Document qk4vqMBbmj1ZxqXOGEroLmvxk

G E N E R A L ^ ELECTRIC C O R P O R A T E n7 R E S E A R C H 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 O D E V E L O P M E N T C E N T E R . P O. B O X 3 S C H E N E C T A O Y . NEW Y O R K 12301. Phone l518> 335-2211 DEVELOPMENT Dr. James D. McKinney NIEHS, Chemistry Section POB 12233 Research Triangle Park North Carolina 27709 Building February K-l, Room 13, 1978 3B35 c 37 Dear D r. M cK inney: My apologies for not getting to you faster with the backgroL information on the samples of used and unused transformer-type PCS formulations, or the issues of concern. The story turned out to be a bit more complicated than I had realized. Basic Concerns. To*my knowledge, every reported episode of chronic human poisoning (e.g., chloracne, etc.) associated with expo sure to PCB's has involved a PCB preparation that was either known tc contain toxic contaminants or else had been subjected to considerable heating, or both. In the most dramatic and recent episode, the 1968 Yusho incident in Japan, the heating occurred in a heat exchanger, considerable alteration of the PCB resulted, and sizeable amounts of the highly toxic PCDF's are known to have been formed. However, heat exchangers are not the only devices where P C B `s are exposed to heat. For forty years, a few percent of all transformers manufactured in this country have contained PCB-based dielectric fluids (another few percent were dry-type, and the remaining 90% filled with a special grade of mineral oil). The PCB-filled transformers were those designee for indoor use in factories, public buildings, laboratories, etc. (you've probably got one somewhere in your own lab) and also on rail roads. Both types are designed for a maximum operating temperature around 100C - far less than might have been encountered in, e.g., he exchanger applications - but in some instances such temperatures may have been frequently attained over prolonged periods. Eventually, a m transformer is likely to show up at a service shop for maintenance a m repair - operations in which exposure of workers to the dielectricsoaked component parts is almost inevitable. To date, I am aware of no reports of human toxication resulting from such exposure (or, for that matter, to any other PCB-containing product made in the US since 1950) ; however, in view of the serious effects seen in Yusho, and the large numbers of warm PCB-filled transformers scattered around the country, it seems prudent to find out whether any strange PCB conver sion products are being found in those transformers. 783212 K) Dr. James D. McKinney - 2- February 13, 19 /O Source of "Used" PCB. According to my informant. Dr. R.C. Osthof the used PCB I sent y.ou came from an industrial transformer that had been subjected to fairly heavy duty for 20 years, i.e., probably 3 hours a day, 5 days d week close to its design limit of 65 over ambie or nearly 100C. I'l'l comment on ways of making sure it had that length of service later on. The resulting severity of exposure prob ably represents close1 to the upper limit for transformer PCB's in the field. Railroadj transformers are designed for a 75 rise over ambient but the ambient averages a lot lower, the maximum loadings are only rarely attained, and they get beat up and leak, so that the PCB doesn' stay in them as long. Fluid loss can occur in industrial transformers too, for reasons that are not well understood; one known route is peri odic sampling toj check on dielectric properties, which can be quite a drain if practiced over-zealously. Composition of RCB-based Transformer Fluids. Now for the compli cated part: the' nitty-gritties about what's actually in those samples The labels on th'e containers say "Fyranol," but that is merely a gen eric trade name |for GE-purified and blended non-inflammable dielectric fluids (Askarels), arid the 6-digit number that indicates the specific composition has been lost in both cases. So a bit of background and sleuthing is required. The GE "Pyranols" were all based on commercial PCB's (Aroclors from Monsanto) riht ad been given quite intensive treatment with "fuller's earth"j (specifically, attapulgite clay from the FloridaGeorgia border region) to remove all traces