The mechanistic understanding and control of an API’s behavior stems from the characterization of its polymorphic landscape, particle size distribution and the knowledge of such attributes’ relationship with the nonlinear interplay of thermodynamics, kinetics, and multivariate process parameters. Automated platforms leveraging high throughput screening are reported for polymorph and solubility, while their adoption for particle size control during process development is yet on the rise. This poster outlines the challenges and highlights strategies for effective automation in particle engineering workflows to gain accelerated access to the system mechanistic understanding.
We present a case study in (semi-)automated process development and optimization of a continuous anti-solvent crystallization process for a fast-nucleating active pharmaceutical ingredient (API). Crystallization is performed via high-intensity mixing in a customized impinging jet mixer (NanoScaler, Knauer GmbH). The platform is augmented with a fourth feed stream and integrated process analytical technology (PAT) for real-time monitoring of particle size distribution. All system components—including pumps and PAT—are interfaced via OPC-UA and orchestrated through a SCADA-based control architecture, enabling automated execution, data logging, and closed-loop optimization. Using Bayesian optimization the target particle size can be achieved rapidly unsupervised. This case study illustrates how automation and real-time analytics can accelerate and enhance the development of continuous crystallization processes with minimal human intervention.