/MS and electron capture GC analysis. Concerning charcoal grill fumes, only trace quantities of OCDD were found near the detection limit. New York strip steaks grilled over a charcoal fire were sampled at 13, 18, and 25 minutes (rare, well-done and overdone samples). Each of the three samples was immediately wrapped in aluminum foil, placed in a plastic bag, and put in a freezer to await analysis. Sample extraction and clean-up procedures were state-of-the-art. Analyses of the extracts were performed by high pressure liquid chromatography and low resolution GC/MS. The data is certainly tenuous. Published reports on PCDDs and PCDFs found in the fumes from accidental fires are more numerous. The first reference to PCDDs found in an accidental fire was a report by Harvanjet al, 34 in 1981, who collected air filter samples from an industrial fire in Elizabeth, NJ. The samples were extracted and the extracts cleaned up in an acceptable manner. Analysis was by HRGC/HRMS with MID and CID techniques. Analytical complications arose due to chemical interferences. The researchers concluded that, of the nine samples taken, no sample contained more than 40 pg of TCDDs and that one sample contained ^ 20 pg of a specific TCDD isomer. Four of the samples contained less than 5 pg of TCDDs. From 1982 to the present, several reports have been published on an electrical fire that occurred in the State Office Building in Binghamton, NY. A transformer, which contained about 1100 gallons of Pyronox (a 55% mixture of PCBs and 35% PCBzs) was damaged by the fire. About 180 gallons were lost. How much was burned and how much leaked before it was cleaned up is not known. As a result of the fire, a large amount of soot was formed. Kim 51 reported the level of 2.3.7.8- TCDD in the soot was - 3ppm. This data appears to come from a more scientific article published by Smith, et al. 98 A composite sample of soot from a stairwell of the building was analysed. The extraction and clean-up procedures were state-of-the-art. Analysis was by RP-HPLC followed by capillary column GC/HRMS with ion monitoring techniques. Analysis of duplicate samples gave 2.8 and 2.9 ppm of 2.3.7.8- TCDD, and 270 and 120 ppm of 2,3,7,8 -TCDF. des Rosiers, 42 of the EPA Office of Res. & Dev., reported, however, that the analysis of soot samples found in the building shows PCDDs as high as 10-20 ppm while PCDFs were present as high as 2,160 ppm. He said the most toxic PCDDs and PCDFs were the major components in the soot, including 2.3.7.8- TCDD, 2,3,7,8 -TCDF, 1,2,3,7,8 -PeCDD, and 1,2,3,7,8 - and 2.3.4.7.8 - PeCDFs. des Rosiers also said that even after parts of the building were decontaminated using steam and detergents, trace amounts (ng/m2) of PCDDs and PCDFs were still present. CONFIDENTIAL SU B JE C T TO PROTECTIVE ORDER. 5 C2389 1 ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE The data in the report by des Rosiers seem to have come from some outstanding work by Rappe and co-workers. 9 4 123 In addition to analysing soot from the-Binghamton, NY fire, these researchers analysed samples from PCB fires at Stockholm, Skovde, and Surahammar, Sweden as well as an explosion of a capacitor filled with PCB in an electrical railway locomotive. The toluene Soxhlet extraction of the acid treated samples and the clean-up of the extracts were state-of-the-art. The analyses were performed by HRGC/MS using El or MNCI sources and SIM technique with reference standards. In the Binghamton, NY soot, the total level of PCDFs was 2,160 ug/g(pPm ) The most toxic isomers, as mentioned above, were found to be the major constituents. In addition, a series of PCDDs were found with the highly toxic isomers again dominating (2,3,7,8-TCDD at 0.6 ppm and 1,2,3,7,8-PeCDD at 2.5ppm). Most interesting was the finding of polychlorinated biphenylenes (PCBPs). The PCBPs are closely related to the PCDFs and PCDDs and the 2,3,6,7-TCBP has biological activity in the same range as 2,3,7,8-TCDD 124. In the Stockholm capacitor fire, the wipe samples were found to contain PCDFs and PCBPs but no PCDDs. In the Skovde capacitor fire, the capacitor fluid was a mix of mineral oil and PCBs. The wipe 1 samples contained PCDFs but no PCDDs or PCBPs. The Surahammar explosive fire and the electric locomotive fire also generated only PCDFs. Table 50 gives the analytical results. Table 50 PCDFs, PCDDs and PCBPs in Soot from PCB Capacitor Fires PCDFs L C l4 2378 i c i5 1 C16 = C17 Bingnamton (.g/g) Stocunolm SfcOvOe (i.g/a^ ) 2 Suraftanmar (wg/ra ) (before cleaning ) 2 Syranammar (ijg /n -J (a fte r cleaning) 2 Hallsidnam m ar (^g/m ) (before cleaning ) Hal 1itanammar (^g/o^) (a fte r cleaning) E le c t r ic a l locomotive (ug/o2) 2G 1 .2 0 .6 1.2 < 0.02 1 .6 0 .0 0 2 4 .9 12 0 .1 5 0.1 0 .3 670 0.175 0.1 0 .3 965 - 0 .0 6 0 .1 5 < 0.004 < 0.01 0.01 0 .5 4 0 .3 6 0 .6 0 0.0002 0.0005 0 .0 0 0 3 1.2 1.5 0 .4 460 - 0 .0 0 8 0 .0 7 < 0.015 o .e o 0.0007 0 .1 7 CONFIDENTIAL SUBJECT TO PROTECTIVE ORDER. C1a l PCDDs 40 0 .0 0 6 0.01 20 - < 0.002 - 1.34 - 0 .0 0 1 l - 0 .0 4 - PCBPs - 02389b 51 * ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE Also, Nestrick, et al, 7 0 >71 in their excellent research on emissions from residential wood combustion (RWC) units admitted that none of the combustion sources in Dow's "Trace Chemistries of Fire" paper truly demonstrates situations representative of "common combustion" in a time scale which includes the "advent of fire". These researchers reasoned that if PCDDs can be detected from wood combustion today, PCDDs have been present in the environment from at least as far back as the appearance of trees. With the finding of PCDDs in RWC emissions, they concluded, "Because it has been estimated that particulate emissions from forest fires in the U.S. alone range between 0.5 x 106 and 54 x 106 tons/year, we find it difficult to believe that CDDs are not ubiquitous in the environment." Finally, Czuczwa and Hites have recently 119l2 reported on a classic study of PCDDs and PCDFs in sediment cores. The results are most significant. PCDDs and PCDFs were found in sediment samples in an urban area as well as in remote sites, the concentration at the remote sites being about 100 ppt. This indicates that PCDDs and PCDFs are stable in the environment and "indeed, may be ubiquitous". The PCDD and PCDF isomer profiles were similar in all sediment samples and closely followed those found in combustion samples. This suggests that combustion is the major source of the compounds. When PCDDs and PCDFs were measured in sediment cores, the depth-concentration profiles in these cores showed that the input of PCDDs and PCDFs increased dramatically at a depth corresponding to about 1940 and remained high to the present time. In Figure 2, the time-dependent concentration profiles of OCDD and HpCDF in a Lake Huron core are compared to the trends for U.S. coal production and production of chlorinated hydrocarbons for that time period. It seems clear that coal use has not been a major source of these materials at least since 1940. The sedimentary historical record agrees well with the production of chlorinated aromatic compounds. According to the authors, this suggests that chlorinated aromatics are precursors of PCDDs and PCDFs, and these compounds are present in combustion fuels. The following statement made by the authors when presenting this information at the 186th National ACS Meeting in Washington, D. C. in September, 1983 caused much argument: "Low or undetectable levels of PCDDs and PCDFs at depths corresponding to times before 1940 show that PCDDs and PCDFs were not with us since the advent of fire." This statement was challenged for several reasons: 1. The authors did not demonstrate that PCDDs and PCDFs on the partiCjles had always been there. 2. The sample population was very small. 3. Precision values were not available. 4. The data given showed a background level of PCDDs negating the claim. 5. The rise of the chemical industry parallels the rise in use of municipal incinerators, industrial waste incinerators, fireplaces, etc. SUBJECT TO PROTECTIVE ORDER. 53 * attorney W ork PRODUCT ATTORNEY-CLIENT PRIVILEGE Figure 2 U.S. Coal and Chloro-aromatics Production Compared to the Concentration of PCDDs and PCDFs in Lake Huron Core Sample as a Function of Time After considering all of the information and evidence, we believe that PCDDs as well as PCDFs are indeed ubiquitous now and have been so for a long time. CONFIDENTIAL SU BJECT TO PROTECTIVE ORDER. 54 G23898 ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE Trace chemistries of fire consist of numerous chemical reactions occurring during combustion at very low concentrations. Recent advances in analytical methodology now permit the products of these reactions to be measured at concentrations as low as 0 . 0 0 0 0 0 0 0 0 0 1 percent (10-11*) in favorable cases. With this sensitivity, products resulting from reactions with yields as low as 1 0 *1" percent can be identified and measured. The chemistry is such that low concentrations of either inorganic or organic chlorides in the fuel can be expected to produce traces of chlorinated dibenzo-p-dioxins. Table 12 (p. 14)13 gives the data which Crummett and co-workers claim is evidence in support of their hypothesis and the ubiquitous nature of PCDDs. Eminent scientists in the U.S. have suggested other ways to test the hypothesis. Among the suggestions were; 1. Determine if chlorinated hydrocarbons and chlorinated dioxins are present in soil high in carbonaceous matter taken from drill cores at depths corresponding to 5-, 12-, and 3 5 , 0 0 0 years from under ancient lake beds. 2. Determine if chlorinated hydrocarbons and chlorinated dioxins are present in ice taken from the center of an ancient glacier. 3. Determine if chlorinated hydrocarbons and chlorinated dioxins are present in ash taken from the mouth of a volcano. 4. Determine if chlorinated hydrocarbons and chlorinated dioxins are present in sea breezes from islands remotely located in the South Pacific. 5. Determine if chlorinated dioxins can be formed by burning fossil fuel in the presence of chlorine or inorganic chloride. 6 . Determine if chlorinated dioxins are present in fish taken from rivers remote from pesticide manufacturing facilities but close to incinerators and fossil-fueled power houses. Crummett indicated that suggestions 1 through 4 would require detection limits of 1 ppq and possible contamination would make the analysis very difficult. However, Czuczwa and Hites 119,120 repQrted finding PCDDs and PCDFs in sediment core samples, particularly from remote sites, as far back as 1868. They said that "these data indicate that PCDDs and PCDFs are stable in the environment and, indeed, may be ubiquitous". Also the isomer profiles were similar in all core samples and closely followed those in combustion samples, suggesting that combusti is the major source of these materials. Concerning suggestion 6 , Crummett reported on the analysis of fish for TCDDs. The results are shown in Table 51. 23 CONFIDENTIAL SU B JE C T TO PROTECTIVE ORDER. 56 C2389!) ATTORNEY W ORK PRODUCT -CLIENT TM.!V!IEGE chloride may form PCBzs which, in turn, form PoCPs which proceed on to form PCDDs. Such transformations can be expressed in simple chemical form by the following reaction path: pNaCl IA salt + Si02 sand CH.-CH,* A zx a polyethylene bottle + Cl -- atomic chlorine V p * Cl PVC chlorocarbene HCB(a PCBz) R-H 2 _A PCDDs PCPPs ("predioxins") x*2 to 5 PoCPs * = high heat (550-1050C.) Much more about the chemistry of formation of PCDDs and precursors is coming up in the next section. B . Formation of CDDs The production of CDDs from the pyrolysis of known chlorinated chemical compounds has been known for some time. By reviewing the findings from such experiments, one can obtain a rather clear picture of the various chemical transformations that can occur in all combustion processes. Probably the first known production of a PCDD was in 1957 by Sandermann,et al, 126 who heated pentachlorophenol (PCP) in a tube. OCDD and hexachlorobenzene (HCB) were formed. CONFIDENTIAL SU BJEC T TO PROTECTIVE ORDER. 58 C 2390II ATTORNEY w o r k p r o d u c t ATTORNEY-CLIENT PRIVILEGE With the finding by 01ie and co-workers 73 in 1977 of PCDDs as well as PCBzs and PoCPs in the emissions from a municipal incinerator, many researchers have performed laboratory studies over the years to determine the chemistry of formation of PCDDs. Four excellent review articles by Esposito,et al, 31, A.D. Little, Inc. 6S, Lustenhouwer, et al, 84, and Choudhry and Hutzinger 1 S ) 3 0 > S 5 > 8 6 > 1 2 7 have summarized these findings up to 1982. The articles by Lustenhouwer, et al, and Choudhry and Hutzinger are particularly good. Lustenhouwer's article is probably the first to try to discuss the possible mechanisms for the thermal generation of PCDDs and related compounds theoretically. The review by Choudhry and Hutzinger is truly a monumental piece of work covering all mechanistic aspects of the thermal formation of PCDDs an-: related halogenated organic compounds. Part I 18 gives the theoretical background and thermochemical decompositions of monomeric organic compounds that may take place in the "Trace Chemistries of Fire" hypothesis. Part II 30 deals directly with the thermochemical generation of PCDDs and PCDFs from closely related precursors. Part III 127 deals with the thermodegradation of organometallies which relates to the catalytic effect that certain metals may have on the formation of PCDDs, PCDFs and their related precursors. We will not discuss this area to any degree; however, such chemistry plays a role in the "Trace Chemistries of Fire" hypothesis. Part IV 85 discusses the incorporation of inorganic chlorine into organic matter and thermochemical formation of aromatic compounds like PCDDs from aliphatic organics like chloroform, polyethylene or chlorinated ethylenes. Part V covers hypotheses for the formation of PCDDs and PCDFs in combustion processes incorporating all elements discussed in the four previous articles. It discusses the so-called "de novo" syntheses of PCDDs and PCDFs, which in essence is the same as Dow's "Trace Chemistries of Fire" hypothesis. Theoretically, the presence of a wide variety of PCDDs and PCDFs in combustion processes can be attributed to three possibilities. It is possible that PCDDs and PCDFs: 1 . are trace components of refuse and do not undergo effective thermal destruction. 