Sensitive Measurement of Per- and Polyfluoroalkyl Substances in Drinking Water

Significance of the topic

The quantification of per- and polyfluoroalkyl substances (PFAS) in drinking water is of high regulatory and public-health importance because regulatory parametric values and lower reporting limits require highly sensitive, reproducible methods. Traditional sample preparation workflows for low ng/L detection typically rely on large-volume solid-phase extraction (SPE) followed by solvent evaporation and reconstitution to achieve high concentration factors. These steps are time-consuming, introduce contamination and analyte loss risk, and increase environmental footprint. The application of Feed Injection for high-volume LC injection offers a route to maintain sensitivity while reducing sample-prep time and associated risks.

Objectives and overview of the study

The study compared two workflows for PFAS analysis by triple-quadrupole LC/MS: (A) classic low-volume (2 µL) flow-through injection after SPE concentration and evaporation/reconstitution (concentration factor ~250) and (B) high-volume (20 µL) Agilent Feed Injection of SPE eluate (concentration factor ~25) without evaporation/reconstitution. Main goals were to evaluate analytical performance (sensitivity, accuracy, precision), method feasibility for 100% organic SPE eluates, blank behavior, and time savings.

Methodology and sample preparation

• Sample enrichment: WAX SPE (Agilent Bond Elut PFAS WAX, 200 mg) of 100 mL water samples (tap water and ultrapure water) to give concentrated eluates. Several blank types were processed: cartridge blank, double blank, and solvent blank.
• Spiking levels: low spike 0.5 ng/L (threefold below typical LOQ target and tenfold below the EU parametric value per compound) and high spike 5 ng/L (at the parametric value).
• SPE elution and extract handling: elution with 2 × 2 mL 2% ammonium hydroxide in MeOH; for the flow-through path the combined eluate was evaporated to dryness and reconstituted to achieve a larger concentration factor; for Feed Injection the eluate was diluted 1:10 to match 20 µL injection calibration.
• Calibration: nine-point calibration for both injection modes. For 2 µL injection standards covered 20–20,000 ng/L (corresponding to measured sample concentrations of 0.08–80 ng/L with 250× factor). For 20 µL Feed Injection standards covered 2–2,000 ng/L (same sample concentration range, 25× factor). All calibration curves had R2 ≥ 0.998.

Instrumental setup (Used instrumentation)

• LC: Agilent 1290 Infinity III Binary Pump and 1290 Infinity III Hybrid Multisampler with Feed Injection capability; InfinityLab PFAS HPLC conversion kit and an Agilent InfinityLab PFC delay column to reduce PFAS background.
• Analytical column: Agilent InfinityLab Poroshell 120 Aq-C18, 3.0 × 100 mm, 2.7 µm.
• MS: Agilent 6495D triple-quadrupole LC/TQ with Agilent Jet Stream electrospray in negative ion mode; data acquired in dynamic MRM (dMRM) using Agilent PFAS MRM database settings (retention times adjusted).
• Software: Agilent MassHunter for acquisition, qualitative and quantitative analysis.

Main results and discussion

• Sensitivity and calibration: Both injection strategies produced linear calibration with R2 ≥ 0.998 across the set ranges; comparable measured sample concentrations were obtained for the same target sample concentration range (0.08–80 ng/L).
• Accuracy and precision: For all PFAS, Feed Injection (20 µL) gave recoveries within 80–120% and calculated concentration %RSDs below 10% across compounds. Overall, no statistically significant differences in measured concentrations between 20 µL Feed Injection (no evaporation) and 2 µL flow-through after evaporation/reconstitution were observed.
• Low-level performance: At the low spike (0.5 ng/L) and at the parametric 5 ng/L spike, reproducibility (area RSD) was slightly improved for certain short-chain analytes (e.g., PFBA, PFDS, PFDoS) with the 20 µL Feed Injection.
• Blanks and background: Cartridge blanks showed trace PFAS—low levels of PFBS, PFHxA and PFBA—requiring blank subtraction; other PFAS were negligible in blanks. In tap water samples several PFAS were found, mostly < 0.2 ng/L, except PFBA and PFBS at ~0.8–1.0 ng/L.
• Practical implication: Feed Injection successfully accepts SPE eluates in high organic strength (up to 100% MeOH) by online dilution into the mobile phase via a metering valve (adaptive feed speed at 10% of flow), avoiding adverse solvent effects on chromatographic peak shape.

