Until 1985, gold was widely believed to be chemically inert. But once researchers discovered that nano-sized gold particles can act as remarkable and selective catalysts, a world of possibility opened up.
Today, gold is used in many industrial catalytic processes, such as the removal of carbon monoxide from exhaust at low temperatures or even the replacement of mercury-based catalysts in the production of PVC plastics — both good steps for the environment. However, gold is expensive and scarce.
Virginia Tech researchers aim to maximize the power of every atom of the particles without relying on time-consuming trial and error. This long-lasting problem may have a solution in the near future, thanks to the recently published work of Hongliang Xin, an assistant professor in the Department of Chemical Engineering in the College of Engineering at Virginia Tech, and Xianfeng Ma, a postdoc fellow in Xin’s research group.
In a new study published in the peer-reviewed Physical Review Letters (Phys. Rev. Lett. 118, 036101), Xin and Ma propose a new model that can rationalize reactivity trends of a variety of gold nanoparticles with different sizes, shapes, and compositions — meaning, the model can potentially predict just the right formula of gold catalysts to achieve a desired outcome for a given chemical reaction.
According to Xin, this model demonstrates that s-electrons, which are not permanently attached to any atoms like localized d-electrons, govern the reactivity of surface atoms. This challenges the conventional wisdom of the standard d-band model, which is the theory widely used to explain catalytic activity.
“This model can be easily understood through an analogy to ballroom dancing: if you are dancing with many friends who are attractive to you, you are less likely to interact with strangers,” Xin said. “The same can be said of catalyst atoms, which will be more active to reactants if they are not surrounded by many attractive neighboring atoms.”
Xin’s research group focuses on computational modeling for energy solutions, which is mainly supported by Advanced Research Computing at Virginia Tech.
“With respect to computational modeling, this is incredibly important because catalytic processes are complex and the information at that smallest length and fastest time scales is not easily accessible with experimental techniques,” Xin said. “Our work and many others in the field may offer unique capabilities to discover and design better catalysts through an understanding of structure-reactivity trends of model catalysts in computers.”
The finding has significant practical applications, especially in chemical industry and renewable energy technologies. Because of the model’s general nature, it can be adapted to use with other catalytic materials, such as nickel, platinum, and palladium, which are commonly used in industrial catalytic processes.
Xin’s work is funded by the American Chemical Society Petroleum Research Fund, the National Science Foundation, the Army Research Office, and the Institute for Critical Technology and Applied Science.
Written by Erica Corder