Case study: Targeted HPLC-MS/MS analysis of TFA and other USC-PFAS in beverages and waters from Italy and Asia

Case study: Targeted HPLC-MS/MS analysis of TFA and other USC-PFAS in beverages and waters from Italy and Asia Alex Affricano a,*,1 , Alberto Asteggiano a,1 , Alice Di Bernardo a , Orapan Apirakkan b , Angkhana Khachonwongwattana b , Savarin Sinaviwat b , Claudio Medana a a Department of Molecular Biotechnology and Health Sciences (MBC), University of Torino, Via Nizza 52, 10125, Torino, Italy b Department of Science Service (DSS), National Institute of Reference Laboratories, 75/7 Rama VI Road, Ratchathewi, Bangkok, 10400, Thailand ARTICLE INFO Keywords: Ultra-short chain PFAS Beverages HPLC-MS/MS Environmental Food ABSTRACT Trifluoroacetic acid (TFA) is an ultra-short chain perand polyfluoroalkyl substance (PFAS), persistent, highly soluble, and increasingly detected in the environment. This study reports the first systematic screening of TFA and other USC-PFAS in Italy, covering surface, mineral, spring and tap waters as well as alcoholic and nonalcoholic beverages. A total of 172 liquid samples were analyzed with a validated high-performance liquid chromatography–tandem mass spectrometry (HPLC-MS/MS) method. Additional bottled tap and surface waters from China and Thailand were included for international comparison. Among eight target analytes, only TFA was detected. In Italian waters, concentrations were generally low (0.10–2.02 μ g/L; mean 0.39 μ g/L), with slightly higher levels in well and surface waters. In contrast, beverages showed markedly elevated concentrations, especially wines (45–407 μ g/L; mean 138.8 μ g/L), with several red wines exceeding 300 μ g/L. Furthermore, the analysis of several wine heritages showed an interesting increase in TFA concentration from 1997 to 2024. These findings demonstrate a distinct contamination profile of TFA, likely linked to atmospheric deposition, and reveal unexpectedly and high accumulation in wines, suggesting dietary exposure routes beyond drinking water. The work supports the inclusion of TFA in PFAS monitoring and regulatory frameworks and provides a preliminary comparison with Asian regions, where Chinese waters showed higher levels (0.56–1.85 μ g/L) and Thai waters were consistently below detection. Finally, the preliminary analysis on wine vintages shows an increase in TFA concentration during the years and paves the way to a future and more throughout systematic study on the increasing presence of TFA in this matrix. 1. Introduction In recent years, increasing attention has been directed towards emerging pollutants that pose a potential risk to environmental and human health. Among these, perand polyfluoroalkyl substances (PFAS) are of particular concern due to their extreme persistence, mobility, and widespread occurrence in various environmental compartments. Within this broad class, USC-PFAS, including TFA, perfluoropropanoic acid (PFPrA), perfluorobutanoic acid (PFBA) and their sulfonic analogs (trifluoromethanesulfonic acid [TFMSA], perfluoroethanesulfonic acid [PFEtSA], perfluoropropanesulfonic acid [PFPrSA], perfluorobutanesulfonic acid [PFBSA]), have emerged as substances of high environmental relevance, yet they remain underrepresented in both regulatory frameworks and environmental monitoring programs (Cousins et al., 2020). TFA is a small, highly mobile fluorinated organic acid that is extremely water-soluble and chemically stable. It is primarily formed through the atmospheric and microbial degradation of various fluorinated compounds, including hydrofluorocarbons (HFCs), hydrofluoroolefins (HFOs), and various fluorinated agrochemicals (Boutonnet et al., 1999; Ellis et al., 2001). Once released into the environment, it shows resistance to further degradation and tends to accumulate in water bodies where it can persist for years (Zhai et al., 2015). Similarly, USC-PFSAs such as TFMSA and PFBSA exhibit high water solubility, low sorption to soils or sediments, and strong chemical stability. These compounds can either be used directly in industrial * Corresponding author. E-mail address: [email protected] (A. Affricano). 1 These authors contributed equally to this work. Contents lists available at ScienceDirect Food Control journal homepage: www.elsevier.com/locate/foodcont https://doi.org/10.1016/j.foodcont.2025.111779 Received 17 July 2025; Received in revised form 24 September 2025; Accepted 10 October 2025 Food Control 181 (2026) 111779 Available online 10 October 2025 0956-7135/© 2025 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ ). processes (e.g., TFMSA as a catalyst or solvent) or arise from the degradation of longer-chain sulfonic PFAS. Their environmental pathways and persistence characteristics are still not fully understood, but recent studies have detected them in surface and drinking water, suggesting long-range mobility comparable to TFA (Tao et al., 2023; Wenbin Zhu, Wenbo Liu, e Hangbiao Jin 2024). Environmental modelling studies have recently highlighted the widespread presence of TFA in European water bodies especially via long-range atmospheric transport and deposition. A 2023 report by the German Environment Agency (Shang et al., 2023) highlighted the accumulation of TFAs in European river basins-including the Rhine-primarily through atmospheric transformation of HFOs, raising concerns even in remote or non-industrial areas. These findings underscore the need to expand environmental monitoring programs beyond the traditionally considered long-chain PFASs. Despite its environmental persistence and potential ecotoxicological impacts, USC-PFAS (including TFA) are largely excluded from current regulatory frameworks in most of matrices. In Europe, recent revisions to the Drinking Water Directive (EU Directive 2020/2184) set limits for total PFAS (0.5 μ g/L) and for a selected group of 20 substances (0.1 μ g/ L). The only notable limit for USC-PFAS in EU has been only recently set on TFA (10 μ g/L) by the Italian government in July 2025 under the dlgs n.102 of June 19, 2025 and in effect from January 2027 (Gazzetta ufficiale della Repubblica Italiana 