Abstract
Using the high-entropy alloys (HEAs) CoCuGaNiZn and AgAuCuPdPt as starting points we provide a framework for tuning the composition of disordered multi-metallic alloys to control the selectivity and activity of the reduction of carbon dioxide (CO2) to highly reduced compounds. By combining density functional theory (DFT) with supervised machine learning we predicted the CO and hydrogen (H) adsorption energies of all surface sites on the (111) surface of the two HEAs. This allowed an optimization for the HEA compositions with increased likelihood for sites with weak hydrogen adsorption{to suppress the formation of molecular hydrogen (H2) and with strong CO adsorption to favor the reduction of CO. This led to the discovery of several disordered alloy catalyst candidates for which selectivity towards highly reduced carbon compounds is expected, as well as insights into the rational design of disordered alloy catalysts for the CO2 and CO reduction reaction.