CFD–DEM–PBM coupled model development and validation of a 3D top-spray fluidized bed wet granulation process

Abstract

In pharmaceutical manufacturing, fluidized bed granulation is one of the common processing options available to achieve better flowability of powders through size enlargement of primary particles. In fact over the last 50 years, various fluidized bed operations including freezing, drying, impregnation, coating, etc. have become a common place in the chemical processing industry due to the high level of contacts between fluids and solids attainable in a fluid bed system. These complex interactions between the fluid and particles also mean that simulating fluidized beds are still a challenging endeavor. Generally, Computational Fluid Dynamics (CFD) packages are employed to model the pressure drops in fluids; however, the presence of high concentration of solids and the complexity of granulation behavior require more advanced particle models than are available with CFD software. As a result, coupled frameworks that utilize the strength of particulate simulations such as Discrete Element Method (DEM) and bulk granulation modeling such as Population Balance Model (PBM) in conjunction with CFD information are the next steps to developing practical fluid bed granulation models.This paper aims to provide a comprehensive description of the development and validation of a coupled CFD–DEM–PBM framework for a fluidized bed wet granulation operation. A two-way coupled CFD–DEM model is developed for a 3-dimensional, lab-scale, top spray fluid bed granulator to study the effects of process parameters such as inlet air flow rate and inlet air temperature on the particle flow dynamics and the residence time in the spray zone. A one-way transfer of data from CFD–DEM to PBM is then applied to relate the effects of particle-fluid interactions to granulation behavior occurring within the fluidized bed system. Mechanistic rate expressions were developed in the PBM to create links between CFD–DEM results and PBM rate kernels which can express the effect of critical process parameters (CPPs) such as air flow rate, inlet air temperature, binder spray rate, etc. to experimentally measured critical quality attributes (CQAs) including granule size distribution and liquid content values. From comparison with experimental results, the framework presented shows good accuracy at capturing the dynamics of the system. The presented framework demonstrates a practical process model development methodology by efficiently coupling multi-phase simulation techniques which can be used for effective process design, development and scale-up purposes.