Document RXRKQ8v6ygVNpmoX65o52dVk

R.A. Lidgett, Huabon ^13th April, 1972 Aroclor 1242 Analvaig HAL:eh 15.M. Emery, St, Louis /c'U (<-/_/ Jl, E, Kellor,_ St, Louie R.A. Baxter, Huabba II.A. Vodden, Huabon Many thanks for your recent telexes and comments. We ore holding issue of any method for determination of penta Cl and components in 1242 until MCL and MC agree on all aspects of the analysis, Receipt of your last telex prompted us to re-examine our gc/m.s. work and has resulted in a number of observations which I am sure are important in both the method development side and in possible plant control. O.C.Al.S, Peal: Identification Tarry Gay did identify peak 30 as containing a mainly 5C1 component but has not detected significant amounts in the other peaks particularly peak 31 which he allocates to C14. However, when he examined 1248 there was a definite increase in C13 components in the peaks mentioned by yon and,more significantly, I think, C15 peaks appeared at shorter retention times than have been detected in 1242. While we accept that tills analysis is prone to difficulties and it is possible that we have missed C13 in some of the small peaks in 1242 we think that there is a good possibility that the composition of 1242 is very dependent upon chlorination level and that slight increases in chlorine content may affect the C15* concentration more than we might have expected. This is at present supposition but could explain differences in MC and MCL results and in the differences found between various batches of MCL material. We are at present examining a range of samples from Newport to determine if this is in fact the case. We will also be discussing this with Newport TSD to see whether examination of samples at the low and high ends of the 1242 specification can be carried out. Because of the critical effect that g.c./m,s. has on the determination of composition of 1242 we havo also arranged to meet the Unilever people to discuss the general difficulties associated with P.C.B. analysis. Examination of their mass spectra may help to clear up peak identification differences. O08')4<>5 Absolute Level of Penta CI Components In 12'i2 gent to Prodolec As discussed above It may not be possible to analyse a 1242 sample by allocating Chlorine numbers to the various poults and assuming that these do not change composition with small changes in chlorination level. However, in the sample sent to 1'rodelec 1 do not think that the ei'Xoot of peaks 50, 32 und 39 will greatly change tho result since apart from peak 38 these are present in 124213 which has a very low C15+ content. IOI(e Peak 31 remains a problem since If it does contain 2yft CIS then I agree it will increase the total Cl5+ concentration slgnlficuntly. Perhaps this anomaly will be clurlfied when you have looked at the sample of Prodeloc material - if C15 is present in that sample we will have to re assess our g.o./m.s, results. Further Developments In,the lust few days we have been testing a 300' x 64 thou. S/S Api'zon L column and results are promising (chromatogram attached). If stable we will be able to use this column without split und with a loading of l-3nl of sample. We should therefore increase sensitivity to high boilers and ior ra.s. applications. I will let you know what happens. R.A. Liduett 000 94 96 TOWOLDMONOQ57101 TOWOLDMONOQ57102 TOWOLDMONOQ57103