Document R2L7YyYGkj2QwjQv6d7vwzLDB

/& V- 4* G E N E R A L ELECTRIC CORPORATE RESEARCH AND g e n e r a l e l e c t r ic c o m p a n y , r e s e a r c h a n d d e v e l o p m e n t CENTER, P.O.I BOX 8 SCHENECTADY, NEW YORK 12301, Phono (518) 385-2211 DEVELOPMENT Building K-l, Room 3B35 N o v e m b e r 22, 1978 Dr. James D. McKinney Head, Chemistry Section, EBCB National Institute of Environmental Health Sciences P.O. Box 12233 Research Triangle Park, N.C. 27709 a Subject: Chemical Constituents of Heavily Used PGB-Based Transformer Oil D e a r D r. M cKinney, O a Thank you for your letter of November 15, and even more for the e x t r a work that yo u a n d y o u r collaborators di!d to |pr o v i d e at least p a r t i a l resolution of the qu e s t i o n s I raised |on October 17. If I understand your findings correctly, as reported in your letters of September 22 and N o v e m b e r 15, one can now d r a w the following c o n c l u sions: 11 Exposure H i s t o r y . W e st a r t e d ou t with dielectric fluid from a transformer that had seen 20 years of heavy duty service, b u t no solid evidence that the fluid h a d b e e n in there- all that time. Yo u r findings that the PCB was predominantly Aroclor 1260 and the stabilizer pre dominantly phenyl tin puts the date of manufacture of the fluid back in the early 1950's, thus supporting the presumption that it had been i n the transformer all the time. Additional |s u p p o r t for the pr e s u m e d exposure history is provided by your observed conversion of the origi nal tetraphenyl tin to triphenyl tin hydroxide. 2. Composition of E x p o s e d F l u i d . Your analyses turned up several extraneous contaminants, such as aliphatic hydrocarbons, polycyclic aromatic hydrocarbons, p h t h a l a t e esters, thiazols;!, and brominated a r o matics, but all at the low levels expected for normal industrial cross-contamination with non-Askarel dielectric fluids, plasticizers, antioxidants, flame retardants, etc. There is no| evidence that any sizeable amount of a non-Askarel fluid was ever added as make-up. in short, the fluid exposed in the transformer was indeed a conventional 1950-era Aroclor 1260 - trichlorobenzene-tetrachljorobenzene mixture. The latter two components are prepared by random chlorination of benzen and are known to contain a b r o a d range of isomers; and homologs. 783257 * ' J I Chemical Constituents of Heavily Used PCB-Based Transformer Oil J - 2- 3. Identifi c a t i o n of Poly c h l o r i n a t e d D i p h e n y l Ethers in both U s e d and Unused F l u i d s . I believe this is the m o s t novel finding to come out of your work, and one that is still enigmatic. At least some of the P CDPE m u s t have been p r e s e n t f r o m the sjhart,| b u t w e cannot exclude the possibility that some was also formed in service until we can get a hold of a control sample of 25-year o l d u n u s e d -fluid. M o n santo da t a indicates that there w a s n ' t any P C D P E in their p'cB's of recent m a n u facture. W e ' v e checked over some of our Gcj data' here and found one trichlorobenzene spectrum showing peaks i n jthe C ^ C l x range d i f f e r e n t from those resulting from photochemical coupling^ , but no mass spec was done at the time to determine whether the contaminants were PCB's or PCDPE's. We've also just run across a 'chlorodiphenyl ether con taminant in a -chlorobiphenyl s p e c i m e n thajt we-were using for. b i o degradation studies. M y guess is th a t the PCDPE1s found in the dielectric fluids got in there as original contaminants in the tri- or tetrachlorobenzenes (the latter w o u l d e x p l a i n tlie high degree of chlorine substitution seen) bu t w e d o n ' t halve *any definite proof as yet, 4 U n d e t e c t a b i l i t y of P o l y c h l o r i n a t e d Dinenzofurans in Either Used or U nused Tra n s f o r m e r O i l s . I g a t h e r f r o m |your j November 15 letter tha the previous identification of a hexachlorodibenzofuran in unused oil now looks p r e t t y doubtful: the ch r o m a t o g r a p h i c ( b e h a v i o r (binding weafc to silica, strong to alumina*) d o e s n ' t fit, |nor does the mass spec. Tf presence of PCDPE's at the 2000 p p m level in the old oil certainly complicates the detection of trace levels of PCDF's .therein, b u t it's obvious that that used oil could not contain anything like the 10,000 pp m of PCDF reported for Yu s h o oil. In addition, y o u found it to lac} the quaterphenyls that were apparently formed by pyrolysis in the YusI oil at levels some 50-fold greater than those of the PCDF's. I think we must conclude from this that although one cannot exclude the possi bility that the level of PCDF in the transformer fluid may have in creased slightly over the 0.5 ppm level reported for Aroclor 1260, there is also no evidence that'it actually|did;Jand further no evident that usage in the transformer ever subjected the PCB to sufficient thermal or chemical stress to cause any structural alteration. I1 Please let me know if the above summary misstates your views in any way. As I indicated in my letter of Febr u a r y 15, we are concernei over the question of PCDF levels in old PCB-typW transformers because tnese devices ao snow up at our service snops rrom time to time C tt- > 783258 ( Chemical Constituents of Heavily Used PCB-Based Transformer Oil -3- whether for service, rebuilding, or dismemberment for disposal - and e x p osure of w o r k e r s to the P CB-soaked parts |is a l m o s t inevitable. Existing guidelines for handling PCB's apparently presume that all specimens are as toxic as Yusho oil, and hence probably provide safety factors of m a n y thousands. If there is any e v i d e n c e that this is not the case, we need to know about it. I gather fro m the final paragraph of y o u r N o v e m b e r 15 letter that you*ve already done everything you consider appropriate as regards the detailed chemical analysis of transformer fluids 1 and would be willing to turn over the raw data (chromatograms, spectr, etc.) if we'd like to have it. W e w o u l d indeed very much like to h a v e this information? please pack it up and send it to me here. A g a i n let m e thank y o u for y o u r help w i t h aj job that obviously turned out to b e f a r m o r e complex than any o'f us 'suspected at the .outset. : 1' Sincerely yours, v M - ~V- John F. Brown, Jr. Mgr - Life Sciences Branch PHYSICAL CHEMISTRY LABORATORY cc: I. Selikoff (Mt, Sinai) RC Osthoff (Pittsfield) 783259 a O JP* 00, I