Evaluation of extraction methodologies for PFAS analysis in mascara: a comparative study of SPME and automated μSPE

Significance of the topic

The widespread and largely unregulated use of per- and polyfluoroalkyl substances (PFAS) in consumer products raises direct human exposure concerns. Cosmetics, including mascara, are complex formulations where PFAS are used for emulsification, surfactancy and film‑forming properties; however, analysis is challenging due to matrix complexity, low concentrations, and risk of contamination during sample preparation. Developing robust, selective and sustainable extraction protocols for PFAS in cosmetics is essential for accurate monitoring, exposure assessment and regulatory decisions.

Objectives and study overview

This study compared two microextraction approaches—solid‑phase microextraction (SPME, including thin‑film SPME) and automated micro solid‑phase extraction (µSPE)—for quantitative analysis of eight anionic PFAS (short and long chain, including PFBS, PFHxA, PFHpA, PFOA, PFOS, 4:2 FTS, 8:2 FTS and 6:2 diPAP) in mascara products. Key goals were to (i) optimize dispersive medium and extraction/elution conditions for each technique, (ii) assess recoveries and matrix effects across multiple commercial mascaras (waterproof and non‑waterproof), (iii) validate methods with LC–MS/MS quantification, and (iv) compare greenness, throughput and practical suitability for routine analysis.

Methodology

Summary of analytical workflow and optimization strategy:

• Samples: Six different mascaras used for method development and nine commercial mascaras for real sample analysis (various countries of origin).
• Analytes: Eight anionic PFAS spanning carboxylates, sulfonates, fluorotelomer sulfonates and polyfluoroalkyl phosphate diesters (6:2 diPAP).
• SPME: TF‑SPME and coated fibers with HLB‑WAX/PAN binder; applied previously optimized desorption (80:20 MeOH:H2O + 2% ammonium formate), desorption volume 250 µL, desorption time 20 min; extraction times profiled (10–120 min) and 60 min chosen as operational compromise.
• µSPE: Automated PAL System with µSPE WAX cartridges; cartridges conditioned with water, sample loaded from MeOH dispersions, elution optimized (100% MeOH + 0.5% ammonium formate), elution volume 400 µL, elution rate 1 µL/s.
• Dispersive media and pre‑treatment: Systematic evaluation of CH3OH:H2O ratios and sonication effects to maximize recovery and avoid degradation (notably for 6:2 diPAP).
• Validation: Matrix‑matched calibration using a chosen low‑matrix mascara (P4W); linearity 0.025–25 ng/g (method dependent), assessment of LOD/LOQ (S/N ≥3 and accuracy/RSD criteria), accuracy/precision over 7 days.

Instrumentation used

• Liquid chromatography tandem mass spectrometry (LC–MS/MS) for targeted PFAS quantification.
• SPME devices and thin‑film SPME coatings (HLB‑WAX on PAN binder).
• PAL System autosampler with PAL System µSPE WAX cartridges (CTC Analytics AG) for automated µSPE.
• Vortex and agitator platforms for sample mixing; filtration with Whatman filter paper for µSPE sample clarifications; 0.22 µm nylon syringe filters for final cleanup.

Main results and discussion

Key experimental findings and interpretations:

• Sorbent chemistry: WAX sorbent outperformed C18 for µSPE when targeting a broad set of anionic PFAS. C18 exhibited breakthrough of shorter‑chain PFAS; WAX retained analytes more comprehensively but required optimized elution for hydrophobic species.
• Elution optimization (µSPE): 100% MeOH containing 0.5% ammonium formate provided significantly higher recoveries than 80:20 MeOH:H2O. Increasing ammonium formate beyond 0.5% reduced elution efficiency. A 400 µL elution volume minimized carryover (<2%) and maximized recovery; an elution rate of 1 µL/s was selected for robustness.
• Dispersive medium and sonication effects: For µSPE, 100% MeOH as dispersive medium maximized recovery for hydrophobic PFAS (PFOS, 8:2 FTS, 6:2 diPAP). For SPME, 100% H2O was optimal (hydrophilic PFAS extraction favored aqueous medium). Sonication decreased recoveries for 6:2 diPAP—likely due to hydrolysis/transformation—so it was avoided in optimized protocols.
• SPME kinetics: Equilibrium in mascara dispersions was approached at ~90 min, but 60 min extraction time was adopted to balance throughput and recovery.
• Matrix effects: SPME produced minimal matrix effects overall due to the biocompatible PAN binder limiting co‑extraction of interferents. µSPE showed larger matrix effects for specific analytes (notably signal enhancement of 6:2 diPAP), and washing steps prior to elution did not reliably mitigate matrix interferences unless internal standard correction was used.
• Method performance and LOQs: Both methods exhibited wide linear ranges and good reproducibility. SPME achieved lower LOQs for the most hydrophilic analytes (e.g., PFBS LOQ = 0.025 ng/g). µSPE attained superior LOQs for hydrophobic PFAS: PFOS 0.2 ng/g, 8:2 FTS 0.1 ng/g, and 6:2 diPAP 0.5 ng/g, reflecting the advantage of performing extraction in 100% MeOH.
• Real samples: Using both methods, PFOA was detected in six of nine mascaras (highest by SPME: 3.04 ± 0.33 ng/g in P2W). 6:2 diPAP was quantified in four of nine mascaras with concentrations by µSPE ranging from 1.26 to 3.48 ng/g (comparable SPME detections where above LOQ). Differences between methods stress the need for matrix‑matched calibration or standard addition when exact quantification across diverse formulations is required.