2 . are produced during the combustion and pyrolysis of chlorinated precursors like PCBzs, PoCPs, chlorinated 2-phenoxyphenols (PCPPs) ("predioxins")( chlorinated diphenyl ethers (PCDPEs), PCBs, and 2,4,5-T and its esters. This includes the dechlorination of higher PCDDs and PCDFs to lower chlorinated isomers. CONFIDENTIAL SU BJECT TO PROTECTIVE ORDER. 59 C23901 ATTORNEY WORK PRODUCT ATTORNEY-CLIENT PRIVILEGE 3 . are formed as a consequence of a complex array of pyrolytic processes of chemically unrelated organic compounds (Mde novo" syntheses) including: (a) anthropogenic (man-made) chlorinated organic chemicals such as polyvinyl chloride (PVC), DDT, or chloroform. (b) anthropogenic non-chlorinated organic such as polyethylene or polystyrene and inorganic chlorine such as salt (NaCl) or hydrochloric acid (HC1). (c) autochthonous (naturally occurring) organic chemicals like tobacco, coal, lignin and petroleum coke and inorganic chlorine (NaCl or HC1). 1 . Ineffective combustion of PCDDs present in refuse Concerning the hypothesis of PCDDs being in the emissions of combustion processes due to incomplete combustion of PCDDs present in the refuse, it is now apparent that PCDDs, per se, are destroyed by incineration above 800C. 1323 . however, PCDDs bound to particulate matter appear to be largely unaffected by incineration temperatures as high as 1150C. 29 Incineration tests conducted for the U.S. Air Force on Agent Orange, which contained small quantities of 2,3,7,8-TCDD, showed no TCDD in the emissions at 300 and 2 ppb detection limit. The tests were run with firewall temperatures at 1150, 1400 and 1600C.13 and 30% excess air. A typical municipal incinerator, such as the Chicago Northwest incinerator, operates at a flue gas temperature of 1100C with 50% excess air, which may be borderline for combustion of any PCDDs bound to particulates in the refuse. Industrial waste incinerators are usually operated at higher temperatures by the use of supplementary fuel (coal, oil or gas), and, if operated properly, should not release significant levels of PCDDs in its emissions. The New York State Dept, of Environmental Conservation Handbook in 1980 indicates that "incineration of 2,3,7,8-TCDD at 1000C for two seconds is believed adequate. " 66 To date, no systematic, thorough study, to our knowledge, has been carried out to determine if PCDDs or PCDFs are minor constituents of domestic refuse. These materials could enter the refuse via herbicide formulations or treated wood. However, there are three reports which suggest that PCDDs and PCDFs are not present in ordinary refuse. Liberti and co-workers 58,e^ crushed, homogenized, dried and extracted a large sample of garbage to be incinerated. Analysis of the extract showed no PCDDs or PCDFs to be present. However, upon incineration, the emissions did contain PCDDs (see Table 19, p. 21). Lustenhouwer, et al, 84 analysed a sample of compost and found CONFIDENTIAL 023902 SUBJECT TO PROTECTIVE ORDER. 0* W SUBJECT TO PROTECTIVE ORDER. ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE From the above study by Rappe, et al, 87 , three possible routes to PCDDs from chlorophenols in combustion processes seem to be: 1 . condensation of chlorophenates directly to PCDDs which includes Smiles rearrangement 2. dechlorination of higher chlorinated PCDDs 3. cyclization of chlorinated 2-phenoxyphenols (PCPPs) ("predioxins") already existing as contaminants in chlorophenates. Route 1 is by far the most important. c. Chlorinated 2-phenoxyphenols (PCPPs) and chlorinated diphenyl ethers (PCDPEs) Both of these type compounds have been shown to form PCDDs upon pyrolysis. The PCPPs are also called "predioxins". Nilsson and co-workers 128 investigated the thermal degradation of 5-chloro-2-(2,4-dichlorophenoxy) phenol (I) and 4,5,6-trichloro-2-(2,4-dichlorophenoxy) phenol (II) at various temperatures. Even at 980C., a temperature at which most organic compounds decompose completely, MS showed only the starting materials and the products 2,8-DCDD (III) and 1,2,3,8-TCDD (IV). In a typical pyrolysis (770C., 15 sec.), it was observed that 70% > 510C. 15 sec. Cl ^ Cl HO II Cl > 358C0 15 Sec. Ill Cl Cl of II degraded and 30% appeared as IV (3%) and unchanged II (27%). Table 58 gives the results of this experimentation. Table 58 Pyrolytic Formation of PCDDs from PCPPs Temp., C PCDD formed (%) * from I II 358 0 0.4 510 1.7 0 . 8 770 6 . 0 3.0 980 6 . 2 2 . 8 ^ C a l c u l a t e d on in i t i a l a m o u n t of I or II 69 C23903 ** ATTORNEY WORK PRODUCT ATTORNEY-CLIENT PRIVILEGE Lindahl, et al, 64 studied the laboratory pyrolysis of eight PCDPEs with three to eight chlorine atoms at 500, 600 and 700C. No isomerization or dechlorination to lower chlorinated PCDPEs was observed. Among the decomposition products, PCDDs and PCDFs were observed with PCDFs predominating. The highest yields were obtained at 600C. At 700C. most of the PCDDs and PCDFs decomposed. In addition, chlorobenzenes were observed from all , PCDPEs pyrolysed. Table 59 gives the results of these pyrolysis experiments. T a b le 59 P y r o ly s is o f PCDPEs a t 600C Compound D e co m p o sitio n % PCDDs fo rm ed * (u g /1 0 0 u g ) 3 , 5 , 4 ' - t r i CDPE 2 ,4 ,5 ,4 '- te tra 2 ,4 ,6 ,3 ',5'-penta 2 , 3 , 4 , 2 * , 3 ' , 4 ' -hexa 2 , 3 , 4 , 2 ' , 3 ' , 5 ' -hexa 2 , 3 , 4 , 2 ' , 4 ' , 5 ' -hexa 2 , 4 , 5 , V , 4 ' , 5 ' -hexa 2 ,3 ,4 ,5 ,6 ,2 ' , 3 ' ,4 ' -octa 70 30 40 50 60 75 40 99.9 4 .5 4.4 3.5 1.4 1.0 0.7 1.3 3.8 PCD Fs form ed (u g /1 0 0 ug) 0.1 0.8 0.3 0.8 1.3 0.6 As an e xa m p le o f th e c h e m is t r y , th e p y r o ly s is O f hexa CDPE gave one HCDF, one PeCDF and two TCDD isomers identified as shown below. No TCDFs were found. Minor amounts of two other 2,3,4,6,7,8-HCDF 1,2,6,7,8-PeCDF CONFIDENTIAL SU B JE C T TO PROTECTIVE ORDER. + ci 70 1,2,8,9-TCDD 1,2,6,7-TCDD (Smiles rearranged) G239U4 i* attorney w o rk product a t t o r n e y -c lien t pr iv ile g e PeCDFs were observed, but the identity and source of these compounds is unknown. The 2,3,4,6,7,8 -HCDF was formed by loss of ortho - H 2 ; the 1,2,6 ,7,8 -PeCDF, by loss of ortho - HC1; and the 1,2,8 ,9- and 1 ,2 ,6 ,7 -TCDD by the loss of ortho - Cl2 (normal product and Smiles rearranged product, respectively). To sum up the reaction mechanics (see Figure 3), the formation of PCDFs from the pyrolysis of PCDPEs was found to follow two main reaction routes involving the loss of ortho H2 and ortho - HC1. In one case (OCDPE), the formation of two PCDFs via the loss of Cl2 (involving a rearrangement) was also observed. In addition to PCDFs, the formation of PCDDs was also observed. In this case, the formation is primarily via the loss of ortho Cl2 with Smiles rearrangement occurring. Figure 3 Reaction Mechanics of Formation of PCDFs and PCDDs by Pyrolysis of PCDPEs Cl Cl Ct, Cly -h 2 -HCI CIx Cl Cl* Cl Cl Cly Cly -Cl2 A -Cl2 Cli Cly Finally, Janssens and co-workers 45 have found PCPPs and PCDPEs in emissions from a municipal incinerator in Beveren, Belgium. Since significant quantities of PCDDs and PCDFs were also found in the emissions (see Tables 23 and 24, pp. 24 and 25), this provides further evidence for PCPPs and PCDPEs being precursors of these materials in combustion processes. CONFIDENTIAL SUBJECT. TO PROTECTIVE ORDER. 71 C2390O i$ ATTORNEY WORK PRODUCT ATTORNEY-CLIENT PRIVILEGE many reference compounds. The researchers found the cyclization processes in these pyrolyses are intramolecular and can be described by the following 4 reaction routes leading to different PCDF isomers from the same PCB: Route 1 Loss of ortho - CI2 -Cl- Route 2 Cl* Cl Cl Cly Cl, ' 0 ' Loss of HC1 involving a 2 ,3 -chlorine shift Route 3 C l, Loss of ortho - HCI t l Cl H Cl, --HCI -H C I Cl Route 4 C l, Cl H Cl C l, ' 0 ' Cly The intramolecular nature of these processes is indicated by the fact that pyrolysis of a known mixture of PCBs does not result in the formation of PCDFs other than those formed during pyrolysis of the PCBs separately 131. As an example, let us look at the results of the pyrolysis of 2,3,4,5,21,4',5'-heptachlorobiphenyl (HpCB) shown in the following scheme: CONFIDENTIAL SU BJECT TO PROTECTIVE ORDER. 73 1,734679- 02390  attorney w o rk product a t t o r n e y -c lie n t pr iv ileg e ci-Zo^-O'c h 2-cooh Cl 2,4,5-T 5Q0-6Q0oc. C l* \ O V " 0 CH- COONa c P -iH3 Na 2 ,4 ,5 -T P YpT nV\ 2,3,7,8-TCDD A In 1977, Stehl and Lamparski 99 pyrolyzed grass and paper coated with 2,4,5-T, its sodium salt, and the butyl ester. All gave 2,3,7,8-TCDD in yields of 1.2 x 10 " 5 to 16 x 10'5%. Grass impregnated with 2,4,5-T (equivalent to 12 lbs. per acre) gave the higher values. The purpose of this work was to find out the extent of the thermal formation of 2,3,7,8-TCDD from precursors under the conditions as "natural" as possible. Thus the burning was carried out at its own rate with good air supply. The most comprehensive study to date was performed by Ahling, et al, 2 in 1977, who performed laboratory combustion studies on herbicide formulations containing the 2 -butoxyethyl ester of 2,4,5-T. PCDDs and PCDFs were formed. The conditions in the 2-butoxyethyl 2,45-T PCDDs n = 4 to 8 PCDFs burning experiments together with the emitted amounts of PCDDs and PCDFs bearing 4 to 8 chlorine atoms are given in Table 61. The results show that, with few exceptions, only TCDDs are formed. This emission decreases with increased temperature and transit time. The very low temperature under "open fire" conditions (experiment 8 ) does not give large amounts of TCDDs. According to the researchers, the explanation is that the formation of PCDDs is CONFIDENTIAL S U B J E C T T O P R O T E C T IV E O R D E R - 76 C2387 ** J ATTORNEY W ORK PRODUCT a t t o r n e y -c l ie n t p r iv il e g e Table 61 PCDDs'and PCDFs from Combustion of a Herbicide Formulation Containing 2-Butoxyethyl 2,4,5-T Experiment no Temperature lC) Transit time (s) Oj in flue gas {%) COj in flue gas (%) 00 in flue gas (1) 1 500 0.7 15.0 2.6 0 .0 23 675 0.6 12.0 4.0 0 .0 850 0.7 10.5 6.1 0 .0 45 500 O.B 13.0 4.1 0 .0 625 0.B 12.0 *1 4 0 .0 678 850 0.9 7.5 7.7 0 .0 750 3.7 0 .0 9.0 5.5 100 1.0 17.5 0.9 0 .0 Bnitted amounts fcejAg fomulation) Dioxins 4CEO 5CXD 6CEO 70 8CED 3,0 <0.4 2.0 0.4 <0.4 2.3 <0.5 <0.5 <0.5 <0.5 C .1 <0.2 <0.2 <0.2 <0.2 1.3 <0.6 <0.6 0.6 <0.6 0.4 0.3 0.3 0.3 0.3 0.2 0.3 0.3 0.3 1.4 1.4 1.2 0.8 0.3 0.05 2.5 0.5 0.5 0.5 0.5 F\irans 4CEf 5CDF 6CDF 7CXF 8CEF <0.2 1.8 1.3 <0.7 <0.7 B.O 7.1 2.8 <0.7 <0.7 1.3 1.9 1.8 1.4 <0.6 0.6 1.3 <0.6 <0.6 <0.6 0.2 0.3 0.3 0.3 0.3 2.9 7.6 1.0 4.2 5.7 0.5 4.9 1.7 0.4 3.4 0.2 2.3 3.9 0.1 0.4 insufficient at this low temperature or that the formed PCDDs are effectively destroyed during the relatively long transit time. No explanation for the formation of the higher chlorinated PCDDs' in some cases, especially under oxygen deficient conditions in experiment 7, could be given by the researchers. Background levels of PCDFs were unexpectedly high. The PCDF results in Table 61 are not corrected for these background levels, thus the low values must be considered questionable. In most experiments, the PCDF emissions were higher than those of the PCDDs. These amounts of PCDFs cannot be explained by the formulation contents. Also the emissions of PCDFs can not be correlated with temperature or transit time, whereas the PCDD emissions decrease with increased temperature and transit time. Based on other studies, the minor amounts of chlorophenols (2.42 x 10"4 g/g) and PCDDs (~ 20 x 10*9 g/g) present in the formulation cannot give the amount of PCDDs found. The 2,4,5-T impurity (4.7 x 10"2 g/g) could possibly have an impact on the results serving as a precursor. The largest formation of TCDDs found in this study corresponds to a formation of about 1 ug per m2 in a forest fire. CONFIDENTIAL SUBJECT TO PROTECTIVE ORDER. 