Document mpKZVnG7NZJYYmLV4ppxZKbG4

From The De9k Or DN DANLY ;JH 1 9 7 1 O \ c C_ vO H oIa - 'O y O \.'T t < , V ir- S'V/XVwO r \. r- c<riu*~% C CO X T p V -^ o * ~ S ^ i "-> Cu 1 CLOD vS r r tZ co^ : O .IL \ - u V > --c A ^ VV^ f ) ( / 'r-^yV'^ V ^ v ^ V o*r- ? C07408 CONFIDENTIAL  i ' ,, ' Ce FROM (N AM S-LO CATiO N -P H O N E ) ^ H. M. Keating - G4WA (4-8171) ' tQ . V J>ATE Jonfe 2 REFERENCE \^ ^ L Q C .- w C-TD c.<?\ \w^c_ o ^ " %- f% -w ^V cV ___ y J. T. Garrett - A2SG S. L. Hunter - G4WA P'nO G. E. Paget - A2SC W. B. Papageorge - G4WA TO W. G. Jv W. R. Richard - R2E M - COMPANY CONFIDENTIAL c/' V * /j (L e m u t,h. 6 u u S fu c/ <2^ /( 2 f t tJ 'e r e * *}'/ , C a + see K y / ^ to tS oa This is a request to identify all processes that in the light of st+R y feasible chemistry might give rise to CDDfs in the workplace, product, by-product, or effluent. The request stems from an OSHA visit to tbe W. G. Krummrich plant, and a subsequent allegation by OSHA that CDl were detected in measurable quantities in one product sample and o n e ^ wipe sample. A citation followed, alleging violation of the OSB Acte General Duty Clause. S To be certain that we are using a common base of understanding, I am including below excerpts from a draft prepared by Dr. G. E. Paget, DMEH. "Nomenclature It is common for the class of compounds under discussion to be referred to as "dioxins" (c.f. Krummrich citation). This is incorrect. We are specifically considering chlorinated dibenzo dioxins, and to avoid falling into the trap of loose nomenclature, this whole class will be referred to here as CDD's. Of particular concern within that generic class are the tetrachlorinated species and this group (of 22 isomers) will be referred to as TCDD's. Of this group of 22 isomers, one is of particular; concern. This is the. 2*3.7.8. isomer and this will be referred to as 2.3.7.8. TCDD. Background Biology Several? CDDVs are extremely toxic experimentally and 2.3.7.8. TCDD is especially so. In various species, single doses of 2.3.7.8. TCDD of the order of 1 - 6.>fcg/kg. produce serious and irreversible changes, ranging from death through teratogenesis and carcinogenesis to longlasting biochemical anomalies. V! IN - IO (REV. 8 / 7 7 ) Knowledge of the toxicology of CDD's and 2.3.7.8. TCDD in particular is still far from comprehensive. Even more incomplete is knowledge of human response to these C07409 ^ materials. It is known that the most obvious response is chloracne and that CDD's are powerful causes of this condition. Other effects of exposure are less well established. However, evidence CONFIDENTIAL D. E. Danly, et al -2- June 22, 1979 Background Biology (continued) suggests inconclusively that humans may be less susceptible to CDD toxicity than most laboratory species. In the experimental animal, the half-life of TCDD's is of the order of 30 days. In the environment TCDD1s are more stable, with a half-life of circa one year. They are firmly bound to soil and do not readily migrate. In the absence of precise knowledge of human susceptibility, and in view of the extreme sensitivity of some species, it is obvious that very small amounts of TCDD in the human environment, in certain circumstances of intimate exposure, or of abuse or misuse, or casual worker procedures, may (we repeat may) , have significance. Background Chemistry It appears that in the very small concentrations of which we speak of CDD's (p.p.b. and ppt, o r ^ g / m 2) CDD's can be formed under a wide variety of conditions. They appear to be very frequently formed by combustion of many materials and may, therefore, be of ubiquitous appearance at low levels in the environment. Of importance to Monsanto is the fact that it appears that they may be formed when any aromatic substance is chlorinated under industrial conditions where, for example/ some water usually contaminates so-called dry intermediates although some chemical processes (chlorination of phenols under alkaline conditions) are more likely to produce CDD's than others. Because of Monsanto's involvement with such processes at several locations, CDD's are of particular concern to us. Analytical Background It is important to distinguish between detection of CDD's generically, the detection of TCDD's generically, and the specific identification of 2.3.7.8. TCDD. Many laboratories, including Monsanto, can perform the first two identifications at high levels of sensitivity. Few can perform specific 2.3.7.8. TCDD analysis at high sensitivity. Some that claim to be able to do so are mistaken. It is probable that Dow can