Automated, Sensitive, and Robust Analysis of PFAS in Soil and Fish

Significance of the topic

Per- and polyfluoroalkyl substances (PFAS) are persistent, widely used synthetic chemicals that raise serious environmental and human-health concerns. Regulatory efforts such as U.S. EPA Method 1633A set standardized criteria for monitoring many PFAS in aqueous, solid, biosolid and tissue matrices. Reliable routine analysis of PFAS in complex solid matrices (soil, sand, tissues) is challenged by labor-intensive manual extraction, variable recoveries, and matrix effects. Automated extraction platforms combined with sensitive LC–MS/MS can improve throughput, reproducibility and compliance with regulatory workflows.

Objectives and study overview

This study evaluates an integrated workflow that uses the CEM EDGE PFAS automated extraction system coupled to the Shimadzu LCMS-8065XE for quantitative analysis of an extended PFAS panel aligned with EPA 1633A. The aims were to demonstrate automated extraction protocols for soil, sand, chicken and fish tissue; to validate analytical performance for a broad set of target analytes (48 native PFAS targets), and to assess recoveries, calibration metrics and continuing calibration verification (CCV) performance across low and mid spike levels representative of method limits of performance.

Methodology

• Targets and standards: 48 native PFAS analytes were monitored using 28 extracted internal standards (EIS) and 7 non-extracted internal standards (NIS). Calibration solutions were prepared in methanol/water with low percentages of ammonium hydroxide and acetic acid.
• Sample preparation and automated extraction: Soil and Ottawa sand samples were processed as 5 g aliquots and chicken and fish tissues as 2 g aliquots, with final extract volumes normalized to 5 mL. The CEM EDGE PFAS system executed matrix-specific extraction cycles at 65°C using basic methanolic solutions and acetonitrile where indicated, interleaved with isopropanol (IPA) washes. Key cycle parameters (examples): for soil/sand: 0.3% KOH in MeOH, 10–15 mL, 65°C, 3 min cycles with short 0.5 min washes; for tissue: 0.05 M KOH in MeOH and ACN cycles at 65°C.
• Solvents and reagents: LiChrosolv-grade MeOH, ACN, water, formic and acetic acids, ammonium solution (25%) and potassium hydroxide were used both for extraction and LC mobile phases.
• Spiking strategy: For each sample set, two low-level spikes and one mid-level spike were prepared to assess method performance across relevant concentration ranges.
• LC–MS/MS analysis: Chromatography was performed on a Shim-pack Scepter C18 column (50 x 2.1 mm, 3 µm) using an LC-40 Nexera X3 system, 12 min total run time, flow 0.3 mL/min, oven 35°C, injection 7 µL. Mobile phases consisted of 2 mM ammonium acetate in water (A) and acetonitrile (B). Detection used the Shimadzu LCMS-8065XE in ESI negative mode with MRM acquisition; MS source and gas parameters were set to enable sensitive PFAS detection.

Main results and discussion

• Calibration and instrument performance: A multi-point calibration range of 0.125 to 2000 µg/L was established for native PFAS targets. Relative standard error (RSE) for calibration points was <15% across compounds. Continuing calibration verification (CCV), performed after every 10 injections per EPA 1633A guidance, showed accuracies within the 70–130% acceptance window.
• Recoveries in soil and sand: Percent recoveries for spiked soil and sand samples varied by analyte; overall values ranged from approximately 38% (e.g., PFBSA) up to 174% (e.g., PFOA). Of the 40 PFAS regulated under EPA 1633A included in this assessment, all met acceptance criteria at both LLOPR and OPR spike levels except PFOA in the soil LLOPR condition.
• Recoveries in tissue (chicken and fish): Tissue recoveries spanned roughly 32% (PFBSA) to 225% (TFSI, an unregulated analyte). Thirty-eight regulated compounds met EPA 1633A acceptance criteria for both low and mid spikes in chicken and fish matrices; PFPeA and PFHpS did not meet criteria in tissue LLOPR and OPR spikes.
• Interpretation of outliers: Several compounds that are not currently regulated under EPA 1633A (e.g., FBSA, TFSI) exhibited wide recovery variability, highlighting matrix-dependent extraction efficiency or potential matrix interferences. A small number of regulated analytes also showed noncompliant recoveries in specific matrix-spike combinations, indicating compound‑ and matrix‑specific optimization may be necessary.

Benefits and practical applications

• Automation advantages: The CEM EDGE PFAS automated extraction reduced manual handling, standardized extraction cycles, and improved throughput suitable for routine environmental and tissue monitoring laboratories.
• Regulatory alignment: The workflow demonstrated compatibility with EPA 1633A QC practices (e.g., CCV frequency and calibration expectations) and provided acceptable performance for the majority of regulated PFAS across tested matrices.
• High-throughput capacity: Short extraction cycles, combined with a 12-minute LC–MS/MS method and robust internal standardization, support processing larger sample batches while maintaining data quality.

Used instrumentation

• CEM EDGE PFAS automated extraction system for solvent-based extraction of solid and tissue matrices.
• Shimadzu Nexera X3 LC (LC-40) for chromatographic separation using Shim-pack Scepter C18 columns.
• Shimadzu LCMS-8065XE triple quadrupole mass spectrometer operated in ESI negative mode with MRM acquisition for PFAS detection.

Limitations and considerations

Some analytes exhibited poor or excessive recoveries in specific matrices; these outliers underscore the need for compound-specific checks, potential additional clean-up, or recovery correction when reporting regulatory data. The study used standards and reagents tied to specific suppliers and the authors disclosed affiliations with instrument vendors, which should be considered when interpreting applicability across different lab setups.

Future trends and potential applications

• Broader analyte panels: Continued expansion of target lists to include emerging PFAS and precursors will require ongoing method adaptation and validation.
• Automation and sample throughput: Further integration of automated sample handling, online clean-up and robotic sample logistics will improve lab efficiency and reduce analyst variability.
• Miniaturization and green chemistry: Reducing solvent volumes and developing greener extraction chemistries will be priorities to lower costs and environmental impact.
• Interlaboratory harmonization: Wider adoption of automated workflows aligned to EPA 1633A could enhance comparability of PFAS monitoring data across labs and jurisdictions.

Conclusion

The integrated CEM EDGE PFAS automated extraction with the Shimadzu LCMS-8065XE provides a viable high-throughput workflow for quantitative analysis of a broad PFAS panel in soil, sand and tissue matrices. The approach met EPA 1633A performance criteria for the majority of regulated analytes, demonstrated robust calibration and CCV performance, and reduced manual variability. Some compound‑ and matrix‑specific recovery issues remain and warrant targeted optimization prior to routine regulatory reporting.

Reference

1. Method 1633A Analysis of Per- and Polyfluoroalkyl Substances (PFAS) in Aqueous, Solid, Biosolids, and Tissue Samples by LC-MS/MS.