Addressing Matrix Complexity in PFAS Analysis: Chromatographic Strategies for Soil, Serum, and Food

Per- and polyfluoroalkyl substances(PFAS) are anthropogenic fluorinated compounds characterized by multiple carbon–fluorine bonds that confer exceptional chemical and thermal stability.¹ They include perfluoroalkyl substances, with fully fluorinated carbon backbones, and polyfluoroalkyl substances, which retain at least one non-fluorinated carbon. Major subclasses such as perfluoroalkyl sulfonic acids (PFSAs) and perfluoroalkyl carboxylic acids (PFCAs) differ in head-group chemistry and chromatographic behavior.¹ Widely used in industrial and consumer products, their environmental persistence leads to accumulation in soil, groundwater, wildlife, and ultimately in the human body, particularly in blood serum.² Increasing evidence of adverse health effects has prompted global regulatory action, driving demand for robust, highly sensitive analytical methods capable of quantifying PFAS at trace levels in complex matrices.^3,4

However, the analysis of PFAS is analytically demanding for several reasons. PFAS are often present at very low concentrations, with regulatory limits requiring detection at trace levels in the parts-per-trillion (ppt) range.^5,6 Methods must therefore provide high sensitivity and low limits of detection while maintaining quantitative reliability. PFAS are encountered in highly complex matrices. Soil extracts contain humic substances and diverse organic contaminants; human serum contains proteins, phospholipids, and endogenous metabolites; and food samples such as meat or eggs contain lipids and bile acids. These coextracted components can cause ion suppression, chromatographic interferences, or coelution.^7,8 In addition, PFAS encompass a wide polarity range. Short-chain PFCAs and PFSAs are highly polar and tend to elute early under reversed-phase conditions, sometimes with insufficient retention. In contrast, long-chain PFAS are strongly retained and may require high organic solvent content for elution. Achieving adequate retention and resolution of both groups within a single chromatographic run remains a central method development challenge.⁹ Isomerism further complicates PFAS analysis. Many PFAS occur as mixtures of linear and branched isomers or positional variants. These isomers may differ toxicologically and can exhibit distinct chromatographic and fragmentation behavior.¹⁰ Full, or at least partial, chromatographic resolution enhances the interpretability of MS/MS spectra and improves structural elucidation in both targeted and non-targeted workflows. Finally, PFAS analysis is uniquely susceptible to background contamination originating from laboratory materials, solvent lines, and system components.¹¹

Liquid chromatography coupled to mass spectrometry (LC–MS) has become the primary technique for PFAS analysis due to its sensitivity and selectivity.^4,11 Negative electrospray ionization is particularly suitable for acidic PFAS such as PFCAs and PFSAs.¹¹ Two complementary analytical strategies are commonly employed. Targeted approaches rely on predefined analyte lists and selected reaction monitoring (SRM) transitions for quantification. These methods are essential for regulatory compliance and routine monitoring. In contrast, non-targeted and suspect screening approaches use high-resolution mass spectrometry combined with mass-defect analysis and homologous series detection to identify unknown or emerging PFAS.^11,12 Both strategies benefit significantly from optimized chromatographic separation. However, mass spectrometric sensitivity alone does not resolve the underlying chromatographic challenges. Method performance depends fundamentally on stationary phase selectivity, mobile phase composition, gradient design, sample preparation strategy, and contamination control. In the approaches described here, these requirements are addressed using a consistent chromatographic platform based on a C18 stationary phase. The hybrid silica material provides high robustness and surface area, enabling stable retention and resolution of structurally diverse PFAS across different matrices. The following sections illustrate how these considerations are applied in practice through representative applications in soil, human serum, and food matrices.

Analysis of PFAS in Soil: A Non-Targeted Perspective

Soil acts both as a repository and a secondary source of PFAS contamination. Especially at industrial or firefighting sites, soil may contain mixtures of legacy and emerging PFAS, many of which lack analytical standards. The complexity of soil extracts poses substantial analytical challenges because of coextracted organic matter and unknown PFAS species.

A non-targeted LC–MS approach was used to demonstrate reproducible total ion chromatograms across triplicate measurements of contaminated soil extracts, despite the matrix complexity (Figure 1). To accommodate the wide polarity range of PFAS species, gradient conditions were optimized to ensure sufficient retention of short-chain compounds while allowing elution of longer-chain homologues within a single analytical run. Careful control of mobile phase composition was essential to maintain stable peak shapes and minimize ion suppression from coeluting matrix components.

Kendrick mass defect analysis and evaluation of CF₂ repeating units were used to identify homologous PFAS clusters, enabling recognition of compound series even in the absence of reference standards.