perform this analysis, although they do not do so routinely. It is also important to bear in mind the matrix in which the CDD of interest is found. Some matrices - water for example - present fewer problems than others and Monsanto analysts are already capable of performing analysis of TCDD in such matrices at very high levels of sensitivity. Other matrices, particularly products, present very many more difficulties, and we have to accept that in some circumstances the state of the art will not permit low level CDD analysis, and particularly 2.3.7.8. TCDD identification. Additionally, achieving the ultimate in analytical sensitivity for TCDD*s is extremely demanding of time and skilled labor. It cannot be undertaken on a routine "quality assurance" basis with any acceptable commitment of resources, but must be reserved for special circumstances." C_ O_ .N,.F..tI_DaE- *NTIAa iL coP / v D. E. Danly, et al -3- June 22, 1979 To give guidance on what might be considered "presence of CDD*s", OSHA alleged having found 300 ppb of total CDD's in a product sample, and 642 nanograms of CDD's in a wipe sample probably taken on a surface of 100 sq. CM. It is probable they would have made the same allegations even if the quantities they report were much lower. As further background information, attached is a copy of a Dow article from C St E News. Response by July 13 is requested. O J 1 7 ) H. M. Keating HMKrku Attachment C O N F ID E N T S L. C0741J. .Nevys Feature 9 Scientists and regulators exam ine Dow Chem ical's claim that most dioxins enter the environment from combustion, not pesticide manufacture Rebecca L. Rawls C&EN, Washington According to the ancient Greeks, Pro metheus stole fire from the gods and brought it to mankind* thereby bringing civilization, culture, and benign technol ogy. But the effects of this deed were not all beneficial. Zeus, angered that humans now had fire as the gods did, sent Pandora with a box, containing all the plagues and disasters of mankind, to take away some of the advantage that having fire had brought to humans. There may be some chemical tru th in "vis Greek myth. In a report it calls "The /race Chemistries of Fire," Dow Chemi cal recently has said th at several chlori nated organic compounds that have been causing increased concern as environ mental pollutants may not be exclusively the products of the synthetic chemical industry. They are also being formed in trace amounts in all forms of combustion. Among the chemicals they find in the emissions of everything from power plant stacks to home fireplaces to automobile exhausts are polychlorinated biphenyls, polychlorinated dibenzofurans, and dioxins. It is the possibility th at dioxins--and particularly 2,3,7,8-tetrachlorodibenzop-dioxin (or simply 2,3,7,8-TCDD)-- might be entering the environment from a wide variety of combustion sites th a t is focusing the attention of scientists, envi ronmentalists, and government officials on the Dow report. 2,3,7,8-TCDD is one of the most toxic substances known for some laboratory animals, causing death at levels as low as 0.6 microgram per kg body weight in cer tain species. Its presence as a contaminant in Agent Orange, an herbicide used by the Air Force to defoliate sections of Vietnam in the late 1960's, raised ethical questions th at eventually resulted in discontinua- n of the herbicide spraying program in *0. These questions form the basis of a class action suit brought this winter by Vietnam veterans who claim their health has been jeopardized by exposure to this chemical. 2,3,7,8-TCDD is also present in several herbicides used in the U.S. The question of whether these herbicides with 2.3.7.8- TCDD cause unacceptable risk to human health and the environment has been a matter of concern to the Environ mental Protection Agency practically since it was established. EPA is currently in the Rebuttable Presumption Against Registration (RPAR) stage of its review of the safety of 2,4,5-trichIorophenoxyacetic acid (2,4,5-T), the first pesticide containing 2.3.7.8- TCDD to come Up for review. EPA requires 2,3,7,8-TCDD concentrations in 2,4,5-T to be less than 0.1 ppm. Current manufacturing practices result in about one tenth this level. The RPAR process involves a year-long period of scientific ' evaluation and public comment on the benefits and risks associated with use of the pesticide. A decision based on this evaluation is expected from EPA in April. Dow's interest in this area, too, centers on dioxins. The company is one of three major producers of 2,4,5-T in the U.S. (the other two