Document ExaK5oNvrkRM7RGNKBQMQ4MOg

.`i * .-C* * `7 , fcvHAkt ^ f i P<J*V*4**tl* H Q | V* i POLYCHLORINATED DIOXINS AND DIBENZOFURANS '*'* i 1 tout 5. B. H&rniuoi* C h ris to ffe r Rappe INTRODUCTION Polychlorinated dibenzo--dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs) are two series o f t r ic y c l ic almost planar aromatic compounds which e x h ib it s im ila r physical, chemical and b io lo g ic a l prop erties and have been the subject o f much concern in recent years. They have been involved in accidents lik e th a t in Seveso, I t a l y in 1976, the f i r e in the Binghamton State O ffic e B uilding in 1981, the Love Canal incident in Niagara F a lls in 1979, the in to x ic a tio n s a t horse arenas in Missouri in 1971 and Times Beach, Missouri in 1982-83, the Yusho accidents in Japan in 1968 and in Taiwan in 1979 and also in the herbicide spraying program in Vietnam in the la te 1960s. The chemical stru ctu res and numbering o f these hazardous compounds are given in Figure 1. The number o f chlo rine atoms in these compounds can vary between one and e ig h t producing up to 75 PCDD and 135 PCDF p o sitio n al isomers as shown in Table 1. oo ^ O Animal studies and in v itr o experiments have in dicated th a t there is a pronounced d iffe re n c e in to x ic and b io lo g ic e ffe c ts among d iffe r e n t PCDD and PCDF isomers. A fa c to r o f 1 000 - 10 000 in d iffe re n c e can be found fo r so clo sely re la te d isomers as 2 ,3 ,7 ,8 - and 1,2 ,3 ,8 -te tra -C D D and 1 ,2 ,3 ,7 , 8 and 1 , 2 , 4 , 7 , 8-penta-CDD (Bradlaw and C asterlin e, 1976; Poland et h , 1976). The isomers with the highest a c t iv it y and the hiahest acute to x ic ftv are 783496 G B b fP 2 those having 4-6 chlo rine atoms and a ll la te r a l (2 ,3 ,7 and 8) positions substitu ted fo r c h lo rin e , see Table 2. The LDgQ-values fo r guinea pigs fo r 2 ,3 ,7 ,8 -te tra -C D D , 1 ,2 ,3 ,7 ,8-penta-CDD, 2 ,3 ,7 ,8 -te tra -C D F and 2 ,3 ,4 ,7 ,8 -penta-COF are a l l below 10 pg/kg (Me Connell e t _aK, , 1978; Moore e t a l . , 1979; Huff et a K , 1980). ANALYTICAL PROCEDURES Because o f the extreme t o x ic it y o f some o f the isomers o f PCDDs and PCDFs, highly s e n s itiv e and s p e c ific a n a ly tic a l techniques are required fo r the measurements. Detection le v e ls in human and ecological samples should be orders o f magnitude below the usual detection le v e ls obtained in pesticid e analysis. A detection level o f 1 pg (10 -12 g) or less might be required to detect 2 ,3 ,7 ,8 -te tra -C D D and the other isomers lis te d in Table 2 in 1 g sample (1 ppt or sub p p t). Any analysis at such low le v e ls are complicated by the presence o f a m ultitude o f other in te r fe r in g compounds. Another complication is the large number o f isomers (Table 1) and the large v a ria tio n in to x ic ity and b io lo g ic a l potency between closely re la te d isomers. A ris k evaluation should be based on the le v e ls o f the highly to x ic isomers found in isomer s p e c ific analyses. 783497 In recent years a great number o f a n a ly tic a l methods have been developed fo r the analysis o f trac e amounts o f PCDDs and PCDFs and e s p e c ia lly fo r the most to x ic 2 ,3 ,7 ,8 -te tr a -C D D .in environmental samples. The most s p e c ific o f these methods are based on mass spectrometry (MS). P re re q u is ite s fo r a good analysis are: rep resen tative sampling, good storage, e f f ic ie n t e x tra c tio n and sample p u rific a tio n (c le a n -u p , containment enrichment) followed by good isomer separation, u ltr a s e n s itiv e detectio n and useful confirm atory in f o r mation. Several review a r tic le s incorporating the a n a ly tic a l methods of GENP 010686 PCDOs and PCDFs have appeared re ce n tly: McKinney (1 9 78 ), Rappe and Buser (1981), Esposito e t , (1 9 80 ), Karasek (1 9 82 ), Tiem an (1983) and Tosine (1983). Most o f the old er methods have been c r i t i c a l l y reviewed by a panel o f experts in the National Research Council of Canada (1981). Samgle^Clean-UB Two d iffe r e n t procedural trends can be recognized: a) a ll PCDD and PCDF isomers can be analyzed in one single fra c tio n by the containment enrichment procedure (S ta llin g e t a l .,1 9 8 3 ; Norstrom e t _al_., 1982;. and Tiernan, 1983). b) s p e c ific isomers are analyzed in d iffe r e n t fra c tio n s mainly a fte r reverse phase and normal phase High Pressure Liquid Chromatography (HPLC) separation (Lamparski et ial_- 1979). The la t t e r method allows the id e n tific a tio n o f only a few PCDD isomers, mainly 2 ,3 ,7 ,8 -te tra rC D D . However, fo r a monitoring program a broad and general method is p re fe rre d . 783498 A series o f PCDDs and PCDFs have now been id e n tifie d in technical form ulations and in various environmental samples such as human blood and adipose tis s u e , fis h tis s u e , seal f a t , b ird eggs, beef f a t and deer tis s u e , sediments, s o il, f l y ash and other products from in c in e ra tio n o r combustion processes. The clean-up-methods vary fo r the d if fe r e n t m atrices. As one example, the method fo r the e x tra c tio n and containment enrichment o f a fis h tissu e sample is discussed in d e ta il by S ta llin g e t (1983). This method has also been GENP 010687 used fo r other samples such as sediments and human tissue samples. In th is method fis h tissu e (up to 100 g) is ground w ith ^ S O ^ and applied to a f i r s t chromatographic system containing a segment o f potassium s ilic a t e and another segment o f s ili c a g e l. This system is follow ed by a column containing cesium s ilic a t e and s ilic a gel followed by a column o f carbon dispersed on glass fib e r s . Before e lu tio n the f is h - s a lt m ixture was spiked w ith 13C12-Z ,3 ,7 ,8 -te tra -C D D , 13C12* or 37C l4-2 ,3 ,7 ,8 -te tra -C D F and 37Clg- 13 -octa-CDD or C^-octa-CDD. The carbon adsorbs most planar chlorinated (and unchlorinated) polynuclear aromatic compounds and the major portion o f b io lo g ic a l c o -e x tra c tiv e s is not retain ed . The PCDDs and PCDFs along with s im ila r chemicals are then removed from the carbon by reverse e lu tio n with toluene. The toluene is evaporated slow ly, O'Keefe e t a l . (1 9 8 2 ), and the sample is re-d issolved in hexane and applied to two columns in tandem. The f i r s t column contains both s u lfu ric acid dispersed on s ilic a gel and cesium s ili c a t e . The e lu a te from th is co lumn passes d ir e c tly in to the second column containing alumina. J?