of polar impurities. The} also contained races of added HCl-scavengers as "stabilizers." Be yond this, the ijyranols used in capacitors were essentially straight PCB's (Aroclor 1254 up to 1954; mostly Aroclor 1242 from 1954 to 1970; and Aroclor 1016 from 1971 to 1977), while those used in transformers a l w a y s 'contained some added tri- and tetrachlorobenzenes for viscosit} control and decreased flammability. The historical pattern on trans former-grade Pyranols runs_mainly as follows: 1. Prior to the early 1950's, Monsanto'used a different PCB production process from their ultimate one, and hence in very old spe cimens (probably not]what you've got here) the distribution of trace impurities at the time of manufacture would be different from that in modern materials; ancl possibly like that in the European and Japanese PCB's. GE also[used| tin tetraphenyl as the stabilizer at that time, so if you find any tin in a GE transformer fluid you know you've got a real antique. 783213 / 'JCN`>010403 Dr, James D. McKinney -3- February 13, 1 5 /8 2. During the period 1953-1963, when your "used" Pyranol was probably put in service (assuming that there was no make-up, and it . really is all 20 years old) the composition used was 45% Aroclor 1260, 40% trichlorobenzene (commercial mixed isomers; about three-quarter 1,2,4- and one-quarter 1,2,3-), 15% tetrachlorobenzene (also commercia mixed isomers) and 0.125% 3,4-epoxycyclohexyl-3,4-epoxycyclohexane carboxylate as stabilizer. 3. In the late 1960's the favored blend was 80% Aroclor 1242, 13% trichlorobenzene, 7% tetrachlorobenzene, with the same epoxide stabilizer. 4. Since 1970, GE usually used 65% Aroclor 1254, 25% trichloro benzene, 12% tetrachlorobenzene in manufacturing, but the service shoj: often used a 60% Aroclor 1254-4C& trichlorobenzene blend, or even a Westinghouse "Inerteen" (often a 70%-30% blend) for make-up. The same epoxide stabilizer has remained in the GE transformer-type Pyranols; I don't know what W. puts -in their Inerteens. 5. Regardless of original date of manufacture, it is important realize that no one, least of all the manufacturer, really knows for sure what a customer's maintenance crew does to a transformer once th get it. If they lose fluid for any reason, they may use just about anything for make-up: old "Pyranol" they had in storage; new "Pyrano or Westinghouse "Inerteen" of different composition, or even (rarely) mineral oil. Don't panic if your search for tracecontaminants and conversion products turns up something wild; just compiain politely an we'll go after another sample. In short, the bad news is that there's a high probability that t nice new unused "Pyranol" we sent you has a different composition frc the used fluid; if it doesn't,- it's because somebody drained and re filled the transformer from which the oil came sometime within the pa 7 years. Also, there's no real chance of duplicating the original 1953-63 fluid; it hasn't been used in 15 years, and Aroclor 1260 it self hasn't been made since 1970. The good news is that because of the progression in Aroclor usei (1260 to 1242 to 1254), and the early switch in stabilizers, it shou. be possible to confirm the ages of transformer fluids from their cher compositions. As a minimum, I would strongly recommend that you che< 783214 Dr. James D. McKinney - 4- February 13, 19 the gross G C on both the used and unused Pyranol specimens in order to verify their al .eged ages. If for any reason you'd rather we do it, let me know and I'll send the other gallon of used oil back to the ^ Transformer Lab people M y information sources are Drs. Robert C. Osthoff, Manager Transformer Programs], Tel {413) `494-4381 and William Fessler, an Organic Chemist| in the Transformer Laboratory, Tel (413) 494-4391, bot located at the GE Large Transformer Dept, 100 Woodlawn Ave, Pittsfield Mass. 01201. They've asked me to handle discussions of toxic impuri ties in PCB forj them, but both are readily available by phone if you need solid informati'on in a hurry and cen't get ahold of me. cc: RC Osthoff - Pitt;sfield RE Henry - Hudson Falls RE Brooks Sincerely yours, y. } .e > . ~ - John F. Brown, Jr. Mgr - Life Sciences Branch PHYSICAL CHEMISTRY LABORATORY t'Ofinrnjk 783215 c