A Computational Study of Hydrogen-bonded Molecular Crystals
The current climate in research has focused on the rational design of new materials with desirable characteristics. The demand for their full characterisation has in turn placed a new importance on structural chemistry, and important developments have taken place as a consequence. For instance an important probe to understand the interactions between molecules is to use variable pressure, and this has been exploited experimentally through the design and implementation of the diamond anvil compression cell (DAC). Using a DAC to study molecular materials at high pressure can result in problems, however: in X-ray diffraction the physical presence of the cell restricts access to reciprocal space, resulting in experimental structures of lower precision and often missing hydrogen atom location data. Traditionally the solution has been sought in neutron diffraction where hydrogen (deuterium) atoms scatter more intensely and so contribute more to the scattering pattern. This introduces another set of problems, however, in that the assumption is made that the isotope substitution does not alter the overall structure. In addition expense and time delays are incurred through this protracted experimental route. This thesis reports the development of a computational technique which can be used to reliably locate hydrogen atoms without the need for neutron diffraction data. The project reports rigorous testing on cases of varying difficulty, from the simple to the more complex. The test cases selected were also of industrial and environmental importance, so determining their complete structures under high pressure conditions was in itself a desirable outcome. Computationally completed structures were then compared to neutron diffraction results or used as the model to be refined against the neutron diffraction pattern.