are Thompson-Hayward Chemical Co. and Transvaal Inc.) and the only U.S. producer of 2,4,5-trichlorophenol, the chemical intermediate whose production inevitably results in some formation of 2,3,7,8-TCDD. It is 2,3,7,8TCDD from this source that eventually is carried over into 2,4,5-T and other prod ucts. Dow began the four-and-a-half-month study that resulted in its report as an ef fort to find the source of parts-per-billion levels of "a compound that satisfies the analytical criteria of 2,3,7,8-TCDD" in fish taken from the Tittabawassee River on which Dow's Midland, Mich., chemical complex is located. This is the plant where Dow makes 2,4,5-T, 2,4,5-trichlorophenol, and another dioxin-containing com pound, pentachlorophenol. "If you were standing outside Dow Chemical Co., looking in at us, you'd say, `Aha! You've got dioxins in the T ittaba Dow rsearchers take samples fr m wassee River, and therefore, they are Tittabawassee River near Dow's p ^ n _ oto rsEtf- 73 Bumb: our motives were suspect coming from 2,4,5-T, trichlorophenol, and pentachlorophenol. The way they are getting into the river is you are discharg ing them in the waste streams, they are going through your collection pond, and they are coming out into the river,1" says Dr. Robert Bumb, director of research for Dow's production unit, called the Michi gan division. "And, I've got to admit, that makes good sense, if you are standing outside and looking in. If you are inside Dow, you wouldn't say that. We monitor those plants very, very closely, and we didn't think at the time the dioxins were coming from those plants," he con tinues. Dow reported its findings of 2,3,7,8- TCDD in fish to Michigan's Department of Natural Resources last June. The company made a commitment to the state authorities that it would find the source of the dioxins in the fish. A task force was put together, headed by Bumb, which averaged about 30 members. This group worked full-time on the project from early July until mid-November, when the "Trace Chemistries of Fire" report was released. Altogether, Dow spent about $1.8 million on the project, primarily to collect samples and analyze them. Dow announced its findings at a press conference instead of taking the more conventional route of publishing them in the scientific literature (C&EN, Nov. 20, 1978, page 7). It concluded that the diox ins in the river did not get there from the discharges of its plant, but rather from "normal combustion processes that occur everywhere." "We now think dioxins have been with us since the advent of fire," Bumb said at the time. "The only thing that's different is our new ability to detect them in the environment." Not surprisingly, the report was re ceived with skepticism by most of the press and by some government officials. The conclusions were so clearly favorable to the company that, as Bumb puts it, "Our motives were immediately sus pect," However, Dow's report does not exactly say that there is no connection between the Dow plant and the dioxins found in the river. "Dow has never said that the dioxins in the Tittabawassee didn't come from Dow," says Bumb. "We say they probably are coming from this plant. But there is a difference. We say that the dioxins that are in the Tittabawassee, we believe, are due to combustion and not due to discharges from the pesticide or herbicide plants." Dow also doesn't say, according to Bumb, that because dioxins are being produced throughout the environment and are ubiquitous, their contamination in pesticides is not important. "If you think it's our position to say it's fine that pesticides contain dioxins, we're producing them everywhere, so let's not worry about them, that's not what we're saying," he insists. "T he concentration of dioxins in the environment from combustion and the concentration from the use of pesticides are separate issues, and we have never tried to make them the same issue." Taken as a separate issue, Dow insists, its study of dioxins from combustion is a fascinating and significant piece of envi ronmental chemistry. One .of the first things that it did after setting up the task force was to take dust samples from its own analytical laboratories to calibrate the "baseline" level for its analytical techniques. Many studies of 2,3,7,8- 2 ,3 ,7 ,8-TC D D 's toxicity has been widely studied on a variety of anim als 2 ,3 , 7, 8-Tetrachlorodibenzo-p-dioxin is, without question, very toxic stuff. But a consensus has not yet developed among scientists as to just how toxic it is, especially to humans. Toxicologically, 2,3,7,8-TCDD