Chromatographic selectivity played a critical role in resolving PFAS clusters and partially separating structural isomers. This selectivity is provided by the used hybrid silica C18 stationary phase, whose robust particle and balanced hydrophobic interactions help maintain stable retention even in the presence of coextracted soil components. Extracted ion chromatograms of different PFAS groups showed effective separation across the chromatographic run, thereby reducing coelution and improving the reliability of subsequent tandem mass spectrometry (MS/MS) acquisition (Figure 2). Notably, isomers of SF₅(CF₂)₉SO₃^- were separated sufficiently to allow comparison of their fragmentation patterns. Differences in fragment ion distributions suggested structural variation between branched and linear isomers, probably reflecting differences in bond cleavage along the fluorinated carbon chain. Such partial resolution is particularly valuable in non-targeted workflows, where structural interpretation depends on acquiring clean, interpretable MS/MS spectra avoiding composite fragmentation from overlapping peaks.

To support trace-level detection, acquisition parameters were optimized to balance mass resolution, scan speed, and sensitivity. Consistent retention times and peak areas across replicate injections confirmed method robustness in the presence of matrix-derived components. The method achieved high sensitivity and accurate detection of PFAS clusters at trace levels while maintaining reproducibility in complex, real-life environmental matrices taken in western Germany. This illustrates how chromatographic resolution, combined with structured data evaluation strategies, enhances both qualitative identification and semi-quantitative assessment in environmental PFAS screening.

Analysis of PFAS in Human Serum: A Targeted Biomonitoring Approach

Human serum is a central matrix for biomonitoring and exposure assessment, as PFAS accumulate in blood over time.¹³ However, serum analysis presents challenges, including strong protein binding, adsorption to surfaces, and carry-over risks, necessitating sensitive and reliable quantification across multiple compound classes.

A targeted LC–MS/MS method was developed for the quantification of 28 PFAS in human serum. The method integrated automated solid-phase extraction (SPE) with an online column-switching configuration to allow large injection volumes and matrix cleanup prior to analytical separation (Figure 3). The SPE step enabled preconcentration of analytes and removal of bulk matrix components before transfer to the analytical column, supporting trace-level detection. The used hybrid silica-based column provided high mechanical and chemical robustness, maintaining consistent retention and peak shape in complex serum extracts. Its large surface area increased interaction capacity, improving tolerance to higher matrix load while helping reduce adsorption and carryover during PFAS analysis.

Chromatographic separation was optimized to accommodate both neutral and ionizable matrices by adjusting eluent conditions during the gradient.
The method achieved separation of 28 PFAS within a single run, including carboxylates, sulfonates, sulfonamides, and fluorotelomer derivatives (Figure 4). Early elution of highly polar short-chain PFAS was controlled through gradient programming, while long-chain PFAS were eluted at higher organic solvent composition to avoid excessive retention.

To mitigate hardware-derived contamination, anion-exchange scrubber columns were placed downstream of the LC pumps to remove background PFAS signals originating from the solvent delivery system. This configuration minimized system-related interference in trace-level analysis.

The method demonstrated reproducible chromatographic performance and stable retention behavior, supporting its applicability for quantitative biomonitoring of PFAS in human serum.

Analysis of PFOS in Food: Resolving Matrix Interference

Food analysis presents a distinct analytical problem, as PFAS must be determined at trace levels in matrices rich in lipids and other endogenous compounds. The coextraction of matrix constituents can influence chromatographic behavior and ionization efficiency, increasing the likelihood of interference during LC–MS analysis.⁸ In the case of perfluorooctanesulfonic acids (PFOS), taurocholic acid isomers can produce the same transition (m/z = 499 → 80) under negative ESI conditions. Without chromatographic separation, these compounds are indistinguishable in tandem mass spectrometry (MS/MS), as their precursor and product ions overlap completely, eliminating the selectivity typically expected from MS/MS spectrometry and increasing the risk of false-positive identification.

The same stationary phase mentioned previously was used, but it was packed in bio-inert coated column hardware to prevent unwanted interactions between analytes and metal ions, thereby preserving analyte sensitivity. Under methanol-based mobile phase conditions, PFOS coeluted with taurocholic acid isomers, preventing selective detection and compromising quantitative accuracy. Substituting acetonitrile as the organic solvent resulted in complete chromatographic separation of PFOS from these endogenous interferences. This demonstrates how solvent choice can dramatically influence selectivity and resolution by altering analyte–stationary phase interactions, hydrogen-bonding behavior, and overall elution strength. The optimized acetonitrile-based gradient enabled separation of a 21-component PFAS mixture, covering both carboxylate- and sulfonate-type species across a broad hydrophobicity range. Recovery studies in meat, fish, vegetable, and egg matrices showed acceptable accuracy and precision at low spike levels, confirming method robustness despite varying matrix composition. Importantly, chromatograms of unspiked samples confirmed that taurocholic acid-related interferences eluted earlier, while PFOS appeared later in the run, ensuring analytical specificity.

This case underscores the importance of chromatographic resolution in preventing false positives and ensuring reliable quantification in complex food matrices.

