Forever Means Forever: PFAS and the Limits of Regulating One Molecule at a Time

In January 2026, new European rules on per- and polyfluoroalkyl substances in drinking water came into full application across all member states — the first time systematic, harmonized monitoring of these compounds has been required across the bloc. Under the recast Drinking Water Directive, member states must keep the sum of twenty specified PFAS below 0.1 micrograms per liter and total PFAS below 0.5, and must act when those limits are exceeded: adding treatment, restricting supply, or closing contaminated wells. It is a genuine regulatory milestone. It is also, on its own, a monitoring regime rather than a solution.

PFAS earned the label "forever chemicals" honestly. The carbon-fluorine bond is among the strongest in organic chemistry, which is precisely why these compounds work so well in non-stick coatings, waterproof fabrics, firefighting foams, food packaging, and semiconductor manufacturing. The same property means they do not meaningfully degrade in the environment. They move through groundwater, accumulate in soil and organisms, and have been detected in human blood serum in populations worldwide. Unlike a spill or a smokestack plume, the contamination has no expiry date.

The regulatory history here is a case study in what happens when you chase compounds one at a time. Restrictions on the best-studied long-chain molecules — PFOA and PFOS — produced not an exit from fluorochemistry but a migration within it. Manufacturers substituted shorter-chain and structurally modified alternatives that fell outside the existing rules, and researchers then spent years demonstrating that several of these were also mobile, persistent, and toxicologically concerning. The term for this pattern is regrettable substitution, and PFAS is its textbook example. There are thousands of compounds in the family; regulating them individually is a race the regulator cannot win.

That is the argument behind the universal restriction proposal submitted to the European Chemicals Agency in 2023 by five national authorities, which would treat PFAS as a class rather than as a list. In March 2026, ECHA's scientific committees signaled support for the broad restriction while proposing extensive time-limited derogations for sectors where substitutes do not yet exist. The compromise is instructive: even a class-based approach concedes that some uses — certain medical devices, some industrial applications — currently have no alternative. The question is whether derogations function as a bridge or as a permanent exemption.

The costs of delay are not evenly distributed, and they are not recoverable. The European Food Safety Authority based its tolerable intake for key PFAS on immune effects in children, including reduced antibody response to vaccination. Prenatal exposure has been associated in cohort studies with lower birth weight and altered immune development. Those windows close. A cleanup completed in 2040 does not undo exposure that occurred in 2026, which makes PFAS an unusually clear case where the precautionary principle is not merely a philosophical preference but an arithmetic one.

There is a broader lesson here that extends well past fluorochemistry, and it echoes what we have written about microplastics. Both are cases where a material's most commercially valuable property — durability, persistence, resistance to breakdown — is identical to its environmental liability. We selected for permanence and are now surprised by it. Until chemical regulation evaluates persistence as a hazard in its own right, rather than waiting for toxicity to be demonstrated compound by compound, we will keep inventing substances whose defining virtue is that they will outlast the institutions that approved them.