Surface steric effect in heterogeneous catalysis as the origin of the high activity induced by strong metal-support interactions.

Supported nanoparticles offer unique opportunities for enhancing catalytic activity via strong metal-support interaction (SMSI). Even with state-of-the-art experimental techniques, the atomistic origin of this enhancement remains unclear, while current computational limitations make it difficult to provide a theoretical explanation. This study focused on clarifying the atomistic mechanism of SMSI by investigating N dissociation from Ru/LaCeO catalysts. Fast calculations using a neural network potential enabled the analysis of 328 complex nanoparticle models with varying degrees of site heterogeneity, encompassing over 25,768 adsorption sites. Our findings were validated against infrared spectra and helped identify catalyst configurations with enhanced catalytic activity, driven by SMSI. Specifically, the dissociation path of N molecules sandwiched between decoration cations on a nanoparticle near the support exhibited a low activation barrier. Our theoretical approach represents a major advancement in bridging the gap between simulation and empirical data and in our understanding of complex supported nanoparticle catalysts.