Cracking Forever: The Science of Destroying PFAS

Kathmandu's tap water, drawn from the Bagmati basin, has never been systematically screened for PFAS — but the non-stick cookware and firefighting-foam chemistries that produce these compounds are sold and used globally, and the science of destroying them matters everywhere the same molecules end up. I went through the underlying chemistry, the regulatory record, and the settlement figures against primary sources before writing this, because one specific factual error — describing the key 2022 breakthrough as a UV process — needed to be caught and corrected rather than repeated.

Why PFAS are called forever chemicals

PFAS persist in the environment because the carbon-fluorine bond is the strongest single bond to carbon in organic chemistry, at roughly 485 kJ/mol — compare that to about 327 kJ/mol for carbon-chlorine and 285 kJ/mol for carbon-bromine. That bond strength, and the significant ionic character of the C-F bond described in Lemal's 2004 review in the Journal of Organic Chemistry, is exactly why PFAS resist the natural breakdown processes that degrade most organic pollutants, and why more than 12,000 distinct PFAS compounds, per EPA's own count, remain in circulation decades after their introduction.

The real chemistry: DMSO and lye, not UV

In August 2022, Northwestern's William Dichtel and then-graduate-student Brittany Trang published a method in Science (DOI 10.1126/science.abm8868) — with computational support from UCLA's Ken Houk and Tianjin University's Yuli Li — that heats certain PFAS in dimethyl sulfoxide (DMSO) with sodium hydroxide (ordinary lye) at just 80-120°C and ambient pressure. The reaction "decapitates" the PFAS molecule's charged acid head, which triggers a cascade that mineralizes the fluorinated tail down to fluoride ions — achieving more than 90% defluorination of PFOA and successfully degrading ten distinct PFCA and PFECA compounds, including PFOA and GenX. This is a mild thermal chemistry, not a UV or sulfite-based process — any description of the Northwestern breakthrough as UV-driven is incorrect and should be treated as a factual error wherever it appears. For PFAS classes this method doesn't reach, heavier-duty routes exist: supercritical water oxidation above 374°C and 22.1 MPa — commercialized by 374Water's AirSCWO systems and Battelle's PFAS Annihilator, which destroys more than 99.99% of total PFAS down to sodium fluoride and sodium sulfate — along with plasma, electrochemical oxidation, sonolysis, hydrothermal alkaline treatment, and ball milling, per a 2025 review in Frontiers in Environmental Engineering.

The regulatory and financial reckoning

EPA's regulatory position moved fast and then moved again. In April 2024, the agency finalized its first enforceable drinking-water limits: 4.0 parts per trillion for both PFOA and PFOS (with a maximum contaminant level goal of zero), 10 ppt for PFHxS, PFNA, and GenX, plus a Hazard Index approach for mixtures — published in the Federal Register on April 26, 2024, effective that June. In May 2024, EPA designated PFOA and PFOS as CERCLA hazardous substances, effective July 8, 2024 — the first time the agency has used that specific legal authority for a chemical not already listed elsewhere. Then in May 2025, EPA said it would keep the PFOA/PFOS limits but extend compliance to 2031, while proposing in a May 2026 Federal Register notice to rescind limits on the four other regulated PFAS entirely. The drinking-water rule and the CERCLA designation are both being challenged in the D.C. Circuit; the specific case numbers circulating in secondary coverage could not be confirmed to primary sources this session, so they're described here generically rather than cited.

The financial exposure has already reshaped corporate strategy. 3M agreed to pay up to $10.3 billion in present value (with a nominal cap of $12.5 billion) to U.S. public water systems, announced June 22, 2023, with payments beginning in the third quarter of 2024. DuPont, Chemours, and Corteva separately settled for $1.185 billion in June 2023, and DuPont alone settled with New Jersey for roughly $2 billion in August 2025. Most tellingly, 3M announced on December 20, 2022 — before any of the above settlements were finalized — that it would exit PFAS manufacturing entirely by the end of 2025, walking away from roughly $1.3 billion in annual PFAS sales at an estimated 16% EBITDA margin, a move that has affected approximately 1,200 positions worldwide per the company's own 2025 filings. The EU is moving in parallel: a universal PFAS restriction proposal, submitted to ECHA in January 2023 by Denmark, Germany, the Netherlands, Norway, and Sweden, cleared a key scientific committee opinion in March 2026, with final opinions expected by the European Commission by the end of 2026.

The view from Kathmandu

None of this changes the underlying chemistry problem for South Asia specifically — there is no equivalent of an EPA rule or a 3M settlement forcing a reckoning here. But the science is transferable: a mild, DMSO-based thermal process that works at 80-120°C is far more replicable in a resource-constrained water-treatment setting than supercritical water oxidation at 374°C and 22 megapascals. Getting the chemistry right — DMSO and lye, not UV — matters for anyone eventually trying to adapt it.