Revisiting the stable structures of gold clusters: Au n (n = 16–25) by artificial neural network potential

Abstract Identifying the stable structures of gold (Au) clusters is a huge challenge in cluster science. In this work, we have searched the ground-state structures of neutral Au n ( n = 16–25) clusters using the potential of an artificial neural network (ANN) trained with density functional theory (DFT) data. Compared with the DFT data, the root mean square error of binding energy predicted by the ANN potential is about 8.66 meV/atom. Applying the ANN potential to search the ground-state structures by comprehensive genetic algorithm, we have found several new candidates of Au 18 , Au 22 , and Au 23 , which have not been previously reported. Au 18 has a hollow cage structure, whereas Au 22 and Au 23 are flat cage structures. From the electronic analysis, we elucidate the stability mechanism of the newly found structures that are associated with the electronic shell closure of superatomic orbitals. Additonally, we also clarified how to clean a database to train an efficient ANN potential in detail. Overall, this work proves that applying machine learning to the description of atomic interactions can accelerate the search of ground-state structures of clusters and help to find new candidates for stable cluster structures.