77 C23908 I ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE f , a Snnw"ation of Precursor Chemistry and Other Related Laboratory Experiments In 1982, Janssens and co-workers published an outstanding article 45 . The focus of the experimentation was a systematic monitoring of the PCDDs and PCDFs and their possible precursors emitted by a municipal incinerator in Beveren, Belgium. One of the major purposes of the study was the definition of the role of the PCBzs and PoCPs as precursors for the PCDDs and PCDFs in the actual conditions of the waste burning process. The sampling of the particulate phase and the vapor phase of the emissions, the extractions, and the purification techniques for the various chlorinated compound types were state-of-the-art. Analyses of the PCDDs and PCDFs were performed by HRGC/MS with El source with quantitation done by the MID technique. Table 23 (p.2A) summarizes the whole of the emission data (particulate phase + vapor phase). The findings were in agreement with data from other investigators. Also several PCDPEs and PCPPs as well as some PCBs were found, but only as minor constituents in the flue gas. The researchers pointed out that only a minor fraction of all of the compound types (< 10% in most cases) is found on the particulates. Thus analysis of fly ash samples is in no way relevant for the total emission. Of real significance, the authors postulated a pyrosynthesis mechanism based on data from the many laboratory studies performed by other researchers on the pyrolysis of possible precursors, which we have discussed. They claim that this mechanism is confirmed by their identification of the above mentioned intermediates as well as by the correlation of the concentration levels of the precursors with those of the PCDDs and PCDFs in the actual conditions of refuse burning. Their mechanistic scheme for the formation of PCDDs and PCDFs is shown below. CONFIDENTIAL, SU BJEC T TO PROTECTIVE ORDER. 78 C2 3 9 0 9 * CONFIDENTIAL SUBJECT TO PROTECTIVE ORDER ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE The essential role of PoCPs and PCBzs as precursors is obvious. This study does seem to provide strong experimental evidence for the proposed overall mechanics. Janssen and co-workers also showed PVC going to PCBzs. We omitted this reaction step, because we consider such a conversion as definitely part of the "de novo" synthesis mechanics, which will be discussed in the next section. The authors suggest such a possibility in their discussion of the experimental results. Related to this precursor mechanistic approach, laboratory studies by Olie, et al, 77 on the burning of lignin sulfonate (a polyphenolic) in the presence of HC1 or PVC (a chlorine source) to give PCDDs and PCDFs tend to support the Janssen scheme. Also, Liberti, et al, 60,63 performed laboratory combustion experiments on vegetable materials (chestnut extract, fresh fruits and vegetables, tannic acid and mimosa extract), per se, and in the presence of chlorine in the air stream or added PVC. In the per se case, specific phenols (phenol, ra-cresol and p-cresol) were formed. In the case where a chlorine donor was added, PCDDs and PCDFs were found as well as di-, tri-, tetra- and pentachlorophenols. Their mechanistic model is simple: Precursor A + (organic matrix) Phenols or polyphenols Precursor B -- > PCDDs + PCDFs (CI2 or HC1 donor) PVC 3. PCDDs and PCDFs by "de novo" syntheses In 1980, Lustenhouwer, et al, 84 published the first review article dealing with some possible mechanisms for thermal generation of PCDDs and PCDFs from a theoretical viewpoint. According to these researchers, very little direct evidence existed at that time for the "de novo" formation of these or related compounds from carbon sources and natural chlorine donors. The presence of PCDDs and PCDFs from pyrolysis or incineration processes without sources of anthropogenic (man-made) chlorine compounds (Dow's "Trace Chemistries of Fire") and chlorobenzenes in fly ash from coal fired installations suggest that this process is operating. Chlorine does occur at low level's in most fuels and material used for combustion, as Table 62 shows. Table 62 Chlorine Content of Fuels and Combustible Material Fuel Chlorine(ppm) coal refuse paper leaded gasoline unleaded gasoline 1300 2500 300-1600 300 1-6 79 8 ATTORNEY WORK PRODUCT ATTORNEY-CLIENT PRIVILEGE Lustenhouwer and co-workers indicated that the scarcity of data in this area does not mean that this mechanism is not operating, but only indicates that not enough work has been carried out. This seems to have spurred researchers into action. In 1982, Janssens and co-workers 45 cited some evidence that this mechanism may be operating in their study of chlorinated organics in emissions from a municipal incinerator. They found that the contribution of the individual PCBzs to the total PCBz emission was nearly constant, and was not influenced by the concentration levels of the individual PCBzs in the flue gas or by the composition of the treated refuse. They indicated that this strongly suggests the occurrence of an alternate formation process (to their proposed "precursor" scheme) for PCDDs and PCDFs, similar to that for polynuclear aromatic hydrocarbons (PAHs), which has been studied extensively. This appears to involve the recombination of small initial radical fragments (usually containing two carbons). The different analogs are formed in characteristic concentration ratios, depending on the combustion temperature and, to some extent, on the composition of the material burned. In view of the above findings, the authors attempted to relate the PCBz formation to the content of the organic chlorinated compounds of the burned material. Several analyses were performed on fly ash samples, obtained from wood, coal, and heating oil. The results indicate that the chemical nature of the chlorine containing compounds in the substance to be incinerated is only of minor importance for the formation of PCBzs, which again suggests an alternate mechanism. Such a suggestion sounds very much like the "Trace Chemistries of Fire" hypothesis or "de novo" synthesis. It could be that both the "precursor" scheme proposed by Janssens and co-workers and the "de novo" mechanics are operating, either independently or concerted more likely concerted. Also, in 1982, Choudhry, Olie, and Hutzinger 109 published an outstanding review of the literature dealing with experimentation which provides substantial indirect evidence for the "de novo" syntheses of PCDDs, PCDFs and related compounds, and the possible mechanisms involved. Like Lustenhouwer, they say that little experimental evidence is available to show thermal synthesis of chloroaromatic compounds from non-chlorinated aliphatic precursors and inorganic chlorine. This review highlights particular points of a far more detailed treatment of the "Mechanistic Aspects of the Thermal Formation of Halogenated Organic Compounds Including Dibenzo-p-dioxins", which was published in 5 parts by Choudhry and Hutzinger 18>308586i127 CONFIDENTIAL SU BJEC T TO PROTECTIVE ORDER. 80 023911 * attorney w o r k product ATTORNEY-CLIENT PRIVILEGE They divided their review into three parts. The first part deals with the therraochemical decomposition of chlorinated monomers (aliphatics and aromatics) and polymers, such as polyvinyl chloride (PVC) and lignin as well as the thermal production of PCDDs and PCDFs. In the second part, the thermolysis of organic matter is discussed along with incorporation of chlorine from inorganic chlorides with the formation of chloroorganic compounds, such as methyl chloride, carbon tetrachloride and hexachlorobenzene (HCB). In the final part, hypotheses about the thermal formation of PCDDs, PCDFs and related compounds in incinerators are formulated based on well known experimental literature. A capsule summary with pertinent examples follows. a . Thermochemical decomposition In aliphatic hydrocarbon chemistry, certain reactions appear to play a role in the formation of PCDDs and PCDFs. The pyrolysis of chloroform has been shown to yield the following products: 109(P ` 277 Cl) CHC13 ------ > :CC12 + HC1 ("41%) (2) I CHC13 c h c i 2c c i 3 (3;| Cl + C2HC14 + CC13 + CHC12 >! CC14 + C12C=CC12 + C12HC-CC13 + C2H2C14 + C2H3C13 + C13C-CC13 (-13%) (-0 .7%) (-0.3%) (~3%) Notice that tetrachloroethylene (C12C=CC12) is formed in relatively high yield. Also, by heating any perchlorocarbon compound or mixture of compounds having an overall mole ratio of chlorine to carbon between 1 and 4, an equilibrium is reached between hexachlorobenzene (HCB)(CeCls), carbon tetrachloride (CC14) and hexachloroethane (C2C16) . 1S* P `23 Related to this-, 9 C?C1R --- CClfl + 12 CC14 the results of the kinetic study of the pyrolysis of octachloropropane (a perchlorocarbon compound^ at 350C. shows that the following reactions occur. CONFIDENTIAL SUBJECT TO PROTECTIVE ORDER. 81 ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE S1 Cl3 C-(j-CCl3 CI3CCCI2 + c i3c C I 3 C C C I 2 + ci3C c > eioC=nnio + CCI4 ^__________ a perchlorocarbon +2 C 1 2 C = C C 1 2 --- ClCgCCl + CCI3 CCI3 C12 C=CC12 + CCI3 CCI3 \----- > C1CECC1 + 2CC14 3 C1CHCC1 clr V cl Jo X Cl^S^Cl Cl HCB Notice that tetrachloroethylene (Cl2C=CCl2 ) and dichloroacetylene (CICe CCI) are intermediates in this sequence of reactions, and that the end result is HCB (a PCBz). It has been known for a long time that dichloroethylene even at room temperature will form HCB. 18 27 This supports the intermediate nature of dichloroacetylene. Very recently, the decomposition of acetylene has been conclusively proven to be primarily responsible for the growth of soot particles in combustion processes 1'*5 . The reaction of carbon tetrachloride CCC14) and hydrogen in a hot tube at 600-650C. gives the following products 18 P *29 CC1. H2 (CC13) (CC12) (CC1 ) C H 3CI CI3C-CCI3 5% nJ' C12C=CC12 35% C6C16 (HCB) 60% CC1 ' CH3CC13 0 .01% Notice the formation of HCB in high yield from the chlorocarbene radical. Also, it is known that the pyrolysis of chloroform yields HCB, HC1 and Cl2 . 18 ^ ' 30 Here again the obvious intermediate seems to be the chlorocarbene radical (*CC1). 6 HC1 + 3 Cl2 CONFIDENTIA F SUBJECT TO PROTECTIVE ORDER 6 *CC1 Concerning synthetic aliphatic hydrocarbon polymers, a large number of these materials are commonly used in daily life. The major commercial polymers include plastics, such as polyvinyl chloride (PVC), polyethylene and polyesters. PVC is one of the three most important plastics in use. Some 82 G2 3 8 1 S ATTORNEY WORK PRODUCT ATTORNEY-CLIENT PRIVILEGE of its applications are cable insulation, leathercloth (car seat covers), packaging, and toys. Many researchers have studied the pyrolysis of PVC. Of particular significance was the work by Ahling and co-workers. 1 0 9, 1 32 Eleven pyrolysis experiments were performed at temperatures between 570-1130C. at different oxygen levels. Chlorinated benzenes (PCBzs) from dichloro- to hexachlorobenzenes including several isomers of dichloro-, trichloro-, and tetrachlorobenzene were identified. In addition, PCBs and octachlorostyrene (OCS) were identified. Cl 570-1130 C. -fCH2 -CH>- --* Cl \c=cci. 0^ Cl PVC x= 2 - 6 OCS 0.04-21.9mg/g PVC .0-2.3mg/g PVC Notice the trichloroethylene moiety (a two carbon radical) present in OCS. From our discussion of possible precursors to PCDDs and PCDFs, PCBzs and PCBs appear to be two of them. Concerning natural polymers, lignin is an extremely complex one. It is biosynthesized by plants and it is the second most abundant organic material in the environment. It is classified as a polyphenolic. Pyrolysis of this material at 475C. resulted in the formation of 26 compounds as shown below 109 P*291. All of the phenols generated could certainly represent precursors for PCDDs and PCDFs if a chlorine source were present. LIGNIN 3? -C2H5 4J -CH3 3 n"C3H7 42 -C2H5 35 -CH=CH2 45 35 -CHj-CH=02 44 -CH-CHj 37 -CH=OOH3 > -CH?-CH=CH2 (cis.Irans) 35 < P <3 35 -C CONFIDENTIAL SU BJECT TO PROTECTIVE ORDER. 83 29 R.H 8 -CB3 30 -CjHs 1211 023814 ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE You will recall, in the discussion of the chemistry of PCDD precursors (page 79), the laboratory studies by Olie, et al, 77 on the burning of lignin sulfonate in the presence of HC1 or PVC to give PCDDs and PCDFs, and the work of Liberti 60 * 63 on the combustion of vegetable materials in the presence of Cl2 or added PVC to give polychlorinated phenols (PoCPs) as well as PCDDs and PCDFs. Similar studies by Ahling and Lindskog 110 on the pyrolysis of concentrated bleaching effluents from wood pulp (contains chlorinated degradation products of lignins) and wood pulp black liquor (prior to bleaching), which contains -- 6 g/liter of inorganic chlorine but no organic chlorine compounds, gave PCBzs. The formation of PCBzs was greatest from the pyrolysis of black liquor. b Incorporation of inorganic chlorine into an organic molecule We have already discussed the studies by Ahling and Lindskog 110 on the pyrolysis of black liquor to form PCBzs and by Mahle and Whiting 67 on the combustion of bituminous coal in the presence of NaCl, HC1 and Cl2 resulting in PCDDs (see Table 43, p. 42). Also petroleum coke has been chlorinated and pyrolyzed in a potassium chloride, sodium chloride melt. 109 ^ 292. The following equation shows the results. Table 63 gives the results of the coke pyrolysis 700C. Petroleum coke Cl2 , KCl/NaCl CC1. + HC1 + C0C12 + CO 3 5 o a: CT u LU P O 0 u. p f- a 5 g + C02 + H 2S + H2 + CH4 at various temperatures. The formation of the chlorocarbons Table 63 Chlorination and Pyrolysis ^of Cokeb in a KCl/NaCl MeltC Tem perature ( C) Products (r*g/g) ccr 2hd 4hd 2hd i 3" 700 B00 900 1000 2 .0 2 .3 0 .5 1.5 l.S 0 .0 0.1 0.1 0 .0 0 .0 0 .0 0 .0 0 .7 0 .0 0 .0 0 .0 *0uring the p rocess,, the h e llu a flo w (fre e o f oxygen tra c e s and d rie d e n to n n e (lO O t) f i o * w ere 2 and 7 .5 1 / h r , r e s p e c 1 t l v e l y . tn a l l the e x p e rim e n ts , 80 gm o f the n e l t and 1 gn of petroleu e coke rt u t iliz e d ^The com position tu t. S) was: H, 3 . S I ; C , 9 4 .5 ; ash , 0 .5 ; S , 0 .8 5 ; and (0 N ). 