(597f) Attainable Regions of Particle Sizes for Continuous Milling-Crystallization Processes

Vetter, T., University of Manchester
The properties of particulate products are not only defined by their composition and crystal structure (polymorph, salt form, solvates, etc.), but also by their size and shape. The influence of particle size and shape is particularly felt in downstream processing operations, such as filtration, drying, granulation, and blending of one particulate product with another. For instance, one can surmise that thin needle-like crystals are more prone to break in an agitated dryer than compact crystals of the same material or that the time required to separate mother liquor from crystals by cake filtration must â?? among other factors â?? depend on how tightly crystals in the filter cake are packed, which in turn depends on the crystal size and shape. Therefore, crystallization processes should be designed to deliver the high purity and yield, the right crystal form, as well as to deliver a reproducible particle size distribution that avoids undesired effects in downstream processing.

Within the quality by design (QbD) initiative[1], there is an increased drive towards a deeper understanding of manufacturing processes, aiming at identifying a design space for a given combination of product and process in which the product can be reliably produced to specification. Recent works within the crystallization community established a design methodology, which allows to deduce what particle sizes are attainable from a given crystallization process based on knowledge of the thermodynamics and kinetics[2]. This was formulated in regions of mean particle sizes against processing time (residence time for continuous crystallizers or batch time for batch crystallizers), while fulfilling constraints such as yield considerations and maximum/minimum allowed temperatures. The methodology was subsequently extended to take into account additional product characteristics, such as the width of the particle size distribution[3], as well as additional operational constraints[4]. The frameworkâ??s robustness against uncertainty in the kinetic parameters was assessed as well[5]. However, these efforts were exclusively focused on crystallization processes that are dominated solely by crystal growth and nucleation, i.e., secondary kinetic processes, such as crystal breakage and agglomeration were neglected.

In this work, it will be shown that attainable regions can also be obtained for processes that include extensive breakage processes. The study is motivated by the realization that processes that intricately combine crystallization and milling devices often allow to tune particle size distributions in a more extensive way than crystallization processes alone[6]. For instance, they allow to obtain smaller particles, or allow reaching a desired yield in shorter processing times (through an increase in surface area of the crystals). This work specifically investigates the combination of a continuous mixed suspension mixed product removal crystallizer (MSMPRC) with a high shear wet suspension mill that is connected to the crystallizer in a recycle loop. After determining breakage kinetics for this mill were determined, attainable regions for this process were calculated using a population balance equation model including breakage kinetics. The methodology is applied to the cooling crystallization of paracetamol crystallized from a mixture of isopropanol and water (1:4 w/w).


[1] Food and Drug Administration, â??Pharmaceutical Quality for the 21st Century A Risk-Based Approach Progress Reportâ?, http://www.fda.gov/aboutfda/centersoffices/officeofmedicalproductsandtob..., accessed: May 6th, 2016.

[2] Vetter, T., Burcham, C.L., Doherty, M.F., â??Regions of attainable particle sizes in continuous and batch crystallization processesâ??, Chem. Eng. Sci., 2014, 106, 167-180.

[3] Yang, Y., Nagy, Z.K., â??Advanced control approaches for combined cooling/antisolvent crystallization in continuous mixed suspension mixed product removal cascade crystallizersâ?, Chem. Eng. Sci., 2015, 127, 362-373.

[4] Power, G., Hou, G., Kamaraju, V.K., Morris, G., Zhao, Y., Glennon, B., â??Design and optimization of a multistage continuous cooling mixed suspension, mixed product removal crystallizerâ??, Chem. Eng. Sci., 2015, 133, 125-139.

[5] Vetter, T., Burcham, C.L., Doherty, M.F., â??Designing robust crystallization processes in the presence of parameter uncertainty using attainable regionsâ??, Ind. Eng. Chem. Res., 2015, 54, 10350-10363.

[6] Yang, Y., Song, L., Zhang, Y., Nagy, Z.K., â??Application of Wet Milling-Based Automated Direct Nucleation Control in Continuous Cooling Crystallization Processesâ??, Ind. Eng. Chem. Res., 2016, 55, 4987-4996.