2025). In Asia, PFAS regulation is generally more fragmented and often targets legacy compounds like PFOS and PFOA, sometimes with stricter restrictions or even total bans. However, systematic monitoring of newer compounds like TFA is still limited or absent. As a result, although some Asian countries may apply more stringent thresholds for known PFAS, neither Europe nor Asia has yet implemented comprehensive strategies to address the specific risks posed by ultra-short-chain PFAS. However, both European and International regulatory authorities (European Chemicals Agency (ECHA), European Food Safety Authority (EFSA), and national authorities in Asia (e.g., Japan’s METI, China’s MEE, or Taiwan’s EPA)) are becoming increasingly aware of their presence in food and water and are currently working to update their guidelines and regulations to better manage these substances (ECHA Europe, 2023; EFSA Panel on Contaminants in the Food Chain (EFSA CONTAM Panel) 2020). Building on these concerns, increasing attention has been paid to TFA also outside Europe, particularly in Asia. In China, several monitoring studies have documented its persistent accumulation in aquatic systems, highlighting a trend of growing environmental presence over time. Initial investigations already in the early 2000s reported detectable levels of TFA in rainwater, rivers, and wastewater, suggesting longrange atmospheric transport and deposition from industrial emissions and fluorinated compound degradation (Zhang et al., 2005). A subsequent study conducted in the Beijing region revealed a 17-fold increase in TFA concentrations in landscape waters over a single decade, confirming the rapid environmental buildup of this compound in highly urbanized and industrialized areas (Liang, Steimling, e Chang 2023; Zhai et al., 2015). In Italy, regional PFAS monitoring programs have so far focused mainly on long-chain compounds, driven by contamination episodes such as the one that occurred in Veneto (Giglioli, Colombo, e Azzellino 2023). Additional studies have reported PFAS occurrence in riverine and groundwater systems, for example the biodistribution of PFOA in Paracentrotus lividus as a potential biomonitor (Savoca et al., 2021) and detailed surveys of PFAS levels in Piemonte (Binetti et al., 2019) and Lombardy regions (Barreca et al., 2020) using hyphenated HPLC–MS/MS techniques. The presence and distribution of USC-PFASs on the Italian territory, however, remain largely unexplored. In particular, data on their presence in drinking water, surface water and groundwater are very limited, and systematic monitoring in liquid food products is completely lacking, despite growing international evidence suggesting potential food-related exposure pathways. An interesting insight on the USC-PFAS contamination on Italian crops has been led (Schiavone et al., 2025) and shows the presence of these contaminants in tomatoes, highlighting the importance of investigating also the upstream potential contamination that comes from water irrigation. This confused regulatory environment about most of USC-PFAS may be originated from the fact that the toxicity of mediumand long-chain PFAS is well established (Dickman e Aga 2022; Fenton et al., 2021), while, for now, USC-PFAS, such TFA, remain poorly studied and understood. In this regard, most studies report that TFA accumulates in aquatic environments and can be absorbed by both aquatic and terrestrial plants, with some evidence of bioaccumulation. Despite this, reported toxicity is generally low, and TFA appears biologically inert and readily excreted by mammals. However, recent biomonitoring studies have detected TFA in human serum and urine, suggesting potential chronic exposure. Further research and improved biomonitoring strategies are needed to fully assess its environmental and health risks (Garavagno et al., 2024). Recent investigations, such as the PAN Europe study on TFA in European wines (PAN Europe, 2025), have shown the presence of TFA in bottled wine samples from several countries. These results suggest that USC-PFAS may enter the human food chain not only through water, but also through food processing, agricultural chemicals or atmospheric deposition. To date, no study has systematically assessed the levels of USC-PFAS in different classes of food and beverage products at the national level. This study aims to fill this knowledge gap by providing a preliminary screening of USC-PFAS and USC-PFSA in a range of environmental and commercial matrices in Italy. The investigation focuses on two main categories: - Water samples, including tap water, mineral water, well water, surface water and rainwater, collected or purchased from different sites across Central and Northern Italy. In this case, to further contextualize the data within a wider global framework, the study also included a set of bottled, tap and surface water samples from Thailand and China. - Food products, including wines, fruit juices, vinegar, and other representative commercial beverages, purchased from different Italian regions and screened for the presence of TFA and related ultra-short PFAS. Quantification was performed using a high-performance liquid chromatography–tandem mass spectrometry (HPLC-MS/MS) method optimized and validated for ultra-polar analytes in complex aqueous matrices. The method ensures low detection limits (0.05–0.1 μ g/L, depending on the compound), high reproducibility, and robustness in both environmental and food contexts. By coupling environmental monitoring with a case-study approach, this work aims to (i) provide a first spatial snapshot of USC-PFAS distribution in Italy, framed within a broader international context through comparison with selected Asian water samples; (ii) evaluate the potential for combined human exposure through both environmental and dietary pathways, and (iii) contribute to ongoing international efforts—both in Europe and Asia—toward the recognition, regulation, and analytical standardization of USC-PFAS in water and food monitoring programs. 