The development of a process model for the batch cooling crystallization of an anticancer drug from Takeda Pharmaceuticals is presented. The compound forms needle-like crystals that exhibit significant variation in shape (aspect ratio) during the crystallization. The parameters of a two-dimensional population balance model were fit using experimental concentration, particle size, shape, and relative number density information. The benefits of including focused beam reflectance measurement data in the parameter estimation to significantly enhance the robustness of the model are demonstrated. A new formulation of the parameter estimation optimization problem was used with augmented objective function and constraints to achieve improved convergence. The model was used for in silico experimentations to determine the achievable particle shape space in a broad batch time domain and for the digital design of the optimal operating temperature. The results were validated experimentally providing improved product quality in terms of size distribution and smaller aspect ratio.

• Purdue University

• Modeling

1. Digital design
2. modeling
3. Morphological population balance
4. Needle-like crystals
5. parameter estimation
6. Pharmaceutical crystallization
7. Research Article