Impact of method parameters on the performance of suspect screening for the identification of trace organic contaminants in surface waters

Abstract

The performance of a suspect screening method to detect diverse small-molecule trace organic contaminants (TOCs) was systematically evaluated using a set of 39 model compounds. Experiments showed that ionization efficiency, ion transfer parameters, and chromatography could affect the detection of TOCs. As expected, compounds with low ionization yields and poorly retained compounds in chromatographic columns are more difficult to identify in the samples at environmental concentrations. Similarly, TOCs with large deviations from the average mass of the compounds screened were not transmitted efficiently in the mass spectrometer thus negatively affecting their detection. The suspect screening method was validated in terms of recovery and limits of identification of the model compounds using three different types of solid-phase extraction cartridges (reversed phase with polar groups, mixed-mode anion exchange, and mixed mode cation exchange). Experiments showed that more than two-thirds of the model compounds had recoveries >75% with each of the three cartridges, and comparison of limits of identification showed that more than one-half of the model compounds could be identified at concentrations between 6 and 100 ng L−1. However, it was observed that the amount of co-extracted compounds was higher in mixed-mode ion exchangers compared with the reversed-phase cartridge. Application of the suspect screening method using the three different cartridges to surface water samples showed that between 0 to 3% of the positive matches found by the peak identification algorithm were classified as probable structures. Solutions to improve suspect screening of TOCs are proposed and discussed.