2. Materials and methods 2.1. Materials 2.1.1. Sample collection A total of 172 samples, comprising 124 water samples and 48 liquid food products and beverages was analyzed. The mineral water samples (n =31) were commercially available bottled products, purchased from various Italian retailers and selected to include springs from different A. Affricano et al. Food Control 181 (2026) 111779 2 Italian region. The tap water samples (n =64) were collected directly from household or public-use sources located in different regions of Italy, while well water (n =9) and surface water samples (n =7) were collected in-situ using clean, PFAS-free sampling containers. Surface waters included: rivers, small lakes, and collected rainwater. In addition to the Italian samples, 11 water samples from Asian countries were included. These consisted of three Chinese waters—one bottled sample purchased from a convenience store, one urban tap water sample, and one river water sample—as well as eight samples from Thailand, including five bottled mineral waters purchased from local retailers and three tap water samples collected from residential and institutional buildings (see Table S1 for additional information about geographical and regional origin of the tested samples). Among the liquid food samples, wine (n =36) was the most represented category, including both red (n =21) and white (n =15) wines. Additional food liquids (n =12) consisted of fruit juices, vinegar, beer, and other alcoholic beverages, all purchased from local retail markets (see Table S1 for additional information about all the sample types). All samples were stored at 4 ◦C before the analysis to minimize degradation or volatilization of ultrashort-chain PFAS. 2.1.2. Chemicals and standards All solvents and reagents used were of analytical or ultrapure grade, suitable for LC-MS analysis. Acetonitrile (Ultrapure grade, HPLC-MS quality) and glacial acetic acid (≥99 %, ARISTAR® ULTRA) were obtained from VWR Chemicals (Milan, Italy). Ammonium acetate (ACS reagent, ≥97 %) was purchased from Sigma-Aldrich (St. Louis, MO, USA). Ultrapure water was obtained using a VWR apparatus (Milan, Italy) equipped with a RO (reverse osmosis) cartridge. The analytical standards Ultrashort-chain PFAS-7 (C1–C4) (DRE-A30000064MW) and perfluorobutanoic acid13 C 3 (DRE-A15986523MW-50) were purchased from LGC Standards (Teddington, Middlesex, UK) and were used for method development and validation. 2.2. Methods 2.2.1. Sample preparation Non-water samples, such as food liquids were first diluted 1:10 (v/v) in ultrapure water to reduce matrix complexity and viscosity. The diluted samples were then centrifuged at 16,000×g for 15 min at 4 ◦C to remove suspended particles that could interfere with chromatographic separation and analyte detection. The resulting supernatant was transferred into PFAS-free polypropylene vials and spiked with 10 μ g/L of the isotopically labelled internal standard 13 C 3 -PFBA. For water samples, a 1 mL aliquot was directly transferred into PFAS-free polypropylene vials and spiked with the same internal standard ( 13 C 3 -PFBA, 10 μ g/L). All samples were then analyzed by HPLC-MS without any further pretreatment. 2.2.2. HPLC-MS analysis All analytes were separated and quantified using a high-performance liquid chromatography system coupled to tandem mass spectrometry (HPLC-MS/MS). The chromatographic platform consisted of a Shimadzu Nexera X2 UHPLC system (Shimadzu, Kyoto, Japan), connected to a QTRAP® 5500 triple quadrupole mass spectrometer (Sciex, Darmstadt, Germany). The mass spectrometer was equipped with a Turbo V™ electrospray ionization (ESI) source operating in negative ion mode (ESI-), using nitrogen as nebulizer and desolvation gas, and air as the source of heat. Chromatographic separation was performed using a Luna Polar Pesticides column by Phenomenex (Bologna, Italy) 100 ×2.1 mm, 3 μ m particle. A delay column with the same stationary phase (50 ×2.1 mm, 3 μ m) was positioned upstream, directly after the HPLC mixer. The mobile phase consisted of solvent A (ultrapure water containing 0.1 % acetic acid and 10 mM ammonium acetate) and solvent B (acetonitrile). The flow rate was maintained at 0.400 mL/min, with a total run time of 12 min. The gradient program was scheduled as follows: 0–2 min, 0 % B; 2–5 min, linear increase to 95 % B; maintained at 95 % B until minute 7; returned to 0 % B at minute 8; and held at 0 % B until minute 12 for column re-equilibration. The injection volume was set at 50 μ L. The ion source parameters were optimized as follows: ion source gas 1 at 20 psi, ion source gas 2 at 25 psi, curtain gas at 20 psi, and collisionactivated dissociation (CAD) gas set to 9. The source temperature was kept at 300 ◦C. The MS analysis was conducted in MRM mode, the transitions and the relative parameters for TFA, PFPrA, PFBA, PFBA 13C3(IS), TFMSA, PFEtSA, PFPrSA, and PFBSA are summarized in Table 1. Representative extracted ion chromatograms (XIC) for the target analyte in standard solution and of TFA in real sample matrices are provided in the Supplementary Materials (Fig. S1). 2.2.3. Method validation The analytical method was validated in accordance with standard criteria for quantitative LC–MS/MS analyses. Validation parameters included linearity, accuracy, precision, limits of detection (LOD) and quantification (LOQ), and calibration ranges in relevant matrices. Linearity was assessed across multiple concentration levels using matrixmatched calibration with isotopically labelled internal standards. Accuracy (expressed as BIAS% between the found and the calculated value) and precision (RSD% of measurements) were evaluated at low, medium, and high concentrations in triplicate on different days. Method sensitivity (LOD, LOQ) was determined from signal-to-noise ratios and adapted to both environmental water samples and diluted food matrices. Detailed validation results are presented in Section 3.1 and Table 2. Matrix effects were evaluated by comparing the slopes of calibration curves prepared in spiked Milli-Q water and in diluted beverage matrices (red wine, white wine and fruit juice). The relative difference between these slopes was expressed as an absolute percentage. Matrix spike experiments were carried out by adding levels up to approximately 100 % of the native concentration of TFA, in order to reflect realistic sample conditions and remain within the calibration range. 