1 .2 cThe a o lt c u la r r a t io o f the t i t was one to one ime o f p y r o ly s is C2391 8A ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE is believed to be the result of interaction of CI2 with the irregular side structure of the coke macromolecules. At temperatures of 900-1000C., the decomposition of the side structure probably takes place faster than the direct chlorination, thus no chloroorganics are formed at - 1000C. Another interesting study involves the combustion of tobacco. 1 9 >P* 294 When tobacco is burned methyl chloride (CH3CI) is formed. The methyl chloride can come either from the organic or inorganic chlorine present in the tobacco (~0.25% and ~0.88%, respectively). When tobacco impregnated with radioactive inorganic chloride was burned, radioactive methyl chloride (CH3 3 $C1) was formed. Tobacco + 3 6C1" smoking----- > CH336C1 Concerning the formation of molecular chlorine (Cl2) or atomic chlorine (Cl) necessary for chlorination of an organic molecule, the formation of chlorine from salt, sand and water has been known for many years. The following reactions are involved at temperatures of 600-1000C. 85#p*280 The first reaction occurs in dry air, whereas all three occur in moist air, with the second reaction A 4NaCl + Si02 + 02 salt sand > 2Na20 Si02 + 2C12 A 2NaCl + Si02 + H20 ---- Na20-Si02 + 2HC1 4HC1 + 02 ^ > 2H20 + 2C12 predominating. The last reaction is the basis for several industrial chlorination processes. c. Hypotheses about the thermal generation of PCDDs, PCDFs,. and related compounds in incinerators ("de novo" syntheses) After reviewing all of the literature, Choudhry, et al, consolidated the information into a reaction flow chart to illustrate many of the hypothetical possibilities for the "de novo" syntheses of PCDDs, PCDFs, and related compounds. These are shown in Figure 4. 86,109 CONFIDENTIAL SU B JE C T TO PROTECTIVE ORDER. 85 023916 ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE Figure 4 Complex Array of PCDD and PCDF Syntheses from Chemically Unrelated Organic Matter ("de novo" Syntheses) They also published another reaction flow chart on some possible sources of molecular and atomic chlorine (Clg and Cl) in incinerators 86* ^*303. This is shown in Figure 5* Figure 5 Possible Sources of Cl2 and Cl 86 SUBJECT TO PROTECTIVE ORDER. ATTORNEY WORK PRODUCT ATTORNEY-CLIENT PRIVILEGE Within the past year, Ballschmiter and co-workers 5 published preliminary results of their study on the emissions from a municipal incinerator. Their goal is to correlate the operating conditions of an incinerator with the organics found in the particulate emissions, primarily PCDDs and PCDFs. As yet, they have found no correlation of the PCDD content in the fly ash with other measured parameters. They compared HC1, S02 , C02 , H20, 02 emissions in the stack gases with PCDD formation. They were particularly interested in HC1 where there appears to be no correlation at all. Comparing the sum of bond dissociation energies of reactions involving the reactants HC1 and chlorine, thermodynamics seem to back these experimental results, according to the authors. The authors obviously are advocates of the "de novo" synthesis hypothesis, since their introduction in this publication discusses the chemistry in flames being the chemistry of radicals in complex reaction pathways. They suggest that the chemistry in flames of incinerators, due to the variation in input and operating conditions, will be an extreme in complexity and can be dealt with only in very general terms. In addition to oxidation as the main type of reaction in a non-reductive flame, addition elimination, reduction and chlorination will be further reaction pathways, at least as intermediate routes. Products formed will be the object of pyrolytic reactions in the gas phase as well as in the adsorbed state on particles. Metal-catalyzed reactions on particles also must be considered. The chemistry in flames, they say, is governed by formation of stable radicals and olefinic and acetylenic compounds. They indicate that chloroethenes (chloroethylenes) and chloroethynes (chloroacetylenes) appear to play a central role as intermediates in the formation of PCDDs, PCDFs, and other chlorinated aromatic compounds. They also published a series of theoretical radical reactions that possibly occur in an incinerator. These reactions involve radicals of the C=C, C=0, C=N, 0, 00H and Cl type. From the results obtained thus far, these researchers believe that municipal incinerators can in practice be run under conditions of very low level of emissions of organic compounds. The authors did not elaborate on this point. Back to Choudhry, et al,and their mechanistic review of the thermal formation of PCDDs and PCDFs, 109 they conclude that there is no doubt that "de novo" syntheses play a significant role in incinerators and other fires. They say it is evident that the PCBzs and PoCPs are key intermediate compounds of prime importance in the "de novo" synthesis of PCDDs and PCDFs from any organic matter. In other words, both Choudhry, Olie and Hutzinger as well as Ballschmiter and his co-workers support Dow's "Trace Chemistries of Fire" hypothesis. After digesting all of the literature to this point in time, there is no question in our minds that "de novo" syntheses (chemistry in flames, "Trace Chemistries of Fire" - call it what you may) play a significant role in the formation of PCDDs and PCDFs in incinerators, probably in conjunction with ordinary chemical reactions involving the described "precursor" chemistry. 87 ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE of PeCDDs and TCDDs from two isomeric HCDDs, 1,2,3,6 ,7,8 - and 1 ,2 ,3 ,7 ,8 ,9 -HCDD (XII and XIII, respectivly) again from photolysis in benzene-hexane (5:95 by vol.) solution (see Figure 6 ). All photoproducts were identified by comparing retention times and mass Figure 6 Photolysis of 1.2,3,6,7,8 - (XII) and 1,2,3,7,8,9-HCDD (XIII) in Benzene/n-Hexane (5:95 by vol) Solution aa 1,3,6,8 -TCDD 1,3,7,9-TCDD spectra with those of reference compounds. The formation of 1,3,6,8and 1,3,7,9-TCDD as the principal isomers from the photolysis of 1,2,3,6 ,7,8 - and 1,2,3,7,8 ,9-TCDD, respectively, clearly shows a preferential dechlorination from the lateral positions (2,3,7 and 8 positions) as opposed to dechlorination from the peri positions (1,4,6 and 9 positions). Dobbs and Grant 114 observed another pertinent fact. Compounds possessing the same number of chlorine atoms in the 2 -positions (lateral positions) photolyse more rapidly (T^ is shorter) as chlorine atoms are removed from the 1-positions (peri positions). Extrapolation of these results led them to predict that 2,3,7,8 -TCDD should be the most photolabile of the PCDDs. According to Dobbs and Grant, their observations are reassuring from the environmental viewpoint that the PCDDs most likely to be the most toxic are also the most susceptible to photodegradation, and the photodegradation of PCDDs is not a facile route for the generation of highly toxic PCDDs. Nestrick and co-workers 113* ^ ` 144 studied the photodegradation of all 22 TCDD isomers. Table 66 summarizes the data for dilute n-hexadecane solutions and for material exposed on a clean, soft-glass surface. These data show that 2,3,7,8 -TCDD is a very unusual isomer with respect to photolytic behavior. It has the CONFIDENTIAL SUBJECT TO PROTECTIVE ORDER. 92 * 0 1 YJK jV\ " + V JL.r c l C l ^ ^ C 1 N a t T ^ ^ 'ci NaTCP ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE Cl Cl 2,3,7,8-TCDD * ci + 2NaCl `Cl overchlofinated benzenes in the 1,2,4,5-TCB, the sodium salts of di- and tetrachlorophenols present in the NaTCP might produce trace quantities of lower or higher chlorinated PCDDs. In 1976, an explosion associated with the production of NaTCP at the ICMESA plant at Seveso, Italy sprayed the reactor contents over a densely populated area. Realizing that considerable 2,3,7,8 -TCDD would likely be present, Baser and Rappe 140, in 1980, analysed soil samples from the contaminated area in Seveso for TCDDs. They used a HRGC/MS technique capable of separating and identifying all 22 possible TCDD isomers. Figure 7 shows the mass spectroscopic analysis of the soil extract. Notice that only two TCDD isomers are present, with the Figure 7 Selected Ion Chromatogram of Seveso Soil Extract Showing TCDDs Present 2,3,7,8 -TCDD overwhelmingly dominant. For comparison, the researchers analysed the extract of a fly ash sample from a municipal incinerator in Switzerland using the same technique. Figure Q shows the mass spectroscopic analysis of the fly ash extract. The chromatogram shows a complex isomeric mixture CONFIDENTIAL SUBJECT. TO PROTECTIVE ORDER. 96 023921) ATTORNEY WORK PRODUCT ATTORNEY-CLIENT PRIVILEGE Figure 8 Selected Ion Chromatogram of Fly Ash Extract Showing TCDDs Preseat with at least 17 of the possible 22 TCDD isomers present. The major isomers are 1,3,6 ,8 -, 1,3,7,9- and l,2,3,7-(or 1,2,3,8 -) TCDD, which make up more than half of the total TCDD amount present. Notice that the 2,3,7,8-TCDD isomer is present at a very low level ( M % of the total TCDD amount). The authors said that this same isomeric pattern was observed in other fly ash extract samples. Since only the mass/charge ratio (m/e) of 320 was measured, higher chlorine number homologs are not shown in Figures 7 and 8 . Another example of the limited number of PCDD congeners generated by industrial processes is the production of 2,4-D and its formulated amine salts and esters. 2,4-D is produced by the condensation of 2,4-dichlorophenol (2,4-DCP) with monochloroacetic acid under basic aqueous conditions. The 2,4-dichlorophenol is produced by the chlorination of phenol. The commercial material usually contains some 2,6-dichlorophenol as well as 2,4,6-trichlorophenol as impurities. CONFIDENTIAL SU BJEC T TO PROTECTIVE ORDER. 97 C23S2 ATTORNEY W ORK PRODUCT a t t o r n e y -c u e n t PRIVILEGE Table 67 D i - , Tri- and Tetra CDDs in 2,fr~D Amine Formulations Sam pk So. Ingredient label gvarantt* (oz. 2,4-D amine per gallon) 1 80 2 80 3 80 4 80 3 80 6 80 7 80 8 80 9 80 10 80 11 80 12 80 13 80 14 80 13 80 16 80 17 80 IS 80 19 80 20 80 21 80 22 80 23 96 24 80 23 80 26 80 2,4-D a c id % (\rjw ) Dioxin* (ppb) 2.7-dU e q u iv 0 44.6 - - 41.3 41.7 37.9 42.44 43.10 - - 42.9 42.9 - - 42.55 42.66 42.74 42.74 49.59 42.74 - -- - - - - 316 275 - - 409 - 140 - 5 33 - - - -- - - - - - 490 587 - - 551 - 230 38 584 54 533 - - - * All dioxin results ire based on 2.4-D acid Indicates none detected above 1 ppb. tw 3 e q u i v * IJ .6 .3 tetra- -- -- - - - 132 136 - - 210 - 96 54 278 20 208 -- - - the isooctyl ester (IOE), mixed butyl ester (MBE) and propylene glycol butyl ester (PGBE) types. The PCDD levels ranged from 104 ppb to 23.8 ppm for 2,7-DCDD, 35 ppb to 2.45 ppm for 1,3,7-TrCDD, and 120 ppb to 8.7 ppm for 1,3,6,8-/!,3,7,9-TCDD. No other PCDD congeners were found in these samples. HPLC analysis of technical grade 2,4-DCP showed that, in addition to 2,4-DCP (86.22%), it contained 2,6-DCP (5.64%), 2,4,6-TCP (3.23%) and o- and p-chlorophenols (1.85%). The CONFIDENTIAL SU BJECT TO PROTECTIVE ORDER. 99 02392z < ATTORNEY W O R K PRODUCT ATTORNEY-CLIENT PRIVILEGE ST.10UIS POST-OISPiTCH Tu., k n . 10. 1M4^ Municipal Incinerators Appear To Be Spouting Dioxin, Swedish Researcher Says A Swedtok retttrcb cr who found dioxin In mother's mUk and In Human fat U a w says the discovery shows that m unicipal Incinerators are m otam im Hiig the envtroom eot with the toxic chem ical. C h risto ffer R appe, o f th e department of organic chem istry at the University of Umea In Sweden, reported his findings Monday In S t Louis at the national m eeting of the American Chemical Society. Rappe seld that 2,3,73-TCDD had been found la all the sam ples In his study. That type of dioxin Is Uw sam e found at Tim es Beach and at 36 other sites in M laourt. Seven sam ples of mother's m ilk and 23 sam ples of f it H ans w ere collected at random In Wert Germany and northern Sweden. Tests showed that the levels of dioxin In the sam ples had ranged from about 4 parts of dioxin for each trillion parts of fat or mUk to about 40 parts for each trillion, Rappe said. He said this was not a dangerous level, but because dioxin can accum ulate in the body, co n tin u ed ex p o su re to In cin erators that produce the chem ical could prove to be harmful. Rappe said he did not know how much the subjects In the study had been exposed to Incinerator pollution. But he said that the types of dioxin found In the sam ples included soma produced only through lncloerattan. Indicating that the more dangerous type of dioxin also cam e from the sam e source. Asked what could be done to _ graven^ problems, Rappe replied: * `Construct better Incinerators.