Gregory Beran at the University of California, Riverside is supported by an award from the Chemical Theory, Models and Computational Methods (CTMC) Program in the Chemistry Division and the Computational and Data-Enabled Science and Engineering Program (CDS&E) to develop state-of-the-art computational strategies for predicting the most stable molecular crystal structures and for coupling these predictions with experiments to determine crystal packing, The efficacy of pharmaceuticals, novel organic electronic materials, and many other crystalline solids depends critically on the manner in which the molecules pack together to form crystals. In the longer term, these studies of pharmaceutical crystal packing will help the formulation of medicinal drugs with improved shelf-life, solubility, and other important properties. Software resulting from this project will be made widely available to the community. The principal investigator also engages in outreach activities to bring awareness of scientific computing to the a broad range of students and to the general public.
To achieve these goals, new electronic structure algorithms that provide more accurate descriptions of intermolecular interactions at lower computational cost are being developed that will enable reliable crystal structure modeling. Improved techniques for predicting nuclear magnetic resonance chemical shifts are being developed to allow mapping between experimentally measured spectra and three-dimensional structural models. High-quality chemical shift predictions are needed to discriminate among different crystal packing forms. These techniques are being integrated into a workflow that combines crystal structure prediction, nuclear magnetic resonance spectrum simulations, and experimental measurements to identify the structures for several unknown crystal forms of two important pharmaceuticals. The new methods are incorporated into software which will be released to under open-source licensing terms. The code will be integrated into the Q-CHEM package, a commercial code with a very large user base.