-!!r2!D3i99r32!!i-i!?l..BI2ii2D-DiLI25r-I^Diiiiii2Q The p u rifie d ex tra c ts are used d ir e c t ly fo r the fin a l analyses using a Gas Chromatograph - Mass Spectrometer (GC/MS) equipped w ith a glass c a p illa ry column coated w ith e ith e r o f S ila r 10 C, SP 2330, OV-17 or 0 V -101 or a fused s ili c a column coated w ith e ith e r SP 2330, SP 2340, SILOV, SE 54 o r OB-5. The column leads d ir e c tly in to the ion source o f the MS instrum ent, which is operating e ith e r in the ele c tro n impact (E l) or the negative chemical io n iz a tio n (NCI) mode. GENP 010688 783499 In view o f the large v a ria tio n in to x ic o lo g ica l and b io lo g ical e ffe c ts of the PCDD and PCDF isomers, i t is im perative th a t the isomers, p a r tic u la r ly those having the highest t o x ic it y jid e n t if ie d . For an unambiguous isomer id e n ti f i c a t i o n i t is necessary to have access to a ll isomeric standards w ithin a s p e c ific group o f congeners, a ll the the 22 tetra-CDDs and a ll the 38 tetra-CDFs. A ll the 22 tetra-CDD have been prepared , and using a S ila r 10 C glass c a p illa ry column the highly to x ic 2 ,3 ,7 ,8 -te tra -C D D can be separated from a l l the other 21 t e t r a isomers (Buser and Rappe, 1980). Recently a ll the 14 penta-CDDs and the 10 hexa-CDDs have been prepared. Using the S ila r 10 C column the to x ic 2 ,3 ,7 ,8 -s u b s titu te d isomers can be separated from a ll the other isomers (Buser and Rappe, 1983). The SP 2330 fused s ili c a column can also be used fo r th is separation (Rappe e t al_, 1984) and Figure 2 illu s tr a te s the separation o f t e t r a - , penta- and hexa-CDDs. In the PCDF series Mazer e t (1983) have synthesized a ll the 38 po sitio n al tetra-CDF isomers. The products were mixtures o f isomers, and each o f these isomers could be id e n tifie d using both a SP 2330 and a SE 54 c a p illa ry column. L a te r, B ell and GarS (1983) have is o la te d and characterized a l l t e t r a - , penta- and hexa-CDFs. The SP 2330 column can separate most o f these isomers (Rappe e t al_., 1984) see also Figure 3. The to x ic 1 , 2 , 3 , 7 , 8-penta-CDF co-elutes with the 1 , 2 , 3 , 4 , 8 - isomer and the to x ic 1,2 ,3 ,4 ,7 ,8 -h e x a -C D F with the 1 ,2 ,3 ,4 ,7 ,9 - - isomer, but these isomers can be separated on less p o la r columns lik e OV-17 and DB-5. A very lim ite d number o f in ve s tig a tio n s have been performed using these complete sets o f synthetic standards. GENP 010689 783500  # 5yn iifictio n Mass s p e c ific detection (mass fragmentography) has been used to q u an tify PCDDs and PCDFs in the samples by s e le c tiv e monitoring M, M+2 and/or M+4 ions. The q u a n tific a tio n is based on the peak area measurements and com parison of these areas using e ith e r is o to p ic a lly la b e lle d in tern al standards ( 13C or 37 Cl) or c a lib ra tio n curves o f external standards. At a f i r s t approach i t has been g e n e ra lly assumed th a t a l l isomers o f a p a r tic u la r congener o f PCDD or PCDF (e .g . the te tra c h lo ro isomers) have the same re sponse fa c to rs using the MS q u a n tific a tio n technique. However, an in v e s ti gation using 13 w ell defined tetra-CDF isomers have shown a 3 -fo ld v a ria tio n in the response fa c to rs using the El mode and up to a 2 0 -fo ld v a ria tio n using the NCI mode. For the higher chlo rinated homologues (C l^ , Clg) the d iffe re n c e was found to be sm aller (Rappe e t al_., 1983 a ). LEVELS OF PCDDs and PCDFs IN COMMERCIAL PRODUCTS Phenoxy_herbicides 783501 The dioxin problem was f i r s t recognized in connection w ith teratog enic e ffe c ts found fo r the phenoxy herbicide 2 ,4 ,5 -T . These e ffe c ts were shown to be caused by 30 yg/g o f 2 ,3 ,7 ,8 -te tra -C D D present in th is p a r tic u la r sample (Courtney and Moore, 1971). The le v e ls o f 2 ,3 ,7 ,8 -te tra -C D D in drums o f the herbicide "Agent Orange" placed in storage in the USA or in the P a c ific before 1970 have been found to vary between 0.02 to 54 yg/g . More than 450 samples have been analyzed in th is study (F ire s to n e , 1978; Esposito e t al_., 1980). Since "Agent Orange" was formulated as a 1:1 m ixture o f the butyl esters o f 2 ,4 ,5 -T and 2 ,4 -D , the le v e ls of 2 ,3 ,7 ,8 -te tra -C D D in in d iv id u a l 2 ,4 ,5 -T preparations th a t were manufactured and used in the 19601s could be as high as 100 yg/g. GENP 010690 As a re s u lt of governmental re g u la tio n s , e ffo rts were made during the 1970's to control and minimize the formation of 2 ,3 ,7 ,8 -te tra -C D D during the 2 ,4 ,5 -T production, and now a ll producers claim th a t t h e ir products con ta in less than 0.1 yg/g of 2 ,3 ,7 ,8 -te tra -C D D . The a n a ly tic a l methods used in these studies o f phenoxy herbicides are not isomer s p e c ific , however more recent studies using isomer s p e c ific methods have confirmed th a t the 2 ,3 ,7 ,8 -is o m e r is the major isomer in 2 ,4 ,5 -T form ulations (Buser and Rappe, 1980). Sixteen samples o f 2,4-D esters and amine s a lts from Canada have been ana lyzed fo r the presence o f PCDDs. Eight out o f nine esters and four out o f seven amine s a lts were found to be contaminated, the esters showed s ig n ific a n tly higher le v e ls than the s a lts . The tetra-CDD observed was the 1 ,3 ,6 ,8 -is o m e r as v e r ifie d by s y n th e tic a lly prepared authentic stan dard (Cochrane e t _al_., 1982). In other studies i t has been found th a t no other tetra-CDD than the 1 ,3 ,6 ,8 -is o m e r elu tes in th is window. H---e-x--a-c--h--l-o--r-o-peh--e--n--e 783502 The b a c te ric id e hexachlorophene is prepared from the same s ta rtin g m aterial as used fo r the manufacture o f 2 ,4 ,5 - T , but there are very few data a v a ila b le in the lit e r a t u r e on the le v e ls o f contaminants in th is product. Three samples analyzed by Baughman (1974) showed 0 .2 -0 .5 ng/g o f 2 ,3 ,7 ,8 -te tra -C D D . However, hexachlorophene also contains 100 yg/g o f a hexachloroxanthene, the 1 ,2 ,4 ,6 ,8 ,9 -isomer (Gothe and Wachtmeister, 1972). In th is connection i t could also be pointed out th a t a p u rific a tio n step ( d is t illa t io n , or re c ry s ta lliz a tio n ) resu lts in a remarkable increase o f im p u rities in the residues. GENP 010691 Chlorgghenojs Chlorophenols has been used exten sively since the 1930s as fun g icid es, mold in h ib ito r s , a n tis e p tic s , d is in fe c ta n ts and in s e c tic id e s . The annual world production volume is estimated to be in the order o f 150 000 tons. The most important use o f 2 , 4 , 6 - t r i - , 2 ,3 ,4 ,6 - t e t r a - and pentachlorophenol (o r t h e ir Na or K s a lts ) is fo r wood perservation. Pentachlorophenol or s a lts are also used fo r slime control in the manufacture of pulp or in synthetic c u ttin g f lu id s , fo r tanning le a th e r, in p a in t, glues and out door te x tile s . Chlorophenols may contain a v a rie ty o f contaminants, including PCDDs and PCDFs. Table 3 c o lle c ts a number o f re le va n t analyses o f these products. The le v e ls o f PCDDs and PCDFs are much higher fo r these products than fo r the phenoxy h erbicides. I t has also been reported th a t several p o sitio n al isomers o f PCDDs and PCDFs are present in the chlorophenols. However, a complete set o f iso meric standards has not been used in these in v e s tig a tio n s , and more research is necessary to id e n tify a l l the isomers present fo r a r is k evaluation o f these products. 