The OER is one of the main processes occurring in metal-air batteries, solar water splitting and solar-powered fuel cells in general. It can be used to produce oxygen from water and as such could be a clean and renewable way to generate energy. Unfortunately, the reaction is rather slow and researchers are trying to find efficient catalyst materials to help speed it up. Although oxides based on ruthenium and iridium are good in this respect, these metals are expensive, which means they cannot realistically be used to make devices on the large scale.

First-row (3d) transition-metal oxides could be good alternatives to Ru and Ir but they are usually made using electrodeposition techniques that cannot produce highly porous structures with large numbers of electrochemically active sites, needed for reactions like the OER.

Iron, cobalt and tungsten mixed together homogenously

Researchers led by Edward Sargent at the University of Toronto and Aleksandra Vojvodic of Stanford University and SLAC say they have now discovered a new class of efficient catalyst for the OER based on the metals iron, cobalt and tungsten mixed together homogenously in an oxy-hydroxide framework that do have a nanoporous structure with large numbers of electrochemically active sites. The gelled multimetal structures can also be prepared via a simple room-temperature process, so they could easily be produced in large quantities.

The team began by dissolving precursors for all the metals in a polar organic solvent to form a homogenous solution. "By choosing suitable reagents, auxiliary agents and reaction conditions, we coordinated the hydrolysis of the different metal precursors, such that they were programmed to occur at similar rates to one another," explains Sargent. "This approach allowed different metal oxy-hydroxides to form simultaneously in an atomically homogenous distribution. Combining the metals in this way is challenging, since metals and metal oxides tend to phase-separate in crystalline oxides."

"The key advance at an experimental level was to produce metal oxides in which multiple metals could be homogeneously dispersed," he tells "This approach allows the active site metal oxy-hydroxides to be modulated by additionally including the tungsten. Only by developing the sol gel-based method for making such catalysts, were we able to modulate energy levels in one set of atoms (the active sites) using another (the modulator, tungsten).

Predictive materials science

"The other key (and closely related) point was that the beautiful theoretical work by Vojvodic's team allowed us to account for this modulation in full detail, and, in particular, how it specifically impacts the chemistry of water splitting. We believe that this kind of predictive materials science will be an ever-more powerful tool in the discovery of future generations of even more efficient and versatile catalysts."

"Using state-of-the-art computational tools and powerful computers, we were able to understand, on the atomic scale, why and how the multimetal component oxide works," explains Vojvodic. "For example, an active site, where the OER takes place, containing cobalt, iron and, importantly, tungsten was found to be crucial."

The researchers say they are now busy trying to develop more multimetal compounds with optimized OER energetics for use in water splitting and perhaps even for reactions that chemically reduce carbon dioxide.

The present work is detailed in Science DOI: 10.1126/science.aaf1525.