Researchers at the University of Rhode Island have detected trifluoroacetic acid (TFA), a byproduct of some f-gases, in 66% of 320 drinking water samples collected in the U.S. state, with median concentrations of 580ng/L and a maximum concentration of 2,400ng/L (2.4mcg/L).
The findings were reported by Rainer Lohmann, Professor at the University of Rhode Island’s Graduate School of Oceanography and Director of the University’s Superfund research program STEEP (Sources, Transport, Exposure & Effects of PFAS). Lohmann was a member of the U.S. Environmental Protection Agency (EPA) Board of Scientific Counselors from 2017 to 2023 and serves as editor of the Environmental Toxicology and Chemistry Journal.
Lohmann shared the research findings during a panel discussion on PFAS (per- and polyfluorinated substances) at the ATMOsphere (ATMO) America Summit 2025, held June 11 and 12 in Atlanta, Georgia. The conference was organized by ATMOsphere, publisher of NaturalRefrigerants.com.
TFA forms in the atmosphere from the complete oxidation of fugitive HFO-1234yf gas over about two weeks, and the partial oxidation of other HFOs and HFC-134a. Many f-gases and TFA fall under the definition of PFAS pollutants set by the Organisation for Economic Co-operation and Development (OECD), which says that PFAS have at least one fully fluorinated carbon atom.
While exposure to long-chain PFAS like perfluorooctanoic acid (PFOA) and perfluorooctanesulfonic acid (PFOS) have been linked to cancer and other adverse health effects, the toxicity of ultrashort-chain PFAS like TFA is “unclear,” said Lohmann. But the finding of hundreds of nanograms per liter of TFA was “a large number – enough to cause health concerns if it was PFOA or PFOS.”
“It is a concerning trend if everything is going up.”
Rainer Lohmann, Professor at the University of Rhode Island’s Graduate School of Oceanography
Most recently, the German government officially proposed to the EU that TFA be classified as reproductively toxic based on rabbit experiments. Germany has issued a drinking water advisory for TFA of 60 mcg/L while the Netherland’s advisory is 2.2 mcg/L, he noted. But there have been competing contentions about the potential toxicity of TFA made by the fluorochemical industry, which refers to a 2020 Assessment Report by the UN Environment Programme that says TFA is unlikely to cause adverse effects in terrestrial and aquatic organisms.
In addition to TFA, Lohmann and his team looked for the presence of other ultrashort-chain PFAS, including perfluoropropionic acid (PFPrA), trifluoromethanesulfonic acid (TFMS), perfluoroethanesulfonic acid (PFEtS) and perfluoropropanesulfonic acid (PFPrS). All but PFPrS were detected, though TFA had by far the highest concentrations.
Lohmann attributed the source of TFA in Rhode Island drinking water to “mostly f-gases” that form TFA in the atmosphere as well as to direct production and the breakdown of polyfluorinated alkyl substances such as fluorinated pharmaceuticals. It also comes from the breakdown of certain fluorinated pesticides and other sources. Still, he said that additional work on TFA would include identifying “potential point sources” and links to other PFAS. It’s also not apparent where the other ultrashort-chain PFAS in the samples originated.
TFA dominates
While TFA had a median concentration of 580ng/L in 66% of the 320 samples, the second most detected chemical, PFPrA (similar to TFA except with one additional fluorinated carbon atom), had a median concentration of 11ng/L in 35% of the samples. TFMS’s median was just 2.8ng/L in 40% of samples, while PFEtS was seen in only 1%.
The highest concentration for TFA, 2,400ng/L, dwarfed the next-highest maximum concentration, 96ng/L for PFPrA. Lohmann concluded that “TFA dominates PFAS in water in the mcg/L range,” adding that the EPA targets long-chain PFAS like PFOA and PFOS at 4ng/L. He noted that ultrashort-chain PFAS are found in water sources around the world, but TFA “is always the highest concentration.”
Lohmann pointed out that the range of TFA in Rhode Island drinking water (no detection to 2,400ng/L) was similar to that found in surface or ground water elsewhere in the world, such as the San Francisco Bay Area (290 to 2,800ng/L), Denmark (0 to 1,500ng/L) and Germany (300 to 3,000ng/L).
He also referenced data showing “a strong increase in TFA in different matrices” – not just water but biological sources like plants, adding, “it is a concerning trend if everything is going up.” Another study, of Denmark groundwater, underscored this trend by finding that “the younger the groundwater, the higher the TFA,” he said.
The 320 drinking water samples analyzed by Lohmann and his team came from a variety of water sources in Rhode Island collected between April 2024 and February 2025. These range from “the really big drinking water providers of Providence to small wells that serve schools somewhere in the not-so-urban areas of the state” and from Massachusetts rivers, he said. The researchers analyzed the samples in a SCIEX X500 QToF liquid chromatography mass spectrometry system.
Government-supplied samples
The samples were provided by Rhode Island government agencies. “We asked them whether there was a chance they could share with us some of the drinking water samples that Rhode Island collects to measure PFAS, and it turns out that they were more than happy [to do so],” Lohmann said.
Rhode Island, the smallest U.S. state in geographic area, has a shallow aquifer in most places, with recharge times of typically a year or so, but there are also some much deeper wells, said Lohmann. The state tests for PFAS because of its own rules and because the EPA “likes these samples on a fairly regular basis,” he said.
However, given that the human effects at this level of TFA are not known, Lohman does not expect the Rhode Island government “to lead the charge on regulating it, just based on the history of my small state.” On the other hand, he added, if enough other New England states push something forward, Rhode Island could follow them. “That’s what typically seems to happen.”
Lohmann was part of a group of 20 PFAS scientists who recently published a statement supporting the OECD definition of PFAS that includes f-gases and TFA and refuting an effort to change that definition. That definition says that with a few exceptions, any chemical with at least one perfluorinated methyl group (–CF3) or one perfluorinated methylene group (–CF2–) is a PFAS. This includes thousands of compounds, hundreds that are produced and around 10 that are monitored.
TFA has been detected in other parts of the U.S. A 2024 study of PFAS in rainwater in Michigan found TFA to be the largest contributor in two out of three sample locations and one of the largest in the third location. A 2023 study of households in Indiana found TFA in samples of dust, drinking water, human blood serum and, to a lesser degree, urine.
