Developing strategies for assessing structural assemblies

A novel method (XPac) for the identification of packing similarity, developed by the appli-cant, enables the systematic evaluation of large sets of related molecular crystal structures. A main aim of this project is to apply this strategy to a large set of structures of model compounds (barbiturates) and further develop the method by employing complementary ex-perimental and theoretical methods. The 5,5-substituted derivatives of barbituric acid display an ideal set of compounds for this approach because of their rigid geometry and the special H-bond donor and acceptor functionalities. Their aggregation in the solid state is thus mainly directed by H-bonding and will produce only a limited number of distinct extended patterns, which makes the structures more predictable. Barbiturates are also known for their general tendency to crystallize in multiple solid forms (polymorphs, hydrates and solvates) and to crystallize reasonably well, which enhances the probability of accessing many new crystal structures. Approximately 30 different barbiturates from a library at the host institute will be used in crystallization experiments. It is expected that subsequent single crystal structure determinations and full polymorph screenings for selected candidates will result in a sizeable set of barbiturate structures that will be available for the comparison study. Using the XPac method, the structures are classified in terms of hydrogen bonding and general packing of their molecules, and possible courses for and consequences of the observed packing similarities will be explored. Moreover the relative thermodynamic and kinetic stability of polymorphs of the individual model compounds and other relevant data will be established. By the combination of experimental investigations and computational analyses of crystal structures it should be possible to achieve a basic understanding of the structure-property relationships of the selected model compounds. It is expected that the results will enable us to predict features of unknown and inaccessible crystal structures of barbiturates and that we learn to understand the nucleation and growth principles of individual polymorphic forms and the reasons for the differences in the homologous series of molecules. This project has a high potential to expand our general knowledge in supramolecular aggregation and crystal engineering of molecular crystals and provides a valuable pool of information and data for theoreticians, who aim at the improvement of algorithms in ab-initio structure predictions. Moreover the project should result in improved cognition and suitable tools to quickly judge the novelty of polymorphs with very similar structural features - a relevant and growing problem that often causes scientific disputes and also affects intellectual property issues of industrially relevant organic compounds.