Analysis of toxaphene and its eight congeners in sediment and fish tissue by gas chromatography-negative ion mass spectrometry

Abstract

Toxaphene quantification incorporating gas chromatography/negative chemical ionization mass spectrometry (GC/NCI-MS) offers improved sensitivity and specificity. The U. S. Environmental Protection Agency (USEPA) recently released a GC/NCI-MS method (Method 8276) for the measurement of technical toxaphene and eight specific congeners (Hx-Sed, Hp-Sed, P26, P41, P40, P44, P50 and P62). However, there is still lack of a practical and complete analytical method including sample extraction, clean up, instrumental analysis, and data analysis. The goal of this work was to develop a ready-to-use method for the quantification of total toxaphene and the eight congeners. Sediment and salmon fish tissue were selected as sample matrices and extracted with methylene chloride using an accelerated solvent extraction system. The sample extracts were cleaned up with active copper powder or gel permeation chromatography, and finally silica/alumina combination column. Separation was performed on a DB-XLB column. GC/NCI-MS was operated under selected ion monitoring mode with an identical set of confirmation and quantitation ions for total toxaphene and the eight congeners. Oxygen reaction of polychlorinated biphenyls (PCB) was monitored by PCB204, an internal calibration standard, and the reaction level was kept below 1%. Average relative response factors were used in quantitation. Quantitation of total toxaphene employed the sum of all detectable (S/N > or = 3) 6-C1 to 10-Cl homolog peak areas, while the individual congeners were quantified followed the standard procedures for single analytes. Multi-point calibration solutions ranged from 0. 5 (5 for P62) to 500 microg/L for the individual congeners, and 50 to 500 microg/L for technical toxaphene, with the lowest calibration levels as lower limits of quantitation. Average congener recovery was (90.8 +/- 17.4)% (n =10) in spiked sediment with relative standard deviations of 5.4% - 12.8% (n =10), underscoring an excellently accurate and precise method. The method was applied to analyze sediment and fish tissue sample.