Benefits and practical applications of the method

• Time savings: Eliminating the evaporation/reconstitution step reduced sample-preparation time by about 65% (total prep time approximately 5.5 h for the classic workflow vs 2.5 h using Feed Injection for the same batch scenario).
• Lower contamination and analyte loss risk: Fewer handling steps reduce opportunities for contamination and losses associated with concentration by evaporation.
• Greener workflow: Reduced solvent usage and fewer consumable steps lower environmental impact and operational cost.
• Operational flexibility: Feed Injection allows injection of larger volumes of high-organic eluate, enabling simpler workflows for routine PFAS monitoring in drinking water and potentially higher throughput laboratories.

Limitations and considerations

• Concentration factor trade-off: Feed Injection requires a lower concentration factor (25×) compared with evaporation/reconstitution (250×); therefore, initial SPE enrichment volume or sensitivity needs must be considered for extremely low target limits.
• Blank control: Laboratory PFAS background (cartridge and system) must be characterized and subtracted; suitable PFAS-free consumables and delay columns remain important.
• Current procedure dependency: Although Feed Injection eliminates evaporation, the study still used a reconstitution/dilution step to match calibration; removing this step entirely would require optimization of SPE elution volumes and concentration strategy (for example, enrichment of 25 mL sample to ~1 mL final eluate).

Future trends and opportunities for application

• Wider adoption of online-dilution injection strategies (Feed Injection) will likely increase as laboratories aim to balance sensitivity with throughput and greener workflows.
• Method evolution could focus on reducing initial sample volumes (e.g., 25 mL) coupled with optimized elution to enable direct high-volume injection without any reconstitution, further decreasing solvent consumption and turnaround time.
• Extension to broader PFAS panels and non-target screening via high-resolution MS combined with high-volume injection could expand routine environmental monitoring capabilities.
• Standardization and inclusion of Feed Injection approaches in official methods and guidelines would promote inter-laboratory comparability and reduce method-related variability for low-ng/L PFAS analysis.

Conclusion

This application study demonstrates that Agilent Feed Injection with the 1290 Infinity III Hybrid Multisampler allows reliable 20 µL injections of SPE eluates in high organic solvent without detrimental chromatographic effects, producing comparable accuracy and precision to a classic workflow employing evaporation/reconstitution and a 2 µL flow-through injection. Key advantages are substantial reduction in sample-preparation time (≈65%), lower risk of contamination and analyte loss, and a more sustainable workflow. For routine PFAS monitoring in drinking water at low ng/L levels, Feed Injection is an effective, time-saving alternative that preserves analytical performance while simplifying laboratory operations.

Reference list

1. Zarębska M.; Bajkacz S. Poly- and Perfluoroalkyl Substances (PFAS) - Recent Advances in the Aquatic Environment Analysis. Trends in Analytical Chemistry, 2023.
2. Per- and Polyfluoroalkyl Substances (PFAS): Final PFAS National Primary Drinking Water Regulation. U.S. Environmental Protection Agency (EPA), accessed 2026-05-28.
3. Kamuf M.; Wälz M.; Borowiak A. Reduce PFAS Background with the Agilent PFC-Free HPLC Conversion Kit. Agilent Technologies technical overview, publication 5994-2291EN, 2024.
4. Giardina M. Determination of Per and Polyfluoroalkyl Substances in Drinking Water Using Agilent Bond Elut PFAS WAX SPE and LC/MS/MS. Agilent Technologies application note, publication 5994-4960EN, 2025.
5. Bende J. L.; et al. Greenness assessment of 174 CEN, ISO, and pharmacopeia standard methods and their sub-methods used for environmental, food, trace elemental and pharmaceutical analysis. Advances in Sample Preparation, volume 14, 2025, article 100180.