Tuning crystal structures with weak interactions: From solid solutions to stoichiometric cocrystals
This thesis shows that when two organic compounds A and B are cocrystallized and the A∙∙∙B interactions are neither completely isotropic nor anisotropic enough to cause the formation of a stoichiometric cocrystal AxBy, the result is a solid solution AxB1–x that takes the structure of neither A nor B, but rather alternate forms of A/B depending on the relative proportions of A and B. Thus, by using A∙∙∙B interactions that are neither completely isotropic nor strongly anisotropic, one may gain access to normally inaccessible regions of the crystal structure landscape. However, there is a limit to how strong the anisotropy can be made. Even weak hydrogen bonds like C–H∙∙∙N can facilitate the formation of a stoichiometric cocrystal. For obtaining a solid solution one must certainly have weaker interactions. The tunability associated with solid solutions allows for the generation of a large number of structures over a wide range of composition, all of which are in fact data points in a potential energy well corresponding to a particular structure type, leading to a more comprehensive charting of the crystal structure landscape. This may be considered as the experimental equivalent of computational crystal structure prediction, where a large number of data points are generated on the basis of force fields and energy-density considerations. Solid solutions can thus lend access to an alternate packing mode of a molecule which may be associated with a particular property which can be tuned by varying the composition. Applications have already been shown by using human insulin as a model system. This could have implications in the context of the pharmaceutical industry and also in tuning properties of materials.