(468c) Parameter Range Optimization and Modeling for a Reactive Crystallization Process | AIChE

(468c) Parameter Range Optimization and Modeling for a Reactive Crystallization Process

Authors 

Dummeldinger, M. - Presenter, Bristol-Myers Squibb Co.
Treitler, D., Bristol-Myers Squibb Co.
Tabora, J., Bristol-Myers Squibb Company
Rogers, A., Bristol-Myers Squibb Co.
Chemical purity control is an essential component of quality risk management in active pharmaceutical ingredient (API) manufacturing. Impurity control typically relies on optimization of reaction parameters to minimize impurity formation, with downstream unit operations such as extractions and crystallizations serving as additional means of purity upgrade. In cases where crystallization occurs during the reaction, process performance can be more difficult to predict as the nucleation and uncontrolled crystallization of the product introduces stochastic variability with respect to impurity entrapment in the crystal lattice. This presentation details an industrial case study focused on process analysis and design of a robust impurity control strategy for an API step. In this step, due to a large solubility difference between API and penultimate starting material, high supersaturation conditions are achieved in the reaction, resulting in API self-nucleation and crystallization. Variability in nucleation and crystallization dynamics observed during process parameter space development resulted in variable levels of starting material entrapped in the isolated product. Extensive experimental studies were conducted to determine factors that influenced residual starting material impurity levels in isolated API, including reagent charge rate, solvent composition, and mixing conditions. A model was produced to assess process robustness over wide parameter space and quantify risks associated with operation on edges of or outside of the recommended ranges.