(260d) Experimental and Theoretical Study of Tablet Flow in a Lab Scale Pan Coater

Authors: 
Toschkoff, G. - Presenter, Research Center Pharmaceutical Engineering
Böhling, P., Research Center Pharmaceutical Engineering GmbH, Graz, Austria
Johnson, D., Merck & Co., Inc.
Sun, J., Merck & Co., Inc.
Silverman, A. R., Merck & Co., Inc.
Scharrer, G., RCPE GmbH
Khinast, J. G., Graz University of Technology
Rajniak, P., Merck & Co., Inc.



The coating of tablets using a film
coating operation has become a common process in the pharmaceutical industry to
add color distinction, physical protection, or taste masking properties to the
final product.  Given the delicate nature of some tablets, it is important to
know how certain process parameters may affect the tablet motion and
interactions within a coating pan.  In recent years, in silico computer
simulations have become a common trend for trying to predict this behavior
using techniques like Discrete Element Methods (DEM) to compare/calibrate to
actual experimental/physical conditions.

The aim of this work is to develop DEM
simulations for a laboratory film coating pan (Labcoat 24? O'Hara Technologies, USA) and calibrate the model using experimental measurements from the same
instrument.  The model was developed around a 12-16 kg fill of the coating pan
with 10/32? Round Standard Curve placebo tablets, with comparisons being
performed on the distribution of spray zone residence time of selected tablets
and the bed cycle time. 

Experimental data was generated using
tracking tracers and a video imaging system that was placed inside the coater. 
The images used a rectangular region of the tablet bed surface as the spray
zone, and the frequency of the tracer in the spray zone was extracted using
automated image processing, see Fig.1.

0e562d8e-af5a-49e6-aa81-3a74c3d5da65

Fig. 1: Schematic of the experimental
setup; the same size and location of the spray zone is also used in the DEM
simulations.

Simulations were performed using the
commercial software EDEM 2.4 (DEM Solutions, UK) where the tablet shape was
approximately reproduced using 8 partially intersecting spheres (?glued sphere?
approach).  A post-processing algorithm was developed featuring the same zone
as was imaged as the spray zone in the experiments to track the spray and bed
residence times of selected tablets.

The simulation results showed good
agreement with the experimental data, helping to calibrate and validate the
model and its parameters.  The validated DEM simulation can be used to show
local tablet velocity and collision forces, and how they relate to the
operating conditions of the coating pan.  With further development of this DEM
approach and understanding the hardness of different tablet systems, it may
become possible to design and optimize coating pans and process conditions for
ideal coating operations with minimal tablet damage.