2 l t)!2 riD $ L !iP t) Q y l Vos e t l_. (1970) were able to id e n tify PCDFs ( t e t r a - and penta-CDFs) in samples o f European PCBs (Phenoclor DP-6 and Clophen A 60) but not in a sample of A roclor 1260. They found the to x ic e ffe c ts o f these PCB products to be p a r a lle l to the le v e ls o f PCDFs present in the products. Bowes e t a l. GENP o1 (1975) examined a series o f Aroclors as well as the samples o f Aroclor 1260, Phenoclor DP-6 and Clophen A-60, th a t had previously been ana lyzed by Vos e t ^1_. (1970). They used packed columns and very few stan dard compounds and they reported th a t the most abundant PCDFs had the same chromatographic re te n tio n time as th a t o f 2 ,3 ,7 ,8 -te tra -C D F and 2 . 3 . 4 . 7 . 8 - penta-CDF. The q u a n tita tiv e re s u lts o f t h e ir in v e s tig a tio n are sunmarized in Table 4. More research' is needed to prove the existence of these to xic isomers. J?lE!iQl_ibr-!ir!?iid In 1980 Yamagishi e t l_. reported on the occurrence o f PCDDs and PCDFs in the commercial diphenyl eth e r herbicides CNP, NIP and X-52. The values are c o llected in Table 5. Very few synthetic standards were used, but the major tetra-CDDs were id e n tifie d as the 1 ,3 ,6 ,8 - and 1 ,3 ,7 ,9 -is o m e rs , the ex pected im p u ritie s in the s ta rtin g m aterial 2 ,4 ,6 -tric h lo ro p h e n o l. No 2 . 3 . 7 . 8 - tetra-CDD was found in these samples. FORMATION BY PHOTOCHEMICAL REACTIONS 783504 The photochemical formation o f 2 ,3 ,7 ,8 -te tra -C D D under labo ratory con d itio n s by dim erizatio n o r c y c liz a tio n reactions has been studied by Akermark (1 9 78 ). Although i t was possible to d etect very small amounts o f 2 ,3 ,7 ,8 -te tra -C D D , i t could be expected th a t i f th is compound was formed under environmental cond itio ns, i t should be e f f ic ie n t ly photo- dechlorinated to much less to x ic products, e s p e c ia lly in the presence o f potent hydrogen donors lik e the phenoxy herbicide e s te rs . (Akermark, 1978; see also Sundstrbm e t 1_., 1979). However, w ith in the s o il photochemical reaction are not very lik e ly to occur. A 1 AzTAO lu Another photochemical process o f p o te n tia l environmental importance is the dechlorination o f higher chlo rinated PCDDs, by which process a tetra-CDD can be formed by d echlo rination o f the octa-CDD. This reaction was studied by Crosby e t , (1971). The tetra-CDD isomer formed in th is reactio n has la t e r been id e n tifie d as the 1 ,4 ,6 ,9 -is o m e r (Buser and Rappe, 1978) a tetra-CDD expected to be s u b s ta n tia lly less to x ic than the 2,3,7,8-isom er. Crosby and Moilanen (1973) studied the photolysis o f a series o f PCBs dispersed in water. For two isomers, 2 ,5 -d ic h lo ro - amd 2 , 2 ' , 5 , 5 '- t e t r a chlorobiphenyls, small amounts (0.2%) o f 2-mono-CDF was found among the products. FORMATION BY THERMAL REACTIONS The formation of 2 ,3 ,7 ,8 -te tra -C D D as a re s u lt o f thermal reactions of 2 .4 .5 - T and 2 ,4 ,5 -T d e riv a tiv e s has been the subject o f controversy. Heating 2 ,4 ,5 -T s a lts a t 400-450C f o r 30 minutes or longer y ie ld s approximately 1 g o f 2 ,3 ,7 ,8 -te tra -C D D per kg o f 2 ,4 ,5 -T s a l t , w hile no dioxin was id e n tifie d from the same treatm ent o f 2 ,4 ,5 -T acid or esters (Langer e t , 1973; Baughman, 1974). Using a more s e n s itiv e a n a ly tic a l method Ahling e t^ l_ . (1977) reported th a t 0 .2 -3 mg of 2 ,3 ,7 ,8 -te tra -C D D was formed per kg o f 2 ,4 ,5 -T esters during combustion a t 500-850C. Two reports (Stehl and Lamparski, 1977; Andersson et a l . , 1978) have shown th a t 2 ,3 ,7 ,8 -te tra -C D D could not be found on burning samples o f spiked or sprayed vegetation a t 600C. The combustion gases, soot, p a rtic le s and ashes were analyzed and the detection li m i t was 4 mg/kg 2 .4 .5 - T burned, or less. 783505 GENP 010694 Rappe e t ^1- (1978) have studied the burning o f m aterial impregnated w ith various s a lts o f chlorophenols. Very c a re fu lly p u rifie d 2 , 4 , 6 - t r i - and pentachlorophenate were studied in addition to a commercial form ulation o f 2 ,3 ,4 ,6 -te tra c h lo ro p h e n a te . The a n a ly tic a l method used in th is study was not isomer s p e c ific , but the follow ing conclusions could be drawn concer ning the formation o f PCDDs by thermal reactions: the expected dim erization products and products formed in the "Smiles rearrangement" are the major PCDDs, no other thermal isom erization o f the PCDDs formed can be observed; no formation o f higher chlo rinated PCDDs can be observed, octa-CDD and other higher chlo rinated PCDDs y ie ld lower chlorinated dioxins in a nonspecific dechlorination reactio n . This is also a major reaction pathway, a series of PCDFs was also observed I t has been found th a t PCBs can be converted to PCDFs under p y ro ly tic conditions. The pyrolysis o f commercial PCBs in sealed quartz ampoules in the presence o f a i r y ie ld e d about 30 major and more than 30 minor PCDFs. The optimal y ie ld o f PCDFs was about 10% calculated on the amount o f PC8 decomposed. Thus uncontrolled burning o f PCBs can be an im portant environ mental source o f hazardous PCDFs. Therefore, i t was recommended (Buser e t a l . , 1978 a; Buser e t _al_., 1978 b) th a t a ll destruction o f PCB contaminated waste using in c in e ra to rs must be c a re fu lly c o n tro lle d . In the temperature range 300-400C, the y ie ld o f conversion seems to be in the p a rt-p e r-m illio n range (M orita e t a l . , 1978), but Nagayama e t a l. (1981) have reported a dramaii increase o f PCDFs at these temperatures in the presence o f s ta in le s s s teel and n ic k e l. 783506 GENP 010695 m 12 Buser and Rappe (1979) studied the pyrolysis o f 15 in divid ual synthetic PCB isomers. This study showed th a t the formation o f PCDFs can follow several competing reaction pathways, as seen in Figure 5. In another study, where a series o f chlorobenzenes were pyrolyzed in the same way, Buser (1979) found th a t s ig n ific a n t amounts (~ 1%) o f PCDDs and PCDFs were formed. A complex m ixture o f isomers o f PCDDs and PCDFs found suggesting several reaction routes. Using the same technique as above Lindahl e t K , (1980) studied the th e r mal decomposition o f polych lorin ated diphenyl ethers. Both PCDDs and PCDFs were formed in vo lving several pathways. The temperature range was 500-600C and the y ie ld s varied from 0.1% to 4.5%. I t has also been reported th a t pyrolysis o f PVC y ie ld s higher chlo rinated benzenes (Ahling _et _al_- 1978). The data discussed in th is section is summarized in Table 6. OCCURRENCE IN FLY ASH AND INCINERATION PRODUCTS Emissions from municipal in c in e ra to rs , heating f a c i l i t i e s and thermal power plants have f o r long time been the subject o f much concern. Whereas pre v io u s ly , the emission o f dust, smoke, to x ic metals and noxious gases were o f prime concern, the presence o f p o te n tia lly hazardous organic compounds from these emissions has been recognized only re c e n tly . In 