(719e) High Shear Wet Granulation Scale-up Study Using Discrete Element Modeling

Authors: 
Sen, M., Eli Lilly and Company
Wade, J. B., Eli Lilly and Company
Miesle, J. E., Eli Lilly and Company
Schrad, M., Eli Lilly and Company
High shear wet granulation is a complex unit operation, since there are several rate processes occurring simultaneously that affect physical characteristics of the granulation (e.g., granule size distribution, granule density, wet mass consistency, etc.). A poorly understood system and inappropriate scale-up strategy can result in a poor process at the commercial scale.

This work illustrates the application of Discrete Element Model (DEM) to gain an understanding of the flow and force field distributions across the small scale, pilot scale and commercial scale granulators. It is well-known that granule density plays an important role in tablet tensile strength and drug product performance [1]. The granule density is highly influenced by the binder liquid quantity, and the magnitude of force inside the granulator relative to the dynamic yield strength of the wet granules. The total force field includes the force imparted by the impeller, force due to gravity, or weight of the granulation mass, and force due to other particle-particle and particle-wall (equipment boundaries) interactions. It was anticipated that at the commercial scale the effects of increased granulation mass would yield granules of reduced porosity, which are unsuitable for tableting, if the process design did not consider the total force field contributions across the different scales.

The DEM of the granulators of different scales was developed. A qualitative analysis of the variation of force field across the different scales was conducted. The process knowledge obtained from the DEM was used in directional adjustment of the operating conditions at the commercial scale that moved the process towards a lower risk operating space, which was confirmed through regime map and dimensionless spray flux analysis.

Reference

[1] van den Ban and Goodwin, 2017. Pharmaceutical Research, 34, 1002-1011