Enhanced Fingerprinting Strategy for Rapid and Accurate Quantitative Source and Contaminant Apportionment via Online SPE-UHPLC-QTOF-MS

This study refined the nontarget fingerprinting approach for accurate source apportionment by combining online solid-phase extraction with liquid chromatography-quadrupole time-of-flight mass spectrometry in data-independent acquisition mode, enabling both source concentration estimation and compound-level attribution. We established dilution curves (100 to 0% v/v) for three representative point sources and applied the nontarget source apportionment approach to five laboratory-prepared mixed samples. A total of 2489, 5750, and 6585 qualitative fingerprints (including both MS¹ and MS² features) were established for each source; their peak areas showed statistically significant and consistent associations across the dilution samples (r = 0.51-0.98, p < 7.2 × 10^-10) and mixed samples (r = 0.38-0.91, p < 3.1 × 10^-5). Linear and nonlinear regression models were applied across multiple dilution ranges to establish quantitative fingerprints, combined with a two-step approach for source apportionment. Incorporating MS² data alongside MS¹ increased the number of quantitative fingerprints by 29.1-39.6% across three representative sources and enabled accurate source concentration estimates (0.79-1.32-fold of actual values) in mixed samples. Analysis of compounds with peak areas >10⁴ in mixed samples showed that the majority compounds (63.9-81.5%) had estimated peak areas closely aligned with observed values (0.5-1.5-fold), indicating reliable and meaningful interpretation in compound-level source apportionment. This study enhances the efficiency and robustness of source estimation and deepens environmental insights, underscoring the potential of nontarget methodologies to enhance our understanding of contaminant distributions in environments.