A systematic review of methods for the analysis of total per- and polyfluoroalkyl substances (PFAS)

This manuscript systematically reviews 156 peer-reviewed articles on methods for estimating total per- and polyfluoroalkyl substances (PFAS), following preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines. Direct and indirect methods of estimating total PFAS include targeted analysis, total fluorine (TF), total organic fluorine (TOF), extractable organic fluorine (EOF), absorbable organic fluorine (AOF), and total oxidizable precursor (TOP) assay. Combustion ion chromatography (CIC) was the most utilized method (>50%), followed by particle-induced gamma-ray emission (PIGE, 9%) and high-resolution-continuum source graphite furnace molecular absorption spectrometry (HR-CS-GFMAS, 6%). Techniques like instrumental neutron activation analysis (INAA) and nuclear magnetic resonance (NMR) were less common. A geographic bias was evident, with 69% of studies from the US (33%), Sweden (12%), China (12%), and Germany (11%). Most research targeted environmental samples (water, soil, sediments), while significant data gaps were noted in South America, Africa, and atmospheric PFAS. Challenges in inter-laboratory comparisons arise from inconsistent reporting units (e.g., mg/L, μg/m³, %, etc.). About 75% of studies involved pre-treatment (e.g., solvent extraction, sorbents), while 25% did not. PFAS detection limit and observed concentrations varied widely, from low concentrations in water (ng/L) to higher levels in soil, biota, and products (mg/L). Limitations of total PFAS methods include contradictory results when complementary techniques are applied to the same sample, potentially leading to over- or under-estimation. Across studies, a substantial fraction of TF remains unaccounted for, highlighting the need for non-targeted screening (NTS) to identify unknown PFAS (UPFAS or UOPFAS). Bridging these gaps is critical for advancing PFAS research and environmental risk assessment.