In addition to purity, the “quality” of a crystalline product is determined by such factors as the crystal size distribution, crystal shape and polymorph. Excessive fines or needle-shaped crystals can cause problems during operations such as filtration, powder handling and tableting. Metastable crystal polymorphs are undesirable because they can convert to the stable polymorph over time. For the case of a drug manufactured in crystal form, the polymorph change can potentially change the in vivo performance of the drug. In conventional crystallization processes, the stochastic nature of nucleation and the effects of growth rate dispersion can lead to a crystalline product with distributions of crystal size, shape and polymorph. Localized inhomogeneities in conditions within a crystallizer can broaden these distributions or lead to batch-to-batch variability.

The ability to closely control crystal properties functions as an enabling technology that can be used, for example, in the design and production of inhaled or controlled-release drug delivery systems. By dispersing the mother liquor as small drops within a continuous phase, it is possible to achieve a greater degree of control over the crystallization conditions and minimize the negative effects of stochastic phenomena on the product properties. My research uses experimental and computational techniques to support the rational design of drop-based crystallization processes for the production of engineered crystals of proteins and small organic molecules. Population balance and Monte Carlo methods are used to drive experimental design to rapidly screening a wide parameter space to identify crystallization conditions that are likely to produce crystals with the desired properties. Microfluidic techniques are employed to experimentally investigate the crystallization process and to validate model predictions.