Benefits and practical applications of the methods

• SPME advantages: Higher greenness score, fewer consumables (reusable devices), minimal sample handling (direct immersion into dispersive media), reduced matrix effects, high sample throughput (40–50 samples per hour on an agitator), and integration of extraction/preconcentration into a single step—making SPME attractive for screening and routine monitoring where hydrophilic PFAS predominate.
• µSPE advantages: Automation and improved sensitivity for hydrophobic PFAS due to extraction from 100% MeOH, delivering lower LOQs for compounds like PFOS and 6:2 diPAP; suitable where automation and robustness for hydrophobic targets are priorities.
• Common practical notes: Both methods minimize solvent volumes (<10 mL) and avoid laborious evaporation/reconstitution steps. For accurate quantification in diverse cosmetic matrices, matrix‑matched calibration or standard addition is recommended.

Greenness and practicality assessment

Objective greenness metrics favored SPME: AGREEprep scores — SPME 0.51 vs µSPE 0.44; BAGI scores — SPME 75 vs µSPE 70. Main drivers were throughput, reusability and lower consumable waste for SPME, while µSPE’s single‑use cartridges and slower sample processing reduced its greenness rating despite automation advantages.

Future trends and potential applications

• Method extension: Apply optimized SPME and µSPE workflows to a wider range of cosmetic categories (foundations, powders, lipsticks, nail products) and other complex personal‑care matrices.
• Target expansion: Include neutral and polymeric fluorinated species (and fluorotelomer precursors) using suspect and non‑target workflows coupled to high‑resolution mass spectrometry to better characterize total extractable organic fluorine and unidentified PFAS fractions.
• Standardization: Development of interlaboratory validated protocols and certified reference materials for cosmetic matrices to enable harmonized monitoring and regulatory compliance.
• Green improvements: Design of more durable/biodegradable extraction phases and reduced‑waste automated platforms to further improve sustainability.
• Exposure and risk assessment: Integrate analytical data with dermal bioaccessibility and toxicokinetic studies to refine human exposure estimates from cosmetic use.

Conclusion

This comparative study shows that both SPME and automated µSPE can be successfully optimized for determination of anionic PFAS in complex mascara matrices. SPME offers higher throughput, lower matrix effects and superior greenness, particularly suited for hydrophilic PFAS and routine screening. Automated µSPE provides greater sensitivity for hydrophobic PFAS through extraction in 100% MeOH and is appropriate when low LOQs for those targets are needed and automation is desirable. Accurate quantification across diverse cosmetic formulations requires matrix‑matched calibration or standard addition. Continued monitoring and methodological refinement are warranted given the detection of PFAS (notably PFOA and 6:2 diPAP) in commercial mascaras and the public‑health implications of direct human exposure.

