(535f) Development of a Coupled CFD-DEM Multiphase Model for a Fluidized Bed Dryer Process

Authors: 
Sansare, S., University of Connecticut
Duran, T., University of Connecticut
Mao, C., Genentech
Chaudhuri, B., University of Connecticut
Purpose: Drying is an essential unit operation in pharmaceutical dosage development process and is extensively used after wet granulation. Among all the drying processes, fluidized bed drying is the most popular technique because of the advantages that it provides such as excellent solid-gas mixing and enhanced heat - mass transfer which leads to efficient drying. However, a lack of fundamental understanding of the drying process of wet granules can potentially hinder a manufacturing process. A fluidized bed has multiphase flow in the system which can be difficult to control/ scale up. Thus, it is critical to understand the effect of process parameters on the quality attributes of the granules such as the moisture content. Understanding these parameters using experiments can be demanding, in terms of time and money. Therefore, studying such effects with a numerical model can prove to be useful and economical. Thus, in this study, a 3D coupled model has been proposed to study the effect of process parameters on the flow and moisture content of granules. This model would be validated with experimental data collected from a Glatt fluidized dryer.

Methods: Drying process of wet granules contains a continuous phase (air), modeled by Computational Fluid Dynamics (CFD) and a discrete phase (solid), modeled by Discrete Element Method (DEM). Two commercial software, ANSYS FLUENT for CFD and EDEM for DEM were coupled for the CFD-DEM approach. Geometry and meshing were implemented using ANSYS SpaceClaim and FLUENT, respectively. Heat transfer was implemented to look at the particle-particle and particle-fluid interaction. Different air flow velocities were implemented to study the pressure and velocity contours of the fluid phase and solid phase.

Results: Preliminary study was conducted using dimensions from a reference paper. Different air temperatures and velocities were used. Once the coupling was confirmed, dimensions of the actual Glatt equipment to be used was created. Different velocities were used in the new geometry. The effect of heat transfer was studied through the velocity and pressure contours and product temperature. After studying the flow patterns, the next step was to create wet particles. Liquid bridge model was used for this purpose to create particles with varying liquid contents. The mass transfer ability through the ‘Species Transport’ capability is currently being explored to look at the transfer of species between two phases to look at the change in moisture content. The developed model would be validated against the experimental data carried out in Glatt dryer with pharmaceutical raw materials.

Conclusions: Such a 3D coupled CFD-DEM approach can provide a good predictive model for studying the multiphase fluidized drying process. It can contribute towards faster optimization of manufacturing process with lesser batch failures.