Enhanced selectivity for PFAS analysis in complex biological matrices by LC-MS/MS: Application to mouse feces and
Gong Zhang1, Amélie Blais1, Azam F Tayabali1
1Environmental Health Science and Research Bureau, Health Canada, 251 Sir Frederick Banting Driveway, Ottawa, ON K1A 0K9, Canada.
Abstract:
A sensitive and robust micro-solid phase extraction LC-tandem mass spectrometry (μSPE-LC-MS/MS) method was established for quantifying 23 per- and polyfluoroalkyl substances (PFAS) in mouse feces and liver. Method development identified interferences due to the complex fecal matrix, especially at unit-mass resolution detection, where co-eluting endogenous compounds caused transition-specific interferences. Conventional C18 stationary phases failed to adequately resolve these matrix components, particularly for perfluoroalkyl sulfonic acids (PFSAs), resulting in compromised selectivity. To reduce these matrix effects, alternative stationary phases, including Phenyl-Hexyl (Phe) and Pentafluorophenyl (PFP) columns, were systematically evaluated. The PFP column, operated with acetonitrile in the mobile phase, demonstrated superior chromatographic selectivity and effectively eliminated co-eluting interferences. Notably, transition-specific interferences previously observed for PFOS on C18 columns were resolved, allowing accurate quantification using both m/z 499 → 80 and 499 → 99 transitions. Using the improved selectivity of the PFP column, automated μSPE cleanup combined with isotope-labelled internal standards achieved high extraction efficiencies (96-104%), ensuring reliable accuracy and precision (recoveries 96-99%, RSD <5%) for most analytes despite matrix complexity. These findings demonstrate the importance of chromatographic selectivity, rigorous cleanup, and isotopic correction for quantifying PFAS in highly complex biological matrices using unit-resolution mass spectrometry. The developed workflow provides a reliable analytical platform for toxicokinetic studies and may be applicable to other challenging biological matrices.