is a puzzling compound. It has been one of the most extensively studied of all chemicals, but so far its mechanism of action has not been established. Acute toxic doses vary as much as 1000 times from one animal species to another, indeed, it doesn't seem to affect even the same organs in all animal species. Many of its oncogenic and teratogenic effects are both contested and confus ing. And what its toxicological effect is on man, based on animal studies and limited human exposure, is currently the subject of fiery debate. Much of this debate is being focused by the Environmental Protection Agen cy's pending review of the continued registration and use of pesticides that contain trace amounts of 2,3,7,8-TCDD as an impurity. EPA is attempting to pull together all available data on human and animal toxicity to evaluate the possible hazards involved in use of those pesti cides. Its report, based on a year of in tensive scientific review and public comment, is due in April. Meantime, some of what's known about 2,3,7,8-TCDD toxicity follows: Acute toxicity. The most sensitive laboratory animal to 2,3,7,8-TCDD is the guinea pig, for which the LD50 level, or the dose needed to cause death in half the animals receiving it, is 0.6 microgram per kg of body weight given as an oral dose. This toxicity level is compa rable to that of some of the most toxic natural substances, such as mold toxins, and it makes 2,3,7,8-TCDD one of the most acutely toxic substances known. Other lab animals are much less sensitive than the guinea pig, however. Male rats have an oral LD50 of 22 mi crograms per kg; female rats are only about half as sensitive with an LD50 of 45 micrograms per kg. Recent data on monkeys set their oral LD50 for 2,3,7,8-TCDD at less than 70 micro grams per kg. Rabbits are less sensitive, at 115 micrograms per kg, and dogs are even less sensitive. In some animals TCDD leads to death by causing lesions in the liver. Dr. Richard J. Kociba of Dow Chemical Toxicology Research Laboratory ex plains. But in some other species, in cluding the very sensitive guinea pig, no single organ appears damaged enough, to be responsible for the animal's death. Instead, the animal seems to waste away in a syndrome that looks much like starvation. Estimates of acute toxicity to humans a re undergoing revision based on evi dence collected following a chemical plant release in Seveso, Italy, in July 1976. The explosive release spread 2,3,7,8-TCDD, contained as a contam inant in trichlorophenol, along with other contained substances over an area of several hundred acres near the plant (C&EN, Aug. 23, 1976, page 27). Esti mates vary as to how much 2,3,7,8TCDD was dispersed over this area but center on several kilograms. " If you calculate it out, that dose should have killed every human being within miles," based on projections from acute toxicity in guinea pigs, says Ko ciba. "But, to the best of my knowledge, the only thing that was observed in hu mans following the Seveso incident was some chloracne [a skin condition as sociated with 2,3,7,8-TCDD exposure] and some transient liver toxicity that has pretty well reversed itself in the months following the episode," he says. No human deaths have been directly at tributed to 2,3,7,8-TCDD exposure at Seveso, Kociba says, ''When we ex trapolate to man, it appears that TCDD * 24 C&EN Feb. 12. 1979 UJTt t t 3 ---------------------------------------- TCVQ in the environment indicate that it bfeiks down fairly rapidly via photo degradation in the air. Where it persists, it usually does so in soil, water, or animal tissues. Dow found, however, not a background `evel of dioxin in the lab dust samples but evels in some samples of more than 8700 ppb for all dioxin species combined. These samples contained up to 2.6 ppb of 2,3,7,8-TCDD and the other 11.isomers that the analytical procedure used could not separate from the 2,3,7,8 isomer. It is the 2,3,7,8 isomer that contami nates trichlorophenol and its derivatives. These dust samples, however, contained an array of dioxins including octachloro, heptachloro, and hexachloro species and tetrachloro isomers that could be distin guished from the 2,3,7,8 isomer. "That told us that the problem was not a waterborne problem, but an airborne problem, because where we collected the samples in the laboratory had never been and could never be under water," Bumb explains. He goes on, "So we then said, all right, if the soil or air in Midland has dioxins in it, then we have to find control areas where there are not dioxins. So we went around to other cities in Michigan, cities outside of Michigan, and the upper part of the lower peninsula of Michigan and took samples. . . . We found varying amounts of the dioxin isomers, some of which included TCDD, some hexa, some hepta, and some octa species. These were dust and soil samples. We found chlori nated dioxins everywhere we looked, in k ' : tor taken at the Dow plant Holes: Table shows data from only a few of the samples taken Indude sample -jV* :rfwth the highest total dioxin concentration taken from each type of site. NLD. =.no detectable peak at 2 S times ^ mm .