3. Results and discussion 3.1. Method validation results The method was validated in matrix using the isotopically labelled compound (perfluorobutanoic acid13 C 3 ) as internal standard for quantification. Calibration curves were performed in triplicate across three separate days (n =3, x =3) using 5 to 8 concentration levels depending on the analyte. The performance of the analytical method used for the quantification of USC-PFAS was evaluated following standard validation criteria. Validation parameters included linearity (R 2 ), slope variability (expressed as DIFF%), accuracy (BIAS%), and precision (RSD%) both intra-day and inter-day (Table 2). Linearity was excellent for all target compounds showing correlation coefficients (R 2 ) between 0.9930 and 0.9994 and a standard deviation of the slopes (DIFF%) between 3.74 % and 10.38 %. Accuracy, measured as the difference between measured and expected values (BIAS%), was evaluated (together with precision) at low, medium, and high concentrations. Inter-day accuracy was generally good, with BIAS% values below ±15 %, and often below ±10 %. Intraday accuracy was similarly consistent, though slightly higher variability was observed at the lowest levels for some compounds (e.g., PFPrA: 14.84 %) but it was always below ±15 % even in this case. Precision, expressed as the relative standard deviation (RSD%), was also acceptable across all analytes. Inter-day RSD values ranged from 0.77 % to 8.13 %, while intra-day RSD values were between 1.93 % and 10.11 %, with most analytes showing values below 7 %, confirming stable performance over time. In terms of sensitivity, the method showed detection limits (LOD) A. Affricano et al. Food Control 181 (2026) 111779 3 between 0.03 and 1.55 μ g/L, and quantification limits (LOQ) between 0.10 and 5.18 μ g/L, depending on the analyte. The Lower Limits of Quantification (LLOQ) were adapted for each compound and matrix, ranging from 0.05 to 2.00 μ g/L, while Upper Limits of Quantification (ULOQ) extended up to 20 μ g/L, suitable for samples with higher contamination levels. The method shows a low-enough LOQ to comply to the 10 μ g/mL limit established from EU guidelines for TFA. In addition, the evaluation of matrix effects confirmed the robustness of the method for most analytes in both wine (red and white) and juice matrices. The calculated ME% values were consistently within the generally accepted threshold of <20 % for TFA, PFPrA, PFBA, PFEtSA, and PFBSA, indicating minimal ion suppression across the tested matrices. By contrast, TFMSA and PFPrSA exhibited higher matrix effects (>20 %) in some matrices. Although this parameter has not been formally validated for these two compounds, it is noteworthy that they were never detected in the analyzed samples. Importantly, the only PFAS observed was TFA, for which ME% values remained within the acceptable range. For TFA—which was later found to be the only USC-PFAS detected consistently across all samples (see Sample Analysis)—the method was Table 1 MRM transitions and optimized instrument parameters used for the quantification of USC-PFAS by HPLC-MS/MS. For each compound (including isotopically labelled internal standards), the table reports: retention time (RT), precursor ion (Q1) and product ion (Q3) masses, dwell time, declustering potential (DP), entrance potential (EP), collision energy (CE), and collision cell exit potential (CXP). Multiple transitions were monitored for structural confirmation and quantification, * indicates the transition used for quantitation. Compound RT (min) Q1 mass (Da) Q3 mass (Da) Dwell time (ms) DP (V) EP (V) CE (V) CXP (V) TFA 2.78 112.90 69.00* 25.0 −40.0 −10.0 −16.0 −10.0 112.90 112.90 25.0 −40.0 −10.0 −5.0 −10.0 PFPrA 3.16 162.90 119.00* 25.0 −40.0 −10.0 −14.0 −10.0 162.90 162.90 25.0 −40.0 −10.0 −5.0 −10.0 PFBA 4.03 212.90 169.00* 25.0 −40.0 −10.0 −14.0 −10.0 212.90 212.90 25.0 −40.0 −10.0 −5.0 −10.0 TFMSA 3.11 148.90 79.90* 25.0 −60.0 −10.0 −31.0 −10.0 148.90 98.90 25.0 −60.0 −10.0 −34.0 −10.0 148.90 82.90 25.0 −60.0 −10.0 −25.0 −10.0 PFPrSA 5.39 248.90 98.90* 25.0 −60.0 −10.0 −30.0 −10.0 248.90 79.90 25.0 −60.0 −10.0 −54.0 −10.0 248.90 118.90 25.0 −60.0 −10.0 −30.0 −10.0 PFEtSA 3.75 198.90 79.90* 25.0 −60.0 −10.0 −30.0 −10.0 198.90 84.90 25.0 −60.0 −10.0 −30.0 −10.0 198.90 82.90 25.0 −60.0 −10.0 −27.0 −10.0 PFBSA 6.09 298.90 118.90* 25.0 −60.0 −10.0 −33.0 −10.0 298.90 98.80 25.0 −60.0 −10.0 −37.0 −10.0 298.90 168.90 25.0 −60.0 −10.0 −32.0 −10.0 298.90 79.90 25.0 −60.0 −10.0 −58.0 −10.0 PFBA 13 C 3 (IS) 4.02 215.9 172.00* 25.0 −40.0 −10.0 −12.0 −10.0 215.9 215.9 25.0 −40.0 −10.0 −5.0 −10.0 Table 2 Validation parameters of the HPLC-MS/MS method for the quantification of ultra-short chain PFAS (USC-PFAS) in environmental and food-grade liquid matrices. Results include linearity (R 2 ), slope variability (DIFF%), accuracy (BIAS%) and precision (RSD%) assessed intraand inter-day, matrix effect (ME%) in the given matrices, as well as limits of detection (LOD), quantification (LOQ), and the validated calibration range (LLOQ–ULOQ) for each analyte. Calibration was performed using the standard addition method in matrix with isotopically labelled internal standards. Parameter TFA (0.1–3) TFA (0.5–20) PFPrA PFBA PFBSA PFEtSA PFPrSA TFMSA Linearity (R 2 ) 0.9930 0.9970 0.9961 0.9940 0.9986 0.9988 0.9992 0.9994 Slope variability (DIFF%) 7.81 5.8 4.09 5.49 10.38 7.49 6.05 3.74 BIAS% Inter-day (Low) 9.05 7.37 13.91 9.22 13.06 11.02 2.7 12.99 BIAS% Inter-day (Medium) 1.55 5.42 2.4 7.88 3.17 3.13 2.36 4.75 BIAS% Inter-day (High) 6.1 4.56 2.37 1.65 3.86 3.1 1.86 3.43 RSD% Inter-day (Low) 4.42 1.11 0.77 8.13 5.15 4.59 2.69 2.28 RSD% Inter-day (Medium) 1.21 5.07 2.59 7.02 2.58 3.8 2.14 4.85 RSD% Inter-day (High) 5.14 4.77 0.93 1.85 4.34 3.31 2.01 3.62 BIAS% Intra-day (Low) 7.33 7.77 14.84 9.66 5.82 2.77 11.45 8.92 BIAS% Intra-day (Medium) 7.17 5.04 2.96 9.96 6.03 6.21 8.41 4.9 BIAS% Intra-day (High) 3.96 2.83 3.81 