Stability and Phase Transformations in Au–Pd Nanoparticles Studied by Means of Combined Monte Carlo and Molecular Dynamics Simulations

Surface morphology, element distribution, shape, and stability of bimetallic nanoparticles (NPs) play crucial roles in determining their catalytic activity and performance. However, ab initio molecular dynamics (MD) can treat only systems with tens or few hundreds of atoms on moderate time scales. In this study, we investigate the structural stability and phase behaviors of Au–Pd NPs with sizes up to several thousand atoms using combined Monte Carlo (MC) and MD simulations. By employing realistic ReaxFF interatomic potential models [J. Phys. Condens. Matter 2023, 35, 065901], we examine the effects of composition, temperature, and NP size on phase separation, alloying, and surface segregation phenomena in NPs. Our potential allows treating Au–Pd alloy NPs of arbitrary shape, size, and composition. ReaxFF parameterization can be further extended to include interactions with organic molecules. The simulations demonstrate the dynamical transition from the cuboctahedral to the icosahedral shape of Au–Pd NPs and reveal that as the Pd content increases, the NPs tend to exhibit phase separation into Au-rich and Pd-rich regions. The volumes of the resulting monometallic regions and the extent of segregation depend on the size of the NPs and the temperature. We observed the lability character of surface adatoms and their migration, which explains why smaller NPs demonstrate a higher propensity for alloying and surface segregation. The results provide insights into the structural stability and phase transformations in Au–Pd NPs, shedding light on the factors that influence their properties upon annealing and guide the rational design of bimetallic catalysts.