(Invited) Genetic Algorithms and DFT for Accelerated Design of Nanoalloys

Abstract

Binary and ternary alloys often show improved activity, selectivity or durability as electrocatalysts compared to elemental catalysts. For practical applications, nanoparticle catalysts are required to maximize utilization of precious metals such as platinum. Theoretical simulations based on density functional theory (DFT) have helped rationalize experimental observations and suggested new design directions for experimentalists.1 Commonly, theoretical studies are, however, limited to small clusters or extended surfaces under periodic boundary conditions. Although DFT simulations of nanoparticles with hundreds of atoms have become possible in recent years,2 prediction of properties such as morphology, atomic distribution and catalytic activity of nanoalloys remains a challenge. Here, we report recent theoretical work on the global optimization of atomic distribution within nanoalloy catalysts using genetic algorithms and we predict their activity using density functional theory. We investigate the activity of Pt-Pd-Au binary and ternary alloys for the oxygen reduction reaction (ORR) and the stability of these ternary alloys under ORR conditions. Our findings suggest subsurface Pd, in binary Pt-Pd films, results in improved activity over pure Pt, while addition of small amounts of Au will hinder Pt dissolution and might also enhance activity. We study the ORR activity of 2 nm ternary Pt-Au-M Mackay icosahedral core-shell nano-particles in order to identify core elements, which increase activity and particle stability through strain and ligand effects. Acknowledgements Financial support from the European Commission under the FP7 Fuel Cells and Hydrogen Joint Technology Initiative grant agreement FP7-2012-JTI-FCH-325327 for the SMARTcat project is gratefully acknowledged.