Control of batch and semi-batch crystallization

Crystallization is a multi-phase process that includes nucleation, growth, agglomeration and breakage. The objective in the operation of a crystallization process is to achieve high purity and to form desirable mean size crystal with a unimodal and narrow size distribution. Many factors such as seeding, mixing, impurities, and type of solvent can affect the product quality at the end of the process; however the supersaturation is the most important in its influence in both batch and semi-batch processes. It is important to control the extent of supersaturation during crystallization since nucleation, growth, agglomeration depend on the supersaturation profile achieved during the crystallization process. In this research work, some of the most important operational techniques and control approaches that affect the supersaturation and the product quality have been studied. Different cooling and seeding policies are employed to investigate the effect of cooling and seeding on the final CSD of ammonium sulphate grown in a batch-cooling crystallizer. A new cooling policy developed and tested. This new policy, the impulse changes in natural cooling policy, results in the same quality product as controlled cooling policy. However, the ease of implementation of the new policy makes it particularly attractive. In the semi-batch mode, crystallization of paracetamol in isopropanol-water mixtures has been investigated. An attempt has been made to measure the concentration and solubility of paracetamol in different solutions using an in-situ ATR-FTIR device and chemometric technique. Using three predictive models, the solubility is predicted and compared with the corresponding experimental values. The UNIQUAC model results in the best predictions. Then a fuzzy logic control methodology is developed for the control of supersaturation in an anti-solvent seeded semi-batch crystallization of paracetamol in isopropanol-water mixtures. It is found that the fuzzy logic controller can ensure tracking of the concentration within a desirable zone close to the solubility curve leading to substantial improvement of the end product quality.; Measurement of the differential heat of crystallization at any stage of nucleation or growth phenomena can offer some valuable information for the design or improvement of the operating conditions. The available on-line calorimetric techniques fail to deal with viscous and heat sensitive systems. Using flexible heat flux sensors, a new approach is developed for the isothermal calorimetric technique to overcome the disadvantages of heat flow calorimetric methods. The capability of the system to deal with the overall heat transfer coefficient variations has been demonstrated.; Keywords. Batch and Semi-Batch Crystallization; Cooling and Seeding Policy; Crystal Size Distribution; Supersaturation; Ammonium Sulphate; Solubility; Paracetamol; Co-solvent; ATR-FTIR; Chemometrics; Partial least Squares; Ideal solubility; Activity Coefficient; Regular-Solution Theory; UNIFAC; UNIQUAC; Fuzzy Logic Controller; Anti-Solvent Crystallization; FBRM; Heat Flux Sensor; Reaction Calorimetry; Isothermal Heat-Flow Calorimetry; Acetic Anhydride; Overall Heat Transfer Coefficient; Detection Limit.