Mechanistic crystal size distribution (CSD)-based modelling of continuous antisolvent crystallization of benzoic acid

Since crystallization is one of the leading separation, isolation, and purification methods, the present mechanistic work investigated its continuous approach, presenting the precipitation of the benzoic acid molecules in a plug flow crystallizer. In some crystallization experiments, seeding was used in an attempt to eliminate spontaneous nucleation phenomena to allow better control over crystal size distribution by promoting crystal growth. At the same measurement time, a systematic mathematical model, based on mass varying expressions, kinetics and particle balance equations, was applied to predict the properties of the product crystals. Benchmarking the results of validation experiment, the established physical representation sufficiently described operation, which is confirmed by comparing various experimentally and theoretically determined process and product characteristics. Since liquid static mixers were used, a nearly perfect defined regime formation was achieved, which could be demonstrated by comparing simulated, calculated and experimentally-determined differential functional size distribution; therefore, a developed general morphology independent modelling could be built to optimise unit satisfactorily. Although crystallization of benzoic acid was investigated, the methodology can be easily applied to crystallization of other active pharmaceutical ingredients. The utilization of process analytical technologies and simulations can aid optimisation, intensification and automation, which can improve the quality of formulations. (c) 2021 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

Automated summary

The study investigated the continuous approach of crystallization, particularly focusing on the precipitation of benzoic acid molecules and promoting crystal growth through seeding. A systematic mathematical model and liquid static mixers were utilized to predict product properties and achieve a nearly perfect defined regime formation, allowing for the development of morphology independent modeling to optimize unit operation.