Crystal Structure Prediction of Energetic Materials

Abstract

The synthesis and experimental testing of energetic materials can be hazardous, but their many industrial and military applications necessitate their constant research and development. We evaluate computational methods for predicting the crystal structures of energetic molecular organic crystals from their molecular structure as a first step in computationally evaluating materials, which could guide experimental work. Crystal structure prediction (CSP) is evaluated on a test set of 10 energetic materials with known crystal structures, initially using a rigid-molecule, anisotropic atom-atom force-field approach, followed by reoptimization of predicted crystal structures using dispersion-corrected solid-state density functional theory (DFT). CSP using the force field was found to provide good results for some molecules, whose known crystal structures are reproduced by one of the lowest-energy predictions, but are more variable than typical results for other small organic molecules. Reoptimization of predicted crystal structures using solid-state DFT leads to reliable predictions, demonstrating CSP as an approach that can be applied in the area of energetic materials discovery and development.