* He said an Incinerator of the type being considered in an experim ent to burn dloxln-contam lnated soil in BUsmuri would present no problems because It presumably would have p ollu tion -control equipm ent to prevent the spread o f dioxin. European scientists have been concerned for several years by tests showing dioxin, dlbenzofurnn and heavy m etals coming from municipal Incinerators. The scien tists are uncertain about why the toxic ch em ica ls ir e form ed . Som e speculate that the dioxin is formed by burning plastics cnatslnlng chlorine. B . Atmospheric and Surface Photodegradation As mentioned in the Photochemistry discussion in the Chemistry section (p. 8 8 ), photodegradation of PCDDs and PCDFs in the atmosphere (gas phase) is believed to occur but there is very little information available. It has been reported 66 that a "half-life" for 2,3,7,8-TCDD in the gas phase under environmental sunlight conditions would be on the order of 5 to 24 days. In contrast, the same researchers suggest that oxidation of 2,3,7,8-TCDD by the hydroxyl radical in the atmosphere would be rapid. Mill 137 indicates that the low vapor pressure of 2,3,7,8-^TCDD suggests that some of the PCDDs emitted to the atmosphere from incinerator sources will adsorb to particulate matter and either remain in suspended particulate or return to the soil and surface waters. That fraction of PCDDs remaining in the vapor phase, Mill says, could undergo photochemical degradation in the same way as 2,3,7,8-TCDD dissolved in water. He says kinetic rate laws governing photolysis of 02392-5 103 a* ATTORNEY WORK PRODUCT ATTORNEY-CLIENT PRIVILEGE chemicals in the atmosphere are virtually identical to those for photolysis of dilute aqueous solutions, and the photolytic "half-life" for 2 ,3 ,7 ,8 -TCDD vapor in summer sunlight could be as low.as ^ hour. More than likely, the majority of the PCDDs emitted to the atmosphere will be strongly adsorbed to particulate matter, and the photolysis rate may be quite different than that in the gas phase. Townsend 112 has investigated, theoretically, the fate of PCDDs in the atmosphere in terms of the change in ` isomer distribution at various atmospheric sites. His results suggest that airborn microparticulates transport and mix the lower volatility CDDs, and that photodegradation may be a major loss process for them. Janssens, et al, 45 also discusses the adsorption of PCDDs on particulate matter in the atmosphere. It can be expected that during the cooling of the flue gases in the atmosphere, the PCDDs and PCDFs in the gas phase will probably adsorb rapidly and almost quantitatively onto the particulate matter, especially the very small, respirable particles. This is especially true for the less volatile PCDDs and PCDFs. These researchers indicate that further investigations of ambient aerosols are necessary. Mill 137 goes on to say that the major transformation process in the environment for most organic chemicals is oxidation by the OH radical. Using chlorine and phenoxy substituents, he estimated the "half-life" for 2,3,7,8-TCDD in the atmospheric vapor phase to be 319 hours. He says we have no data to evaluate the rate of oxidation of 2,3,7,8-TCDD adsorbed to atmospheric particulate matter, but suspects that the rate will be much slower,. Thus, according to'Mill, these estimates show that photolysis is the dominant process in the atmospheric vapor phase. In our opinion, this is awfully thin data and there is a definite need for gas phase and microparticulate adsorbed phase photodecomposition rate studies on PCDDs and PCDFs, together with the determination of the photolytic products. Such studies can have great significance with regard to the risk of human exposure. Data on the photolytic decomposition of PCDDs on solid surfaces is also limited. In contrast to the rapid photodecomposition of 2,3,7,8-TCDD in organic solvents, Crosby, et al, 144 irradiated this compound on dry as well as wet soil (sandy or silty clay loam soil) supported on a glass plate for 96 hours, and on the surface of a glass plate as a thin dry film for 14 days. No observable photodecomposition occurred. Methanol was used as the carrier solvent for the preparation of samples of 2,3,7,8-TCDD adsorbed on solid surfaces for these photolytic experiments. No photodecomposition in the soil experiments was probably due to the fact that when the methanol solution of 2,3,7,8-TCDD was applied to the soil, it soaked in and deposited much of the 2,3,7,8-TCDD within the soil where light could not reach it. Also another reason for no photodecomposition of 2,3,7,8-TCDD in all of these cases is SUBJECT TO PROTECTIVE ORDER. 104 rt $4 ATTORNEY W ORK P R O U U V . 1 a t t o r n e y -c l ie n t p r iv il e g e VII. ANALYTICAL TECHNIQUES Analysis of PCDD and PCDF emissions from combustion sources is a very sophisticated operation for several reasons. First of all, we are dealing with trace quantities of materials usually in the ppb or ppt range. Secondly, many organic type compounds are liberated, some of which are closely related to PCDDs and PCDFs (such as the PCDPEs and PCBs) and cause interferences with the analysis for PCDDs and PCDFs unless sufficiently removed from samples. Finally, because we are dealing with both vapor and solid phase emissions, the number of operations involved in the analysis for total quantities of PCDDs and PCDFs emitted are considerable. This, coupled with the ppb or ppt concentrations with which we are dealing, makes the margin for error appreciable. For these reasons, it is imperative that we employ the most effective procedure known in each of the analytical phases - sampling, extraction, purification of the extracts, and identification and quantification. A. Sampling To analyze for total PCDD and PCDF emissions from a combustion source, such as a municipal incinerator, five types of samples must be obtained. These are (1) bottom ash - the ash present in the furnace after total combustion, (2 ) fly ash particulates removed from the flue gas by an electrostatic precipitator (ESP), (3) smaller sized particulates which escape collection by the ESP and are present in the stack emissions, (4) resin (such as XAD-2) which adsorbs chemicals present in the gas phase of the hot stack gases, and (5 ) condensate, which is collected by passing the hot residual stack gas that has been filtered (to remove fine particulates) and passed through the adsorbent (to partially remove chemicals in the gas phase) through an ice-cooled trap (impinger) containing water or a water miscible organic liquid such as ethylene glycol. The bottom ash and fly ash samples are usually obtained with a thoroughly cleaned spoon or scoop and stored in a clean, dark-colored glass or plastic container in a cool, dark place until ready for analysis. The method of collecting the samples associated with the flue gas has varied over the years. Currently, the most accepted collection method in the U.S. is the EPA Modified Method 5 sampling train. Figure 9 shows this sampling train. 33 The sampling probe is inserted in a sampling port in the stack wall. Stack emissions are drawn through the sampling train by means of an air pump. The stack emissions are kept hot (^120C.) at least through a heated filter, which collects the fine particulates. The vaporous material conducted through the filter is carried by Teflon tubing to a water cooled (<20C.) XAD-2 sorbent resin cartridge, which adsorbs gas phase compounds. The gas exiting from the resin cartridge is bank of 4 ice-cooled C23925 SUBJECT TO PROTECTIVE ORDER. 109 ATTORNEY WORK PRODUCT ATTORNEY-CLIENT PRIVILEGE impingers. The first 2 impingers contain ^*100 ml of glass distilled water for condensate trapping. The third impinger serves as a condensate overflow reservoir. The final impinger contains ^500 g. of silica gel to protect the air pump from moisture. Much of the early work (1977 to 1980) on analysis for PCDDs and PCDFs from municipal incinerators or other combustion sources were performed by analyzing extracts of the fly ash. This was done because it was easy to obtain fly ash samples. We now know that this was not a very reliable measure of the total PCDD and PCDF missions. Janssens and co-workers 45 indicate that only a minor fraction of the total Figure 9 EPA Modified Method 5 Train for Organics Sampling in Combustion Processes Tn*rmocoupl* fiw trw Typ* Piro i Tub* ------- * ^ Condenter I / Cartridge / Th*rmomr*r. Conto I* Im pinen 1 ,3 ond 4 o r* of )H* M o d ified G r**n b u rg -S m ith Typ* Inp ing er ? | of *k* G r* * n b u ry Sm llh D #iign lopingor 1 ond 2 C ontoin 100 ml W ot*r Im ping *' 3 Eir^jty Im ping' 4 C ontain* 200*300 Gram * S tlic o G * l CONFIDENTIAL SU B JE C T TO PROTECTIVE ORDER. 110 C2392G Al lU K N tr WUKK PRODUCT ATTORNEY-CLIENT PRIVILEGE Table 70 Measured Concentration of PCDDs in Extracts from Fly Ash and Efficiencies of Extraction Methods method * TCDD 1 1 2 2 3 3- 4 4 6 6 6 6 7 ,7 110 121 6 0.3 0.2 10 68 19 29 17 91 98 107 94 R ttults in ppb (nfi/i fly ah). P. HCDD H,CDD 488 1199 457 1204 902 943 11 22 19 3 19 18 4 63 90 220 330 251 624 417 67 155 90 114 349 226 81 196 156 463 1089 935 393 1063 966 443 1145 837 403 1036 836 6 For description of mathod aa text. OCDD r REE* 384 478 3083 3203 10 2JlAioUoU 4 61 2 4 44 1.5 7 19 0 .5 130 780 25 151 1511 48 31 362 12 50 768 24 37 487 15 279 2857 91 304 2824 90 355 2887 92 341 2724 87 c REE = relativa extraction efficiency ( 1 = 100%)- gave the best results, with the highest extraction efficiency found on acid treatment of the fly ash prior to extraction. Eiceman and co-workers 27 compared benzene' Soxhlet extraction with benzene ultrasonic extraction of fly ash for PCDD recovery efficiency. Although ultrasonic extraction has a reduced recovery efficiency for PCDDs in comparison with Soxhlet extraction, the authors claimed their results showed that ultrasonic extraction of fly ash when combined with a rapid filtering device has acceptable precision for most routine or survey analysis for PCDDs at the ng/g (ppb) concentration levels. Morselli, et al, 69 compared the efficiencies of the Soxhlet extraction of the PCDDs in fly ash with hexane, benzene, toluene, and xylene (mixed isomers). Table 71 shows the efficiencies of the various solvents in extracting specific PCDD isomers from fly Table 71 Relative Efficiency of Soxhlet Extraction of PCDDs Detected in Fly Ash with Various Solvents (xylene assigned 1Q0) Paak No. Extraction with Soxhlet (250 cycles) hexane benzene toluene xylene Haxa-CDD 1,2,4,6,7,9 Haxa-CDD Haxa-CDD 1,2,3,6,7,8 Hapta-CDD Hapta-CDD Octa-CDD 43 1.28 45 3,14 46 1.65 52 0.95 54 0.57 57 0.73 CONFIDENTIAL SUBJECT TO PROTECTIVE ORDER. 112 12.84 16.89 16.07 14.19 15.51 17.48 __________ 26.79 39.18 36.4 7 37.40 41.73 63.90 100 100 100 100 100 100 | C23827 e ATTORNEY WORK PRODUCT ATTORNEY-CLIENT PRIVILEGE average maximum stack concentration of TCDDs is 3500 times that in the A.D. Little study, yet the estimated ground level concentration is the same. Comparing these results with those of the Haile 33 study, the average maximum stack concentration is 8.5 times greater. If the conversion ratio from stack concentration to ground level concentration is 0.1, as used in the A.D. Little study, the degree of risk would be considerable - much greater than motorcycling. If we accept the average maximum ground level value of 0.092 pg of 2,3,7,8-TCDD/m3 , and apply the "toxic equivalents" factor of 80 for the other toxic PCDDs and PCDFs, the degree of risk is still only slightly greater than that associated with floods, tornados, and earthquakes. Also, in 1983, Olie, et al, 77 did a risk assessment on the emissions from a municipal incinerator in Zaanstad, Holland. This is probably the best risk assessment performed to date. Using a proven dispersion modeling technique, the points of minimum and maximum concentration of emissions in the vicinity of the incinerator were determined. The air was sampled at these two points and the filters extracted and analyzed for PCDDs and PCDFs. The results are given in Table 76. In Table 77, the average concentrations of PCDDs and PCDFs in the flue gas are given together Table 76 Air Concentrations of PCDDs and PCDFs in the Vicinity of a Municipal Incinerator (Zaanstad, Holland) Group of isomers TCDD TCDF PgCDD P^CDf H6C0D k6cdf H?CDD 0CD0 OCDF location A cone, pg/m^ pq tox. equ/m^ (oO 1.5 5 .12 3.4 .17 4.8 .24 4.0 .28 5.2 39 5.2 .26 13.3 - 4.8 - 1.54 location B cone, pq/,m3 pn tox eau/m3 . 1 .005 .4 .009 .5 .025 .6 .030 2.0 .140 2.1 .158 2.1 .105 .9 _ .4 .472 with the "toxic equivalents". Using dispersion modeling, the 119.2 ng/m3 "toxic equivalents" in the flue gas translates to a maximum average air concentration of 38 pg/m3 of "toxic equivalents" in the vicinity of the incinerator. Working from the daily average inhalation of a human, a maximum of 3.6 pg/day of "toxic equivalents" is inhaled. Since the acceptable daily intake (A.D.I.) is 250 pg of "toxic equivalents"/day according to Netherlands C O N F lO tm -fS *esearchers conclude that inhalation of air in 023828 SUBJECT" TO. PROTECTIVE ORDER. 122 IX. SUMMARY, CONCLUSIONS. AND RECOMMENDATIONS ATTORNEY W ORK PRODUCT ATTORNEY-CLIENT PRIVILEGE A. Relative Toxicity of CDDs 1. The toxicity of the various CDD isomers and homologs range all the way from the relatively non-toxic OCDD to 2,3,7,8-TCDD, which is one of the most toxic substances known as measured by animal studies. Most unusual is the fact that the toxicity of 2.3.7.8- TCDD varies greatly from one species to another. The degree of toxicity of 2,3,7,8-TCDD and the other toxic isomers and homologs to humans remains controversial. 2. Toxicological research must continue, particularly on humans." Much remains to be learned, especially regarding other potentially toxic isomers and homologs. This is very important, because 2,3,7,8-TCDD is only a minor constituent in the emissions of combustion processes; however, the "toxic equivalents" of the other PCDDs and PCDFs relative to 2.3.7.8- TCDD is 50 to 80 times greater in these emissions. 3. The structure - biological activity relationship of PCDDs is rather well elucidated; however, the ultimate target organ, whose dysfunction leads to death, is unknown. There is excellent correlation between the potency of PCDDs to induce certain enzyme systems and their toxicity. Continued histopathological research is needed. 