Composition and activity of CuAu/Pt core/shell electrocatalytic nanoalloys using first-principles calculations and machine learning

Abstract

Noble metal nanoparticles continue to be of great interest to the catalysis community for their characteristics, and extensive research has been conducted into bimetallic nanoparticles involving these metals. Inclusion of another metal to form trimetallic particles could further enhance the characteristics of these particles by increasing the possible metal compositions and the associated interactions within a particle. With the increase in possible metal combinations and compositions, theoretical methods can provide assistance in understanding reaction mechanisms or pre-screen metal combinations for their catalytic activity. Machine learning methods can additionally reduce the computational cost of such calculations, but the lack of extensive databases for trimetallic particles hampers their application, requiring the calculationof accurate databases using first-principles methods.

In this work, the composition and formation of a trimetallic CuAu/Pt surface, the adsorption of O, H, C and CO on this surface, and the dissociation of O2 and H2 was investigated using density-functional theory and the nudged elastic band method. The Pt-coated CuAu surface was found to form at least one monolayer of Pt in various configurations, ranging from highly ordered to disordered with similar stability. The adsorption of CO and O2 was found to occur on the top and bridge sites, respectively, whilst H2 was found to adsorb dissociatively with no discernible energy barrier. The dissociation of oxygen on the surface resulted in molecular oxygen adsorbing on the bridge site and dissociating to atomic oxygen in the hollow site with an energy barrier of 0.22 eV.