Exploring polymorphism in molecular crystals with a computational approach: phase transitions and energy landscapes

Different crystal structures can possess different properties and therefore the control of polymorphism in molecular crystals is a goal in multiple industries, e.g. the pharmaceutical industry. Part I of this thesis is a computational study at the molecular scale of a particular solid-solid polymorphic transition with the help of different modelling techniques. The studied crystal is the amino acid DL-norleucine and the b $ a polymorphic transition is specifically the topic of research. The cooperative behaviour of molecules during this transition is discussed both from a computational and an experimental perspective. Also the relative energy barriers of the different shifts involved in the transition have been determined. Part II of the thesis is focussed around a new method to describe intermolecular interaction energies in molecular crystals and can therefore also be used to compare different polymorphic forms with each other. This method is named q-GRID and it uses the electron density as obtained within the crystal as an input to calculate a crystal graph that can be used to model nucleation rates, morphologies or as a direct estimation of the lattice energy. q-GRID can be considered as a natural extension of the PIXEL method, that has been developed by prof. Gavezzotti. The last chapter of this thesis describes the participation of the Radboud University (RU) Nijmegen to the Sixth Blind Test on Organic Crystal Structure Prediction Methods that has been organized by the Cambridge Crystallographic Data Centre (CCDC). Within this participation q-GRID has been used