4. The toxic PCDDs are a prototype of a large series of halogenated aromatic compounds (congeners or analogs) which includes the chlorinated dibenzofurans (CDFs), biphenyls (PCBs), biphenylenes (CBPs), azobenzenes (CABs), and azoxybenzenes (CAOBs), all of which show similar biological and toxic responses. Much more toxicity studies on these materials is needed, because some of these analogs may receive considerable environmental attention in the future. B. Non-industrial Sources 1. The presence of significant quantities of toxic PCDDs and PCDFs in the emissions from municipal incinerators, wood-burning fireplaces and furnaces, and industrial waste incinerators has been definitely confirmed. 2. The presence of toxic PCDDs and PCDFs in fossil-fuel fired powerhouse and utility facility emissions, auto/Diesel exhausts, cigarette smoke, and charcoal grill fumes appear insignificant, but additional research is needed to definitely confirm this tentative conclusion. 3. In general, we know that PCDDs and PCDFs are not likely to be released from combustion sources if (1 ) the operating temperature exceeds 1200C., (2) the resident time is sufficiently long (>1.3 sec.), (3) oxygen is in excess during combustion, and (4) material to be combusted is shredded into CONFIDENTIAL 023829 SUBJECT TO PROTECTIVE ORDER. 125 Ve ATTORNEY W ORK PRUUUU ATTORNEY-CLIENT p r iv ile g e E . Environmental PCDD/PCDF Input 1. With the diminished manufacture and use of 2,4,5-T and other chlorophenol derived herbicides, combustion processes, such as municipal incinerators and wood-burning fireplaces, are now suspected of being the major suppliers of PCDDs (including 2,3,7,8-TCDD) and PCDFs to the environment. 2. Although many measurements of the concentrations of PCDDs and PCDFs In emissions from the various combustion processes have been reported, only limited data necessary to estimate mass transport to the environment is available. From the limited data and considering the "toxic equivalents" of the other toxic PCDDs and PCDFs, the 52 municipal incinerators presumed operating in the U.S. may release to the environment considerably more PCDD/PCDF "toxic equivalents" per year than was released in Vietnam over a seven year period. More important, this release is over a more populated area only 2.9% the size of the area sprayed in Vietnam. Additional research is needed to provide better mass transport data from the various combustion sources to better assess potential human exposure. F. Analytical Techniques 1. Analysis of PCDD and PCDF emissions from combustion sources is a very sophisticated operation, mainly because of the large number of organic compounds liberated at very low concentrations (ppb or ppt levels). Research must continue to search for the most effective procedures in each of the analytical phases sampling, extraction, purification of extracts, and identification and quantification. 2. ' Recent studies indicate that >80% of the PCDDs and PCDFs are present in the gaseous emissions of combustion sources, not in the fly ash or bottom ash. Most of the early studies on analysis of PCDDs and PCDFs from combustion sources were performed on fly ash extracts. Earlier results obtained on the various combustion sources should be reevaluated in this light, and additional research performed on those combustion sources where confirmation of PCDD/PCDF emission levels seems warranted. 3. Protocols should be established for each analytical phase. Over the years, various methods of sampling, extraction, clean-up of extracts, and analysis have been used. Currently, the EPA Modified Method 5 sampling train appears to be the sampling method of choice, at least in this country, for stack emissions from combustion sources. For extraction of the emission samples, Soxhlet extraction with benzene or toluene seems to be the most reliable. Liquid chromatography using multilayered packed columns with various types of adsorbents is the method of choice for purification of the extracts. The level -of sophistication of the clean-up procedure is dependent CONFIDENTIAL SU BJECT TO PROTECTIVE ORDER. 128 r0 4o3930 ATTORNEY WORK PRODUCT ATTORNEY-CLIENT PRIVILEGE 132. B. Ahling, A. Bjorseth, and G. Lunde, Chemosphere, 7, 799-806 (1978). 133; C. C. Shih, et al, EPA Report on Contract No. 68-02-3138 (1980). 134. C. F. Cullis and J. E. Manton, Trans. Faraday Soc., 54, 381-389 (1958). 135. E. K. Fields and S. Meyerson, J. Am. Chem. Soc., 8 8 , 3388-3389 (1966). 136. R. H. Munch, "Photochemistry of Dioxin" (Monsanto internal report), April 13, 1984. 137. T. Mill , "Prediction of the Environmental Fate of Tetrachlorodibenzodioxin", SRI, Menlo Park, CA 94025. (Draft Copy) 138. C & E News, Special Issue on Dioxins, June 6 , 1983. 139. A. P. Gray, et al, Tetrahedron Letters, 33, 2873-2876 (1975). 140. H. R. Buser and C. Rappe, Anal. Chem., 52, 2257-2262 (1980). 141. W. P. Cochran, et al, Pergamon Ser. Environ. Sci., 5, 209-213 (1982). 142. W. P. Cochran, et al, J. of Chromatography, 217, 289-299 (1981). 143. W. P. Cochran, et al, Pesticide Chem.: Hum. Welfare Environ., Proc. Int. Congr. Pestic. Chem., 5th, 1982; 4, 341-346 (1983). 144. D. G. Crosby, et al, Science, 173, 748-749 (1971). 145. Scientific American, 250 (4), 8-9 (1984). 146. J. E. Huff, et al, Environ. Health Perspectives, 36, 221-240 (1980). 147. R. W. Bovey and A. L. Young, The Science of 2,4,5-T and Associated Phenoxy Herbicides, John Wiley & Sons, 1980; pp. 389 and 168, respectively. 148. "City Currents", Sept./Oct., 1983 [A publication by the U. S. Conference of Mayors, 1620 I Street, N.W., Washington, D.C. 20006. Phone: (202)293-7330. ion: Ronald W. Musselwhite] 023831 15 ATTORNEY W O RK PRODUCT A ' - ,r W7MEGE Non-Industrial Sources of Chlorinated Dibenzo-p-dioxins I. Citation Author(s): R. H. Stehl and L. L. Lamparski Title: Combustion of Several 2,4,5-Trichloro-phenoxy Compounds; Formation of 2,3,7,8 -Tetrachlorodibenzo -p-dioxin Reference: Science, 197, 1008-1009, (1977) Citation No : 99 II. Source of Dioxins Source: Combustion of the herbicide 2,4,5-T Sampling Techniques: Grass sprayed with 12#/acre 2,4,5-T equivalent, were taken immediately and 7 days later. Herbicide extracted with MeOH or acetone and placed uniformly on filter paper and dried. Complete combustion of paper required 5 minutes. Temperatures of 600-800C. recorded. Combustion flask connected to 4 gas absorber traps (all glass); 1 trap in ice bath other 3 in dry ice. Each trap is filled with glass beads. Air is drawn through system during collection. Flow Data (mass transfer): Not discussed Chemistry of Formation: Not discussed Isomers Identified: 2,3,7,8-TCDD III. Analytical Techniques Low resolution GC/MS. No further details. Specificity: 2,3,7,8 -TCDD Sensitivity: 0.001 ug/0.5g sample of 2,4,5-T IV. Comments (Significant Data, Graphs, etc.) Grass and paper coated with several compounds of 2,4,5-T moiety were subjected to combustion. By using compounds that had been purified to achieve background amounts of 2,3,7,8-TCDD together with an efficient cleanup and analysis of the residue, it was possible to detect as little as 0.001 ug of 2,3,7,8-TCDD in the combustion of 0.5 gms of 2,4,5-trichlorophenoxy material. CONFIDENTIAL SUBJECT TO PROTECTIVE ORDER. 229 02383^ ATTORNEY W ORK PRODUCT -- :'MT poiyij EGE Table 1: Combustion of material containing 2,4,5-T species. Table 2: Combustion of grass treated with 2,4,5-T herbicide (12///acre). CONFIDENTIAL SU B JE C T TO PROTECTIVE ORDER. 230 C23S3S STURGEON LITIGATION FROM THE FILES OF THIS FILE HAS BEEN FOR: - Documents des c r i bi ng the presence or p o t e n t i a l presence of di oxi n in Monsanto products. - Documents r e f e r r i n g to the t o x i c i t y of any Isomer of di o x i ns . - Documents r e f e r r i n g to the t o x i c i t y of any Chlorophenol or Chlorophenol product. - Documents r e f e r r i n g to health e f f e c t s as they relate to the above. May 25, 1984 W einberg Consulting Group Inc. -- ----------- ---------------- -- ------- -- ------------------------------------ (202)342-807 M M PennsyKini* AvenueNW, Suite301 W u tftvo n . D .C 20007 ATTORNEY'S WORK PRODUCT PRIVILEGED AND CONFIDENTIAL Roger Tompkins, Esquire Bowles, McDavid, Graff A Love Commerce Square, Suite 1600 Charleston, WV 25301-1386 Dear Roger: When we visited you in Charleston, we briefly discussed the validity of Alistair Hay's cell transform ation study on dioxin. You noted the need for an audit of this study, and that is the subject of this letter. We have found, by examining writings and depositions, a number of basic problems with Dr. Hay's study. His purported expertise in toxicology (mutagenicity, carcinogenicity) is based on this single experiment; a short term bioassay for |4* carcinogenicity. In actuality, Dr. Hav's only role was to suggest the te s t and provide tfr~ nnmr1^ fra* analysis. He does not understana tne experimental protocol or interpretation. / F o r example, Dr. Hay claims that the results show j TCDD to be a mutagen, when actually, this te st indicates potential c arcin o g en s./ ****"' The cells used in this assay were premalignant (already initiated so only second stage carcinogenesis could be observed). In his test, phenotypic (non-genetic) response was ascertained by growth in soft agar. To show permanent, genetic response, cell colonies should have to be injected into animals and observed for tumor growth. SUBJECT TO v n v u \. 023934 Monsanto Monsanto Company BOO N. Lindbargh Boulevard S t. Louis, Missouri 03107 Phono: (314) 694-1000 June 19, 1984 Weinberg Consulting Group Inc. 2828 Pennsylvania Avenue, N.W. Suite 301 Washington, D.C. 20007 Dear Myron: I've attached some literature which Bo Holmstedt sent Dave Snively last week. We've seen most of the articles he's sent, but haven't seen the monograph by Tuula Thunberg which ties it all together. The more I've gotten into it the more obsorbing its become. The central thesis is that most, perhaps all of the symptoms associated with 2,3,7,8-TCDD toxicity are also induced by hypovitaminosis A. According to Thunberg the symptoms in common between vitamin A deficiency and 2,3,7,8-TCDD toxicity are: 1) Delayed symptoms 2) Failure of normal growth and general kachexia 3) Keratosis 4) Acne 5) Epithelial lesions and infections 6) Immunosuppression 7) Reproductive abnormalities 8) Teratogenesis .9) Skeletal lesions 10) Lesions in the nervous system or neurological symptoms I'm sure that you are aware that only a small fraction of the bodies vitamin A is used in the visual cycle with the majority performing a multitude of. body functions many of which are poorly understood and probably many more unknown. It's inter esting that it is considered a fairly potent anticarcinogen which suggest that its absence would have a promotional effect as seen with 2,3,7,8-TCDD. The importance of these observations is that Thunberg and others have looked at vitamin A storage in rats dosed with 2,3,7,8-TCDD (as well as with other enzyme inducers) . They find that a storage in rat liver is considerably reduced by 2,3,7,8-TCDD. 02393(5 " <5 Page 2 Myron Weinberg Thunberg also indicates that the amount or extent of reduction in various animal species correlates with 2,3,7,8-TCDD toxicity in the various species ie for a given 2,3,7,8-TCDD dose (at the right level) a more severe reduction in vitamin A storage is seen in male Hartley guinea pigs than in Syrian hampsters. This would seem to suggest that not only is "dioxin" toxicity lacking in uniqueness, but that the biochemical effect is one seen to some extent in this country but a chronic problem in many places in the third world; that of hypovitaminosis A Let me know your thoughts. Sincerely, Allan M. Ford, Ph.D P.S. If I haven't discussed this with Dr. Karch by the time you get this, I will shortly. cc: Nathan Karch W. McCarville G. Roush J. Wilson K. Johnson C23837 # Fig, 2. MethyMUoUtion o f hexmch h robenxene (H C B), petuachloeonitrobensene (PCNB), end pensachtorothloamisole (PC TA); oho B it turnover o f C H groups in PCTA m t bnmethylihiotetrachiorobenzrne (bkWTTCfl). Pathways A and B, tn d k* * by the curved errm do net represent mi equilibrium. The forward arrows'represent a C -S iyaxe-mtdiated methylthiolation. The reverse arrows repm ent either enm ohepetk circulation or environmental circulation o f them ethyithio-containm i metabolites (pathway C, Fig. 1). able as substrates for the C-S lyases. This may be possible in the case of the microfloral C-S lyases by analysis of the divalent sulfur present in the bile of animals on various diets. The divalent sulfur of interest would be present in MAP metabolites and 5-glucuronides. NonpolarmethyithMation products A further consequence of the exist ence of the methylthiolation pathway is the generation of nonpolar compounds (CHsS-containing) from the polar con jugates that were formed to facilitate excretion of the original xenobiotic from the cell. These nonpolar methyl thiolation products behave as new xenobiotics in the biosphere. Examples are'the CHjS-, CH3SO- and CHjSOjcontaining metabolites produced in the metabolism of polychlorinated biphenyls13, DDT15, hexachlorobenzene (HCB). and pentachloronitrobenzene (PCNB). Recent reports concerning the meta bolism o f HCB, PCNB, and pentachlorothioanisole (PCTA) indicate that methylthiolation may function to forni CH.iS-containing residues that persist and turnover in the. environment. This process is outlined in Fig. 2. HCB