An optimized method for PFAS analysis using HR–CS–GFMAS via GaF detection

The analysis of per- and polyfluoroalkyl substances (PFAS) via sum parameters like extractable organic fluorine (EOF) in combination with high resolution-continuum source-graphite furnace molecular absorption spectrometry (HR-CS-GFMAS) is highly promising regarding fluorine sensitivity and selectivity. However, the HR-CS-GFMAS method includes several drying and heating steps which can lead to losses of volatile PFAS before the molecular formation step using e.g., GaF formation. Hence, the method leads to a strong discrimination of PFAS within the EOF depending on their physical/chemical properties and is therefore associated with reduced accuracy. To reduce this discrepancy and to indicate realistic PFAS pollution values, an optimization of the HR-CS-GFMAS method for PFAS analysis is needed. Hence, we determined fluorine response factors of several PFAS with different physical/chemical properties upon application of systematic optimization steps. We could therefore improve the method's sensitivity for PFAS analysis using a modifier drying pre-treatment step followed by a sequential injection of sample solutions. The highest improvement in sensitivity of volatile PFAS was shown upon addition of a Mg modifier during drying pre-treatment. Thereby, during optimization the relative standard deviation of fluorine response factors could be reduced from 55 % (initial method) to 27 % (optimized method) leading to a more accurate determination of organofluorine sum parameters. The method provides an instrumental LOD and LOQ of β(F) 1.71 μg/L and 5.13 μg/L, respectively. Further validation aimed to investigate several matrix effects with respect to water matrices. Here, substance-specific behavior was observed. For example, perfluorooctanoic acid (PFOA) which was used as calibrator, showed signal suppressions upon high chloride concentrations (>50 mg/L). Hence, a thorough separation of Cl from analytes during sample preparation is needed for accurate sum parameter analysis.