(272g) A Step-By-Step Strategy for Modeling, Scaling-up, and Control of Aspirin Production in a Continuous Tubular Antisolvent Crystallizer Under Ultrasound

The crystallization of organic nano and microparticles is an important process for several industries, products, and applications [1]. The formed particles are crucial for various applications, such as in the case of organic micron-sized crystals of active pharmaceutical ingredients (APIs) and nanodispersions of coating resins and microcapsules with active ingredients [1]. The crystallization process itself can be performed in batch, semi-batch or continuous mode. Currently, batch crystallization is widely implemented at a large scale to meet market needs. However, batch processes are facing persistent challenges during operation. An indicative example is a hurdle posed during the control of heat and mass transfer; two factors that affect both the average particle size and particle size distribution (PSD), as well as the batch-to-batch variability. Continuous crystallization comes as a promising initiative to enable the maintenance of higher product standards [2].

Continuous processes allow the production of particles with specific particle size, polymorphism and smaller PSD [3]. Additionally, continuous processes produce quantities on demand, minimizing the use of raw materials and equipment. Furthermore, mixing energy and quantity of solvent could be decreased, and, last but not least, they can be scaled up easier as compared to batch systems [4]. Ultrasound can be used for improving continuous crystallization processes because the generated bubbles act as cluster attachment sites, forming finer crystals with a small average size and narrower PSD. However, the nucleation mechanism that takes place during the ultrasound is still unclear [5].

In this work, initially we demonstrate the model development of the continuous antisolvent crystallization of aspirin in water and ethanol in an experimental plug flow crystallization with ultrasound, and its validation with experimental measurements [6]. Concerning the mathematical modeling, population balance models (PBMs) are used extensively because they are available to capture the changes in the number and size of crystals over time, and space. Heat generation due to mixing and ultrasound, which increases crystal solubility and affects the yield of the system subsequently, is also included in the model. A previously validated model is then used for both theoretical scale-up and control. Regarding the scale-up, a CFD model is developed and validated in an effort to study the ultrasound-assisted continuous antisolvent crystallization of aspirin in different dimensions. Therefore, the first tubular crystallizer model is coupled with a validated acoustic field in the experimental set-up which is generated by the ultrasound transducer. The acoustic field in the crystallizer is identified experimentally with sonochemilluminescence experiments. The next part is the consideration of the proper fluid flow model. Although the experimental flow is laminar, the ultrasound bubbles cause turbulent and improved micromixing conditions. The fluid flow pattern is selected by comparing the CFD simulations in both laminar and turbulent flows with the results of the 1-D model by keeping the same previous expressions of solubility, nucleation, and growth rates. The accurate CFD model is used to predict the crystal growth in the presence of ultrasound at an increased diameter [7].

Finally, the last part is focused on a control concept and strategy for the optimization of the volume-based mean crystal size of the PSD. Toward this direction, this work’s core objective is to develop and assess a conventional feedback controller for the ultrasound-assisted continuous antisolvent crystallization of aspirin. In this way a simple, and accurate, controller scheme that controls the volume-based mean crystal size under disturbances in both the active pharmaceutical ingredient inlet concentration, and the inlet temperature (taking into consideration different profiles of disturbances) is presented. During simulated case scenarios, the control scheme is evaluated in extreme conditions of disturbances in the solvent flow rate [8].

This step-by-step strategy, which combines experiments and mathematical modeling, is crucial in a) capturing the crystal growth in continuous crystallizers under the effect of ultrasound, b) studying the continuous crystallizers scale-up with ultrasound, and c) applying control. Subsequently, the steps presented will provide insight into the design of new crystallizers that operate in continuous mode.

References

1. Gao Z, Rohani S, Gong J, Wang J. Recent Developments in the Crystallization Process: Toward the Pharmaceutical Industry. Engineering. 2017;3(3):343–53.
2. Yang X, Acevedo D, Mohammad A, Pavurala N, Wu H, Brayton AL, et al. Risk Considerations on Developing a Continuous Crystallization System for Carbamazepine. Org Process Res Dev. 2017;21(7):1021–33.
3. Sang-Il Kwon J, Nayhouse M, Orkoulas G, Christofides PD. Crystal shape and size control using a plug flow crystallization configuration. Chem Eng Sci. 2014;119:30–9.
4. Da Rosa CA, Braatz RD. Multiscale Modeling and Simulation of Macromixing, Micromixing, and Crystal Size Distribution in Radial Mixers/Crystallizers. Ind Eng Chem Res. 2018;57(15):5433–41.
5. Jordens J, Canini E, Gielen B, Van Gerven T, Braeken L. Ultrasound assisted particle size control by continuous seed generation and batch growth. Crystals. 2017;7(7).
6. Savvopoulos S V., Hussain MN, Jordens J, Waldherr S, Van Gerven T, Kuhn S. A Mathematical Model of the Ultrasound-Assisted Continuous Tubular Crystallization of Aspirin. Cryst Growth Des. 2019;19(9):5111–22.
7. Savvopoulos S V., Hussain MN, Van Gerven T, Kuhn S. Theoretical Study of the Scalability of a Sonicated Continuous Crystallizer for the Production of Aspirin. Ind Eng Chem Res. 2020;59(45):19952–63.
8. Savvopoulos S V., Voutetakis SS, Kuhn S, Ipsakis D. Theoretical Feedback Control Scheme for the Ultrasound-Assisted Continuous Antisolvent Crystallization of Aspirin in a Tubular Crystallizer. Ind Eng Chem Res. 2021;60(17):6221–34.