1977 Lahaniatis e t a i. reported on the fin d in g o f chlo rinated organic compounds in f l y ash o f a municipal in c in e ra to r. The compounds detected were chlo rinated a lip h a t i c s , benzenes, PCBs and p e s tic id e s . OENP 010696 783507 Also in 1977 O lie e t K reported on the occurrence o f PCODs and PCDFs in f l y ash from three municipal in cin e rato rs in the Netherlands. Their re su lts in dicated the presence o f up to 17 PCDO peaks, but not qu an ti fic a tio n or isomer id e n tific a tio n was possible due to lack o f synthetic standards. Buser and Bosshardt (1978) studied f l y ash from a municipal in c in e ra to r and an in d u s tria l heating f a c i l i t y , both in Sw itzerland. In the former the le v e ls o f PCDDs was 0 .2 yg/g and o f PCDFs 0.1 yg/g. In the in d u s tria l in c in e ra to r the le v e ls were 0 .6 yg/g and 0 .3 y g /g , respec tiv e ly . During the period 1978-1982 a series o f papers, reports and reviews have been published confirming the o rig in a l findings o f Buser and Bosshardt. Up to now less data have been reported on the le v e ls o f PCDDs and PCDFs in other in c in e ra tio n products such as p a rtic u la te s and flu e gas condensate, which are the true emissions. As mentioned e a r lie r a ris k evaluation should be based on the le v e ls o f the highly to x ic isomers found in isomer s p e c ific analyses. However, in most studies o f f l y ash the results, are given in terms o f to ta l le v e ls o f t e t r a - , p e n ta -, hexa-, hepta- and octa-CDDs and CDFs. The value o f such studies is lim ite d , e s p e c ia lly in th is case where the number o f isomers is q u ite la rg e . More than 30 PCDDs and 60 PCDFs have been found in f l y ash samples. Other authors make the assumption o f an equal d is tr ib u tio n among the isomers. This seems to be an erroneous approach. Using the isomer s p e c ific a n a ly tic a l method 783508 described above, two samples from municipal waste in cin erato rs were analyzed, the two upper curves Figure 4. The Swedish in c in e ra to r was ope ra tin g by the flu id iz e d bed technique, the Canadian by conventional tech nique. In both samples we could id e n tify a series o f PCDDs and PCDFs. Of special in te r e s t is the observation th a t a ll tetra-CDDs found in these two samples co-elutes w ith the isomers present in a m ixture o f a ll the 22 tetra-CDDs (Figure 4, lower curve). A variatio n in the levels o f the highly toxic 2 ,3 ,7 ,8 -te tra -C D D was also observed. In the Swedish sample th is isomer c o n stitu tes about 3% o f the to ta l level o f tetra-CDDs w hile in the Canadian sample th is isomer is less than 0.33S. This is a te n fo ld v a ria tio n o f th is in te re s tin g isomer. Moreover, in these samples the r a tio between the la rg e s t and the sm allest peak is more than 100:1 (C. Rappe, S. Marklund, unpublished). Another in te re s tin g parameter to in vestig ate fo r various incinerators is the PCDF/PCDD r a tio . At the Cl^ and Cl g le v e ls , the PCDFs are in general much more abundant than the PCDDs, w h ile the s itu a tio n is re versed a t the C l7 and C1Q le v e ls (Rappe e t aU , 1983 b ). 783509 GENP 010698 In a report in 1978 i t was proposed by s c ie n tis ts from Dow Chemical Company th a t PCDDs and e s p e c ia lly 2 , 3 , 7 , 8-tetra-CDD are ubiquitous and formed as trace le v e l byproducts o f any normal combustion (Dow, 1978; Bumb e t a l . , 1980). Consequently dioxins should have been present in the environment since the advent o f f i r e ("The Trace Chemistries o f F ir e " ) . A recent sur vey o f PCDD le v e ls in p a rtic u la te from re sid e n ta l wood combustion units is quoted in support o f the above (N estrick and Lamparski, 1982). The survey showed PCDD le v e ls in the ppt-range,'The major c o n stitu en t was octa-CDD, but tetra-CDDs (in cluding the 2 ,3 ,7 ,8 -is o m e r) was also found. The occurrence o f PCDFs is not discussed in these rep o rts. However, th is hypothesis has been c r it ic iz e d (Anonymous, 1979, 1982). One o f the main arguments is th a t c o a l-fir e d power plants are ruled out as a possible source o f 2 , 3 , 7 , 8-tetra-C D D (Kimble and Gross, 1980; Junk and Richard, 1981). Another argument is th a t data are completely lacking in the Dow studies on le v e ls o f dioxin precursors (see Table 5) in the ma t e r ia l being burned, including the a ir in the flames. LEVELS OF PCDDs and PCDFs IN PCB FIRES In February 1981 a f i r e in the S tate O ffic e B uilding in Binghamton, NY, USA caused a transform er to rupture, releasing soot in the whole b u ild in g . The d ie le c t r ic f lu id in the transform er consisted o f a m ixture o f PCB (65%) and chlo rinated benzenes (35%). The soot was found to be hig h ly contaminated by PCDFs, the to ta l le v e l of PCDFs was more than 2000 ug/g. The re s u lts o f the analysis are compiled in GENP 010699 783510 Table 7. The most to x ic isomers (2 ,3 ,7 ,8 -te tra -C D F , 1 ,2 ,3 ,7 ,8 - and 2 ,3 ,4 ,7 ,8-p en ta-C D F and 1 ,2 ,3 ,4 ,7 ,8 - and 1,2 ,3 ,6,7 ,8 -h e xa -C D F ) were found to be the major c o n stitu en t w ith in each group o f congeners. In addition a series o f PCDDs could also be id e n tifie d including a large amount o f the hig h ly to x ic 2 ,3 ,7 ,8 -te tra -C D D and 1,2,3,7,8-p en ta-C D D . (Rappe e t a h 1983 b; Buser and Rappe, 1983.) I t is assumed th a t the chlo rinated benzenes were the dioxin precursors. During 1981-1983 a series o f cap acito r accidents have occurred in Scan d in avia. The accidents were o f two kinds: capacitor explosions or ca p a c ito r fir e s where mineral o il f i l l e d capacitors were also involved. In a l l cases the accident s ite s were contaminated by PCDFs (see Table 7 ). In the soot from the Binghamton b u ild in g p o lych lorin ated biphenylenes (PCBPs, Figure 1) have also been observed (D. S t a llin g , CNFRL, Columbia, MO, personal communication). Using the MS/MS technique Rappe e t _al_.. (1982) found high le v e ls o f these compounds in the wipes from a capacitor explo sion in Stockholm, Sweden. The 2 ,3 ,6 ,7 -te tra -C B P is c lo s e ly re la te d in s tru c tu re to the most to x ic PCDDs and PCDFs, and th is compound has also been found to have a strong b io lo g ic a l a c t iv it y (Poland e t jil_. 