Conclusion

PFAS analysis represents a convergence of analytical challenges, including trace-level detection, broad polarity range, structural isomerism, matrix complexity, and background contamination. Chromatographic selectivity, along with matrix tolerance and sensitivity, emerges as a central determinant of analytical success across soil, serum, and food applications. Carefully optimized gradients, solvent systems, sample preparation strategies, and contamination control measures enable both targeted quantification and non-targeted structural elucidation.

As regulatory requirements expand and attention shifts toward emerging PFAS and isomer-specific evaluation, robust chromatographic and LC–MS methodologies will remain essential. The approaches discussed here illustrate how scientific method development provides effective solutions to the multifaceted challenges of PFAS analysis.

References

1. Dobrzyńska, E.; Wasilewski, P.; Pośniak, M. Per- and Polyfluoroalkyl Substances (PFASs): A Comprehensive Review of Environmental Distribution, Health Impacts, and Regulatory Landscape. Appl Sci 2025, 15 (22), 11884. DOI: 10.3390/app152211884
2. Dehghani, M. H.; Aghaei, M.; Bashardoust, P.; et al. An Insight into the Environmental and Human Health Impacts of Per- and Polyfluoroalkyl Substances (PFAS): Exploring Exposure Pathways and Their Implications. Environ Sci Eur 2025, 37, 81. DOI: 10.1186/s12302-025-01122-9
3. Yu, R.-S.; Yu, H.-C.; Yang, Y.-F.; Singh, S. A Global Overview of Per- and Polyfluoroalkyl Substance Regulatory Strategies and Their Environmental Impact. Toxics 2025, 13 (4), 251. DOI: 10.3390/toxics13040251
4. Rehman, A. U.; Crimi, M.; Andreescu, S. Current and Emerging Analytical Techniques for the Determination of Per- and Polyfluoroalkyl Substances (PFAS) in Environmental Samples. Trends Environ Anal Chem 2023, 37, e00198. DOI: 10.1016/j.teac.2023.e00198
5. European Parliament; Council of the European Union. Directive (EU) 2020/2184 on the Quality of Water Intended for Human Consumption (recast). Off J Eur Union 2020, L435, 1–62.
6. U.S. Environmental Protection Agency. PFAS national primary drinking water regulation. Fed Regist 2024, 89, 32532–32724
7. Williams, M. L.; Olomukoro, A. A.; Emmons, R. V.; Godage, N. H.; Gionfriddo, E. Matrix Effects Demystified: Strategies for Resolving Challenges in Analytical Separations of Complex Samples. J Sep Sci 2023, 46 (23), e2300571. DOI: 10.1002/jssc.202300571
8. Takayama, T.; Shingu, A.; Kato, S.; et al. Countermeasure for Interfered Monitoring Ion of Perfluorooctanesulfonic Acid (PFOS) From Intrinsic Food Samples Based on LC-MS/MS Analysis of Per- and Polyfluoroalkyl Substances. J Food Compos Anal 2024, 133, 106436. DOI: 10.1016/j.jfca.2024.106436
9. Björnsdotter, M. K.; Yeung, L. W. Y.; Kärrman, A.; Jogsten I. E. Challenges in the Analytical Determination of Ultra-Short-Chain Perfluoroalkyl Acids and Implications for Environmental and Human Health. Anal Bioanal Chem 2020, 412, 4785–4796. DOI: 10.1007/s00216-020-02692-8
10. Londhe, K.; Lee, C.-S.; McDonough, C. A.; Venkatesan, A. K. The Need for Testing Isomer Profiles of Perfluoroalkyl Substances to Evaluate Treatment Processes. Environ Sci Technol 2022, 56 (22), 15207–15219. DOI: 10.1021/acs.est.2c05518
11. 11. Shen, Y.; Wang, L.; Ding, Y.; Liu, S.; Li, Y.; Zhou, Z.; Liang, Y. Trends in the analysis and exploration of per- and polyfluoroalkyl substances (PFAS) in environmental Matrices: a review. Crit Rev Anal Chem 2024, 54 (8), 3171–3195. DOI: 10.1080/10408347.2023.2231535.
12. Jeannot, C.; Macorps, N.; Le Bizec, B.; Parinet, J.; Dervilly, G. Advancing PFASs Monitoring in Food: From Targeted SPE-LC-MS/MS to Non-Targeted QuEChERS-LC-HRMS Approaches. Food Chem X 2025, 29, 102674. DOI: 10.1016/j.fochx.2025.102674
13. Nakayama, S. F.; Isobe, T.; Iwai-Shimada, M.; et al. Poly- and Perfluoroalkyl Mubstances in Maternal Serum: Method Development and Application in Pilot Study of the Japan Environment and Children’s Study. J Chromatogr A 2020, 1618, 460933. DOI: 10.1016/j.chroma.2020.460933