Study of Crystallization Kinetics Within a Generic Modelling Framework

Crystallization is frequently applied in the production of salts and/or active pharmaceutical ingredients (API). In recent years the monitoring and analysis of crystallization operations has gained much attention because of the growing need to control more sophisticated production lines. Crystallization operations involve a combination of different phenomena and the modelling of these can be combined to form the crystallization operation model. For this purpose different kinetic models, for example growth, nucleation, agglomeration and breakage, are required to analyse the crystallization operation. These models require data, which may be available in different forms, in order to describe the kinetics, which determines properties of the process outcome. Growth of the crystals occurs in multiple dimensions and the relative rates of growth in each direction and other kinetic phenomena such as agglomeration and breakage of the formed crystals control the shape and size distribution of the final product. Including more than one dimension in the crystallization will allow a more general description of the kinetic phenomena and the crystallization operation. This requires models for kinetics in multiple dimensions and data to estimate the model parameters. In order to have a full description of a crystallizer, a combination of constitutive (kinetic) models is needed. Such a description is possible within a generic framework where the models can be combined, reused and identified utilizing a kinetic model library for kinetic models and data. From the data in this library new models may also be derived and identified. The framework has been established and tested for simulation of selected crystallization processes.

The objective of this paper is to present a modelling procedure for systematic development, validation and analysis of constitutive (growth, nucleation, agglomeration and breakage) models. In this procedure, data and known crystallization kinetic (constitutive) models representing specific chemical systems are studied with respect to operational variables and model parameters. Based on the analysis results, new kinetic models are generated, and using available data, the model parameters are regressed. The modelling framework allows kinetic models representing different kinetic phenomena to be combined in different ways so that the sensitive kinetic model parameters can be identified and their effect on crystallization operations analysed. That is, the importance of different (multiple) phenomena occurring within any crystallization operation can be studied with and without including the mass and energy balance equations. Also, the identification of any single phenomena occurring within a wide range of crystallizer operations is permitted. A database of kinetic models has been created and combined with the generic mass and energy balance models allowing the validation, use and reuse of the selected kinetic models.

The presentation will highlight a systematic analysis and comparison of kinetic models, their parameters and the involved variables under various crystallization operation scenarios. The simulated results of crystal growth and/or agglomeration/breakage, for example, will be presented for different choices of operational variables, such as, temperature, mass transfer, seeding and present particles. Details of the kinetic model library together with the available data will also be highlighted.

---