Continuous crystallization of pharmaceutical compounds in mixed suspension mixed product removal systems

Continuous pharmaceutical crystallization requires simultaneous control of supersaturation, particle attributes, polymorphic form, fouling, and downstream processability. Accordingly, single-stage and two-stage mixed-suspension mixed-product-removal (MSMPR) crystallization systems were developed and compared for the controlled production of the target crystalline polymorphs: paracetamol Form I and carbamazepine Form III.

Equilibrium solubility and metastable-zone limits were determined before continuous operation. Four critical process parameters, residence time, relative supersaturation, continuous seed loading, and power input per unit volume, were evaluated using independent four-factor Box-Behnken designs. Response-surface optimization, stagewise population-balance modeling, multimodal process analytical technology, and matched configuration comparisons were performed at equal total active volume and overall residence time.

For both APIs, longer residence time increased median particle size toward the prespecified product-size ranges, narrowed particle-size distributions, and improved target-polymorph content, but reduced space-time yield. Higher supersaturation increased recovery and productivity for both APIs while broadening distributions and reducing polymorphic selectivity. Externally validated population-balance models achieved $R^2$ values of 0.894-0.943 and normalized RMSE values of 1.8%-8.8%. Relative to single-stage operation, two-stage MSMPR increased $D_{50}$ by 15.8% for paracetamol and 13.6% for carbamazepine, reduced PSD span by 14.6% and 16.9%, and decreased fouling by 42.4% and 45.8%, respectively. The larger, narrower particles were beneficial because filtration time declined by 26.3% for paracetamol and 34.4% for carbamazepine, while chemical purity and mass-balance closure were preserved.

At matched active volume and overall residence time, staged supersaturation generation suppressed fines, improved particle uniformity and morphology, and reduced fouling and filtration time without compromising chemical or solid-state quality. These results identify controlled supersaturation distribution, not additional crystallizer capacity, as the principal benefit of two-stage MSMPR operation for the two APIs studied.