Developmentof a Nontargeted Algorithm for Per- andPolyfluoroalkyl Substances in the FT-ICR Mass Spectra of Complex OrganicMixtures

The unrivaled mass resolving power and subppm mass error enable the application of Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS) for nontargeted screening of per- and polyfluoroalkyl substances (PFAS). Few automated FT-ICR MS methods exist for nontargeted analysis of PFAS in solutions containing other dissolved organic matter. In this study, isotopic pattern analysis and dynamic homologous series inspection were complementarily employed in the FTMSDeu algorithm to simultaneously assign chemical formulas for PFAS and other organic molecules, with an overall assignment accuracy of 91.2% for 1,802 unique PFAS formulas and their 11,634 isotopic formulas. The negative effects of fluorine-related isobaric isotopes with sub-mDa mass differences on formula assignment for PFAS and natural organic matter were addressed by an empirical F/(H + X) rule and the adoption of a threshold for the isotopic formula number plus the homologous series number. There were 203 unique PFAS formulas containing one novel PFAS formula, C₉H₆O₄S₁I₁F₁₁, identified in a fluoropolymer industry-impacted river water sample using our algorithm for analyzing FT-ICR MS spectra. Overall, the results of this study highlighted the great potential of the updated FTMSDeu algorithm for nontargeted and high-throughput chemical assignment for the FT-ICR MS spectra of complex mixtures containing PFAS and dissolved organic matter from real environmental samples.