6.07 4.81 3.43 6.08 3.81 RSD% Intra-day (Low) 8.03 8.35 2.03 9.87 3.71 3.62 4.44 9.67 RSD% Intra-day (Medium) 7.92 3.52 3.42 10.11 6.71 8.14 8.84 5.79 RSD% Intra-day (High) 3.49 3.11 1.93 6.61 5.36 4.48 6.96 4.31 ME% (wine, red) 5.82 7.86 1.41 3.84 8.22 38.75 9.19 ME% (wine, white) 8.87 17.84 17.74 14.88 4.11 12.06 31.16 ME% (juice) 19.71 14.89 8.54 12.01 9.57 31.73 33.74 LOD ( μ g/L) a 0.04 0.35 0.03 1.55 0.04 0.04 0.03 0.03 LOQ ( μ g/L) a 0.15 1.16 0.11 5.18 0.13 0.14 0.11 0.1 LLOQ ( μ g/L) a 0.1 0.5 0.1 2,00 0.05 0.05 0.05 0.05 ULOQ ( μ g/L) a 3.00 20.0 3.00 20.0 2.00 2.00 2.00 2.00 a Referred to non-diluted samples, for non-water samples (10-fold dilution), the value should be 10x A. Affricano et al. Food Control 181 (2026) 111779 4 validated using two separate concentration ranges to reflect the different contamination levels in the matrices: −0.1–3.00 μ g/L for environmental water samples, −0.5–20.0 μ g/L for wine and matrices where higher concentrations were expected and observed. This tailored approach allowed reliable quantification across both lowand high-level samples, ensuring that TFA could be accurately measured in all relevant matrices. Other methods have previously been developed for the quantification of TFA in environmental waters. These methods have generally shown comparable LLOQ values to ours (Ute Dorgerloh et al., 2025) while others, particularly those employing alternative platforms or derivatization strategies (Scheurer et al., 2017; Scott et al., 2005) have achieved even lower detection limits. However, the method presented in this study enabled quantification of TFA in nearly all the samples analyzed, with only a few cases falling below the LLOQ. Thus, the method proved to be reliable, accurate, and precise not only for environmental water matrices but also for more complex liquid food products, such as wine, juice, and vinegar. The validation results confirm its suitability to detect even low levels of these persistent compounds, thus achieving the main analytical goals of this study: to provide a robust tool to map the presence of USC-PFAS in Italy through both environmental and food-related exposure. 3.2. Sample analysis Among all the target compounds analyzed in this study, TFA was the only analyte detected across the tested samples. In contrast, none of the other ultrashort-chain PFAS or PFSA compounds (PFPrA, PFBA, PFBSA, PFEtSA, PFPrSA, and TFMSA) were observed despite the use of a highly sensitive analytical method. This suggests that TFA contamination occurs through specific pathways distinct from those of other ultrashort-chain PFAS, highlighting its unique environmental behaviour. The main sources of TFA are believed to be atmospheric degradation of fluorinated refrigerants (such as HFOs) and degradation of various PFAS-containing products. In contrast, other USC-PFAS (particularly the sulfonic acids) are less prevalent, likely due to fewer direct release pathways or stricter regulations (Franklin, 1993; Hans Peter H. Arp et al., 2024). TFA was detected in most of tested samples but with substantial differences in concentration levels between water and liquid food groups. Table S1 summarizes all the results. As shown in Fig. 1A, TFA concentrations were markedly higher in liquid food products, with an average of 110.05 μ g/L (SD =87.20 μ g/L), compared to 0.39 μ g/L (SD =0.33 μ g/L) in water samples. 3.2.1. Water samples All water samples were collected or purchased to reflect different geographical areas across Northern and Central Italy providing a preliminary representative view of the presence of TFA across multiple Italian regions. A more detailed overview of water types (Fig. 1B) reveals notable variation across the different categories. Mineral waters generally exhibited the lowest levels, with most samples remaining below the 0.5 μ g/L threshold proposed in the recent EU Drinking Water Directive (Directive 2020/2184). The average concentration was 0.30 μ g/L (SD = 0.26 μ g/L). Tap waters showed slightly higher concentrations, with a mean of 0.38 μ g/L (SD =0.33 μ g/L). In contrast, well waters and surface waters presented higher values, with average TFA concentrations of 0.48 μ g/L (SD =0.21 μ g/L) and 0.63 μ g/L (SD =0.52 μ g/L), respectively. The highest concentration observed among all water samples was 2.02 μ g/L, recorded in a rainwater sample, highlighting the potential contribution of atmospheric deposition to environmental TFA contamination. These findings are consistent with previously reported data across Europe, which showed TFA widespread in tap, mineral and surface waters (Jun Li et al., 2025; PAN Europe, 2024). The values observed in the present study fall within this known range and further support the evidence that TFA is a persistent and ubiquitous contaminant in European water resources. To expand the geographical scope of the analysis and contextualize the Italian data, eleven additional bottled and tap water samples from Asia were included—eight from Thailand and three from China. As shown in Fig. 1C, these samples provided a preliminary international comparison of TFA levels in drinking and surface waters. The Thai samples (five bottled mineral waters and three tap waters) consistently showed no detectable or quantifiable levels of TFA, with all values falling below the LLOQ. This outcome aligns with previous reports indicating the widespread use of advanced purification methods—particularly reverse osmosis—in Thai water treatment, which has been shown to effectively remove even highly soluble and persistent contaminants such as TFA (Scheurer et al., 2017). The near-total absence of TFA in these samples highlights the impact of water treatment practices on final contamination levels and suggests that treatment technologies may play a more critical role than geographic location alone. Fig. 1. Violin plots showing the distribution of TFA concentrations ( μ g/L) across different sample categories. (A) Comparison between water samples (n =113) and liquid food products (n =48). (B) Breakdown of TFA concentrations in four water types: mineral (n =31), tap (n =64), well (n =9), and surface water (n =7). (C) Comparison of TFA levels in Italian water samples and bottled/tap/surface waters from China (n =3) and Thailand (n =8). A. Affricano et al. Food Control 181 (2026) 111779 5 In contrast, the three Chinese samples (one bottled mineral water, one tap water, and one surface water) showed measurable TFA concentrations (0.56, 0.85, and 1.85 μ g/L, respectively), all above the average levels observed in Italian water samples. These values are consistent with prior monitoring studies in China, which have reported increasing TFA accumulation in urban and peri-urban water bodies, likely due to high atmospheric emissions and industrial activity (Zhai et al., 2015; Zhang et al., 2005). These particularly elevated concentration values observed in the Chinese surface water sample further supports the hypothesis that atmospheric deposition is a significant contributor to TFA contamination. 