Finding crystal structures from few diffraction data by a combination of a random search with genetic algorithms.

Area of Science:

• Crystallography
• Materials Science
• Computational Chemistry

Background:

• Structural analysis of crystalline materials is crucial for understanding material properties.
• Traditional methods like direct methods may fail with scarce or incomplete diffraction data.
• A robust computational approach is needed for challenging structural determination cases.

Purpose of the Study:

• To describe a novel computational procedure for crystalline material structural analysis.
• To provide an alternative method for structure determination when direct methods are inapplicable.
• To make the developed routines accessible through the TRY program.

Main Methods:

• Utilizes internal coordinates for structural analysis from diffraction data.
• Employs a Monte Carlo approach with filtering for initial solution selection.
• Applies genetic algorithms (crossover and mutations) for further refinement.
• Requires unit-cell content, space group, chemical formula, and molecular connectivity as starting information.

Main Results:

• The Monte Carlo selection step alone frequently identifies correct structures.
• The procedure is effective with limited, low-angle, or partially missing diffraction data.
• Validation examples demonstrate the algorithm's efficacy.
• The method is particularly useful when direct methods are not applicable.

Conclusions:

• A new, effective computational procedure for crystal structure determination from diffraction data has been developed.
• The method overcomes limitations of traditional approaches, especially with scarce data.
• The developed routines are integrated into the freely distributed TRY program, enhancing accessibility.