Can Qualitative Nontarget Data Be Indicative of PFAS Contamination? First Evidence by Correlation with EOF in Environmental Samples

Abstract

We correlated nontarget screening (NTS) features and extractable organic fluorine (EOF) data for PFAS analysis in environmental samples. Soils, sediments, and ashes from different locations (Germany, France, and Lithuania) (n = 34) were analyzed via liquid chromatography-high-resolution mass spectrometry and high-resolution continuum source-graphite furnace molecular absorption spectrometry. NTS feature prioritization based on mass to carbon (m/C) and mass defect to carbon (MD/C) ratios was applied to discriminate potential PFAS features from common co-extracted organic matrix. Our results show a high correlation between quantitative EOF and the prioritized feature number (R² = 0.87; r_Pearson = 0.93) and the total peak area of prioritized features (R² = 0.97; r_Pearson = 0.98), while commonly applied MD ranges for filtering showed much weaker correlation with EOF. The observed correlations in PFAS-contaminated samples highlight that MD/C–m/C prioritization is associated with a good indication of the overall PFAS burden as evidenced by EOF. Also, semiquantitative data can be extracted as shown by EOF versus prioritized peak areas. The correlations indicate that weakly fluorinated compounds were negligible and that the EOF was mostly composed of highly fluorinated substances. Our method features the potential to automatically screen large NTS data sets for potential PFAS, determine the contribution of PFAS to EOF, and provide preliminary estimates of the overall PFAS burden.