3.2.2. Liquid food samples The higher concentrations of TFA in liquid food products compared to water may arise from multiple factors. Agricultural commodities are directly exposed to atmospheric deposition and irrigation with rain or surface waters (Scheurer et al., 2017), both of which can contain TFA. Some fluorinated agrochemicals are known to degrade into TFA (Joerss et al., 2024), potentially contributing during crop cultivation. In addition, food processing steps such as fermentation, concentration, or blending may further enhance or retain TFA levels relative to raw water. These mechanisms together suggest that dietary matrices can act as amplifiers of environmental contamination, explaining the substantial gap observed between water and beverage concentrations. Among the 48 liquid food samples analyzed, wine exhibited the highest TFA concentrations levels if compared to other product types such as beer, fruit juices, vinegar, and other alcoholic beverages (Fig. 2A). The average concentration in wine was 130.89 μ g/L (SD = 84.95 μ g/L), suggesting both high levels and considerable variability between products. A more detailed focus on wine types (Fig. 2B) shows that red wines generally contained higher concentrations of TFA than white wines, with several red wine samples exceeding 300 μ g/L. The average concentration in red wines was 144.30 μ g/L (SD =98.22 μ g/L), compared to 116.32 μ g/L (SD =58.76 μ g/L) in white wines. The markedly higher concentrations of TFA in wines highlight a potential dietary exposure pathway that may significantly exceed contributions from drinking water. Given the widespread consumption of wine in Italy and other European countries, this finding raises concerns about chronic, low-dose exposure to TFA through common dietary products. Although current toxicological evidence suggests relatively low acute toxicity, recent biomonitoring studies indicate detectable levels of TFA in human serum and urine (Garavagno et al., 2024), underscoring the need for systematic evaluation of long-term health effects. Consistently with the findings of a recent PAN Europe report (PAN Europe, 2025) - suggesting an exponential increase in TFA accumulation in wine over time - the analysis of different vintages of white and red wines (1997–2024) resulted in the finding of increasing TFA concentration over time. Interestingly, a set of red wines coming from the same producer whose vintage spans from 1997 to 2023 showed a markedly increased concentration of TFA ranging from 5.81 to 124.88 μ g/L. A similar increase was evident when considering the entire dataset and remained clearly distinguishable when red and white wines were evaluated separately (Fig. 3). Another relevant observation concerns wine manufacturing: supermarket low-cost wines showed lower TFA concentrations than winery bottled wines. The finding, as a first glance unexpected, paved the way for further speculations: two mains are listed in the following lines. First, these commercial wines are typically produced using large-scale, standardized industrial processes (e.g., intensive filtration or stabilization) which may reduce residual TFA content. In contrast, handcrafted local wines, often produced using biodynamic or low-intervention methods, undergo minimal processing and retain more of their natural composition. This approach, closely related to biodynamic practices, limits technological intervention and typically excludes dilution or aggressive purification treatments (Maioli et al., 2021). Secondly, differences in irrigation practices may also contribute industrial vineyards, especially those producing wine for large-scale distribution, often rely on controlled irrigation systems using treated groundwater or municipal sources, where TFA concentrations are generally low. Local vineyards, Fig. 2. Violin plots showing TFA concentrations in liquid food products. (A) Comparison between wines (n =36) and other beverages including juices, vinegar, and beer (n =12), (B) Comparison of TFA concentrations between red (n =21) and white wines (n =15). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.) A. Affricano et al. Food Control 181 (2026) 111779 6 on the other hand, are more likely to depend exclusively on rainwater. As demonstrated by our findings, TFA levels in rainwater can be considerably higher than in other water sources, with concentrations reaching up to 2.02 μ g/L. This suggests that rain-irrigated vineyards may indirectly absorb more TFA through atmospheric deposition, leading to higher residual levels in the final product. 3.3. Implications and regulatory gaps Overall, our results dwell into a regulatory blind spot: despite growing evidence of persistence and accumulation, TFA and other USCPFAS remain absent from both European and Asian monitoring frameworks for liquid food matrices (e.g., wine). The incumbent adoption for Italy and EU of the limit of 10 μ g/L for TFA in water for 2027 is, for now, the only official parameter to be monitored as regards USC-PFAS. Our study therefore provides timely evidence supporting the inclusion of TFA in PFAS regulatory agendas, with specific additional consideration of its dietary exposure sources such as wine and other beverages. 