Constrained evolutionary algorithm for structure prediction of molecular crystals: methodology and applications.

Area of Science:

• Materials Science
• Computational Chemistry
• Crystallography

Background:

• Evolutionary algorithms are effective for materials discovery.
• Predicting structures of complex molecular crystals remains challenging.
• Existing methods require significant computational resources.

Purpose of the Study:

• To develop an efficient algorithm for predicting the crystal structure of complex molecular crystals.
• To reduce the computational search space by treating molecular units as rigid bodies.
• To enhance the diversity and reliability of crystal structure prediction.

Main Methods:

• A novel evolutionary algorithm treating molecular units as whole bodies.
• Introduction of new variation operators tailored for molecular crystals.
• Generation of diverse initial populations using space-group symmetry and random parameters.
• Testing the algorithm on various molecular and inorganic crystals.

Main Results:

• The algorithm efficiently predicts crystal structures of complex molecular systems, including methane (21 molecules/unit cell).
• Demonstrated reliability across diverse test cases: ice, ammonia, CO2, methane, benzene, glycine, and butane-1,4-diammonium dibromide.
• Showcased potential for complex inorganic materials like Mg(BH4)2 and boron.

Conclusions:

• The developed algorithm offers a significant advancement in evolutionary crystal structure prediction.
• Its efficiency and reliability make it suitable for both molecular and complex inorganic materials.
• This approach holds promise for accelerating materials discovery and design.