Systematic model identification and optimization-based active polymorphic control of crystallization processes

- A model-based active polymorphic control for batch crystallization is proposed.
- A high resolution finite volume method is used to solve population balance equations.
- Optimization was performed to estimate the best temperature profile.
- The dissolution cycle is beneficial to obtain larger crystals of the stable polymorph.

Polymorphism is an important issue in industrial crystallization, since polymorphs of the same compound can present very different properties, such as solubility, melting point or density, influencing considerably the manufacturability and bioavailability of the final product.This work proposes a model-based active polymorphic control strategy that allows obtaining large crystals of the stable polymorph at the end of a batch crystallization process, even in the case of erroneous seeding or in situ nucleation of a mixture of both the stable and metastable forms. A novel systematic experimental design was applied to estimate the kinetic parameters of dissolution, growth and secondary nucleation of the stable and metastable polymorphs of the model compound (ortho-aminobenzoic acid, OABA). Such experimental approach allows the determination of the studied kinetics without any correlation between parameters during the estimation, and without the need of off-line measurements of the crystal size distribution during the experiments.The estimated kinetic parameters were used to build a population balance model for the calculation of the optimal temperature profile needed, during a batch cooling crystallization process, for the (i) elimination of the metastable form crystals nucleated in situ or erroneously seeded and the (ii) maximisation of the size of the crystals of the stable polymorph obtained at the end of the batch process.