Cluster classification by chemi-topology

K Nishio, COMPUTER PHYSICS COMMUNICATIONS, 286, 108659 (2023).

DOI: 10.1016/j.cpc.2023.108659

The atomic-scale structure of disordered materials like liquids and glasses has often been studied using the Voronoi tessellation method. This method divides the space containing atoms into regions called Voronoi polyhedra. Each polyhedron contains one atom. When two polyhedra share a common face, the atoms in the polyhedra are regarded as being connected by a bond. The bonded atoms and their common neighbors form a bipyramidal cluster, whose topology is related to the shape of the common face: when a common face is an n-gon, the cluster is an n-gonal bipyramid. Therefore, the topological order of a material can be characterized by the frequency distribution of the shapes of common faces. However, this approach does not tell anything about the chemical order: how different types of atoms are arranged in the clusters. In this paper, we propose a method for classifying bipyramidal clusters by chemi-topology, namely according to not only the topology of the cluster but also the chemical arrangement in the cluster. Since a bipyramidal cluster is made up of edge-sharing tetrahedral clusters, we also propose a method for classifying tetrahedral clusters. The function to characterize the atomic structures of materials using the proposed methods has been added to the Vorotis software. We describe the usage of this function. As a demonstration, we apply the proposed methods to characterize the structural difference between the liquid and glass states of a model alloy. (c) 2023 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

Return to Publications page