4. Conclusion This study provides the first systematic investigation of ultra-short chain PFAS (USC-PFAS), and in particular TFA, in both environmental waters and liquid food products across different Italian regions. Among the eight analytes investigated, TFA was the only compound detected, confirming its distinct environmental behaviour and dominant role in ultrashort-chain PFAS contamination. The concentrations observed in water samples were generally low but variable depending on the source, with surface and well waters showing the highest levels. When compared to bottled and tap waters from China and Thailand, Italian values appeared intermediate—Chinese waters showed elevated levels consistent with previous reports, while Thai waters showed no detectable contamination, likely due to advanced purification treatments. Those comparative results, however, have to be further investigated by adding more Asian samples since the focus of the present study is to unveil the USC-PFAS distribution in Italy. Perhaps, wine and other beverage products showed markedly higher concentrations, with recent vintage wines displaying the greatest accumulation. In addition, the longitudinal analysis of vintages from 1997 to 2024 revealed a progressive increase in TFA concentrations, consistent with recent international reports and particularly evident in a set of red wines from the same producer, where values rose from 5.81 to 124.88 μ g/L. The absence of other USC-PFAS in all samples, despite the use of a sensitive and validated HPLC-MS/MS method, suggests that TFA contamination follows independent pathways rather than the depolimerization of longer-chain PFAS. These findings raise important considerations regarding dietary exposure, especially in products such as wine that are susceptible to environmental accumulation over time. Finally, the quantitative results showing a high accumulation of TFA in wine samples -most of the times higher than the proposed law level for water-open the frame for a more comprehensive regulation regarding this liquid food matrix. CRediT authorship contribution statement Alex Affricano: Writing – original draft, Validation, Methodology, Formal analysis, Data curation. Alberto Asteggiano: Writing – review & editing, Methodology, Formal analysis, Data curation. Alice Di Bernardo: Investigation. Orapan Apirakkan: Investigation. Angkhana Khachonwongwattana: Investigation. Savarin Sinaviwat: Investigation. Claudio Medana: Writing – review & editing, Conceptualization. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements This work was funded by the European Union’s Horizon 2020 Research and Innovation Programme under the MarieSkłodowska-Curie Grant Agreement No 101007578 (SusWater). Open access publishing facilitated by Universit` a di Torino, as part of the Wiley - CRUI-CARE agreement. The authors are grateful to the winery “Cantina Goggiano”, Refrancore, Italy and to the wine shop “Franco Bongiovanni”, Torino for kindly providing samples of vintage wines. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.foodcont.2025.111779. Data availability Datasets generated during and analyzed during the current study are available from the corresponding author on reasonable request. References Arp, H. P. H., Gredelj, A., Glüge, J., Scheringer, M., & Cousins, I. T. (2024). The global threat from the irreversible accumulation of trifluoroacetic acid (TFA). Environmental Science & Technology, 58(45), 19925–19935. https://doi.org/10.1021/acs. est.4c06189 Barreca, S., Busetto, M., Colzani, L., Clerici, L., Marchesi, V., Tremolada, L., Daverio, D., & Dellavedova, P. (2020). Hyphenated high performance liquid chromatography–tandem mass spectrometry techniques for the determination of perfluorinated alkylated substances in Lombardia region in Italy, profile levels and assessment: One year of monitoring activities during 2018. Separations, 7(1), 17. https://doi.org/10.3390/separations7010017 Binetti, R., Gasperi, J., Grangeia, C., Lucchesi, R., & Montano, L. (2019). Perfluoroalkyl substance assessment in Turin metropolitan area and correlation with potential sources of pollution according to the water safety plan risk management approach. Separations, 6(1), 17. https://doi.org/10.3390/separations6010017 Boutonnet, J. C., Bingham, P., Calamari, D., De Rooij, C., Franklin, J., Kawano, T., Libre, J.-M., McCulloch, A., Malinverno, G., Odom, J. M., Rusch, G. M., Smythe, K., Sobolev, I., Thompson, R., & Tiedje, J. M. (1999). Environmental risk assessment of trifluoroacetic acid. Human and Ecological Risk Assessment: An International Journal, 5 (1), 59–124. https://doi.org/10.1080/10807039991289644 Cousins, I. T., DeWitt, J. C., Glüge, J., Goldenman, G., Herzke, D., Lohmann, R., Miller, M., Ng, C. A., Scheringer, M., Vierke, L., & Wang, Z. (2020). Strategies for grouping perand polyfluoroalkyl substances (PFAS) to protect human and Fig. 3. Trend of TFA concentrations in wines over the years, distinguishing red wines (red), white wines (green), and a series of red wines from the same producer (black). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.) A. Affricano et al. Food Control 181 (2026) 111779 7 environmental health. Environmental Sciences: Processes & Impacts, 22(7), 1444–1460. https://doi.org/10.1039/D0EM00147C Dickman, R. A., & Aga, D. S. (2022). A review of recent studies on toxicity, sequestration, and degradation of perand polyfluoroalkyl substances (PFAS). Journal of Hazardous Materials, 436, Article 129120. https://doi.org/10.1016/j.jhazmat.2022.129120 Dorgerloh, U., Becker, R., Freeling, F., Scheurer, M., G¨ okçe, E., Strack, K., Sommerfeld, T., Sauer, A., Schneider, J., H¨ artel, C., Valkov, V., Moser, L., Geiß, S., Junginger, S., Schlittenbauer, L., Suess, E., & Ruth, K. (2025). Standardising the quantification of trifluoroacetic acid in water: Interlaboratory validation trial using liquid chromatography-mass spectrometric detection (LC–MS/MS). Accreditation and Quality Assurance. https://doi.org/10.1007/s00769-025-01640-2 ECHA Europe. (2023). Annex XV restriction report. European Chemicals Agency. https: //echa.europa.eu/documents/10162/f605d4b5-7c17-7414-8823-b49b9fd43aea. EFSA