If you're despairing at recent reports that Earth's water sources have been thoroughly infested with per- and polyfluoroalkyl substances (PFAS), making even rainwater unsafe to drink, there's a spot of good news.

Chemists at UCLA and Northwestern University have developed a simple way to break down almost a dozen types of these nearly indestructible "forever chemicals" at relatively low temperatures with no harmful byproducts.

In a paper published recently in the journal Science, the researchers show that in water heated to just 176 to 248 degrees F, common, inexpensive solvents and reagents severed molecular bonds in PFAS that are among the strongest known and initiated a chemical reaction that "gradually nibbled away at the molecule" until it was gone, says UCLA distinguished research professor and co-corresponding author Kendall Houk.

The simple technology, the comparatively low temperatures and the lack of harmful byproducts mean there is no limit to how much water can be processed at once, Houk adds. The technology could eventually make it easier for water treatment plants to remove PFAS from drinking water.

Leading PFAS to the guillotine

Northwestern chemistry professor William Dichtel and doctoral student Brittany Trang noticed that while PFAS molecules contain a long "tail" of stubborn carbon-fluorine bonds, their "head" group often contains charged oxygen atoms, which react strongly with other molecules. Dichtel's team built a chemical guillotine by heating the PFAS in water with dimethyl sulfoxide (DMSO) and sodium hydroxide, or lye, which lopped off the head and left behind an exposed, reactive tail.

"That triggered all these reactions, and it started spitting out fluorine atoms from these compounds to form fluoride, which is the safest form of fluorine," Dichtel says. "Although carbon-fluorine bonds are super-strong, that charged head group is the Achilles' heel."

But the experiments revealed another surprise: The molecules didn't seem to be falling apart the way conventional wisdom said they should.

To solve this mystery, Dichtel and Trang shared their data with collaborators Houk and Tianjin University student Yuli Li, who was working in Houk's group remotely from China during the pandemic. The researchers had expected the PFAS molecules would disintegrate one carbon atom at a time, but Li and Houk ran computer simulations that showed two or three carbon molecules peeled off the molecules simultaneously, just as Dichtel and Tang had observed experimentally.

The simulations also showed the only byproducts should be fluoride, carbon dioxide and formic acid, which is not harmful. Dichtel and Trang confirmed these predicted byproducts in further experiments.

"This proved to be a very complex set of calculations that challenged the most modern quantum mechanical methods and fastest computers available to us," Houk says. "Quantum mechanics is the mathematical method that simulates all of chemistry, but only in the last decade have we been able to take on large mechanistic problems like this, evaluating all the possibilities and determining which one can happen at the observed rate."

Li, Houk says, has mastered these computational methods, and he worked long distance with Trang to solve the fundamental but practically significant problem.

The current work degraded 10 types of perfluoroalkyl carboxylic acids (PFCAs) and perfluoroalkyl ether carboxylic acids (PFECAs), including perfluorooctanoic acid (PFOA). The researchers believe their method will work for most PFAS that contain carboxylic acids and hope it will help identify weak spots in other classes of PFAS. They hope these encouraging results will lead to further research that tests methods for eradicating the thousands of other types of PFAS.

The study, "Low-temperature mineralization of perfluorocarboxylic acids," was supported by the National Science Foundation.