Abstract:

Small, supported nanoparticles (NPs) with 1-5 nm diameter play an important role in catalysis and tuning metal nanoparticle-support interactions is a fundamental method to control catalytic reactions. Experimental techniques provide limited insights into how interactions with the support change the structure of nanoparticles whereas density functional theory calculations suffer from system size constraints and rely on idealized constructions. I will show in this presentation that the simulation of supported nanoparticles is possible in quantitative agreement with benchmark microcalorimetric measurements for adhesion, chemical potential, and heat of adsorption. Analysis of coordination number, strain, and crystallinity in optimized Ag nanoparticles supported on graphene/Ni(111) and MgO(100) revealed surface features inconsistent with previously assumed idealized models. Our results demonstrated that the widely used Wulff construction is not satisfactory structural models of supported Ag nanoparticles. The approach developed here based on machine learning interatomic potentials opens up an avenue to achieve modeler’s dreams of simulating industrially relevant supported nanoparticles directly under catalytic conditions.