-fra ; i immri V i r i W t l n rr.~V---^ every sample we took, although not nec essarily TCDD. "We were now coming to the conclusion that finding them in cities and in the soil was riot a function of any use of pesticides or herbicides because these products aren't used in cities. We were coming to the conclusion that they were probably due to combustion." Why combustion? Because of the high concentrations of dioxins found in cities and the fact that the route of entry into the environment was by air. "The only things that have large volume exhausts in cities are combustion sources," Bumb says, "so that's got to be one route." "We then looked at what you've got to consider unusual sources," Bumb ex plains, " To t example, the mufflers of diesel automobiles. We grilled steaks. . . . We scraped in fireplaces and checked home electrostatic precipitators, and the dioxins on an acute basis is probably much less Preliminary data from a study cur toxic than for some of the laboratory rently under way, sponsored by the Na animals," he concludes. tional Cancer Institute, using the same Oncogenic effects. Oncogenicity is animal strain shows a similar increase the property of a substance to produce in the incidence of certain tumors at high or induce benign or malignant tumors in dose levels and a low dose level that can living animals. Most of the work on the be tolerated for the lifetime of the animal oncogenicity of 2,3,7,8-TCDD by itself without any increase in tumors, Kociba (and not as a contaminant in pesticides) says. However, a study in the same rat has been done using the Sprague species conducted by Dr. J. P. Van Miller Dawley strain of rats. Kociba and his and associates at the University of associates at Dow have conducted Wisconsin has found an increased inci lifetime studies using 100 rats at each of dence of tumors in random organs at three exposure levels and 172 control doses as low as 5 ppt. animals. Animals at the highest dose One study finds an elevated incidence level, 2200 ppt in their food, showed of liver cancers among Vietnamese many toxicological effects, particularly following the wide application of Agent in the liver. They had a higher incidence Orange* a defoliant contaminated with of tumors of the liver, lung, and mouth. 2,3,7,8-TCDD, by the U.S. military in that However, this strain of rats has a sig country in 1961 and 1962. The increase, nificant incidence of spontaneous tu from 159 or 3 % of all cancers over a mors and showed a decrease in the seven-year period to 791 or 10% of all number of these tumors in the pituitary, cancers, begins in 1962. Cancer ordi uterus, mammary gland, pancreas, and adrenal gland. narily has a much longer latency period than this, and the study does not rule out At the intermediate dose level of 210 ppt, Kociba found some toxic effects to liver and other organs, but no ab- other possible sources of the increased cancer incidence. Seveso residents have not yet shown an increase in can lal incidence of tumors. The lowest cer incidence following their 2,3,7,8- dose, 22 ppt, showed neither oncogenic TCDD exposure. nor other toxicologic effects for the Teratogenicity. 2,3,7,8-TCDD causes lifetime of the animals. GQ7414'a number of birth abnormalities in some strains of m ice and rats, including cleft palate, and kidney, skeletal, and intes tinal tract abnormalities. These effects can be seen at levels as low as 0 .0 1 microgram per kg in some studies with rats. The compound is also toxic to the fetus at levels as low as 0.125 micro gram per kg in rats and 0.3 microgram per kg in m ice. Because of the narrow range that s found between the no-ef fect level on the fetus and the lethal ef fect on the mother, some researchers classify 2,3,7,8-TCDD as a weak animal teratogen. The evidence from Seveso con cerning teratogenic effects on humans so far has all been negative, according to Kociba. Extensive data from Italian health officials and the World Health Organization gave "absolutely no indi cation of increase in birth defects, abortions, or other teratogenic effects that could be correlated with the Seveso exposure," he says. However, EPA is studying eight women who five in or near Alsea, Ore., near forests that are sprayed periodically with herbicides containing 2,3,7,8-TCDD and who say there is a close correlation between the time the forests were sprayed and 10 miscarriages the women had between 1973 and 1977. .p io x n isomers 9 # ^ chlprodibenzo-p-dioxin.'