7942 - Laserfiche WebLink

Home Browse Search

of 4 % ethyl acetate in benzene. In 1978-79 Silica Gel 60 (E. Merck, 70/230 mesh) was used and the pesticide frac- tion eluant was changed to 25 mL of 25% diethyl ether in hexane. Total volume of the PCB eluant used depended on the activity of the silica gel and was calculated inde- pendently for each batch. In general, 35-45 mL of 0.5% benzene in hexane were used to separate the PCB's from the pesticide fraction. Both eluate fractions were rotoevap- orated to reduce the volume of solvent, and the resulting concentrates were diluted to 5 mL with isooctane prior to GLC. Gas-Liquid Chromatography In 1976 all samples were analyzed with a Tracor 222 gas chromatograph equipped with an electron capture detector. The samples were injected into a glass column (1.8 m x 2 mm i.d.) containing 4% SE-30 + 6% QF-1 coated on 80/100 mesh Gas Chrom-Q support. Operating tempera- tures were as follows: column, 200° C for dieldrin, endrin, and dacthal, and 180° C for the PCB and pesticide frac- tions; injector, 250° C; and detector, 300° C. The carrier gas was argon/methane (95:5) flowing at 30 cc/min. We determined residues by the external standard calculation method, using peak height measurements. All 1977-79 samples were analyzed with a Varian 3700 gas chromatograph equipped with an electron-capture detector. The glass analytical column (1.8 m x 2 mm i.d.) contained 1.5% SP-2250 + 1.95% SP-2401 on 100/120 Supelcoport. We used the following operating temperatures: column, 200° C for the dieldrin-containing and PCB fractions and 180° C for the pesticide fraction; injector, 250° C; and detector, 350° C. The carrier gas was argon/methane flowing at 20 cc/min. We quantitated resi- dues by the external standard method, using peak area calculations. In 1977, peak areas were collected on a Varian CDS-111 digital integrator. In 1977-79 a Perkin-Elmer Sigma 10 Laboratory Data Station was used for data col- lection. Confirmatory analyses were performed with a 3 % OV-1 on 80/100 mesh Chromsorb-W HP column with the same dimensions as those indicated above. Samples that con- tained o, p'-DDE were analyzed on this column to differ- entiate interference from chlordane. Further confirmations or identifications of unknown compounds were performed on this column by comparing results with the U.S. Food and Drug Administration (FDA) chromatography data base (McMahon 1968). Quality Assurance The accuracy and precision of residue determinations depends on four factors: (a) efficient extraction and frac- tionation; (b) analytical methods (e.g., solvents and columns with uniform behavior and low background); (c) proper instrument maintenance and calibration; and (d) accurate standard solutions. To evaluate these factors, we made systematic recovery studies on two types of spiked reference materials: (a) "hot spikes" of 14C-radiolabeled dieldrin and p,p'-DDE were used to examine recovery and separation efficiencies through the extraction and fractionation steps; and (b) "cold spikes," fish tissue blanks to which a representative pesti- cide profile had been added, were used to evaluate residue recoveries at levels approximating realistic tissue concen- trations. Some samples were replicated to evaluate preci- sion, and 10-15% of the samples were analyzed by a second laboratory to corroborate our findings. Residues in selected samples were confirmed by using gas chromatography-mass spectrometry as described by Ribick et al. (1981, 1982). With reference to factors (c) and (d), we closely followed FDA criteria (McMahon 1968) in instrument calibration and maintenance, and in the preparation of standard solu- tions. Recovery efficiency for p,p'-DDT, p,p'-DDD, p,p'-DDE, Aroclor 1254, dieldrin, cis-chlordane, and trans-chlordane exceeded 85 % throughout the study period, based on re- covery of 14C-labeled DDE and dieldrin and non-radio- labeled chemicals in spiked samples. The level of detec- tion was 0.1 µg/g wet weight for PCB's and toxaphene and 0.01 µg/g for all other residues. Concentrations were not corrected for percent recovery. Cross-check Analyses The 1976-77 cross-check analyses were performed at a commercial laboratory, here termed "Lab A," where 25-g subsamples were mixed with 100 g of Na2SO4 and blended with 100 mL of 6% diethyl ether in petroleum ether. The mixture was filtered through a Buchner vacuum funnel. The filtrate was retained and the filter cake placed in a 43- x 123-mm extraction thimble. The thimble was then Soxhlet-extracted overnight at four extraction cycles per hour with 300 mL of 6% diethyl ether in petroleum ether. The extract and filtrate were combined and concentrated by rotoevaporation and diluted to 25 mL with 15% dichloro- methane in cyclohexane. A 5-mL aliquot was loaded into an Autoprep 1001 GPC equipped with a 50-g Biobeads SX-3 column in which 15 % dichloromethane in cyclohexane was used as the solvent. At Lab A the cleaned-up extracts were concentrated just to dryness by rotoevaporation and diluted to 15 mL with petroleum ether; 5 mL were then transferred to a 20-g Flori- sil column. A total of 100 mL of 20% dichloromethane in petroleum ether (eluant A) were used to elute PCB's, DDT homologs, chlordanes, BHC's, aldrin, HCB, heptachlor, and toxaphene; and 180 mL of 50% dichloromethane in petroleum ether (eluant B) were used to elute heptachlor epoxide, dieldrin, and endrin. The Florisil eluates were con- centrated just to dryness and adjusted to a volume of 5 mL with petroleum ether. A 1-mL aliquot of the A fraction was placed on a 5-g silica gel column. The PCB's, HCB, aldrin, heptachlor, and some p,p'-DDE were eluted with 25 mL of 0.5% benzene in petroleum ether. The rest of the com-