and PCNB are assumed to be metabolized in part t PCTA via the methylthiolation pathway. PCTA is, in turn, known to be metabolized to bis-methylthiotetrachlorobenzene (b&MTTCB) by microflora-mediated methylthiolation16. BirMTTCB is metabolized through the same process. However in this case, a CH^S-group, and not a chlorine, is replaced17 which indicates that the CH^S-group is a bet ter leaving group for OSH conjugation than the remaining chlorine groups. This same process, i.e. turnover of CH>S-groups, functions in the meta- holism of PCTA and other methylthio-containing compounds. Turnover of CHjS-groups by the outlined process (Fig. 2 and sequence C, Fig. 1) requires the utilization o f GSH and results in Die transient formation o f thiols that again utilize SAM. Turnover o f methyhhio-groups in hirMTTCB metabolism has been demonstrated only in rats17. However, there is evidence that the processes shown in Fig. 2 can occur, at least in part, not only in mammals, but also in fish, protozoa, fungi, and plants. The ability of a number o f species to cany out any one or all of the steps shown in Fig. 2 indicates that various species could be involved in the formation and turnover of btrMTTCB in the environ ment. Therefore, methylthiolation of HCB., PCNB. and PCTA may be responsible for the high residues of buMTTCB and PCTA present in mus sels collected along the coast of Hol land. Such residues represent another potential source of thiols that could be involved in the toxic processes con sidered earlier. Methylthiolation does function to increase the turnover of residues from HCB in rats. About 70% o f the 14C from oral doses o f [14C]HCB (5 mg kg' 1) were left in the bodies of rats seven days after dosing18 . Three days after . oral doses (IS mg k r 1) o f either [W)PCTA16 or [HC]birMTTCB17, only 5% o f the l*C remained in the bodies. Therefore, methylthiolation can function to increase the turnover of HCB-derived residues in rats. However, the rate of turnover and ultimate fate of these residues in the environment is not known. In another example, however, methylthiolation of polychlorinated biphenyls produces CHjSOj-contain- ing metabolites that persist in lung tissue11. Extension of this pathway to the metabolism o f methylthiolation pro ducts from other xenobiotics Indicates that CHjS*i CH3SO- and CHjSOj-containing compounds can be persistent or terminal residues in the environment: The formation o f sudi products could be a rate-limiting step in the ultimate biodegradation o f many xenobiotics. The wide distribution o f enzymes involved in methylthiolation and the number of spedes capable of being involved in methylthiolation indicate that methylthiolation products are present in the food chain. Coachistons In this review we have described the existence of a novel pathway for the generation o f methylthiolated metab olites of xenobiotics involved in C-S lyase activity in the intestinal'micro flora. Mercapturic add pathway metab olites of xenobiotics have been shown to produce metabolites of toxicological significance during the process of methylthiolation and this process may therefore be regarded as a pathway which involves metabolic activation of xenobiotics by the intestinal micro flora. Other examples of the role of the gut bacteria in generation of toxic metabolites are cleavage of cycasin, reduction o f nitroaromatics such as nitrotoluene, as well as formation of certain mutagenic compounds from unknown precursors. The significant metabolism of xenobiotics by the intes tinal microflora is not surprising in view of the quantitatively important break down of endogenous compounds such as steroid hormones which has been shown to be effected by the gut bac teria19*3?. Conceivably, the metabolic C23950 tiP S - December 1984 517 mic myocardium. The mechanism by which antiangina'l drugs attenuate the coronary occlusion-induced pH decrease, however, remains to be investigated. 1 lchihara, K. and Abiko. Y. (1973). Am: J. P hysiol 229.1385-1399 2 (chihara, K. and Abiko, Y. (1977) Am . J. P hysiol 232.H349-H353 3 lchihara, K. and Abiko. Y. (1975) Ezptrirnaa 3],477-478 4 Icbihan. K. and Abiko. Y. (1973) Expcriaatia 31.1198-1199 5 Abiko. Y., lchihara, K. and lchihara, M. (1979) la International Adaiat Pond D bcuuion (Udukn. P. R.,Kiomra, E. and Taira, H.. edi), pp. 1-6, Excerpt* Medica, Amsterdam A lchihara.K. andAbiko,Y. (1982)Jpn. Heart 7.23,81%*28 7 Abiko. Y. and Sakai. K. (1980) Ear. J. PharmacoL 64,239-248 t lchihara. K. and Abiko. Y. (1962) 7. Pharm acol Exp.Thar.222.720-725 9 Gevcn, W. (1977)7. Mol. Cell. Cardiol. 9, 867-874 10 lchihara. K., lchihara, M. and Abiko, Y. (1979)7.Pharmacol. Exp. Thsr. 209,275-281 11 Shibano,T. and Abiko, Y. (1983)Arch. Ins. Pharmacodyn. Thar. 264,274*289 12 Izumi. T.. Sakai. K. and Abiko. Y. (1982) Naunyn-Schmied. Arch. Pharmacol. 318, 340-343 13 Sakai. K. and Abiko. Y. (1984) 9dt International Congress o f Pharmacology (London) 14 Abiko. Y., Nishunura, T. and Sakai, K. (1984) Br. J.Pharmacol 81. 409uU 15 Picper. G.M.. Todd, G. L,, Wu. S.T., Salhany, J.M., Oaytoo, F. C. and Elliot, R. S.(1980)Cordiovasc. Res. 14,646-633 16 Nakaaawa, M.. Katano, Y., Imai, S., Mauusbiu. K. and Ohudd. M. (1962) 7. Cardiovast. Pharmacol. 4,700-704 17 Sakai. K. and Abiko. Y. (1982) Grtufeun R a . SI.735-742 18 AM A Drug Evaluations (1983), 5th edition (Am. Med. Aaaoc.. ed.), pp. 651-668, Saunden Company, Philadelphia 19 Sashida, H., Miyakawa, K. and Abiko, Y. (1982)Exp. Mol. Pathol. 37,413-426 Yorurhi A biko, M .D ., Ph.D . is a Professor o f Pharmacology, Asahikawo Medical College, the northernmost medical school in Japan. Before joining the Asohikawa Medical College in 1974, he worked at the H okkaido University School o f Medicine (1953-1963), at the University o f Cin* cinnoti College o f Medicine (1963-1964), and at the Fvluuhima Medical College (1964-1974). Kazoo lchihara, Ph.D . is an Assodate Professor o f Pharmacology, Asohikawa Medical College. A fter graduating from the Faculty o f Pharmaceu tics, H okkaido University in 1973, he worked on the myocardial metabolism in the ischemic heart with D r A biko at the Fukushima M edical CoUegt fo r one year and at th Asohikawa Medical Col legefrom 1974to the present He did postdoctoral research training in the Department o f Physiol ogy, H m hey M edical Center o fthe Pennsylvania State University (1979-1961). KenjiSakai Ph.D . is an Instructoro fPharmacol ogy, Asohikawa Medical CoUegt. A fter graduat ingfrom the Tohoku College o fPharmacy (1978), he joined D r A biko in Asohikawa. Mercapturic acid pathway metabolites of xenobiotics: generation of potentially toxic metabolites during enterohepatic circulation Jerome Bakke and Jan-Ake Gustafsson* Metabolism and Radiation Research Laboratory, A R S, VS Departmento fAgriculture. State University Station, Forgo, N D 58/93 USA. 9Department o f Medical Nutrition, Korolinska Institute, Huddutge University Hospital, S-14166 Huddinge, Sweden. No worrmndee are h a rm implied b y the US Department o f Agriculture. The body detoxifies nutny xenobiotics via the mercapturic acid pathway (MAP). Unfortunately, in some cases this processprovides intermediate metabolites that are substrates fo r mutagenltoxin-forming reactions - in particular reactive thiols. Jerome Bakke and Jan-Ake Gustafsson describe this process o f methylthiolation and the novel metabolic pathway involved. They highlight the increasing load o f xenobiotics imposed by industrial processes and the toxicological consequences o f their methylthiolated metabolites in individual organisms and the biosphere in general. Many xenobiotics, including known carcinogens, are metabolized in the mercapturic add pathway (MAP, defined in Fig. 1) after conjugation with glutathione (GSH). This conjuga tion is assumed to effect the detoxifica tion of the xenobiotic upon excretion of the GSH-conjugate from the cell in which it was formed1. However, catabolites of some MAP metabolites have been shown to be mutagens2-3, to be toxic to the kidney3, and to bind to macromolecules, i.e. bound residues4. The toxicological activities o f this cata bolic system are thought to reside in formation o f reactive thiols. These thiols are formed upon cleavage of the cysteine conjugates by cysteine conju gate 7-lyase (O-S lyase5). This catabo lic system is outlined in Fig. 1 (follow pathway A ). Methylation o f the thiols to form methylthio-(CH3S-)contaioing metab olites is another detoxication step. However, methylation is only a tran sient detoxication because the resultant CHjS-, CH3SO- and CHjSOrContaining metabolites exhibit polarities simi lar to the parent xenobiotic; these can therefore be absorbed into txlls where they must once again undergo detox ication (e.gv pathways C , Fig. 1). In essence, the methylation creates new xenobiotics that can add not only to the detoxication burden of the organisms that formed them but also to that of subsequent compartments of the environment. For convenience, the overall process outlined in Fig. 1will be termed methylthiolation. Two metabolic processes are known which result in the methylthiolation of xenobiotics. These are the C-S lyase pathway described above (pathway A . Fig. 1). and the reaction of activated intermediates of xenobiotics with methionine and/or methionyi residues of proteins to form sulfonium ions (sequence B, Fig. 1). These sulfonium ions must then dissociate to form C23947 " --------------- --i f V `i m r r * -- O l i l - M r t W O m . 318 i *fj- t/r e e m w r /w Xtnoblotte (X) activated X (rg . arena oxide or N-hydroxyfation) B -- 7 ------R -S -C H j (e.g. mettilonyl resldueaof proteins) + X-S^CHj binding to macromolecufas other 'phase I and II' reactions Marcapturic Acid Pathway i I X-S-gfutathtona I X-S-cystelnytglydne /I . X-Scystefne I X-S-(W-ac*ty1)cy8tetne nonextractabie residues (tissues and/or feces) Fig. I . PotMmays --ding to the introduction o f metftyhkkh$roupe into xenobioda via ft* mercaptaric pathway (A) m i via lulfonixm ion fo rmation (B). Pathway A ap p et t i to dom nete quantimtvely* Pathway C represent! not only enterohepatic circulation, but obo environmental circulation end fe a o f m ethybhio-conaininf metabolites. Dnshed arrows designate suspected but unproven pathways. SAM S admosybnethionine. CHjS-conUining metabolites. 1 Satfonhnn h pathway The tulfonium iOn pathway has not yet been shown to produce, in vivo, CHjS-containing metabolites that are excreted in either urine or feces. However, methyl sulfides o f 2-acetylaminofluorene were isolated after alka line hydrolysis o f proteins isolated from liven, o f rats that were dosed with 2-acetylaminofluorene6. The quantita tive importance of the sulfonium ion pathway with respect to production of methylthio-oontaining compounds from the strove xenobiotics or other activated xenobiotics has not been determined, but it is considered to be less prevalent than the C -S lyase path way. The sulfonium ion pathway is of toxicological importance because it results in binding of xenobiotics to tis sue mscTomolecuIes. The prevalence o f methylthlolation in xenobiotic metabolism is demon strated by the varied structures that are known to be metabolized to CH3S-containing metabolites (Table I). Most of the xenobiotics that are metabolized to CH3S-containing metabolites are also metabolized in the MAP (noted in Table I). C-S lyase pathway The C-S lyase-mediated methylthio- latlon pathway (sequence A , Fig. 1) is known to function in plants and mamn a ls, and probably functions in fish. protozoa and fungi. The key reaction in this pathway is catalysed by C-S lyases which cleave 5-substituted cysteine conjugates to yield pyruvate, ammonia and the thiols of the original xenobio tics. In vivo, the thiols are subsequently methylated or conjugated with glucur onic add. In the latter case, the 5glucuronides can undergo enterohepatk circulation during which the thiols areregenerated by intestinal glucuroni dases. The thiol-containing aglycones can be absorbed, methylated and excreted. In many cases, the CH*Scontaining metabolites are oxidized to the methylsulfinyl (CH3SO-) and methylsulfonyl (CH3SO r) analogues. In mammals and possibly other bileproducing species, two chemically identical, but physiologically different, C-S lyase-mediated methylthiolation pathways operate. This also applies to species which exist in symbiosis with an intestinal microflora. One pathway is mediated by tissue C-S lyases7'* which c a n . cleave 5-cysteine conjugates formed on the first pass of xenobiotics through the tissues. The other pathway is mediated by C-S lyases located in the intestinal microflora3-9. These latter C S lyases can function on both dietary and biliary MAP metabolites. Specifldty ofC-S lyases The tissue and microfloral C-S lyases exhibit a specificity for 5-cysteine con jugates7. In addition, the tissue C-S lyaies, in vitro, exhibit specificity for C2X9AQ 5-cysteine conjugates with insaturatkm or aromaticity at the carbon of the xenobiotic moiety that is bonded to the cysteinyl sulfur*. This specificity has not been observed with microfloral C S lyases; however, it is not known how many different C-S lyases are present in the intestinal microflora. The differ ing specificities o f the tissue and micro floral C-S lyases are important because they determine, in part, the quantita tive roles played by the tissue and microfloral pathways. Because of the broader specificity o f the microfloral C-S lyases and the presence of the metabolic activities of the microflora