-As jt isconrv-- ? fc?rnonly used, the termdioxinalmost at-v^ -/waysappliesto this compourid,buthow 2 rgir'manyof the other 74 chlorinateddioxins')1?] |||a re included as well is often not clear. ^^fA nd .this^compound also"; is,called &;nie*.troubJeXwitlusi^ Qames'fcc^5frrf^*cbe*^pind l'thisTtte.femviioo,. sometimes means oniy^s f'thy cii be imprecise^eading^'to cd n ^ lth e '2,3,7,8-isomer; and.-sometImes.T:1 ^rnltip!es*irtf;-hi;Qt^ of.y |^numbered;T to 4 and .5 tp 3 inLth0 dK^ separatibn possible wi^P the study was:V? Nagram);:Th^;^^th^eti^iiy-.ther are7S^begun already has been Vrpassed py$J sjpTOfe3^^n^'oc^^fethe time the study is completed.Thus, ^tachloro m 6le^e,two teomreieach of fu s in g consistent nomenclature even `i? ^^e'rnphochlorb'and heptacWoro^ fom s^;w ithin one study is sometimes tricky.- S lb'eah of the dichlorq a n d h e> ^ In :this : article, dioxin ;means any jJ ^ o r m s ^ i 4>^chvbf; thtrichlo'roland^m em bers of the entire family. TCDD ^ oo mans any or all of the tetrachloro iso-v* ^|mers.' The? specific: isomer^2,3,7,8^:-j[ were always there. And, again, we saw an unusual distribution of TCDD, hexa, hepta, and octa forms." At about this point, the Dow group became aware of European investigators who were drawing similar conclusions based on investigations of emissions from municipal incinerators and industrial heating facilities. K. Olie, P. I. Vermeulen, and 0 . Hutzihger of the University of Amsterdam, for example, found organochlorine compounds in the fly ash of mu nicipal incinerators located in Amsterdam and two other Dutch cities. Gas chromatography-mass spectroscopy analysis (GC-MS) detected only hexa, hepta, and octachloro isomers in the Amsterdam in cinerator ash, but these three types of dioxins as well as penta and tetrachloro isomers were detected in the fly ash from Arnhem and AJkaar. The data are not qualitative, but the researchers conclude th at municipal in cinerators may be a source of some of the organochlorine compounds.in the envi ronment. "The amount.of chlorodibenzodioxins and chlorodibenzofurans en tering the atmosphere via flue gases is probably quite small," they suggest. "However, because of the extreme toxicity of some of the components and the fact that most incineration plants are located in densely populated areas, extensive monitoring of such facilities might be advisable." Another group of European researchers vho have made similar findings includes )r. Hans Rudolf Buser and Dr. HansPaul Bosshardt of the Swiss Federal Re search Station and Dr. Christopher Rappe 28 C&ENFeb. 12. 1979 of the University of Umea, Sweden. These researchers have done quantitative studies of the polychlorinated dibenzop-dioxins and dibenzofurans they have isolated from the fly ash of a municipal incinerator and an industrial heating fa cility, both in Switzerland. GC-MS analysis of these fly ashes gives an almost identical pattern for the dioxins formed at each of these combustion sources, although the municipal inciner ator has higher concentrations of the more highly chlorinated dioxins. The pattern is the same as th at found when a mixture of various chlorophenols is burned under similar conditions. However, the munic ipal incinerator burned mainly household and some industrial refuse, not chemical wastes, and the industrial heater burned used industrial oils. Dow, too, has been investigating the ratios of one form of dioxin to another in its environmental samples. It has found that these samples do not match the ratios th at would be expected if their source were the contaminants found in either trichlorophenol or pentachlorophenol. Moreover, dioxins can be reduced step wise from the octachloro compound all the way to the free dioxin. These com pounds appear to exist in equilibrium with one another, and the Dow research ers have calculated the equilibrium pro portions of each compound. Although they can establish the equi librium ratio under experimental condi tions, Dow researchers say that predicting what levels will be found in the environ ment is a much more difficult matter. They suggest that samples taken directly