Construction of crystal structure prototype database: methods and applications

Chuanxun Su,Jian Lv,Quan Li,Hui Wang,Lijun Zhang,Yanchao Wang,Yanming Ma

DOI: https://doi.org/10.1088/1361-648X/aa63cd

2017-02-07

Abstract:Crystal structure prototype data have become a useful source of information for materials discovery in the fields of crystallography, chemistry, physics, and materials science. This work reports the development of a robust and efficient method for assessing the similarity of structures on the basis of their interatomic distances. Using this method, we proposed a simple and unambiguous definition of crystal structure prototype based on hierarchical clustering theory, and constructed the Crystal Structure Prototype Database (CSPD) by filtering the known crystallographic structures in a database. With similar method, a program Structure Prototype Analysis Package (SPAP) was developed to remove similar structures in CALYPSO prediction results and extract predicted low energy structures for a separate theoretical structure database. A series of statistics describing the distribution of crystal structure prototypes in the CSPD was compiled to provide an important insight for structure prediction and high-throughput calculations. Illustrative examples of the application of the proposed database are given, including the generation of initial structures for structure prediction and determination of the prototype structure in databases. These examples demonstrate the CSPD to be a generally applicable and useful tool for materials discovery.

Materials Science

What problem does this paper attempt to address?

### Problems the Paper Aims to Solve

The paper aims to address the issue of constructing a crystal structure prototype database in the field of materials science. Specifically, the authors propose a method based on interatomic distances to evaluate structural similarity and, on this basis, define simple and clear crystal structure prototypes. Using this method, they have constructed a database called the "Crystal Structure Prototype Database" (CSPD).

### Main Contributions

1. **Method Development**: A method based on hierarchical clustering theory is proposed to evaluate the similarity between crystal structures and define crystal structure prototypes accordingly.
2. **Database Construction**: By filtering existing crystallographic data, the CSPD was constructed, which contains a large number of deduplicated crystal structure prototypes.
3. **Statistical Analysis**: Detailed statistical analysis of the data in CSPD was conducted, providing information on the distribution of crystal structure prototypes of different chemical compositions.
4. **Application Examples**: Demonstrated the application of CSPD in material prediction and high-throughput computation, such as generating initial structures for structure prediction and determining the prototype of a given structure.

### Objectives

The overall objective is to accelerate the process of material discovery and improve the efficiency of structure prediction by constructing an efficient, deduplicated crystal structure prototype database. This helps to reduce redundant information and speed up material analysis and discovery.