(668e) Methodology for Prediction and Scale-up of Pneumatic Conveying Systems with Breakage of Granules

This work presents an improvement of methodology for predicting the friability of granules as an extension of our recent work [1]. The methodology of collecting experimental data of granule attrition using a laser diffraction particle size analyzer (Malvern Mastersizer) and the model combining 2-D CFD Eulerian gas-solid model with QMOM methodology for solution of particle breakage was developed and described in [1]. The combined model allowed us to relate the breakage kernel parameters to particular flow properties. It was found that for a given experimental set-up with diluted gas-solid flow, the breakage rate of granules is proportional to the characteristic particle size and to the square of the impact velocity between a granule and the equipment wall (Ghadiri kernel). It was concluded in [1] that the binary breakage assumption available in then existing CFD code might not be fully valid to predict quantitatively the multiple breakage character of the experimental attrition system and more detailed 3-D model of the laboratory scale tester is required for more precise quantitative correlation between the flow properties and the granule attrition. In this extension we present a 3-D CFD Eulerian gas-solid model of flow and breakage in a different (Sympatec) particle size analyzer. The model employs the Ghadiri kernel together with the generalized multiple breakage distribution functions. The model is calibrated by fitting to the moments of experimental particle size distributions (PSDs). The experimental distributions were carefully collected for different narrow sieve fractions of typical pharmaceutical granules from both, the fluid bed granulation (FBG) and high shear granulation (HSG). Parameters of both, the breakage kernel and distribution function are evaluated in the calibration. As expected, evaluation confirms significantly higher friability of the FBG product when compared to the HSG product. The calibrated breakge kernel and distribution function were then employed for prediction of breakage during its transport using existing pneumatic conveying system in the manufacturing scale. The predicted breakage of the FBG granules disallows their conveying in the manufacture. On the other hand, we believe that the calibrated model for the HSG product may allow optimization of the conveying conditions in the large scale.

[1] P. Rajniak, K. Dhanasekharan, C. Sinka, N. MacPhail, R. Chern, S. Fitzpatrick, Modeling and measurement of granule attrition during pneumatic conveying in a laboratory scale system, In press, Powder Technology 185 (2008) 202?210.