C07415 sources probably will show fixed ratios of the various dioxins, as the European investigators find. As the dioxins persist in the environ ment, however, oxidation and reduction reactions occur simultaneously, and dioxins formed at different sources mix. The result is a band of dioxin concentra tions that doesn't match what's expected from any known source. "And that's what we do find," says Bumb. Dow also finds that the ability of diox ins to persist increases greatly when they are bound to particulate matter, com pared to when they are free. For instance, free dioxins are destroyed, or at least re duced to levels that cannot be detected by current methods, by incineration at 800 C with the proper combustion air condi tions. B ut dioxins bound to particulate matter can pass through incinerators op erating at temperatures as high as 1150 C with no change in concentration. It is this particulate-bound dioxin that is probably being emitted by combustion sources. Its environmental fate has not been studied so well as that of free 2,3,7,8-TCDD as it occurs in pesticides and which photodegrades fairly rapidly in the air and on leaves. The Dow report has not convinced Michigan's Department of Natural Re sources {DNR), for whom it was prepared, that the dioxins in Tittabawassee River fish did not come from discharges from Dow's manufacturing plant. Particularly worrisome are .the data showing that dioxin levels around the Dow plant and in Midland are higher than the sample levels found anywhere else, says Dennis L. Swanson of DNR's Office of Toxic Ma terials Control. An alternative explanation for the findings, he suggests, is that the dioxins found were made by Dow in production processes and not destroyed in the com pany's incinerators. In a letter to Dow in mid-January, DNR director Howard A. Tanner outlined what h e considers to be deficiencies in the Dow report and asked for further infor mation in a number of areas. Of first im portance, Swanson says, is the request for a total mass balance for dioxins and re lated oiganochlorine precursors found at the M idland plant. Swanson's other re quests center on the same question: Are the dioadns found in the environment near Midland missing from the Dow plant? To do a mass balance for all organo chlorine compounds, accounting for them to the level that would be needed to see a loss of th e am ount th a t's being found in the environment near Midland, would be a tremendously expensive undertaking, says Bumb. "I'm sure we could do it,*' at a cost, he quickly estimates, of $3 million to $5 million, "but I'm not sure we would gain m uch more insight into what we've discovered and what is being confirmed by other scientists." As for the fact th at the highest dioxin levels are found near the Dow plant, Bumb sees two reasons for this finding. The first is that Dow has more experience v j b ig ^ r iin g in Midland and knows how to l ^ l S S l t h e place where dioxins are most iikely to be. The other is that the Dow plant is a very laige combustion source, comparable to a?city of about 820,000 people. Thus it can be expected to be a source of dioxins. In addition to asking for more information from Dow, DNR is setting up a protocol for its own environmental survey to see if dioxins are indeed ubiquitous in Michigan. The state has no facilities for analyzing dioxins at the parts-per-trillion level, so analysis of samples will have to be done by a contractor, at a cost of about 5800 per sample. Michigan's Department of Public Health also has evaluated the Dow report This department is interested in the issue because edible fish species, are found in the Tittabawassee River. Thus, dioxins in the river pose a potential health hazard. The point is somewhat moot, however, because the fish also contain unacceptable levels of polychlorinated biphenyls, leading the department to issue warnings . not to eat the fish. The health department's conclusions about the Dow report are different from those of DNR. A staff memorandum forwarded to Dow by department director Maurige S, Reizen states the department's agreement with Dow that dioxins are ubiquitous and are produced by a variety of combustion sources. It recommends further research on dioxin-containing particulates. That research, it says, should include studies of their disversion in the environment and an eval- ation of their removal by waste-treatment processes. Dow itself has asked for a panel of outside chemists to review the analytical technique used in its study. The panel is headed by Dr. Lockhart B. Rogers, an