Crystal structure predictions for molecules with soft degrees of freedom using intermonomer force fields derived from...

Area of Science:

• Computational Chemistry
• Crystallography
• Materials Science

Background:

• Crystal structure prediction (CSP) is crucial for understanding material properties.
• Previous CSP methods faced challenges with flexible molecules and complex crystal systems (e.g., salts, cocrystals).
• The seventh blind test provided diverse and challenging targets to evaluate CSP protocols.

Purpose of the Study:

• To present and evaluate a novel molecular crystal structure prediction (CSP) protocol.
• To assess the protocol's performance on diverse targets, including flexible monomers, salts, and cocrystals.
• To investigate the utility of an ab initio force field (aiFF) for CSP.

Main Methods:

• Developed an ab initio two-body rigid-monomer six-dimensional force field (aiFF) for global-minimum conformers.
• Employed a CSP protocol involving conformational searching, packing, and lattice-energy minimization.
• Utilized flexible-monomer CSPs and empirical/reparametrized intramolecular force fields (FFs) for specific targets.

Main Results:

• Achieved success rates of 54% in structure generation and 89% in structure rating in the post-submission phase.
• Demonstrated the protocol's effectiveness for crystals containing flexible monomers.
• Highlighted the potential of the conformer-based approach with rigid-monomer CSPs.

Conclusions:

• The presented conformer-based CSP protocol, utilizing ab initio force fields, is recommended for crystals with flexible monomers.
• The protocol's reliance on first-principles quantum mechanics enables tailor-made force fields for molecular dynamics simulations.
• Further development is needed for empirical intramolecular force fields reparametrized with ab initio data for optimal CSP performance.