MolCrysKit: A Topology-Aware Toolkit for Bridging Experimental Molecular-Crystal Structures and Simulation-Ready...

Area of Science:

• Materials Science
• Computational Chemistry
• Crystallography

Background:

• Modeling molecular crystals presents challenges in maintaining molecular integrity and handling experimental data imperfections like disorder and missing atoms.
• Current atom-centric toolkits are limited for molecular crystals, requiring treatment of bonded entities as rigid units within a lattice.

Purpose of the Study:

• To develop MolCrysKit, a Python toolkit for automated, topology-aware conversion of crystallographic information files (CIFs) into simulation-ready molecular crystal models.
• To enable reliable construction of chemically consistent simulation systems from ambiguous or incomplete experimental crystal structures, particularly for AI-driven workflows.

Main Methods:

• A graph-theoretic approach is used for resolving crystallographic disorder.
• A topology-preserving algorithm facilitates surface slab generation.
• A chemical-environment-aware scheme reconstructs missing hydrogen atoms.

Main Results:

• MolCrysKit automates the transformation of experimental crystallographic data into computationally viable molecular crystal models.
• The toolkit integrates strategies for disorder resolution, surface generation, and hydrogen atom reconstruction.
• It provides an interface for reproducible transformations and supports AI-driven pipelines.

Conclusions:

• MolCrysKit offers a robust solution for generating simulation-ready molecular crystal models from experimental data.
• The toolkit enhances the reliability of computational modeling for molecular crystals, especially when dealing with data ambiguities.
• Its design facilitates the integration with advanced automated workflows and AI agents for materials discovery.