1977). LEVELS OF PCDDS and PCDFs IN ENVIRONMENTAL SAMPLES Soi 1 sedi nn ts_ and_ vgtaXj on 783511 Samples o f s o il and vegetation from Seveso, I t a l y have been analyzed. In 1976, immediately a f t e r the accident in the chemical p lan t .producing 2 ,4 ,5 -c h lo ro p h e n o l, the le v e ls o f 2 ,3 ,7 ,8 -te tra -C D D in vegetation were /^TTK Tn m m A A found to be up to 50 yg/g (F ire s to n e , 1978). In the follow ing years, when there was no d ire c t contact o f the newly grown vegetation w ith the aerosol cloud, the le v e ls o f dioxin in plants dropped by several orders o f magni tude (Wipf e t a l . , 1982). Studies by Young (1983) in d ic a te th a t the h a lf - l i f e fo r 2 ,3 ,7 ,8 -te tra -C D D in s o il is 10-12 years. In 1977 one year a fte r the accident in Seveso no traces of TCDD were found in the fle s h of apples, pears and peaches, nor in com cobs and kernels grown near the fa c to ry . The detection lim it fo r the analysis was reported as 1 pg/g. At the same time about 100 pg/g o f dioxin was detected in the peels o f f r u it s . This stro ngly suggests th a t the contamination is due to dust and not from p lan t uptake. The dioxin level in the s o il was found to be in the order of 10 ppb, about 1000 yg/m^ (Wipf e t _al_., 1982). In 1973 i t was found th a t three horse arenas in Missouri were contaminated by high le v e ls o f 2 ,3 ,7 ,8 -te tra -C D D , the highest value was about 30 ppm (Kimbrough t l_ ., 1977). This contamination resulted from the a p p lic a tio n in 1971 o f waste o il to control dusting a t these lo ca tio n s . The dioxin had o rig in a te d a t a hexachlorophene producing fa c to ry in Verona, MO. A dditional t r i - and tetra-CDDs have also been found, but the major component was 1,2 ,4 ,6,8 ,9 -h exach lo ro xan th en e (Buser, 1978) a compound which apparently can serve as a marker fo r th is type o f contamination. The xanthene is a normal by-product in the hexachlorophene production and has never been associated with the production of 2 ,4 ,5-trich lo ro p h en o l or 2 ,4 ,5 -T deriva tiv e s . E N P 010701 783512 In 1982, numerous s ite s o f p o te n tia l 2 ,3 ,7 ,8 -te tra -C D D contamination were discovered in Eastern M issouri. The contamination o rig in ated from the same waste o il from the fa c to ry in Verona, MO. In the town o f Times Beach, MO, the stre e ts o f the e n tire town had been sprayed. Now, more than 52 00 s o il samples have been analyzed from Times Beach. The dioxin levels were higher than 1 ppb in 22% o f the samples and higher than 100 ppb in 2.2%, the highest value was 350 ppb (Personal communication R.0. K loepfer, US EPA Region V I I , Kansas C ity , Kansas). Another lo cation o f great concern is Love Canal, Niagara F a lls , NY. Here Smith e t a l. (1983) also found high le v e ls o f 2 ,3 ,7 ,8 -te tra -C D D in storm sewer sediments taken from around the Love Canal waste disposal s it e , the highest value was 312 ppb. In a recent study Czuczwa and H ites (1983) have found a series of PCDDs and PCDFs in sediment samples from several locations in Saginaw River and Bay and southern Lake Huron. The le v e ls ranged from 100 ppb in urban areas to 100 ppt at remote s ite s . Although no isomers were id e n t if ie d , the a n a ly tic a l p ro file s in the sediments were reported to fo llo w c lo s e ly to those found in combustion samples, suggesting th a t combustion is the major source o f PCDDs and PCDFs found in the sediments. Analyses o f sediment cores showed a dramatic in crease in the le v e ls a t a depth corresponding to approxim ately 1940 and re maining high up to now. There is no good c o rre la tio n between the trend in these le v e ls and the trend fo r coal burning in the United S ta te s , which can "be predicted from the hypothesis o f "Trace Chemistries o f F ire " . However, the le v e ls in the sediments agree w ell with the production o f chlo rinated aromatic compounds. 783513 19 Very few analyses of sprayed vegetation has been reported. A rough e s t i mate of 2 ,3 ,7 ,8 -te tra -C D D contamination o f 20-1000 pg/g can be made on the basis o f the level of 2 ,4 ,5 -T found in newly sprayed vegetation and the level o f 2 ,3 ,7 ,8 -te tra -C D D in the spray form ulation used, even higher values could be obtained i f Agent Orange is used. Sundstrom e t ^1_. (1979) reported on data in agreement w ith th is estim ation. However the a n a ly tic a l technique used in th is study was not isomer s p e c ific . Vegetation was sprayed with 2 ,4 ,5 -T e s te r contaminated by only 0.06 ug/g o f 2 ,3 ,7 ,8 - t e t r a -CDD. A sample o f leaves c o llected 42-45 days a f t e r the spraying was found to have 170 pg/g o f TCDD, somewhat lower than the expected value (600 pg/g) in d ic a tin g a slow photochemical breakdown. 59yilJ2r93Dl!!! Fish and s h e llfis h taken from areas in South Vietnam, th a t were h eavily exposed to "Agent Orange" during m ilit a r y d e fo lia tio n operations in the 1960s have been reported to contain 18-810 pg/g o f TCDD. (Baughman and Meselson, 1973). The a n a ly tic a l technique o f d ir e c t in le t-h ig h resolution MS used in th is study is not considered isomer s p e c ific . I t did not include any GC separation a t a l l . Rappe t ^1_. (1981) id e n tifie d a series o f t e t r a to octa-CDFs in f a t samples o f a snapping t u r t le from Hudson River and gray seal from B a ltic Sea. The to ta l le v e ls o f PCDFs in these samples were 3 ppb and 40 p p t, re s p e c tiv e ly . In both samples the major PCDFs consisted o f the most to x ic isomers ( 2 , 3 , 7 , 8 - t e t r a - , 2 ,3 ,4 ,7 ,8 -p e n ta - and 1 ,2 ,3 ,4 ,7 ,8 - and 1,2,3,6,7,8-hexa-C D F). GENP 010703 783514 Norstrom e t ^1_. (19821 have analyzed pooled samples of herring g u ll eggs co lle cte d in 1982 from various parts o f the Great Lakes. In a l l samples 2 ,3 ,7 ,8 -te tra -C D D was found in le v e ls ranging from 9 to 90 pg/g. The id e n tity o f the 2 , 3 , 7 , 8 - isomer was confirmed by re te n tio n times on three c a p illa ry columns. In another study S ta llin g e t a h (1983) were not able to id e n tify measurable le v e ls o f tetra-CDDs and o th er PCDDs in fis h samples from Lake Superior (the detection level was 2-5 p p t). The d iffe re n c e could be explained by the m igration o f the herring g u lls during the w in te r. On 't h e contrary a series of PCDFs could be id e n tifie d in the Lake Superior fis h samples, in d ic a tin g a more widespread background levels fo r the PCDFs than fo r the PCDDs. S ta llin g e t (1983) also reported on the analysis o f fis h samples from Lakes Michigan, Huron and O ntario. The to x ic 2 ,3 ,7 ,8 - -s u b s titu te d PCDDs and PCDFs (Table 2) were present in a ll samples, the highest le v e ls being found in samples from Lake Huron, Lake Ontario and Tittabawasee R iver. The residue pattern found in the fis h and lo c a lly high le v e ls suggest a strong influence by point source discharges. The re cent data by O'Keefe e t _al_. (1983) * is also in agreement with th is s ta te ment. LEVELS IN HUMAN SAMPLES Qccugational_Exgosure 783515 Occupational exposure to 2 ,3 ,7 ,8 -te tra -C D D can occur during the production * o f 2 ,4 ,5 -tric h lo ro p h e n o l and the subsequent production and use o f 2 ,4 ,5 -T acid and e s te rs . The 2 ,4 ,5 -T was f i r s t produced commercially in 1944 in United States and the use of 2 ,4 ,5 -T as herbicide in creased .in the 1940's m rw p n 10704 and 1950's. However, the problem o f dioxin contamination in 2 ,4 ,5 -T was not recognized u n til 1957 (Kimmig and Schulz, 1957). The most heavy exposure a f t e r a normal production is during a p u r i f i cation step as the residues are by f a r more contaminated than the p u ri fie d products. Only lim ite d inform