Early prediction of tablet defects during pan coating

Area of Science:

• Pharmaceutical Manufacturing
• Materials Science
• Computational Modeling

Background:

• Tablet appearance is a critical quality attribute, directly impacting patient perception.
• Mechanical stress during manufacturing can cause visual defects like edge chipping.
• Predicting these defects is challenging, especially during technology transfer and scale-up.

Purpose of the Study:

• To evaluate the prediction of tablet defect number and severity under mechanical stress.
• To assess two stress environments: pan coating and friabilator testing.
• To utilize a calibrated breakage model with discrete element method (DEM) simulations.

Main Methods:

• Calibrated a breakage model using drop tests with two placebo formulations (mannitol:MCC and MCC:dibasic calcium phosphate).
• Employed discrete element method (DEM) simulations to estimate collision parameters in a coating pan.
• Used the calibrated model to predict defects in both pan coating and friabilator testing scenarios.

Main Results:

• The model accurately predicted defect severity (Classes I-III) in pan coating but underestimated Class I defects.
• Numerical predictions are highly sensitive to model parameter calibration, especially for minor damage.
• The friabilator overpredicted defects, indicating limitations in high-collision environments and highlighting wear contributions in coating.

Conclusions:

• The developed model shows potential for guiding tablet defect prediction, particularly under low-event conditions with meticulous calibration.
• Differences in damage mechanisms between pan coating (wear) and friabilator (frequent collisions) testing were observed.
• Further refinement is needed to improve prediction accuracy for minor defects and high-collision scenarios.