BHSMS: a new software for searching the most stable structure of clusters based on the improved basin hopping algorithm.

To explore the relationship between the structures and properties of clusters, global-minimum (GM) structures of clusters should be determined in cluster science. In this work, a new efficient global-minimum search package named BHSMS was developed to search for the most stable structure of both atomic and molecular clusters. In this program, the structure relaxation (SR) algorithm, structure similarity recognition (SSR) algorithm, planarity recognition (PR) algorithm, and the constraints of molecules for rotation (CMR) algorithm, combined with the improved basin-hopping (BH) algorithm, were implemented. Furthermore, the interfaces of BHSMS and several quantum chemistry software such as xTB, MOPAC, DFTB+, Gaussian, ADF, ORCA, CHARMM, NWChem, Psi4, PySCF, CP2K, VASP and so on were also developed for the convenience of users. All codes of the BHSMS package are written in Python language and have remarkable compatibility with other quantum-chemistry software. The accuracy and efficiency of BHSMS were tested in metal and non-metal clusters and atomic and molecular clusters, such as Au6, Au20, B8, B40, C20, C60, Be2O2, Si2O2, (CO2)3, (CO2)6, (NH3)4, (NH3)5, (H2O)5, (H2O)8, (HNO3)3(NH3)3 and (HNO3)3(NH3)3(H2O)2 clusters. A detailed overview of the BHSMS codes and several applications are presented in this article.