Abstract Description: The applicability of Environmental Protection Agency (EPA) Method 1633 for analyzing per- and polyfluoroalkyl substances (PFAS) in complex aqueous matrices was evaluated using industry-sourced samples tested across eight Department of Defense (DoD)-accredited contract laboratories. Replicate samples from the pulp and paper industry, including influent, effluent, synthetic wastewater, river water, and blanks, were analyzed using EPA Method 1633, EPA Method 1621, and Total Oxidizable Precursors (TOP). This study aimed to (1) assess the performance and consistency of DoD-accredited laboratories in reporting PFAS results and (2) compare the applicability of Method 1633 with non-targeted PFAS methods in complex industrial matrices. Level IV reports were reviewed to assess adherence to sample preparation protocols and QA/QC measures, revealing reporting biases, false positives, and methodological inconsistencies. Notably, variations in initial wastewater handling protocols, particularly in response to total suspended solids (TSS) exceeding 50 mg, led some laboratories to subsample or dilute samples—practices not recommended under Method 1633. These deviations resulted in higher reporting limits and increased non-detections. In one case, a laboratory’s 20-fold dilution of an influent sample produced a false positive for PFBA, despite non-detection by other laboratories. Comparison with non-targeted methods further highlighted significant gaps in PFAS detection. Non-targeted analysis revealed that the summed mass percentage of fluoride quantified using Method 1633 accounted for only 3% (influent) and 29% (effluent) of the adsorbable organic fluoride quantified using Method 1621, indicating the presence of unidentified fluorinated compounds. These findings underscore the need for standardized sample preparation protocols across certified laboratories to ensure consistent, accurate PFAS reporting. This study provides critical insights for stakeholders and emphasizes the necessity for methodological refinement to enhance data reliability in complex industrial matrices.