analytical chemist at the University of Georgia. Rogers says the panel will try to determine the probable validity of the data, but will not try to arrive at any environmental conclusions. He hopes the review can begin this month. Another chemist, Dr. Michael L. Gross of the University of Nebraska, remarks: "I have difficulty with the statement that dioxins are ubiquitous." Gross has analyzed many environmental samples for 2,3,7,8-TCDD for PA. TCDD's are not always present in the samples he has an- alyzed, down to levels of 5 ppt* he says. He was not looking for other dioxins, "What I'm saying is that the conclusion may be awfully prejnature," he adds. He questions, among other things, where the chlorine is coming from to make chlori- nted dioxins at combustion sites. The compounds probably would be thermo- dynamically favored molecules under fire conditions, he says, but he thinks forma tion of polychlorinated species might re quire more chlorine than is typically present in fuels. "I think that Dow is probably right that you can form dioxins in fires, but I'm sure some fires are worse than others," he says. Says Dr. Thomas O. Tiernan of Wright State University, Dayton: "I think that Dow's conclusions are all possible, but it will take a great deal more work to prove Swanson: total mass balance for dioxins them." Tiernan, too, has been involved extensively in EPA's environmental probably correct," Young says. "I'm now monitoring program for dioxins. He sees wondering if our dioxin monitoring pro no difficulty in finding enough chlorine in ' gram is pointing in the right direction. the environment to supply dioxin-forming Perhaps we should be looking for all the reactions. Chlorine is widely used in water dioxins in the environment instead of just treatment, he says. In addition, he has 2.3.7.8-TCDD. Are the others there in found polychlorinated biphenyls in "ev- concentrations high enough to be a haz erything from Arctic ice to toenails," and ard?" these compounds could serve as a source According to Dr. Matthew S. Meselson of chlorine for dioxin formation, of Harvard University, "Dow is making Dr. Alvin L. Young of the Air Force's more of this interesting information than Occupational & Environmental Health is probablyjustified." It may be true that Laboratory in Texas says, "I was very combustion is putting a wide variety of skeptical when I first saw the Dow re dioxins into the environment, he says, but port." The European work that reaches it is important from the health viewpoint similar conclusions has been changing his to distinguish among the various dioxins. mind, however. Young has been involved By far th e most toxic is 2,3,7,8-TCDD, for the past decade with the Air Force's and "when 2,3,7,8-TCDD has been found extensive environmental monitoring in the environment, it has been found, to program for 2,3,7,8-TCDD. my knowledge, only near chlorinated He says recent work by Dr. B.Jansson phenols," he says. It may be that overall, and Dr. G. Sundstrom of the National combustion sources are introducing more Swedish Environment Protection Board dioxins into the environment than any finds TCDD at levels ranging from 10 to other source, he says, but the "hot spots" 140 ppm in the combustion products of of 2,3,7,8-TCDD contamination are al chlorinated phenols. These levels can be ways associated with chlorinated phenoxy reduced to less than 0.01. to 0.02 ppm compounds. when the combustion occurs in the pres- If the data exist anywhere to refute or ence of copper salts. The TCDD is not substantiate Dow's claim that combustion 2,3,7,8-TCDD, but the much less toxic sources are emitting measurable amounts 1,3,6,3 isomer, he says, of dioxins into th environment, those "I would now say the Dow work is data are at EPA. The agency has spon sored many environmental monitoring programs for dioxins. EPA's programs were not set up to look at combustion sites as possible sources of dioxins, and in many cases the monitoring was done only for 2.3.7.8-TCDD and not the other dioxins that Dow finds more prevalent. However, the data, taken together, may shed some light on Dow's hypothesis. EPA so far has not commented on the Dow findings. It is currently preparing an evaluation of the Dow report, expected to be released in the next few weeks. Among the sources of input for this evaluation are reviews o f the report that have been pre pared by all of EPA's regional offices that have had some experience with dioxins. Several of EPA's advisory committees with expertise in this matter also have been asked for comment. D C07416