Rapid automated screening, identification and quantification of organic micro-contaminants and their main transformation products in wastewater and river waters using liquid chromatography-quadrupole-time-of-flight mass spectrometry with an accurate-mass database

Abstract

In this study we have developed and evaluated an analytical method for a rapid automated screening and confirmation of a large number of organic micro-contaminants (almost 400) and also the quantification of the positive findings in water samples of different types (surface and wastewaters) using liquid chromatography-electrospray quadrupole-time-of-flight mass spectrometry (LC-QTOFMS) based on the use of an accurate-mass database. The created database includes data not only on the accurate masses of the target ions but also on the characteristic in-source fragment ions, isotopic pattern and retention time data. This customized database was linked to commercially available software which extracted all the potential compounds of interest from the LC-QTOFMS raw data of each sample and matched them against the database to search for targeted compounds in the sample. The detailed fragmentation information has also been used as a powerful tool for the automatic identification of unknown compounds and/or transformation products with similar structures to those of known organic contaminants included in the database. The database can be continually enlarged. To confirm identification of compounds which have no fragment ions (or fragments with low intensity/relative abundance) from in-source CID fragmentation or isomers which are not distinguished within full single mass spectra, a "Targeted MS/MS" method is developed. Thereafter, these compounds can be further analyzed using the collision energy (CE) in QTOF-MS/MS mode. Linearity and limits of detection were studied. Method detection limits (MDLs) in effluent wastewater and river waters were, in most cases, lowers or equal to 5 and 2 ng/L, respectively. Only 15 compounds had MDLs between 5 and 50 ng/L in effluent wastewater matrix. We obtained a linearity of the calibration curves over two orders of magnitude. The method has been applied to real samples and the results obtained reveal that most of the pharmaceutically active compounds contained in the created database were present in the water samples with concentrations in the range of ng/L and μg/L levels and in most of the samples between 2 and 15 pesticides of the 300 contained in the database were also detected. In addition to the compounds included in the database, some degradation products were found, thus revealing the method as a useful tool for the analysis of organic micro-contaminants in waters.