GenSym: A tool for random symmetric structure generation and its application in crystal structure prediction...

Area of Science:

• Materials Science
• Crystallography
• Computational Chemistry

Background:

• Crystal structure prediction (CSP) is crucial for materials discovery.
• The efficiency of CSP methods heavily relies on the quality of initial structure generation.
• Incorporating symmetry, prevalent in real materials, can enhance CSP efficiency.

Purpose of the Study:

• To present an open-source package for random symmetric crystal structure generation.
• To provide flexible control over various generation parameters.
• To improve the efficiency of crystal structure prediction frameworks.

Main Methods:

• Developed an open-source package for symmetric structure generation.
• Supported bulk, low-dimensional, and molecular crystal generation.
• Enabled control over cell shape, bond length, symmetry settings, Wyckoff positions, and density.

Main Results:

• The package allows generation based on user-defined constraints and prototype databases.
• Integration with CSP frameworks demonstrated a substantial improvement in search efficiency.
• The tool facilitates accelerated novel materials design and discovery.

Conclusions:

• The developed package offers a significant advancement in symmetric structure generation for CSP.
• Its flexibility and integration capabilities accelerate the materials discovery pipeline.
• This tool holds strong potential for designing and discovering novel materials efficiently.