Unfolding the structural stability of nanoalloys via symmetry-constrained genetic algorithm and neural network potential

Abstract The structural stability of nanoalloys is a challenging research subject due to the complexity size, shape, composition, and chemical ordering. genetic algorithm popular global optimization method that can efficiently search for ground-state nanoalloy structure. However, suffers from three significant limitations: efficiency accuracy energy evaluator algorithm’s efficiency. Here we describe construction neural network potential intended rapid accurate predictions Pt-Ni various sizes, shapes, compositions. We further introduce symmetry-constrained significantly improves viability realistic size nanoalloys. combination two allows us explore space homotops compositions consisting up 4033 atoms quantitatively report interplay composition on dominant ordering patterns.