Panel on Contaminants in the Food Chain (EFSA CONTAM Panel). (2020). Risk to human health related to the presence of perfluoroalkyl substances in food. EFSA Journal, 18(9), Article e06223. https://doi.org/10.2903/j.efsa.2020.6223 Ellis, D. A., Mabury, S. A., Martin, J. W., & Muir, D. C. G. (2001). Thermolysis of fluoropolymers as a potential source of halogenated organic acids in the environment. Nature, 412(6844), 321–324. https://doi.org/10.1038/35085548 Fenton, S. E., Ducatman, A., Boobis, A., DeWitt, J. C., Lau, C., Ng, C., Smith, J. S., & Roberts, S. M. (2021). Perand polyfluoroalkyl substance toxicity and human health review: Current state of knowledge and strategies for informing future research. Environmental Toxicology and Chemistry, 40(3), 606–630. https://doi.org/10.1002/ etc.4890 Franklin, J. (1993). The atmospheric degradation and impact of 1,1,1,2-tetrafluoroethane (hydrofluorocarbon 134a). Chemosphere, 27(8), 1565–1601. https://doi.org/ 10.1016/0045-6535(93)90251-Y Garavagno, M. D. L. A., Holland, R., Khan, M. A. H., Orr-Ewing, A. J., & Shallcross, D. E. (2024). Trifluoroacetic acid: Toxicity, sources, sinks and future prospects. Sustainability, 16(6), 2382. https://doi.org/10.3390/su16062382 Gazzetta Ufficiale della Repubblica Italiana. (2025). Decreto legislativo 19 giugno 2025. GU Serie Generale n.153 del 04-07-2025 (Suppl. Ordinario n. 24), 102. https://www. gazzettaufficiale.it. Giglioli, S., Colombo, L., & Azzellino, A. (2023). Cluster and multivariate analysis to study the diffuse contamination of emerging perand polyfluoroalkyl substances (PFAS) in the Veneto region plain (North-Eastern Italy). Chemosphere, 319, Article 137916. https://doi.org/10.1016/j.chemosphere.2023.137916 Joerss, H., Freeling, F., Van Leeuwen, S., Hollender, J., Liu, X., N¨ odler, K., Wang, Z., Yu, B., Zahn, D., & Sigmund, G. (2024). Pesticides can be a substantial source of trifluoroacetate (TFA) to water resources. Environment International, 193, Article 109061. https://doi.org/10.1016/j.envint.2024.109061 Li, J., Boxall, A. B. A., Swan, C., Wilkinson, J., McGimpsey, N., Evans, H., & GauchotteLindsay, C. (2025). Concentrations of trifluoroacetic acid (TFA) in UK surface waters. Fidra UK. https://www.fidra.org.uk/download/concentrations-of-trifluoroacetic-ac id-tfa-in-uk-surface-waters/. Liang, S.-H., Steimling, J. A., & Chang, M. (2023). Analysis of ultrashort-chain and shortchain (C1 to C4) perand polyfluorinated substances in potable and non-potable waters. Journal of Chromatography Open, 4, Article 100098. https://doi.org/ 10.1016/j.jcoa.2023.100098 Maioli, F., Picchi, M., Millarini, V., Domizio, P., Scozzafava, G., Zanoni, B., & Canuti, V. (2021). A methodological approach to assess the effect of organic, biodynamic, and conventional production processes on the intrinsic and perceived quality of a typical wine: The case study of Chianti DOCG. Foods, 10(8), 1894. https://doi.org/10.3390/ foods10081894 PAN Europe. (2024). TFA: The forever chemical in the water we drink. Only a rapid ban on PFAS pesticides and F-gases can save our water. Pesticide Action Network Europe. https://www.pan-europe.info. PAN Europe. (2025). Message from the bottle: The rapid rise of TFA contamination across the EU. Pesticide Action Network Europe. https://www.pan-europe.info. Savoca, D., Melfi, R., Palumbo Piccionello, A., Barreca, S., Buscemi, S., Arizza, V., Arculeo, M., & Pace, A. (2021). Presence and biodistribution of perfluorooctanoic acid (PFOA) in Paracentrotus lividus highlight its potential application for environmental biomonitoring. Scientific Reports, 11, Article 18763. https://doi.org/ 10.1038/s41598-021-98284-2 Scheurer, M., N¨ odler, K., Freeling, F., Janda, J., Happel, O., Riegel, M., Müller, U., Storck, F. R., Fleig, M., Lange, F. T., Brunsch, A., & Brauch, H.-J. (2017). Small, mobile, persistent: Trifluoroacetate in the water cycle – Overlooked sources, pathways, and consequences for drinking water supply. Water Research, 126, 460–471. https://doi.org/10.1016/j.watres.2017.09.045 Schiavone, C., Romaniello, F., Brizio, P., Suman, M., Rossi, A. M., & Portesi, C. (2025). Reliable and sensitive ultra-short chain perand polyfluorinated alkyl substances (PFAS) analysis in food: A polar reverse-phase chromatography approach. Journal of Chromatography A, 1761, Article 466332. https://doi.org/10.1016/j. chroma.2025.466332 Scott, B. F., Macdonald, R. W., Kannan, K., Fisk, A., Witter, A., Yamashita, N., Durham, L., Spencer, C., & Muir, D. C. G. (2005). Trifluoroacetate profiles in the Arctic, Atlantic, and Pacific oceans. Environmental Science & Technology, 39(17), 6555–6560. https://doi.org/10.1021/es047975u Shang, K. Z., Vijayaraghavan, K., Downes, R., Cush, M., Zhuang, Y., & Connolly, R. (2023). Environmental modeling of trifluoroacetic acid (TFA) originating from hydrofluoroolefins. Ramboll for Honeywell. https://advancedmaterials.honeywell. com/content/dam/advancedmaterials/en/documents/echa-reach-submissions/Ram boll-1620011170-011-Honeywell-TFA-Modeling-Final-Report-Aug2-2023.pdf. Tao, F., Tan, Y., Dai, J., Lu, C., Sha, Y., Liu, Y., Zhou, X., Hasi, Q., Shen, Z., & Ma, Y. (2023). Occurrence of halogenated methanesulfonic acids in water and sediment from the Hangzhou Bay, China. Environmental Research, 224, Article 115463. https://doi.org/10.1016/j.envres.2023.115463 Zhai, Z., Wu, J., Hu, X., Li, L., Guo, J., Zhang, B., Hu, J., & Zhang, J. (2015). A 17-fold increase of trifluoroacetic acid in landscape waters of Beijing, China during the last decade. Chemosphere, 129, 110–117. https://doi.org/10.1016/j. chemosphere.2014.09.033 Zhang, J., Zhang, Y., Li, J., Hu, J., Ye, P., & Zeng, Z. (2005). Monitoring of trifluoroacetic acid concentration in environmental waters in China. Water Research, 39(7), 1331–1339. https://doi.org/10.1016/j.watres.2004.12.043 Zhu, W., Liu, W., & Jin, H. (2024). Sediment-seawater partitioning, bioaccumulation, and biomagnification of perfluorobutane sulfonamide in marine environment. Water Research, 255, Article 121466. https://doi.org/10.1016/j.watres.2024.121466 A. Affricano et al. Food Control 181 (2026) 111779 8