Modeling and simulation of continuous powder blending applied to a continuous direct compression process

Abstract

Continuous manufacturing techniques are increasingly being adopted in the pharmaceutical industry and powder blending is a key operation for solid-dosage tablets. A modeling methodology involving axial and radial tanks-in-series flowsheet models is developed to describe the residence time distribution (RTD) and blend uniformity of a commercial powder blending system. Process data for a six-component formulation processed in a continuous direct compression line (GEA Pharma Systems) is used to test the methodology. Impulse tests were used to generate experimental RTDs which are used along with parameter estimation to determine the number of axial tanks in the flowsheet. The weighted residual from the parameter estimation was less than the χ² value at a 95% confidence indicating a good fit between the model and measured data. In-silico impulse tests showed the tanks-in-series modeling methodology could successfully describe the RTD behavior of the blenders along with blend uniformity through the use of radial tanks. The simulation output for both impulse weight percentage and blend uniformity were within the experimentally observed variance.

Keywords:

• Continuous manufacturing
• powder blending
• residence time distribution
• blend uniformity
• flowsheet modeling

Acknowledgements

A license for the gSOLIDS process modeling software has been provided by Process Systems Enterprise (Ltd.) for the simulation work.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 Process Systems Enterprise, London, UK; www.psenterprise.com

Additional information

Funding

This work was financially supported by the USFDA cooperative research project [5U01FD005294], Process Modeling and Assessment Tools for Simulation, Risk Management and design space development of integrated pharmaceutical manufacturing processes. The project is also partially supported by Merck & Co., Inc., Kenilworth, NJ, USA. The authors would like to thank Merck & Co., Inc., Kenilworth, NJ USA and USFDA for funding the project and providing the dataset used in the modeling work.