Development and validation of a batch fluidized bed dryer model for pharmaceutical particles

Abstract

Batch fluidized bed drying is a common process for water removal in pharmaceutical particles. Multiple phenomenological models for this type of dryer are proposed in the literature, with many of them relying on the two-phase theory. This article elaborates on a two-phase model and extends it for the pharmaceutical case by improving the particle, batch size and heat loss descriptions. An explicit methodology with low computational cost is also developed for solving the fluidized bed equations. Then, based on a grey-box modeling strategy, parameters with known value are fixed, while the unknown parameters are calibrated with experimental data collected on a pilot scale fluidized bed dryer. Lastly, the model is dynamically validated on distinct experimentation datasets. It is shown that the simulations are able to reproduce the pilot data trends, but there is still room for improvements in the heat loss representation.

Keywords:

• Batch fluidized bed dryer
• two-phase theory
• grey-box model
• system identification
• model validation

Acknowledgments

The authors would like to thank Pfizer Montreal and the National Sciences and Engineering Council of Canada (NSERC) for funding this research, and Pfizer Montreal for providing the necessary raw materials and laboratory equipment. They specially acknowledge the valuable time, support, and efforts invested by Pierre-Philippe Lapointe-Garant, Bárbara Santos Silva, Sophie Hudon and Yuwei Zhang (Pfizer Chair, Université de Sherbrooke).

Notes

1 A model sensitivity analysis of ρ_g limiting cases validates this assumption.

2 A three-phase model would include a separate region (thus, separate balance equations) between the bubbles and the emulsion often called the cloud and wake region.^[4,22]

3 A grey-box model combines features of both first-principles (e.g. physical laws, conservation principles) and empirical models (e.g. regressions or transfer functions).

4 A sensitivity analysis on the dependency of parameters on temperature, for limit cases (20 and 100 °C), attests this simplification.