Palladium Membrane Breaks Down PFAS at Up to 83% Defluorination

A catalytic membrane developed at Arizona State University captured and chemically destroyed six regulated PFAS compounds, reaching 96% removal and 83% defluorination (cleavage of carbon–fluorine bonds) over 58 days of continuous operation. The study, published in Proceedings of the National Academy of Sciences, addresses a gap that matters for any lab or utility running PFAS remediation: conventional treatments concentrate the contaminants but leave the destruction problem unsolved.

The work was led by Bruce Rittmann, PhD, director of ASU's Biodesign Swette Center for Environmental Biotechnology, with collaborators at Rice University, Tongji University, the National University of Singapore, and Nanjing University. The paper is titled "Catalytic Reactive-Hydrogen Membranes for Rapid Capture and Hydrodefluorination of Six PFAS of Regulatory Concern."

"PFAS are hard to break down, and normal processes that we use don't work," Rittmann said.

Why conventional treatment stops short

The persistence of per- and polyfluoroalkyl substances comes from their carbon–fluorine bonds, among the strongest single bonds in organic chemistry. Activated carbon, ion exchange, and membrane filtration can pull PFAS out of water, but none of them destroys what they capture. The result is a contaminated waste stream — spent carbon, resin, brine — that still requires disposal or further treatment. For portfolio managers weighing PFAS liabilities, that transfer of risk is the operational cost that never disappears.

The ASU system attacks that step directly. It brings PFAS and reactive hydrogen together on palladium-coated membrane fibers under ambient conditions.

How the mechanism works

Hydrogen gas travels through hollow membrane fibers to a palladium catalyst, where it generates two forms of active hydrogen with distinct roles. Surface hydrogen adsorbs and captures PFAS molecules. Subsurface hydrogen, stored as beta-phase palladium hydride (a distinct crystalline form of the metal with hydrogen embedded in its lattice), supplies the reducing power for hydrodefluorination — the reaction that replaces fluorine atoms with hydrogen.

"The palladium attracts the PFAS molecule, and the palladium attracts the hydrogen and puts them together," Rittmann said.

Solid-state analyses confirmed formation of the palladium hydride phase and the associated lattice expansion. In the paper, the researchers described their findings as "a mechanistic framework for how active hydrogen is generated, stored, and deployed in catalytic membranes." That framework, rather than the specific device, may be the transferable result for other groups engineering catalytic separation systems.

What did the 58-day test show?

The team tested six compounds on regulatory watchlists: PFOA, PFOS, PFNA, PFHxS, PFBS, and GenX. The headline figures — up to 96% removal and 83% defluorination — came from continuous operation across that window. Performance was not uniform:

• Long-chain PFAS were removed and defluorinated faster than short-chain analogs.
• GenX resisted treatment most strongly, which the authors attributed to its ether linkage, weaker adsorption to the catalyst, and higher average carbon–fluorine bond dissociation energy.
• Even so, the membrane defluorinated GenX, producing 2-propoxypropanoic acid as the breakdown product.

The short-chain result deserves attention from anyone specifying PFAS treatment: shorter compounds are typically harder to adsorb, so a system that clears PFOA will not necessarily clear PFBS at the same rate.

What stands between the lab and deployment?

Two constraints frame the commercial picture. First, cost. "Palladium is expensive," Rittmann said, and reducing the metal loading is a stated engineering priority before the system scales. Second, positioning. The researchers see an initial application not as a replacement for existing removal systems but as a destruction step for their concentrated waste streams — a bolt-on that closes the loop on spent filtration media and brines.

The group is also developing hybrid membranes that pair palladium with pollutant-degrading bacteria, though the current study used palladium alone. Whether the hydride-driven mechanism survives contact with real waste-stream chemistry — competing solutes, fouling, variable pH — will determine whether the 83% figure holds outside the bench.