Integrating Computed Crystal Energy Landscapes in Crystal Form Discovery and Characterisation
Over the past two decades, computational methods of crystal structure prediction (CSP) have shown enormous potential in complementing the efforts of crystal engineers to synthesise and characterise new solid forms of organic molecules. This chapter summarises the insights that can be gained from computational methods of CSP when integrated as part of experimental efforts to synthesise and characterise the crystal structures of organic molecules. The value of CSP methods is that they allow us to map the range of packing alternatives that a single-component or multi-component molecular system may adopt in the crystal as a function of the lattice energy. CSP methods can now handle large flexible organic molecules with the sort of complexity typically seen in pharmaceutical drug development pipelines, and it is now not unusual to find the experimentally observed crystal structure at, or close to, the global minimum in the crystal lattice energy landscape with the use of accurate dispersion-corrected density functional theory methods. The fundamental promise of CSP methods is to move us to a point where we can generate a set of plausible low-energy predicted structures for any molecule and be able to target the crystallisation and characterisation of a preferred structure.