Pharmaceutical crystallization design using micromixers, multiphase flow, and controlled dynamic operations

Abstract

Crystallization is a key unit operation in the pharmaceutical industry. Control of crystallization processes can be challenging when undesirable phenomena such as particle attrition and breakage occur. This thesis describes the controlled crystallization of pharmaceuticals and amino acids for more efficient manufacturing processes and better efficacy of products. Crystallization equipment is designed so that (1) the undesirable phenomena do not occur at all, and/or (2) the phenomena that do occur are carefully controlled. One key strategy is to exploit dual-impinging jets and multiphase flow to decouple nucleation and growth so that they can be individually controlled. Various configurations of micromixers were designed to provide controlled nucleation. Based on the dual-impinging-jet (DIJ) configuration, a physical explanation was provided for the discovery that a cooling micromixer can generate small crystals of uniform size and shape. An alternative design replaces the micromixing with the application of ultrasonication to decouple nucleation and flow rates. Based on these nucleation methods, a novel continuous crystallizer is designed where the slurry flow is combined with an air flow to induce a multiphase hydrodynamic instability that spontaneously generates slugs where the crystals continue to grow. These slugs are well-mixed without having the mixing blades in traditional crystallizer designs that induce undesirable uncontrolled crystallization phenomena. Another key strategy is to increase the degrees of freedom in the dynamic operation of the crystallizers. In the slug-flow continuous crystallizer, extra degrees of freedom for control of the crystal growth are created by spatially varying the temperature profile along the tube. In a semi-continuous crystallizer configuration, continuous seeding using a DIJ mixer is combined with growth rate control in a stirred tank to experimentally demonstrate the manufacture of uniform-sized crystals. In addition, temperature-cycling experiments are designed in batch crystallizers to substantially change crystal shape with only a small number of cycles. Experimental validation confirms that the proposed crystallizer designs reduce production time and equipment cost by orders of magnitude while suppressing secondary nucleation, attrition, and aggregation/agglomeration-dominant but undesired phenomena that worsen the ability to control the properties of crystals produced by most existing crystallizer designs.