Unidentified fluorine and TFA dominate the fluorine mass balance in European drinking water

Technical Summary: Unidentified fluorine and TFA dominate the fluorine mass balance in European drinking water

Problem Statement
Human exposure to per- and polyfluoroalkyl substances (PFAS) via drinking water is a significant public health concern, yet current monitoring frameworks are likely underestimating the true extent of contamination. Conventional targeted analytical methods identify only a fraction of the >10,000 PFAS in commercial use, failing to capture emerging substitutes and unknown transformation products. Furthermore, regulatory limits in Europe have historically focused on specific legacy compounds (e.g., PFOA, PFOS) or limited groups (e.g., Σ4PFAS), potentially overlooking the cumulative burden of ultra-short-chain and unidentified fluorinated substances. The revised EU Drinking Water Directive (Directive (EU) 2020/2184) introduces a class-based limit of 500 ng L⁻¹ for "PFAS Total," defined as the totality of PFAS, but systematic surveys characterizing the composition of this total in European tap water have been lacking.

Methodology
The study conducted a comprehensive fluorine mass balance (FMB) analysis on tap water samples collected from 34 major European cities across 18 countries, representing approximately 37 million residents (roughly 5% of the European population). Samples were collected between January 2022 and January 2023.

The analytical approach integrated targeted quantification with non-targeted fluorine mass balance:

1. Targeted Analysis: 40 fluorinated substances were monitored, including 14 perfluoroalkyl carboxylic acids (PFCAs), 10 perfluoroalkane sulfonic acids (PFSAs), fluorotelomer compounds, fluorinated alternatives (e.g., HFPO-DA), and inorganic fluorine (IF) species (BF₄⁻ and PF₆⁻). Quantification was performed using UHPLC-MS/MS with dual chromatographic methods to capture both short/ultra-short and long-chain compounds.
2. Fluorine Mass Balance (FMB): Extractable Fluorine (EF) was measured via combustion ion chromatography (CIC) on samples processed through a polarity-based solid-phase extraction (SPE) fractionation strategy. This separated ultra-short-chain compounds (Fraction I) from longer-chain compounds (Fraction II).
3. Mass Balance Calculation: The study calculated the Unidentified Extractable Fluorine (UEF) by subtracting the fluorine equivalents of all identified targets (Σ38PFAS + IF) from the total EF measured by CIC.
4. Risk Assessment: Two metrics were applied:
   • Concentration-based: Comparison against the EU regulatory limit of 500 ng L⁻¹ for PFAS Total and the 100 ng L⁻¹ limit for the Sum of PFAS.
   • Potency-weighted: A Risk Quotient based on potency equivalents (RQpeq) was calculated using Relative Potency Factors (RPFs) referenced to PFOA, comparing mixture toxicity against the European Food Safety Authority (EFSA) health-based guidance value.

Key Results

• Dominance of TFA and Unidentified Fluorine: Trifluoroacetic acid (TFA) was the most abundant PFAS detected (median 45.7 ng L⁻¹; detection frequency 95%), contributing significantly to the total fluorine load. However, the most striking finding was that >60% of the extractable fluorine (EF) in most sampled cities remained unidentified (UEF). The median UEF concentration was 98.9 ng F L⁻¹, compared to a median of 42.4 ng F L⁻¹ for all identified targets combined.
• Regulatory Exceedances: When recovery-corrected data were converted to the "PFAS Total" metric (defined as the sum of identified PFAS plus a proxy for unidentified species), concentrations ranged from 42.0 to 4020 ng L⁻¹. Ten cities exceeded the EU regulatory threshold of 500 ng L⁻¹, affecting approximately 27 million inhabitants. Notably, in cities like Antwerp, TFA and UEF alone drove exceedances, while targeted legacy PFAS (PFOA, PFOS, PFHxS) remained at low levels.
• Geographic and Compositional Clustering: Hierarchical clustering revealed four distinct contamination profiles. While some clusters were dominated by legacy compounds or specific industrial signatures (e.g., 6:2 FTS in Lyon), many high-burden sites (e.g., Antwerp, Leipzig) were characterized by high levels of UEF and inorganic fluorine (PF₆⁻, BF₄⁻), suggesting diverse anthropogenic sources including industrial discharges and electrolyte usage.
• Decoupling of Concentration and Toxicity: A critical finding was the lack of correlation between total fluorine load and toxicological risk. Cities with the highest EF (e.g., Antwerp, EF = 2,590 ng F L⁻¹) showed moderate potency-weighted risk (RQpeq ≈ 3) because the load was dominated by low-potency TFA and unidentified species. Conversely, Lyon-2 exhibited a much lower EF (226 ng F L⁻¹) but a significantly higher toxicological risk (RQpeq ≈ 11) driven by high-potency legacy compounds (PFOS, PFNA). Twenty-one of the 34 cities exceeded the potency-equivalent risk benchmark.

Significance and Claims
The paper asserts that PFAS contamination in European drinking water has been systematically underestimated by conventional monitoring strategies that rely solely on targeted analysis of legacy compounds. The study demonstrates that:

1. Knowledge Gaps: Current monitoring misses the majority of the fluorine mass balance, with unidentified extractable fluorine (UEF) constituting the dominant fraction in most cities.
2. Limitations of Current Metrics: Concentration-based metrics (like the 500 ng L⁻¹ PFAS Total limit) may flag water with high loads of low-toxicity substances (like TFA) while failing to adequately capture the risk posed by mixtures containing high-potency compounds at lower total fluorine concentrations.
3. Need for Integrated Approaches: The authors conclude that evidence-driven drinking water safety standards require complementing EF-based screening with compositional analyses and potency-based approaches. They emphasize that compliance with concentration thresholds does not guarantee acceptable toxicological risk, and conversely, high total EF may overstate risk if dominated by low-potency species.
4. Source Control: Given the difficulty of removing ultra-short-chain and unidentified fluorinated compounds via conventional or even advanced treatment, the findings underscore the necessity of upstream source control to prevent the release of fluorinated compounds into the water cycle.

The study provides critical baseline data to support the implementation of the revised EU Drinking Water Directive and highlights the urgent need for advanced analytical and toxicological frameworks to address the full spectrum of PFAS exposure.