Illustration of the way the electronic configuration of metal ions can control the activity of metal oxides for oxygen reduction, varying it by a factor of at least 10,000 times. This can serve as a design principle (symbolized as a “volcano plot”) to screen metal oxide candidates and accelerate the development of efficient fuel cells, metal-air batteries and other energy storage technologies. 
Credit: Jin Suntivich, Eva Mutoro and Yang Shao-Horn; MIT News.

Today’s online MIT News reports on work being done by associate professor Yang Shao-Horn on a simple way to identify materials capable of being effective catalysts for oxygen reduction, which is the key electrochemical reaction in solid oxide fuel cells and in metal–air batteries. This new work could change the costly trial-and-error approach to catalyst development.

Palladium and platinum are known to have good catalytic properties, however, like all noble metals, they are pricey and scarce. If alternative and renewable energy technologies are going to become affordable, commodity-scale contributors to the energy portfolio, the materials used will need to be commodities, too.

Shao-Horn’s work builds on earlier (and ongoing) work by Jens Nørskov to understand the relationship between the electronic structure of the catalyst material and its efficacy as a catalyst. A good catalyst must bond with oxygen just enough – neither too much nor too little. Nørskov’s team was able to correlate the average energy of a catalyst’s outermost electron to the binding energy of oxygen to metal surfaces. That is, by looking only at the distribution of electrons in the orbitals that bond metal to oxygen, it is possible to predict which metal-oxide systems are likely to be effective catalysts.

Shao-Horn says in the story, the Nørskov work provides “…a theoretical framework and experimental evidence that explains why” some catalysts work better than others. Shao-Horn’s work, published June 13 in Nature Chemistry, evaluates perovskite oxide catalysts with transition metal B-site ions. (See “Design principles for oxygen-reduction activity on perovskite oxide catalysts for fuel cells and metal-air batteries,” doi:10.1038/nchem.1069)

In the paper, Shao-Horn’s team propose a mechanism for the oxygen reduction reaction, where hydroxide ions are displaced by peroxide ions, followed by the formation of a surface oxide, and finally, regeneration of surface hydroxide. The study was able to show that the bonding orbital (here the sigma-star orbital) and the strength of the transition metal-oxygen covalent bond influence competition between the steps in the ORR. Transition metals investigated in the study include Cr, Mn, Fe, Co, Ni and mixed compounds.

Comparing the electronic configuration of the B-site transition metal ions and their catalytic activity yields a volcano-shaped plot with a sharp peak of high performance and steeply sloped sides signifying poor catalytic performance. The range of performance between the “base” of the volcano-plot and the peak can be a factor of at least 10,000.

This work falls more or less into the category of integrated computational materials design, whereby materials development is accelerated by understanding how structure defines properties, and through processing, the design of formulations and applications. Research advances like this one will be critical to addressing the materials problems in this century’s Energy Race.