Document QJp0Yw221JzyDE80Xa1mDOdq7

tI THE FINDING OF .CHLORINATED DIBENZOFURANS IN A JAPANESE POLYCHLORINATED BIPHENYL SAMPLE by JOHN A. G. ROACH and IRWIN II. POMERANTZ Division of Chemical Technology Food and Drug Administration t Washington D. C. 20204 NV 036456 1 The presence of totre- and pentaclilorinnted dibenxofuratis and hexa- and hcptachJorlnatcd naphthalenes in two samples'of polychlorinated biphenyl, mixtures (PCB) of European origin'has been reported by VOS, ut. al. (1970a). Their interest in a content analysis of the two -PCS samples for impurities wasi stimulated by previous work in which toxic effects observed in chicks were attributed to these two samples (VOS, et. al. 1970b). .| ' , | Kanechlor 400.; (KC-400),., a commercial- PCB preparation of Japanese origin has bcon identified as the causative agent in Yusho, a human poisoning incident resulting from the ingestion of rice oil contaminated with KC-400 (KURATSUNE, et. al. 1972). We have analyzed a sample of KC-400 made available to us from the Kanegafuchi Chemical Industrial C o , , Ltd. through ])r. rf. Blumenthal of the. Food and Drug Administration to determine whether potentially hazardous .chemicals, particularly chlo rinated dibenzofurans (01-DBF), were present as contaminands. It was the objective of thissanalysis.to obtain, where pos-. slble, structural information on these contaminants and an : estimate of their concentration in the KC-400. -> EXPERTMKNTAT. ` ! A hexane solution of the KC-400 was gas chromatographed, using an eletrtFpn capture detector, on a column of 104 DC-200 on Chromosorb VHP (HO-100 mesh) and on a mixed column of 15% QF-1 + 10% DC-200 each coated on Chromosorb WHP (8Q-100 mesh) and mixed in a 1:1 woLght ratio. The KC-400 (chlorine content 48%) showed only ml nor-differences in peak locations and relative- peak heights from Aroelor 1248 (chlorine content 48%), a PCB of American manufacture. Therefore, Aroclor 1248 to whicha mixture of di-% tri- and tocrnehlorinated dibenzofuransl had been added* . was used by Porter to devise a procedure to separate Cl-DBF ' from PCB (PORTER, 1971b). This procedure is outlined in Figure I. ! This reference mixture was kindly provided by Albert pohland,^ ph.D., of FDA, It was. prepared by chlorination of dibenzofurnn in tetrachlomethylene with iron powder as catalyst. NV 036457 / KC-400., in'hexane solution" ** s!** l Florieil Column. FCB in eluate A (discard) i Cl-DBF In eluate B Alumina Micro Column (PORTER, 1971a) FCB in eluate 1 ; (discard) Cl-DBF in eluate 2 ' \ Figure 1. Procedure for Separation pf Chlorinated Dibenzofurans from Chlorinated Biphenyls in KC-400. i The Florisil (8g. of 60/100 mesh PR grade) was activate^ by heating at 130C 12 hr. and cooled to room temperature in a desiccator, prior to the addition of sample, to the column,[the Florisil column was prewashed with about 25 ml-hexane. The KC-400/hcxane solution (< 2-ml) was placed on the Florisil which was contained in a 10 mm i,d. 300 mm glass column fitted with a teflon stopcock and medium porosity fritted.disk The, column was eluted with two elUants in succession at the rate of about 5 ml/min. Eluant A was 150 ml hexane. Eluant B was 100 ml 5% ethyl ether in hexane, Eluate B was concentrated to ! about one ml before placing the sample on the alumina micro \ column. . . ; The micro column was a disposable pipet of roughly 5 mnr bore into which a small wad of glass wool had been inserted ; to retain the aluminum oxide. Merck alumina which had been heat activated exactly like the Florisil-was-poured into the column to a depth of 2.5 cm. Sample or eluting solvent was , then added to the column with the aid of a disposable pipet | or a large glass syringe. Eluant 1 was 10 ml 1% methylene < chloride in hexane.' Eluant 2 was 6 ml 207* methylene chlo ride In hexane. Further FCB removal was accomplished by < concentrating eluate 2 to about one ml and subjecting It to : NEV 036458 J a second alumimt micro column cleanup, simply repeating the alumina column procedure.' .. (* "* i Tills procedure was first applied to.duplicate two mg. por tions of KC-400. Electron capture gas chromatography of \ eluato 2 showed a gas chromatographic peak In each sample that had a relative retention time matching* that of a tctrachlo- rodibonzofuran on both a 10% OV-lOl arid a 1:1 10% DC-200 +' 15% QF-1 column under Pesticide Analytical Manual condi- \ tions.2 Hie reference material for this retention time com parison was the tetrachlorodlbenzofuran component in the mix ture of chlorinated dlbenzofurans (see footnote 1)., 1 i I To obtain sufficient