discussed below, any MAP metabolite excreted with the bile or present in the diet is a potential substrate for micro flora] C-S lyases. Of the two C-S lyase mediated path ways, we believe the pathway mediated by the microflora to be quantitatively the most important in view of the levels of MAP metabolites o f various xeno biotics in the bile and the ability of intestinal systems to produce 5-cys teine conjugates from other MAP metabolites. For example, all the MAP metabolites o f propachlor (2-chloro-Af-isopropylacetanilide, a herbicide), including the sulfoxide of the mercapturic rid, produced varying amounts of the thiol (2-thjofo-AMsopropylacetanilide) upon their incuba tion with the intestinal microflora from pigs3. From these results, it was con cluded that peptidases, acylases and I I C O - y t w r t v o i /v T reductases present in the intestine were functioning to form the 5-cysteine con jugate from the MAP metabolites. Similar tim e systems would not supply these levels of 5-cysteine conjugates to the **** C-S lyases because of excre tion of the other MAP metabolites with bile and urine before metabolism back to the cysteine conjugate. Toxfcolofkal implications The biological function of the C-S lyase is not known. However, toxicolo gical consequences have been attri buted to their action. These conse quences include anemias10, renal mhMiar adenocarcinomas3 and nephro toxicity caused by C -S lyase cleavage of halogensted vinyl-S-cysteine conju gates to reactive thiols. These thiols have been shown to bind to proteins and inactivate the enzyme. C-S lyase cleavage of MAP metabolites has also been proposed to produce nonextractable fecaJ4-3 and plant residues result ing from metabolism of pesticides. The type of chemical bonding present in these residues has not been deter mined. Methylthiolation o f chlorinated hydrocarbons has also been shown to result in methybulfonyl-containing residues in animal tissues11. Here again, the significance of these residues is unknown other than that some drug metabolizing enzymes are induced by the administration of CHjS-conuining metabolites of m-, and p-chloroben zene. Another possibility for an advene role o f C-S lyases is the hemolytic anemia produced in ruminants fed a diet containing high levels o f kale10. The toxic factor is S-methylcystein sul foxide (present in high concentrations in brassicas) which could be reduced by the microflora to 5-methylcysteine and then serve as a substrate for C-S lyases to produce high levels o f roethanethiol. The products from both C-S lyase cleavage and methylthiolation there fore exhibit properties generally asso ciated with toxic compounds, i.e. bind ing to tissue m taom olecules and the formation of persistent residues. Carci nogens are believed to exert their toxic effect in binding o f reactive metabolites to macromolecules, and many carci nogens are metabolized via arene oxides to MAP metabolites. It is there fore possible that the products and/or intermediates of the methylthiolation pathways could be involved in carci nogenesis. Further metabolism The magnitude o f the burden of xenobiotics available to the C-S lyases as cysteine conjugates is now known. However, considering the varied struc tures that arc conjugated with glu tathione (GSH), the prevalence o f bili ary secretion o f the resultant* MAP metabolites, and the activity o f the intestinal microflora, we can assume that there is a high level o f production of thiols in the intestinal lumen. These thiols, because of their reactivity, must be disposed of by further metabolism. There appear to be three general routes of disposal: mthylation and excretion with the feces; mthylation or glucuronidation and excretion with the urine; and incorporation into the feces as nonextractable residues. Two of these routes involve 5-methylation which is presumed to be catalysed by 5-methyltransferases using 5-adenosylmethionine (SAM) as the methyl donor. The physiological locations of the 5-methyltransferases involved in metbylthiolation have not been determined. 5-methyltransferases are present in most tissues including the intestinal mucosa12. However, they have not been shown to be present in intestinal con tents?'12. This indicates that the inte- TABLE I. Xcaobvotics shown to be metaboltred to methylthio-containini metabolites. Polychlorinated biphenyls DDT Biphenyl Naphthalene Pbcnanthrcnc Carbamazepinc Branobcaxana -Phenacetin Acetaminophen Heaachlorobeniene Fcntarhloconitrobenzcpc PentadUomthioamaole Bramaaepaa l-Ailyt-3,$-d}etbyt-6-diiociiundl s K3Jnm Wjaopn'niylnnumlMia Caffeine 2,6-Dichlorobcnzoaiirile 2.6-Dichloroenumide 2,6-Dichlorothiohffinaiirtf 2-Chloro-6-mcthyithiobenzDtutrile l.l.l-TrifluMo-JV-p-mctbyM-tpheoyt- mlfonyOphcnyllmethannulfonainide Acetylaminofluorene O.ODkihyl-CK3.5,5*irictotoK>-2-pyridyi)* pboapborothioate 2-Acetimido-4-chlorowethylthiaiole 3-(5-Nitro-2-furyl)-2-(2-furyl)aayUiude Polychlorinated dibenio-p-dioxin* Hydroxyxanthine Propranolol 1,1Dichlorocthylene * Metfryfthio-mnraining metabolite isolated after a hydrolytic step. Q 2 3 9 4 9 final mucosa would be the first tissue in which thiols formed in the intestinal lumen could be methylated. The con sumption of methyl donon (SAM) by thiols produced in the intestine, together with the production of thiols capable of binding to tissue macromolecules, could work individually, together, or in concert with other carci nogens produced or present in the intestine to produce the toxic effect described below. Carcinogenesis According to Frber13, three condi tions are conducive to the occurrence of chemical carinogenesis, i.e. prolifer ating cells, initiation of the cancer, and promotion of cancerous growth. The three conditions could exist in the inte tine under the stress of a continuous supply of a variety of cysteine conju gates to the intestinal microflora. Gel proliferation is a natural process in tiu intestinal mucosa because the mucosi is constantly replacing cells iluffed of into the intestinal lumen. Initiation o carcinogenesis could be effected by tiu binding o f reactive thiols produced by C-S lyase cleavage of appropriate cys teine conjugates excreted with the bill or present in the diet. Promotion of tiu tumor could be mediate^ by the pro duction of a constant supply of various thiols either by the action of the thiob themselves or, in view of the fact tha diets deficient in methyl donon such as choline and methionine promote car dnogenesis13, by producing a localizer deficiency of SAM required for tbeL methylation. In this model, initiation could also tx effected by compounds formed as j result of the action of the microflora oi other biliary metabolites, such as tin regeneration of the parent xenobiotic as Renwick and Drasar14 have showi for benzo(a)pyrene. Upon incubatioi of biliary metabolites of ben zo(a)pyrene (which would presumably contain MAP metabolites) with intes final microflora, they found ben zo(a)pyrene was formed. Promotion o carcinogenesis need not be attribute to any specific MAP metabolite, but ti the burden of thiol precursora excrete with the bile or present in the diet. Thi description of chemical carcinogenesis could be extended to other tissues, suri as liver and kidney, where C-S lyases are present. Also, thiols produced by the flora which bypass the mucosa 5-methyltransferase could be involved in promotion of carcinogenesis in other tissues. To test this model requires knowledge of both the types and quan tities of thiol precursors that are avail- I JITS - Uecem ua processes occurring in the intestintJ microflora may be influenced by diet* tryhabitsand, consequently, the find* tagspresented inthisarticle may be of flat 1 M euter. A. (1983) S en n et 220, 472-477 ^2 R en n u f. U ., Sundvftll, A. end Hamel, C. ' (1978) C ham .-B iol In ta ra a io fu 20, 1-16 J G reen. T ., Nah. J. A ., O dum , J. and 11 B akke, J. E ., B e rjin a n . . L. and L an e n G. L. (1982) Saance 217, 645-647 12 W eisifer, R. A ., Pm kua. L. M. and Jakoby W . B. (1980) B tocham P harm acol. 29 28&S2887 relevance for the important issue of ' H ow ard, E. G (1983) E anahapaac D rug 13 Frber, E. (1983) E xirabcpauc D rug M etab, utritioQand cancer. Itispossible that M etabolism and C hem ical C arcutogtnasu oium and Chemical Caranogem further investigations concerning the M eeting. Stockholm , Sw eden. A bstract No. Srock hbe. Sw eden, A b stract No L2 influence of dietary components, e g. fiber, on the metabolism of xenobtotics, and perhaps also of endogenous compounds, may help to define a diet tftatcould minimize the formation of B.23. 126. s 4 Bakke. J E .. G uaufuoo. J.-A. and ^ G u au iaao o , B. E. (1980) Science 210, 433- 435 v 5 U n e n . G. L. and Bakke. J. E. (1983) ' X enobioac 13, 113-126 May 17-20 14 Reowicfc, A . G . and D raaar. B S (1976) N ature 263, 234-235 15 Jenson. S. and Jancaoc. B (1976) A m b io 5 257-260 16 B akke, J. E .. A sch b arb er, P. W ., Feil. V. J. toxicmetabolites during enterohepatic 6 D cB aun, J. R-. M ilkr. E. C. and M ilkr, and Guatalaaon, B. E. (1981) X enobioaca 11 circulation. J. A . (1970) Cancer Res 30. 577-595 173-178 7 C o lu c d , D F. and Buyakc, D. A . (1965) 17 B akke, J. E. (1983) C ham otphare 12. 793- B tocham . P harm acol 14. 457-466 798 4 Admcwftedfemeflts 8 Stevens, J. L. and Jakoby, W. B. (1982) Fed. 18 Me b endale, H. M , Fields, M. and \ This investigation was supported by Proc. 41. 1426 M atthew s. J. B . (1973)7. A gric. F ood Cham 9 Suzuki. S.. T oonaw a, H ., Id u tu rs, S.. 23. 261-265 grants from the Swedish Cancer Soci F ukazaw a. H . and T a te n h i. M. (1982) 19 Eriksson, H ., G uaiaftao n , J.- . an d SjOvall, ety and from the Swedish Board for B iochem . Pharm aco l. 11.2137-2140 i . (1968) E ur. 7. B tocham . 6, 219-220 planning and Co-ordination of 10 Sm itli. R. H . (1974) Rap. R ow an M at 30, 20 Erikaaoa, H . and G uau iaao n , J.-A . (1970) Research; 112-131 E ur. J. B tocham . 13, 198-202 New trends in lower urinary tract pharmacology function inmicturition iscontrolled by sn extremely complex neuromuscular mechanism. However, in the last few yean much new information has been compiled describing the detailed anat omy and innervation of this region3-5, opening new ways for further physio logical and pharmacological investiga tions. Karl-Erik Andersson Dpa n n a n t o f C linical P harm acology. L 'n iv en u y H ospital, S-221 B i L u n d . Sw eden. Screaa Incontinence and urethral a-adrenoceptors At least three factors are of impor tanceformaintenance ofcontinence in The very common disorder of urinary incontinence has proved resistant to women; the urethral resting pressure, satisfactory pharmacological therapy. Karl-Erik Andersson describes how recent the intra-vesical pressure, and the developments in understanding ofthe complex neuromuscular mechanisms which transmission of the intra-abdominal control urinary function has opened up new areas for pharmacological investi pressure to the proximal part of the gations. urethra6.In stress incontinence there isprobably adefect inthe transmission In the lastdecade, disturbances of the tions of the bladder (`unstable blad of the intra-abdominal pressure to the function of the lower urinary tract der'). Many drugs have been tried for urethra. Partjy because of this, the have attracted an increasing interest, inhibition of the bladder contractions urethral resistance to flow is not reflected by, e.g., the contributions to in this condition, but their efficacy is enough to prevent leakage of urine the annual meetings of the Interna limited and adverse reactions com when the intra-abdominal pressure is tional Continence Society. These dis mon2. rapidly increased. There is no abnor turbances can manifest themselves as The lack of satisfactory drug treat mal activity in the bladder. As women urinary incontinence. This is a com ment in urinary incontinence depends with stress incontinence have a lower mon condition; in Britain it was esti ultimately on lack of knowledge about than normal intra-urethral pressure, mated that over two million people the basic physiological mechanisms the aim of pharmacological treatment sufferfrom the disorder1.The efficacy regulating the function of the bladder has been to increase this pressure. of available methods of treatment is and urethra, both under normal condi Several investigators have stressed the often unsatisfactory. Only a few tions and in different types of disease importance ofthesympathetic nervous women with urineleakage precipitated processes and injuries. Ithas therefore system tomaintain intra-urethralpres by stress (e.g. heavy lifting, coughing) been one ofthe aims ofresearch inthis sure - it is effectively lowered by a- aresuitable forsurgicaltreatment, and field to increase the amount of infor adrenoceptor blockers, such as phen- pharmacological treatment is restric mation concerning the innervation and oxybenzamine, phentolamine and pra ted to a few drugs with limited effi receptor functions in the bladder and zosin, and increased by a-adrencx-- - cacy. Another form of incontinence, urethra to secure a basis for develop toragonists. Clinically,orally-effec motor urge incontinence, ischaracter ment of rational pharmacological and drugs such as phenylpropanolamine ized by involuntary bladder emptying surgical methods of treatment. It is and ephedrine are the most widely associated with uncontrolled contrac generally accepted that vesicourethral used2.These drugs have no urethral C23951 O I * EJotmi Sana hWWtn