Accuracy and precision of determinations of chlorinated pesticides and polychlorinated biphenyls with automated interpretation of mass spectrometric data

Performance characteristics were determined for an analytical method that includes automated interpretation of mass spectra of polychlorinated biphenyls (PCBs) and 21 chlorinated pesticides. Determination of PCBs was by level of chlorination. After three capillary column injection techniques were evaluated, a splitless, cold-trapping technique was used. Reproducibility of mass spectrometer response to calibration compounds relative to two internal standards was evaluated over a period of 3 months. In the three sets of fortified water extracts, the largest relative standard deviation of replicate measurements was 8.4% for individual pesticides (concentrations of 3-10 ..mu..g/L), 16% for PCB isomer groups (concentrations unknown), and 2.9% for total PCBs (concentrations of 27-130 ..mu..g/L). Method bias was calculated for each pesticide and for total PCB concentrations in each of the three data sets. The highest method bias was observed with river water extracts. The detection limit for each pesticide in reagent water is approximately 1 ..mu..g/L.