Assessing residence time distributions and hold-up mass in continuous powder blending using discrete element method

Abstract

Continuous manufacturing of pharmaceutical products can improve process efficiency and quality assurance. The propagation of material through continuous manufacturing processes can be characterized by residence time distributions (RTDs). Conventional parameter-fitted RTD models rely extensively on experimental measurements that are often expensive and limited to available observation methods and instruments. This limitation can be compensated by discrete element method (DEM) simulations. This work evaluates the predictive capabilities of DEM modeling for a continuous powder mixing process. First, the developed DEM model was validated using experimentally measured RTDs. Then, the effects of the mass flow rate, blender element pattern, and blender speed on mean residence time and hold-up mass were studied. The simulation results showed that the mean residence time increased linearly with increasing number of mixing elements but decreased with increasing mass flow rate and blender speed. The hold-up mass increased with increasing mass flow rate and number of mixing elements but decreased with increasing blender speed. The work successfully demonstrated the capability of the DEM model in improving our understanding regarding the impacts of the blender operating conditions on the RTD and hold-up mass as well as in supporting the assessment of control strategies for continuous blending processes.