Predicting inclusion behaviour and framework structures in organic crystals.

We have used well-established computational methods to generate and explore the crystal structure landscapes of four organic molecules of well-known inclusion behaviour. Using these methods, we are able to generate both close-packed crystal structures and high-energy open frameworks containing voids of molecular dimensions. Some of these high-energy open frameworks correspond to real structures observed experimentally when the appropriate guest molecules are present during crystallisation. We propose a combination of crystal structure prediction methodologies with structure rankings based on relative lattice energy and solvent-accessible volume as a way of selecting likely inclusion frameworks completely ab initio. This methodology can be used as part of a rational strategy in the design of inclusion compounds, and also for the anticipation of inclusion behaviour in organic molecules.