A Two-Step Prioritization Framework for the Nontargeted LC-HRMS Analysis of Chemicals of Environmental Concern and Byproducts during Drinking Water Production

Abstract The growing occurrence of chemicals of environmental concern (CECs) in surface and drinking water requires advanced analytical approaches capable of covering the broad chemical space of known and unknown contaminants potentially present in these matrices. Nontargeted analysis (NTA) based on liquid chromatography coupled with high-resolution mass spectrometry (LC-HRMS) is gaining popularity as it is virtually comprehensive and may provide high data quality. However, NTA has some limitations in terms of sensitivity, which may limit its effectiveness, and offline prioritization processes are usually adopted to bring out the desired features from the vast array of NTA signals. A novel LC-HRMS-based two-stage prioritization framework is proposed to evaluate the combined chemical load by dissolved organic matter and CECs in a drinking water treatment plant (DWTP). For the first time, an experimental design (DoE) approach is adopted to optimize LC–HRMS acquisition parameters for the online prioritization of features responsible for the combined chemical load in the investigated samples. DoE was performed on a training set of 42 regulated CECs, characterized by a wide chemical coverage (e.g., log D at pH = 7 between −2.49 and 7.38), identifying optimal experimental conditions for maximizing their NTA detection. DoE-optimized online prioritization was combined with offline prioritization, performed using multivariate and univariate analyses, capturing significant trends in dissolved organic matter features, known CECs (e.g., perfluoroalkyl substances), and unknown compounds, tentatively identified as potential pesticide byproducts. Targeted analysis, performed on the CEC training set, provided a 100% match with the NTA approach, thus validating the proposed two-step prioritization framework. The two-step prioritization allows for (i) improving method sensitivity thanks to the online prioritization and (ii) revealing both persistent and potential transformation-derived contaminants, thus providing a comprehensive tool for assessing chemical load and treatment performance in DWTP.