Evaluating Competing Intermolecular Interactions through Molecular Electrostatic Potentials and Hydrogen-Bond Propensities

The structural chemistry of six thiazole amides has been explored using experimental crystallographic data, and a combination of calculated hydrogen-bond propensities (HBPs) and hydrogen-bond energies. Both methods correctly identify the main hydrogen-bonded synthon, a pairwise N-H center dot center dot center dot N dimer, which appears in all the available structures. The strength and stability of the homosynthon in these compounds were tested by attempting to co-crystallize all six compounds with 20 different carboxylic acids. A total of 120 attempted reactions were carried out via liquid-assisted grinding experiments; sixty of these reactions produced a co-crystal, as indicated by IR spectroscopy. HBP calculations were employed in order to try to predict which of the 120 reactions would be successful. A multi-component score (MC score) was obtained from the hydrogen-bond propensity calculations, and this score coupled with a cut-off value >0.0 resulted in a 77% agreement between experiment (88% success for aliphatic acids and 67% for aromatic acids). By adjusting the MCcut-off > -0.1, the overall success rate for co-crystal prediction increased to 91% while significantly reducing the number of false negatives. The crystal structures of eight co-crystals were obtained, and all of them contain the synthon that was predicted by both hydrogen-bond energy and HBP calculations. The results demonstrate the importance of using complementary energy-based and structural chemistry methods for identifying which supramolecular pathways are accessible (and which are most likely) as small molecules transition from solution to the crystalline solid state.