material from the KC-400 for combined gas chromatography/mass spectrometry (GC/MS) analysis, the pro cedure was scaled up to permit cleanup of 8 mg portions of ! KC-400. Cleaned up cluates (eluatc 2) were then combined tjo give two composite'samples, each equivalent to approxiinnteljy 100 mg KC-400. These two composites were analyzed by GC/MS (Flnnigan 1015 quadrupole mass spectrometer) and tha re- \ suiting data wereexamined for evidence of the presence of | compounds that were not chlorobiphenyls. A reagent blank cjf equivalent size was also analyzed by GC/MS. The reagent blank contained none of the compounds found in the samples.) The qualitative results of the KC-400 analyses are summe- ! rized in Table I . * The identification-of the compounds 1 listed in Table I.was based on a comparison of the mass | spectra of the sample compounds with the mass spectra of the jchlorinated-dibonzofuran components in the*reference mix- ture described in footnote one and on comparison with mass ! spectral data DiY't'hlurina'ted dibenzqfurans in the published ' literature (VOS ei. al.,1 1970a; FIRESTONE, etV al., 1972; | PL1MMER, et. a l . , 1973). ` The amount of tetrachlorodlbenzofuran contained in the reference mixture,vas approximated by comparing its micro- , coulometric GC response with the response given by a known {quan tity of a tetrachlorodibenzodioxln of known structure. Thej totrachlorodlbenzodioxin was considered an appropriate micrtocoulometric reference ,compound because of Its structural 2 Gas chromatographic conditions; glass column, 6 ft. x 4 ) m a id., @ 200C with 225C injector; carrier gas, nitrogen @ 120 ml /min,; d e t e c t o r t r i t i u m source (concentric type)*, operated In electron capture mode at 210C. Detector t voltago was adjusted so that 1 ng heptachlor epoxide causes 1/2 full nealo recorder deflection at 1 x 10~? AFS. | 'I NEV 036459 * Mim similarity to tctrachlorodfbcnzofuran. The .electron capture! GC responsos of the tetrachl.oro component In the synthetic I mixture and in the KC-400 extracts ere .then compared to I `quantitate the tctrachlorodibenzofuran In the kc-400, The ; KC-400 was estimated to contain approximately.1 ppm tetra- t chlorodibenzofuran." t I TABLE I \' . Compounds present In Kanechlor 400 After Cleanup* *. . . \ *i .* i ; - - \v *- . : . mass (no. of Cl) and . .- molecular weight fragmentation data w- compound present of compound i .r 256(3C1)M-2C1. 270(301) ;M- (COfCl) .o* .* 290(401) ;M-2Cl 298(5C1) M-2C1 `i T 304(401);M-(COfCl), M-(COf3Cl) v 312(5C1) M-C > t trichlorobiphenyl, t *' . trlchlorodibenzofuran \ 256 *I ' 270 ' ;' I tetrachlorobiphenyl 290 '` I pentachloronaphthalene : > 298 ;i tetrach'lorodibenzofuran 304 i unidentified; perhaps i methylpentachloro- i naphthalene or pentachloro- phenyl cyclopentadiefie 312 338(4 or 5C1) ;. M-(COf3Cl) indicated -V possible pentachlorodibenzofuran 338 listed in order* of elution from 3% OV-101, 6- ft. x 2 mm i;d. glass column a t `220Ct carrier flow 20 ml He`/min.; the same MS evidence was found In both -samples examined for all entries but the last (possible pentachlorodiben- tofuran) for which a second determination was not- at tempted. . SUMMARY 1 Chlorinated dlbenzofurans, Including one ppm of tetrachlorlnatod dlbenzofuran(s), and pcntachloro naphthalene have been ; identified by combined gas chromatography/mass spectrometry id a NEV 03 6 4 6 0 f \ polychlorinated biphenyl of Japanese manufacture (KC-400), The . question of whether chlorinated dibenxofurans in KC-400 may have contributed.to the reported Yusho incident is thus raised ACKNOWLEDGEMENT. I ** I The authors wish to thank Sandra Mercier for assistance In the preparation of the samples for GC/MS: analysis., : i` REFERENCES .L FIRESTONE, D . , RESS, J . , BROWN, N, L . , BARRON, R, 3P. and j DAMICO, J. N , : J, Ass. Offic. Anal. Chcm'. 55, 85(1972) iI KURATSUNE, M., YdSHIMURA, T,, MATSUZAKA, J., AND YAMAGUCHI, A.; Env. Health Perapect, Expt. Issue #1, 119 (1972) ' 1*1 PESTICIDE ANALYTICAL MANUAL*,Vol. I, sec. 311.2, Food and.'Drug Administration, Washing ton, D.C., 1968, rev. 1969 and 1970 . I j ! I i PLIMMER, J. R., RUTH, J. M. and W00LS0N, E, A.: J. Agr. Food Chem. 21, 90 (1973) ' PORTER, M. and SURKE, J. A.: J. Asa. Offic. Anal; Chem. 54, (6) 1426 (1971a) .j j 1 PORTER, M . : Private Communication, (Feb., 1971b)* I VOS, J, G., KOEMAN, J, H , , VAN DER MAAS, H. L., TEN NOEVEfe DE BRAUW, M. C., and DE VOS, R. H.: Fd. Cosmet. Toxicol. 8,1 625 (1970a) ,. , <* , ! JVOS, J. G . , and KOEMAN, J. H , : Toxicol, Appl. Pharmacol 17, (3) 656 (1970b)i ' :\ . ! NEV 036461 i