References

• Schultes L, Vestergren R, Volkova K, Westberg E, Jacobson T, Benskin JP. Per‑ and polyfluoroalkyl substances and fluorine mass balance in cosmetic products from the Swedish market: Implications for environmental emissions and human exposure. Environ Sci Process Impacts. 2018;20:1680–1690.
• Meng P, Deng S, Du Z, Wang B, Huang J, Wang Y, Yu G, Xing B. Effect of hydro‑oleophobic perfluorocarbon chain on interfacial behavior and mechanism of perfluorooctane sulfonate in oil‑water mixture. Sci Rep. 2017;7:44694.
• Céline C, Catherine B, Romane C, Laurence C. Per‑ and polyfluoroalkyls used as cosmetic ingredients - Qualitative study of 765 cosmetic products. Food Chem Toxicol. 2024;187:114625.
• Zhang DQ, Zhang WL, Liang YN. Adsorption of perfluoroalkyl and polyfluoroalkyl substances (PFASs) from aqueous solution — A review. Sci Total Environ. 2019;694:133606.
• Brinch A, Jensen AA, Christensen F. Risk assessment of fluorinated substances in cosmetic products. Danish Environmental Protection Agency. 2018;1–116.
• Gagliano E, Sgroi M, Falciglia PP, Vagliasindi FGA, Roccaro P. Removal of poly‑ and perfluoroalkyl substances (PFAS) from water by adsorption: Role of PFAS chain length, effect of organic matter and challenges in adsorbent regeneration. Water Res. 2020;171:115381.
• Pütz KW, Namazkar S, Plassmann M, Benskin JP. Are cosmetics a significant source of PFAS in Europe? product inventories, chemical characterization and emission estimates. Environ Sci Process Impacts. 2022;24:1697–1707.
• Whitehead HD, Venier M, Wu Y, et al. Fluorinated Compounds in North American Cosmetics. Environ Sci Technol Lett. 2021;8:538–544.
• Namazkar S, Ragnarsdottir O, Josefsson A, et al. Characterization and dermal bioaccessibility of residual‑ and listed PFAS ingredients in cosmetic products. Environ Sci Process Impacts. 2024;26:259–268.
• Gaines LGT. Historical and current usage of per‑ and polyfluoroalkyl substances (PFAS): A literature review. Am J Ind Med. 2023;66:353–378.
• Ragnarsdóttir O, Abdallah MAE, Harrad S. Dermal bioaccessibility of perfluoroalkyl substances from household dust; influence of topically applied cosmetics. Environ Res. 2023;238:117093.
• Harris KJ, Munoz G, Woo V, Sauvé S, Rand AA. Targeted and Suspect Screening of Per‑ and Polyfluoroalkyl Substances in Cosmetics and Personal Care Products. Environ Sci Technol. 2022;56:14594–14604.
• Olomukoro AA, DeRosa C, Gionfriddo E. Investigation of the adsorption/desorption mechanism of perfluoroalkyl substances on HLB‑WAX extraction phases for microextraction. Anal Chim Acta. 2023;1260:341206.
• Olomukoro AA, Emmons RV, Godage NH, et al. Ion exchange solid phase microextraction coupled to liquid chromatography/laminar flow tandem mass spectrometry for the determination of perfluoroalkyl substances in water samples. J Chromatogr A. 2021;1651:462335.
• Wu S, Zhu L, Ye Q, et al. Mechanisms for the structural dependent transformation of 6:2 and 8:2 polyfluoroalkyl phosphate diesters in wheat (Triticum aestivum L.). J Hazard Mater. 2023;454:131536.
• Zweigle J, Bugsel B, Schmitt M, Zwiener C. Electrochemical oxidation of 6:2 polyfluoroalkyl phosphate diester — simulation of transformation pathways and reaction kinetics with hydroxyl radicals. Environ Sci Technol. 2021;55:11070–11079.
• Zhu Y, Jia Y, Liu M, et al. Mechanisms for tissue‑specific accumulation and phase I/II transformation of 6:2 fluorotelomer phosphate diester in earthworm (M. guillelmi). Environ Int. 2021;151:106451.
• Cárdenas‑Soracá DM, Singh V, Nazdrajić E, et al. Development of thin‑film solid‑phase microextraction coating and method for determination of artificial sweeteners in surface waters. Talanta. 2020;211:120714.
• Olomukoro AA, Xie R, Paucar FXF, et al. Characterization of a mixed mode fluorocarbon/weak anion exchange sorbent for the separation of perfluoroalkyl substances. J Sep Sci. 2024;47:413.
• Keawmanee S, Piyaviriyakul P, Boontanon N, et al. Concentration and health risk assessment of per‑ and polyfluoroalkyl substances in cosmetic and personal care products. J Environ Sci Health B. 2024.
• Wojnowski W, Tobiszewski M, Pena‑Pereira F, Psillakis E. AGREEprep – Analytical greenness metric for sample preparation. TrAC Trends Anal Chem. 2022;149:116553.
• Armenta S, Garrigues S, Esteve‑Turrillas FA, de la Guardia M. Green extraction techniques in green analytical chemistry. TrAC Trends Anal Chem. 2019;116:248–253.
• Manousi N, Wojnowski W, Płotka‑Wasylka J, Samanidou V. Blue applicability grade index (BAGI) and software: a new tool for the evaluation of method practicality. Green Chem. 2023;25:7598–7604.