Document Ner4jNk63qkMbzXbpxZwgkzGE

Monsanto MONSANTO RESEARCH CORPORATION / DAYTON LABORATORY / DAYTON, OHIO 45407 EASC - Dayton (DfPT./SfCTION) Interim TYPf Of NCPORT REPORT REPORTNO.: JOS/PROJECT NO.: DATE: PERIOO COVERED: TITLE: MDA-291 320.0301 March 16, 1982 REPORT ON ANALYSIS OF PCBs IN HUMAN SERUM AUTHORS: Roy W. Noble OUk ABSTRACT: An interin report'on the analysis of PCB's fn human serum is presented. The problems encountered with the Mound samples and the steps taken to resolve these problems are discussed. 12 o> Technical Approval: Joseph J. ijSroi Research G CC/MS Technology Leader -- Approved by: Jrobeeirtt F. Ivory (7 Environmental Analytical Sciences Center Service Project Manager Approval date: 3 / 6>*&oL. RESTRICTED DISTRIBUTION (see over) COMPANY CONFIDENTIAL TN WW W A, WWW > Mwwi , * m Menu mu tMMi coenotwrui neoenunoH --n** tmMmdnt niwmi* Mw--in mi rt ptcpvr twihwiimin. a fumaw * ime m n,nem. MWMwinCueni HONS 015334 INTERIM REPORT ON ANALYSIS OF PCB's IN HUMAN SERUM INTRODUCTION In September 1981 a report was given to B. Ward and G. Butler on the progress of Monsanto's PCB in human blood analysis pro gram conducted at the Ultratrace Facility of Dayton Laboratory. At that time it was decided to proceed with testing of the me thod on real samples. It was arranged to acquire blood samples from 10 donors. Samples were received from volunteers at the Mound Facility - MRC in mid-November and the analysis commenced. RESULTS AND DISCUSSION The initial analyses using the procedure outlined in St. Louis were completed by the end of November. Fot quality control purposes, spikes were conducted using a pooled human serum which is commercially available that was then foTtificd with PCB's. The results of these analyses are shown in Table 1. The average levels found compare reasonably with previously reported human serum levels of 17 ppb (See Journal of American Medical Association 24S, 2S0S (1981)). The low recoveries were thus attributed to a poor spiking procedure and an alternate method was attempted. This involved spiking the serum using a solui^on of PCBs in acetone. Four of the Mound samples were reprocessed using this technique. Initially it appeared that this spiking technique was sucessful but closer evaluation of the data indicated that.no more than one half of the PCB's were observed. Thus a'series of experi ments were run to test the transfer efficiency of PCBs in serum. These indicated that total recovery of all PCB's was on the order of SOI. The results also indicated that any PCB losses due to transfer of serum to the extraction solvent was less than 101. Nith the high extraction efficiencies of PCB's into hexane we thus suspected that the losses could be a problem related to the lot AgNOa-silica gel column. A complete elution study was then conducted on silica prepared within 3 weeks of the study. A standard PCB mixture was placed on the head of the column and 50 mL of hexane was passed through the column collecting 5 mL fractions. Capillary GC-EC indicated incomplete elution in the first 30 mL of hexane which was the quantity of hexane that had been collected in the Mound studies. This volume was based on the early column elution studies that were analyzed by packed column GC. The more definitive capillary 1 HONS 015335 TABLE 1 MOUND DONOR RESULTS IN ppb Donor 1 2 3 4 5 6 7 8 9 c 10 Sample 11.4 7.9 4.5 36.9 9.0 16.0 14.8 10.0 20.4 2.7 Replicate a 5.4 7.2 12.4 9.4 15.6 14.5 5.8 17.3 3.9 Averaqe 11.4 6.7 5.9 24.7 .9.2 15.8 14.7 7.9 18.8 3.3 Monitrol Spil b 44% 324 b 14% 45% 4% 6% 5% 44% Sample contaminated bNo recovery calculated as Average concentration is greater than Spiked sample total PCB concentration cUsed experimental vacutainer, original donor 10 sample was lost due to breakage x' "OHS 015336 analysis showed a clear fractionation pattern of PCBs on the liquid adsorption chromatography column. It showed that the higher chlorinated biphenyls are eluted off the column first and subsequently elution follows through the sequentially lower chlorinated species. This work was repeated with a freshly prepared batch of AgN03-silica gel. The new batch was prepared in a very strict manner controlling conditions to more accurately follow the published procedures instead of the modified procedures that had been used earlier in the method development. The results of this study showed that all PCB's were eluted off the column by 25 nL of hexane with recoveries of 851. This study also showed that the fractionation of PCB's followed the same pattern as the previous study, but total elution occurred in a smaller total volume more in line with what had been observed in the initial elution studies. Samples of each of these batches of silica gel are being kept to conduct a storage study. Kith this information the chromatograms from the Mound donor samples were re-examined. In particular we examined the acetone spiked samples. They show that the low recoveries are accounted for in the front half of the chromatograms which show the absence of the lower chlorinated PCB's. Unfortunately no sample of thg AfNOj-silica gel used in that study.remains to perform an elubTon study on that batch. We can conclude that the low recoveries are due to incompfeg/te column elution of PCBs with 30 mL of eluant. In order to prevent this problem from reoccurring, we have now instituted some additional quality assurance steps. The first QA procedure we are instigating is to perform a com plete elution study of each new batch of AgNQ3-si`lica gel. We will also now carry through with each work-up procedure a simulated spike that will be applied directly to the head of the column and coluaui recovery will be calculated. Currently we are conducting a time study to discover the elution charac teristic changes of the silica with time. This will enable us to ensure a greater degree of column reproducibility. The PCB in human serum method is now undergoing further testing using blood fTom three donors here at the Dayton Lab. This will give us a better ilea of the status of the method and whether or not it is ready to resume field testing. 3 HONS 015337