ation is a v a ila b le on the le v e ls of dioxin contamination o f products prepared p rio r to the 1970`s , and ab s o lu te ly no inform ation is a v a ila b le on the dioxin levels in the corre sponding residues. Consequently i t is a d i f f i c u l t task to estim ate the le v e ls o f occupational and general exposure during th is period. The dioxin exposure o f so ld iers and the general population during the m i l i ta ry use o f phenoxy herbicides during the Vietnam war has been the subject o f much concern both among the Veterans in the United States and A u s tra lia as fo r the people in Vietnam. Shepard and Young (1983) discuss a study where very low le v e ls o f tetra-C DD , believed to be the 2 ,3 ,7 ,8 -is o m e r, was detected in adipose tissu e from many of the Vietnam Veterans, although the le v e ls are not considered to c o rre la te w ell with known exposure data or w ith health sta tu s . No experimental c o n d itio n s vare given in th is re p o rt. In order to evaluate any study o f trace amounts o f 2 ,3 ,7 ,8 - tetra-CDD in samples from exposed people i t is o f fundamental importance to know the background le v e ls o f th is compound in d if fe r e n t p a rt o f the world. Ryan and W illiam s (1983) re ce n tly reported on the le v e ls o f 2 ,3 ,7 ,8 - t e t r a -CDD in 23 autopsy samples o f adipose tissu e from the Great Lakes area in Canada. The a n a ly tic a l method used was capable o f separating the 2 ,3 ,7 ,8 -isomer with a possible 1-3 other c o -e lu tin g isomers. The 2 ,3 ,7 ,8 -te tra -C D D 783516 was found in 22 o f the 23 samples and values ranged from 4.1 to 130 ppt. Excluding the outlaying high sample the average value was 10.7 - 5 .4 ppt. P relim inary data from the analyses o f human adipose tissu e both from Canada (personal communication J. Ryan, Health P rotection Branch, Ottawa, Canada) and from Sweden (C. Rappe and H. Nygren, manuscript in preparation) seem to in d ic a te background le v e ls (1-600 ppt) in the general po pulation fo r both PCDDs and PCDFs ranging from te tra c h lo ro through octachloro c pounds. The le v e ls have a tendency to increase w ith increasing number of chlo rine atoms, although the octa-CDF was always at very low le v e ls . The highly to x ic 2 ,3 ,7 ,8 -s u b s titu te d isomers were the dominating isomer w ith in each group o f congeners, in d ic a tin g a c o rre la tio n between the isomers found in human tissue and those found in fis h samples, see above. This ra th e r complicated s itu a tio n generates special problems in order to evaluate a certain exposure; in Vietnam, occupational or dietary.The best solu tion seems to analyze fo r a l l PCDDs and PCDFs using a clean-up technique th a t re s u lts in one s in g le fr a c tio n , which can be used to analyze fo r a ll PCDOs and PCDFs. The data should be analyzed using p a ttern recognition methodology. Analysis o f blood plasma can be used to evaluate occupational exposure to PCDDs and PCDFs, which can occur during the production or use of 2 , 4 , 6 - t r i - , 2 ,3 ,4 ,6 - t e t r a - and pentachlorophenol. Rappe e t ]_. (1983 c) have in v e s ti- ' gated such exposure through the analyses of blood plasma o f exposed workers and unexposed c o n tro ls. Good c o rre la tio n was found between the plasma le v e ls and 783517 GENP 010706 1) the nature of exposure 2) the duration o f exposure 3) the isomers present in the commercial form ulation i Yusho^Episodes In 1968 more than 1500 persons in southwest Japan were in to x ic a te d by con sumption o f a commercial ric e o il a c c id e n ta lly contaminated by PCBs, PCDFs and polychlorinated quaterphenyls (Masuda e t j_. 1982, Miyata e t aj_., 1978). In 1979 a s im ila r episode occurred in central Taiwan, the number of persons involved here approaching 2000 (Chent_al_. 1980). Both these accidents have bee re fe rre d to as Yusho episodes, but re cen tly the Taiwanese episode is renamed Yu-cheng. The Japanese ric e o il contained more than 40 PCDF isomers, t r i - to hexa-CDFs. Analyses of li v e r samples taken from the Japanese patients about 18 months a f t e r the exposure showed a dramatic decrease in the number o f PCDF isomers. Apparently most o f the PCDF isomers have been metabolized or excreted during the period between exposure and sampling (Rappe e t al_. 1979). A comparison between the PCDF isomers found in the Yusho o il and in the li v e r samples revealed an in te re s tin g re la tio n s h ip . Most o f the isomers retained had a l l la t e r a l positions ( 2 - , 3 - , 7 - and 8 -) substitu ted fo r c h lo rin e , these isomers have the highest t o x ic it y (see Table 2 ) . The ra te o f excretion o f these to x ic dibenzofuran isomers is very slow, s CD as studied by the Japanese and Taiwanese Yusho p a tie n ts . A 15-20% reduction in the blood le v e ls was found in one y e a r, and PCDFs GFTVTP n i c\Hf\n * 24 - could s t i l l be found 11 years a f t e r the exposure in Japan ( Rappe e t a l . , 1983 c ). This indicates a dramatic d iffe re n c e in the ra te o f excretion between man and rodents, where the h a lf l i f e o f 2 ,3 ,7 ,8 -te tra -C D D is reported to be only a few weeks. c'r . CARCINOGENICITY AND CANCER EPIDEMIOLOGY The carcinogenic e ffe c ts o f dioxins have been studied in a series of a n i mal experiments. The 2 ,3 ,7 ,8 -te tra -C D D , 1 ,2 ,3 ,6 ,7 ,8 - and 1 ,2 ,3 ,7 ,8 ,9 -hexa-CDD are reported to have carcinogenic or co-carcinogenic e ffe c ts . In man the carcinogenic e ffe c ts are re la te d to the exposure to phenoxy her bicides and chlorophenols. I t is d i f f i c u l t to separate the e ffe c ts o f the parent compounds from the e ffe c ts o f the dioxin and dibenzofuran conta minants ( IARC, 1982, NTP, 1980 a and b ). Two Swedish case-control studies have shown an increased ris k o f s o ft- -tis s u e sarcomas in men exposed to phenoxy herbicides and chlorophenols (H ardell and Sandstrom, 1979; Eriksson e t ^1_., 1981). A high incidence o f s o ft-tis s u e sarcomas has also been observed among workers employed in the manufacture o f these products in the United States (Honchar, 1981). Three cases o f th o ra c ic s o ft-tis s u e sarcomas have been reported fo r persons, who served in Vietnam and were exposed to "Agent Orange" (Sarma and Jacobs, 1982). Other studies have f a ile d to demonstrate the r e la tio n between s o ft-tis s u e sarcomas and exposure to phenoxy herbicides and chlorophenols, and the safe diagnosis o f s o ft tissu e sarcomas has been discussed (Smith, 1983; Fingerhut, 1983). 