Crystal structure prediction (CSP) of a given compound from its molecular diagram is a fundamental challenge in computational chemistry that requires accurate and efficient evaluation of energy differences between candidate structures to generate a reliable crystal energy landscape. Planewave/pseudopotential calculations using dispersion-corrected DFT have been found to be sufficiently accurate, in most cases, but are too computationally expensive to be applied to more than a handful of structures in the final stages of a CSP protocol. In this talk, I will highlight how composite methods, employing low-level calculations for geometry optimizations and higher-level calculations for subsequent evaluation of single-point energies, can efficiently provide more expansive crystal energy landscapes. These composite methods are successfully applied to chiral helicenes, with applications in organic electronics, and to four pharmaceutical compounds: 5-fluorouracil, naproxen, carbamanzapine, and olanzapine. Finally, the effect of the density-functional delocalisation error on CSP is highlighted for organic acid-base co-crystals.

Composite methods for molecular crystal structure prediction

ACS Spring Meeting 2021

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.