Artificial neural network potential for Au₂₀clusters based on the first-principles

The search of ground-state structures (GSSs) of gold (Au) clusters is a formidable challenge due to the complexity of potential energy surface (PES). In this work, we have built a high-dimensional artificial neural network (ANN) potential to describe the PES of Au₂₀clusters. The ANN potential is trained through learning the GSS search process of Au₂₀by the combination of density functional theory (DFT) method and genetic algorithm. The root mean square errors of energy and force are 7.72 meV atom^-1and 217.02 meV Å^-1, respectively. As a result, it can find the lowest-energy structure (LES) of Au₂₀clusters that is consistent with previous results. Furthermore, the scalability test shows that it can predict the energy of smaller size Au_16-19clusters with errors less than 22.85 meV atom^-1, and for larger size Au_21-25clusters, the errors are below 36.94 meV atom^-1. Extra attention should be paid to its accuracy for Au_21-25clusters. Applying the ANN to search the GSSs of Au_16-25, we discover two new structures of Au₁₆and Au₂₁that are not reported before and several candidate LESs of Au_16-18. In summary, this work proves that an ANN potential trained for specific size clusters could reproduce the GSS search process by DFT and be applied in the GSS search of smaller size clusters nearby. Therefore, we claim that building ANN potential based on DFT data is one of the most promising ways to effectively accelerate the GSS pre-screening of clusters.