783519 Another case control study in Sweden suggests a c o rre la tio n between GENP 010708 25 and non-Hodgkin's lymphomas (H ardell et a h , 1981) but fu rth e r research is needed to confirm th is . I t has also been pointed out th a t exposure to chlorophenols (and t h e ir im p u ritie s ) could be an exposure in common fo r d iffe r e n t groups with increased ris k o f nasal cancer fo r people occupied w ith wood products, le a th e r, t e x t i le and c u ttin g flu id s (H ardell e t a l . , 1983). CONCLUSIONS Using the a n a ly tic a l standards now a v a ila b le i t has rece n tly been shown th a t isomer s p e c ific analyses can be performed fo r a l l to x ic PCDD and PCDF isomers. Only very few la b o ra to rie s have now access to these standards and c o lle c te d experience to work w ith them (about 5 la b o ra to rie s in the whole w o rld). Due to the p o s s ib ility o f fa ls e p o s itiv e r e s u lts , a ll PCDD and PCDF analyses should be performed using a metodology which has been proven to be isomer s p e c ific . The amount detected in these procedures is in the pg range or lower and containment enrichment methods have been developed to in je c t an a liq u o t corresponding to 1 g o f the sample or more. Thus the detection level is in the ppt range or lower. Using th is a n a ly tic a l technique background le v e ls o f te tra c h lo ro through octachloro PCDDs and PCDFs have been found in ecolog ical samples as w ell as in human adipose tissue in America and Scandinavia. GENP 010709 783520 This s e n s itiv e and s e le c tiv e a n a ly tic a l technique is associated with laboratory problems lik e cross contamination and poor recoveries of in te rn a l standards. More research is needed to specify each step in the complicated a n a ly tic a l procedure. The a v a ila b le standards should be used to optim ize the methods o f sample p u rific a tio n fo r a ll isomers o f PCDDs and PCDFs. There is also a great need fo r more monitoring data, e s p e c ia lly in s o il , sediments, a i r , w ater, aquatic organisms and in the general human population in the in d u s tria liz e d p a rt o f the world as well as in deve loping countries. GENP 010710 783521 - 27* - REFERENCES A hling, B ., Lindskog, A ., Jansson, B. and Sundstrbm, G. Chemosphere, 1977, 461. A hling, B ., Bj/Srseth, A. and Lunde, G. Chemosphere, 192, ]_> 799. Andersson, K ., Bosshardt, H .P ., Buser, H .R ., Marklund, S. and Rappe, C. Ecol. B u ll. (Stockholm) 1978, _27, 24. / Anonymous Chem. Eng, News Sept. 24, 1979, 27. Anonymous Chem. Week A pril 13, 1983, 26. Baughman, R.W. Thesis Harvard U n iv e rs ity , Cambridge, MA, USA, 1974. Baughman, R. and Meselson, M. Environ. Health Perspec. 1973, 5, 27. B e ll, R. and GarS, A. Presented at ACS National M eeting, Washington DC, 1983. Bowes, G.W., M u lv ih ill, M .J ., Sim oneit, B .R .T ., Burlingame, A.L. and Risebrough, R.W. N ature, 1975, 256, 305. Bradlaw, J.A . an d -C a s te rlin e , J .L . 0. Assoc. O ff. Anal. Chem., 1 |Z |* 62, 904, Bumb, R.R. e t a l. Science, 1980, 210, 385. 783522 Buser, H.R. Thesis U n iversity o f Ume3, UmeS, Sweden, 1978. Buser, H.R. and Bosshardt, H .-P . M itt Geb. Lebensmittel unters, u. Hyg. 1978, 69, 191. Buser, H.R. and Rappe, C. Chemosphere, 1978, 7_f 199. Buser, H.R. and Rappe, C. Chemosphere, J979, 157- Buser, H.R. and Rappe, C. Anal. Chem. 1980, 52, 2257. Buser, H.R. and Rappe, C. Submitted to A nal,Chem. 1983. Buser, H .R ., Bosshardt, H .-P . and Rappe, C. Chemosphere, 1Z_> Z* tier Buser, H .R ., Bosshardt, H .- P ., Rappe, C. and L in d ah l, R. Chemosphere, 1|Z 7, 419. Cochrane, W .P., Singh, J . , M ile s , W., Wakeford, B. and S c o tt, J. C hlorinated Dioxins and Related Compounds, Eds. H utzinger, 0 . , Frei., R.W. Merian, E. and P o c c h iari, F . Pergamon Press Oxford 1982, 209. Courtney, K.D. and Moore, J.A . T o xico l. Appl. Pharmacol. 1971, 20, 396.- Crosby, D.G. and Moilanen, K.W. B u ll. Environm. Contam. T o x ic o l. 1?73, JO, 372. Crosby, D .G ., Wong, A .S ., Plinm er, J.R. and Woolson, E.A. Science, 1|Z 1. 173, 748. GENP 010712 783523 Czuczwa, J.M. and H ite s , R.A. Am. Chem. Soc., Div. Environ. Chem. 193, 2 3 :2 , 74 Dow Chemical U.S.A. The Trace Chemistries of F ir e , 1978. Eriksson, M ., H a rd e ll, L . , Berg, N .O ., M lle r, T. and Axelson, 0. B r it . J. Indust. Med. 1 | 1 . 38. 27. Esposito, M .P ., T ie m a n , T.O. and Dryden, F.E. Dioxins EPA-600/2-80-197, 1Q. Fingerhut, M. Chem. Eng. News, October 31, 1|3 p 24. Firesto n e, D. Ecol. B u ll. (Stockholm) 1978, 27, 39. Gthe, R. and Wachtmeister, C.A. Acta Chem, Scand. 1972, 26, 2523. H a rd e ll, L. and Sandstrm, A. Br. J. Cancer, 1979, 39, 711. H a rd e ll, L . , Eriksson, M ., Lenner, P. and Lundgren, E. Br. J. Cancer, 1 ||1 43, 169. H a rd e ll, L ., Axelson, 0. and Rappe, C. Lancet, 1 |8 3 , Vol. 1 , 1157 Honchar, P.A. and H a lp e rin , W.E. Lancet, 1981, 268. H u ff, J .E ., Moore, J .A ., S a ra c c i, R. and Tomatis, L. Environ. Health Perspect. , 1||Q , 36, 221. IARC Monographs Supplement $, 1982, 238. 783524 Junk, G.A. and Richard, J .J . Chemosphere, 1981, 10, 1237.  (1 30 Karasek, F.W. and Onuska, F . I . A nal. Chem. 1982, 54, 309 A. Kimbrough, R .D ., C arte r, C .D ., L id d le , J .A ., C lin e , R.E. and P h illip s , P.E. Arch. Environ. H e a lth , 1977, 32, 77. Kimmig, J. and Schulz, K.H. Dermatologia, 1 ||Z * V15, 540. L a h a n ia tis , E.S., P a rla r, H. and K orte, F. Chemosphere, 1977, , 11. Lamparski, L . L . , N e s tric k , T .J . and S te h l, R.H. A nal. Chem. 1979, 51_, 1463. Langer, H .G ., Brady, T.P. and B riggs, P.R. Environ. Health Perspect. 1973, 5, 3. Lindahl, R ., Rappe, C. and Buser, H.R. Chemosphere, 1980, , 351. Masuda, Y ., Kuroki, H ., Yamaro, T . , Haraguchi, K ., Kuratsune, M. and Hsu, S.T. Chemosphere, 1982, JJ_, 199. M azier, T . , Hileman, F . , Noble, R.W. and Brooks, J. Anal. Chem. 1983, 55, 104. Me Connell, E .E ., Moore, J .A .* Haseman, J.K . and H a r r is , M.W. T o xico l. Appi. Pharmacol. 1976, 36, 65. McKinney, J.D. Ecol. B u ll. (Stockholm), m , 27, 53. 783525 M iyata, H ., Murakami, Y ., Kashimoto, T. J. Food Hyg. Soc. Japan, 1978, j_9v 233. Moore, J .A ., Me Connell, E.E.',' Dalgard, D.W. and H a rris ,, M.W. Ann. N.Y. Acad. S ci. 1979, 320, 151. ________________ _ 31 CTMI M o rita , M ., Nakagawa, J. and Rappe, C. B u ll. Environm. Contam. T o x ic o l. , 1|Z, 19, 665. Nagayama, J . , Kuratsune, M. and Masuda, Y. Fukuoka Acta Med. 1981, 72, 136. National Research Council Canada. P olychlorinated D ibenzo-p-Dioxins: Lim itations to the Current A n a ly tic a l Techniques: P ublication NRCC No. 18576, 1 N e s tric k , T .J . and Lamparski, L.L. A nal. Chem. 1982, 54, 2292. Norstrom, R .J ., H a l le t t , D.J. Simon, M. and H u lv ih ill , M.J. Chlorinated Dioxins and Related Compounds. Eds. H utzinger, 0 . , F r e i, R.W., Merian, E. and P o c c h iari, F ,; Pergamon Press, Oxford 1982, 173. NTP, DHHS P ub lication No. (NIH) 80-1754 (1980). NTP, DHHS P ub lication No. (NIH) 80-1758 (1980). O'Keefe, P ., Meyer, C ., H ilk e r, D ., Aldous, K ., Jelu s-T yro r, B ., D illo n , K ., Donnelly, R ., Horn, E. and Sloan, R. Chemosphere, 1983, 325. O'Keefe, P ., Meyer, C. and D illo n , K. Anal. Chem. , 1982, 54, 2623. 01i e , K ., Vermeulen, P.L. and H utzin g er, 0. Chemosphere, 1977, _6, 455. Poland, A ., Glover, E. and Kende, A.S. J. Chem. B io l., 1976, 251_, 4926 Poland, A. and Glover, E. Mol. Pharmacol. , 1977, 1_3, 924. 