Crystal nucleation in pharmaceutical cocrystals systems

peer-reviewedIn this thesis the nucleation kinetics in three pharmaceutical cocrystal systems has been investigated through induction time experiments, identification of the nucleating solid phase and analysis of the nucleation data by the classical nucleation theory (CNT). An effort is made to elucidate cocrystal nucleation behaviour in general and find learnings on CNT in relation to multicomponent systems. Nucleation kinetics in general are also discussed and a comparison of the nucleation in the binary systems is presented. The nucleation kinetics of cocrystals have never been investigated before in terms of determining nucleation parameters such as interfacial energy and the pre exponential factor. It is interesting to note as a first observation that the multicomponent crystal nucleation was not significantly different to single component crystallisation in terms of driving forces required for nucleation, interfacial energy and pre-exponential factors. For the theophylline:salicylic acid 1:1 (THP:SA) cocrystal system - induction times have been determined in chloroform at different supersaturations at 10 ℃. Approximately 40-80 repetition experiments in 20 mL vials were performed at each condition. Nucleation times, extracted from the median induction times by accounting for a nucleus growth time, have been used to determine the interfacial energy and the pre-exponential factor within the classical nucleation theory. The nucleation behaviour of the cocrystal has been compared with the corresponding behaviour of the pure compounds. The results of driving force required to reach an equal induction time can be interpreted differently depending on how the supersaturation driving force is characterised for the cocrystal. For a pure system the definition of the supersaturation is straightforward, however, for the cocrystal it can be defined per heterodimer assembly or per reactant molecule. Using the prior definition, results show that to reach equal nucleation time the cocrystal requires a higher driving force. Using the latter definition of driving force, the cocrystal is no longer more difficult to nucleate than the pure compounds. Defining the driving force per reactant molecule appears to be a more appropriate definition because it gives the same dimensionality for the pure and the cocrystal systems. It is important to note that how the supersaturation is defined does not affect the interfacial energy since the molecular volume changes accordingly by which the exponential term in the CNT remains unchanged. However, the slope of the CNT graph does change and along with that the experimentally determined pre-exponential factor, thus leading to changes in the order of the difficulty of nucleation. The cocrystal is found to have an interfacial energy in between the respective values for the pure compounds. Similar trends in the value of pre exponential factors are observed from theoretical expressions of volume-diffusion and surface-integration-controlled nucleation respectively. Pure theophylline (THP II) is easier to nucleate than pure salicylic acid (SA), despite the latter having a smaller molecular size, higher solubility, and expected to form dimers already in the solution. The nucleation kinetics in the theophylline:glutaric acid 1:1 cocrystal system (THP:GLU) in chloroform was also investigated under the same conditions as the THP:SA system. Isolation and characterisation of the nucleating phase reveals that β glutaric acid (β-GLU) is the solid phase initially nucleating from the stoichiometric mixture, despite the supersaturation with respect to the cocrystal being much higher. The cocrystal then appears to nucleate on the surface of the β-glutaric acid crystals, and the transformation from the pure metastable crystals into the thermodynamically stable cocrystal takes place. The kinetics of nucleation of β-GLU in a binary chloroform solution and in a ternary stoichiometric solution of GLU and THP, is determined and compared with previous data for nucleation of THP II. The results of eutectic point determination reveal that the THP in the ternary solution increases the β-GLU solubility and greatly facilitates the nucleation of β-GLU. Evaluated within CNT, it is found that compared to the pure system, there is a significant increase in the pre-exponential factor of β-GLU nucleating from the ternary solution which relates to the higher concentration of GLU. β-GLU nucleation from a pure binary solution requires a clearly higher driving force to nucleate at the same time as THP II from a pure solution due to a higher interfacial energy. Nucleation in the p-hydroxybenzoic acid:glutaric acid 1:1 cocrystal (PHBA:GLU) system has been investigated in stoichiometric and non-stoichiometric acetonitrile solutions by induction time experiments at 20 °C. Utilising the ternary phase diagram, the supersaturated non-stoichiometric solutions were created with compositions along the eutectic phase boundary lines. In all cases the PHBA:GLU cocrystal was the nucleating phase, even though the non-stoichiometric solutions were also supersaturated with respect to the pure solid phases. The nucleation of the cocrystal from the ternary solution behaves similarly to the more difficult to nucleate pure compound – GLU. The difference in nucleation difficulty of the cocrystal from stoichiometric and non-stoichiometric solutions is captured by differences in the interfacial energy. The relation between the pre-exponential factors for the different systems calculated according to theoretical expressions for volume-diffusion and surface-integration nucleation mechanisms correlate reasonably well with the experimentally found relation apart from the pure GLU system. As seen for the THP:SA system, the PHBA:GLU cocrystal interfacial energy is also between that of the two individual compounds