(249b) A Digital Twin for the L.B. Bohle Tablet Coater | AIChE

(249b) A Digital Twin for the L.B. Bohle Tablet Coater

Authors 

Iyer, K. S., University of Massachusetts Amherst
Doshi, P., Worldwide Research and Development, Pfizer Inc.
Wanapun, D., Pfizer Inc.
Hodgins, N., Pfizer
Conway, S., Pfizer
Suryawanshi, T., Tridiagonal Solutions Inc.
Saxena, U., Tridiagonal Solutions Inc
Khan, M., Tridiagonal Solutions Inc
Kasat, G., Tridiagonal Solutions Inc.
The coating of pharmaceutical tablets to achieve desired properties is accomplished via a common film coating unit operation. Typically, this unit operation is achieved by coating tablets in a rotating drum, which is equipped with multiple nozzles that spray this coating on to the tablets. Along with such spraying operations, the tablets are also exposed to an environment where hot air is blown onto them to dry the obtained coating. Physically, this combination of tablet bed flows, coupled with spray coating and drying represents a complex process, the effectiveness of which is characterized by the inter-tablet coating uniformity as a function of the number of revolutions of the drum. De-risking the performance of the coater at different scales of process development through experimental evaluations is time and resource intensive, although it is of paramount importance. In order to enable a more efficient process design, we have developed a state-of-the-art digital twin of the tablet coating process.

Our digital twin uses a novel first-principles model by combining Computational Fluid Dynamics to simulate spray coating and air flow, and Discrete Element Method to simulate tablet motion. This provides us with unprecedented levels of precision in virtual design, optimization and scale-up, enabling us to study coaters from laboratory scale to commercial scale. Using this digital twin, we have evaluated the effect of tablet loading, drum rotation speeds and tablet types, and we have computed several measures to characterize the performance of the coater. We have then used the result of these models to guide our scale-up decisions. This digital twin is at the forefront of our digitalization initiatives in process development and extends our predictive capabilities, enabling us to reduce developmental time and resources.