783526 n ttntd m r m ^ 6 Rappe, C. and Buser, H.R. Halogenated Biphenyls, Terphenyls, Naphta!enes, Dibenzodioxins and Related Products. Ed. R. Kimbrough; E lsevier-N o rth Holland, Amsterdam 1981, 41. Rappe, C ., Marklund, S ., Buser, H.R. and Bosshardt, H .-P . Chemosphere, ].97, 7, 269. Rappe, C ., Buser, H.R. and Bosshardt, H.P. Ann . N.Y. Acad. S e i. , 1 9 7 |, 320, 1. Rappe, C ., Buser, H .R ., Kuroki, H. and Masuda, Y. Chemosphere, l | Z i 8 259. Rappe, C ., Buser, H .R ., S t a llin g , D .L ., Smith, L.M. and Dougherty, R.C. Nature, 1 ||1 , 292, 524. Rappe, C ., Marklund, S ., B ergq vist, P .-A . and Hansson, M. Chem. S cr., 1982, 2 0 , 5< Rappe, C ., Marklund, S . , Nygren, M. and Gar,A. C hlorinated Dioxins and Dibenzofurans in the Total Environment. V ol. 1. Eds. Choudhary, G ., K e ith , L. and Rappe, C. Ann Arbor Science P ublishers, Ann Arbor MI. 1983. In press. Rappe, C ., Marklund, S . , B ergq vist, P .-A . and Hansson, M. Ib id . 1983 b. In press Rappe, C ., Nygren, M.and Gustafsson, G. Ib id . 1983 c. In press. Rappe, C ., Marklund, S . , B ergq vist, P .- A ., K je lle r , L .- 0 . and Hansson, M. Chlorinated Dioxins and Dibenzofurans in the Total Environment. V o l. 2 Eds. K e ith , L . , Rappe, C. and Choudhary, G. 1984. In press. 783527 Ryan, J .J . and W illiam s, D.T. Am. Chem. S oc., Div. o f Environ. Chem. 1983 23:2, 157. Sarma, P.R. and Jacobs, J. New England 0. Med. 1982, 306, 1109. Shepard, B.M. and Young, A.L. Human and Environmental Risks o f Chlorinated Dioxins and Related Compounds. Eds. Tucker, R .E ., Young, A.L. and Gray, A.P. Plenum Press New York 1983, 3. Smith, A .H ., F ish er, D .O ., G ile s , H .J. and P ie rc e , N. Chemosphere, 183, ] 2 , 565. Smith, R .M ., O'Keefe, P.W ., Aldons, K .M ., H ilk e r , D .R ., and O 'B rien, J.E. Env. Sci. Techno!. , 1983, ]]_t 6. S ta llin g , D .L ., Smith, L .M ., P e tty , J .D ., Hogan, J.W. , Johnson, J . L . , Rappe, C. and Buser, H.R. Human and Environmental Risks o f C hlorinated Dioxins and Related Compounds. Eds. Tucker, R .E ., Youngi A .L. and Gray, A.P. Plenum Press New York 1983, 3, 221. S te h l, R.H. and Lamparski, L.L. Science, 1977, 197, 1008. .. ~ T ~ Sundstrom, G ., Jensen, S ., Jansson, B. and Erne, K. Arch. Environm. Contam. Toxicol. 1 |7 |, 8, 441. T ie m a n , T. C hlorinated Dioxins and Dibenzofurans in the Total Environment. Vol. 1. Eds. Choudhary, G ., K e ith , L. and Rappe, C .A nn Arbor Science Publishers, Ann Arbor MI. 1983, 211. Tosine, H. Ib id . 1983, 3. 783528 Vos, J .G ., Koemari, J .H ., vand der Maas, H .L ., ten Noever de Brauw, M.C. and de Vos, R.H. Food. Cosmet. T o x ic o l. 1970, 8, 625. ________ ________ W ipf, H.K. and Schmid, J. Human and Environmental Risks of Chlorinated Dioxins and Related Compounds. Eds. Tucker, R .E ., Young, A.L. and Gray, A.P. Plenum Press, New York, 1983, 255. Yamagishi, T . , M iyazaki, T ., Akiyama, K ., M o rita , M ., Nakagawa, J . , H o r ii, S and Kaneko, S. Chemosphere, 1981, JO, 1137. Young, A.L. Human and Environmental Risks o f C hlorinated Dioxins and Related Compounds. Eds. Tucker, R .E ., Young, A.L. and Gray, A .P .; Plenum Press New York, 1983, 173 Akermark, B ., Ecol. B u ll. (Stockholm) 1978, 75. 783529 GENP 010718 Table 1. Number o f p o s itio n e l PCDD and PCDF isomers Level of chlorine substitu tio n Cli (Mono) Cl2 (d i) C13 ( t r i ) Cl 4 (t e t r a ) C15 (penta) C16 (hexa) Cl 7 (hepta) C18 (octa) Number o f isomers PCDDs PCDFs 2 4 10 16 14 28 22 38 14 28 10 16 24 11 75 135 GENP 010719 783530 Table 2. The most to x ic PCDD and PCDF isomers PCDDs 2 ,3 ,7 ,8 -T e tra -C D D 1,2,3,7,8-Renta-CDO 1 ,2 ,3 ,6 , 7 ,8-Hexa-CDD 1,2,3,7,8,9-Hexa-CDD 1 , 2 , 3 , 4 , 7 , 8-Hexa-CDD PCDFs 2 ,3 ,7 ,8 -T e tra -C D F 1 ,2 ,3 ,7 ,8-Penta-CDF 2 ,3,4,7,8-Penta-CDF 1 ,2 ,3 ,6 ,7 ,8 -H e x a -C D F 1 ,2 ,3 ,7 ,8 ,9 -H e x a -C D F 1 .2 .3 .4 .7 .8 - Hexa-CDF 2 .3 .4 .6 .7 .8 - Hexa-CDF 783531 GENP 010720 Table 3. Levels o f PCDDs and PCDFs in commercial chlorophenols (ug/g) (From Rappe e t a l . , 1979) Tetra-CDDs Penta-CDDs Hexa-CDDs Hepta-CDDs Octa-CDD 2 ,4 ,6 - C l3 2 , 3 , 4 , 6 - Cl ^ < 0.1 < 0.1 <1 <1 <1 < 0.1 < 0.1 <1 10 2 PCP PCP < 0.1 < 0.1 <1 0.5 * 4.3 < 0.1 < 0.1 2.5 175 500 Tetra-CDFs Penta-CDFs Hexa-CDFs Hepta-CDFs Octa-CDF 1.5 17.5 36 4.8 <1 0.5 10 70 70 10 < 0.1 < 0.1 0.03 0.5 1.1 < 0.1 < 0.1 < 0.3 19 25 GENP 010721 783532 Table 4. Levels o f PCDFs in commercial PCBs (pg/g) (From Bowes e t a K , 1975) Sample Aroclor 1248 (1969) Aroclor 1254 (1969) A roclor 1254 (1970) Aroclor 1260 (1969) A roclor 1260 (Tot AK3) A roclor 1016 (1972) Clophen A 60 Phenoclor DP-6 Tetra Penta Hexa Total 0.5 1.2 0.1 0.2 0.2 0.4 0.1 0.4 0.2 0.3 ND ND 1.4 5.0 O'. 7 10.0 0.3 1 .4 0.9 0.5 0.3 ND 2.2 2.9 2.0 1.7 1.5 1.0 0.8 8.4 13.6 ND = Not Detectable 783533 GENP 010722 Table 5. Levels o f PCDDs and PCDFs in commercial diphenyl eth er herbicides (u g /g ). (From Yamagishi e t a l . , 1981} Tri-CDDs Tetra-CDDs Penta-CDDs Hexa-CODs Mono-CDFs Di-CDFs Tri-CDFs Tetra-COFs Penta-CDFs Hexa-CDFs ND = Not Detected CNP ND 14.0 37 0.8 ND 0.35 0.41 0.4 1.0 0.2 NIP 0.15 0.38 0.05 ND 0.34 0.12 0.47 0.29 ND ND X-52 0.03 0.03 0.01 ND 0.48 0.21 0.45 0.32 0.08 ND GENP 010723 783534 Table 6. Formation o f PCDDs and PCOFs by thermal processes Precursor 2.4,5-T salt 2,4,5-T (vegetation) Chloro -phenate PCBs Polychloro benzenes Chloro Diphenyl ethers PVC Conditions Pyrolysis Pyrolysis Burning Burning Pyrolysis Pyrolysis Pyrolysis Pyrolysis Products 2,3,7,8-TCDD No TCDD --ll-- PCDDs* + PCDFs PCDFs K* PCDFs + PCDDs * * * PCDFs + PCDDs Polychlorobenzenes x * PCDD formed by d im erizatio n and a n o n -s p e cific d e c h lo rin atio n xx = o th er products: hexa- and penta-chlorobenzenes xxx = other products: PCBs, po lychlorinated naphthalenes 783535 GENP 010724 Table 7. Levels o f PCDFs in samples from PCB f ir e s E C14 2378- L Cl& Binghamton (pg/g) 2 Stockholm (pg/m ) 2 ikovde (pg/m ) iurahamnar (pg/m ) (before cleaning) iurahammar (pg/m ) [a fte r cleaning) 2 lalIstahammar (pg/m ) [before cleaning) 2 lalIstahammar (pg/m ) a fte r cleaning) le c tr ic a l locomotive (pg/m?) 20 12 1.2 0.15 0.6 0.1 1.2 0.3 < 0.02 < 0.004 1.6 0.54 0.002 0.0002 4.9 1.2 670 0.175 0.1 0.3 < 0.01 0.36 0.0005 1.5 E Cl6 E Cl7 C18 965 - 0.06 0.15 460 - 0.008 0.07 40 - 0.006 0.01 < 0.01 < 0.015 < 0.002 0.60 0.80 1.34 0.0003 0.0007 0.0011 0.4 0.17 0.04 t PCDOs 20 - - - PCBPs* + +++ - - - - i polychlorinated biphenylenes +++ high le v e ls , + low levels 909S9Z GENP 010725  a> X +y = 1-8 PC BPs Figure 1. Formulas fo r the compounds discussed. GENP 010726 783537 a h M fK O -C D O a ffO i UtlMttt7S* RJiU 783538 j so nma*ui IU 7U 123**7 i 2920 32.00 J<-40 Figure 2 . Sepa r a t i on o f 22 tetra.-CDDs, 14 penta-CDDs and i a i- . A n r > __ ___ _ <>n a a a a < * 3720Timi* (min) OO 3PI Cf-nO als*fa* s* sa 23 3a fSjlLU OW I Q) *JOD oIf* NY T i i* ? X 7^ 5 3w34 Z<#8 'S o t * E ? <t <v 783539 80K)<IN3O a -i KO OODfu s r reaction route 1 (loss of ortho-CU) reaction route 2 (loss of HCI involving a 2 3-chlorine shift) reaction route 3 (loss of ortho-HCll reaction route 4 ( loss of ortho-Ho) 783541 GENP 010730 Figure 5. Reaction pathways fo r the thermal formation o f PCDFs from PCBs.