(754f) Scaling up Strategy for Continuous Powder Mixing Process
AIChE Annual Meeting
2011
2011 Annual Meeting
Comprehensive Quality by Design in Pharmaceutical Development and Manufacture
Scale-up of Pharmaceutical Manufacturing Processes: Toward a QbD Approach
Thursday, October 20, 2011 - 5:05pm to 5:25pm
Scaling up
strategy for continuous powder mixing process
Yijie Gao, Fernando Muzzio and
Marianthi G. Ierapetritou
Dept. of Chemical and Biochemical Engineering, Rutgers
University, Piscataway, NJ 08854
Continuous
powder mixing has attracted a lot of interest among pharmaceutical industry. Based
on the principle of similarity proposed by Johnstone
and Thring 1,
many works have been done recently that target the scale-up of continuous
powder mixing. In this work, a quantitative scale-up strategy is described that
allows the transition from lab to industrial scale. By using the periodic
section modeling developed in our previous study 2, scale-up of the cross-sectional mixing and scale-up of the
axial mixing are separately considered. To capture the cross-sectional mixing
component, the conception of variance spectrum is used. The Lomb-Scargle Periodogram technique
widely applied in astronomical studies has been introduced to estimate the
variance spectrum of unevenly distributed samples in powder mixing process. The
effect of constant sample size on mixing performance of different scales is thus
elucidated. To capture the axial mixing component, the residence time
distribution is used to characterize mixing of different scales.
Two case
studies are presented to illustrate the applicability of the developed
strategy. For non-segregating mixing case, similar scale-up characteristics are
observed for both cross-sectional and axial mixing. While satisfying the principle
of similarity leads to similar decay contours of variance spectrum, a sampling
size much smaller than scales of both mixers should be used to achieve similar measurements
of mixing performance. A 2.5 power increase of flux rate is generally
obtainable in the scale-up of non-segregating materials. However, in processes
where cohesive materials are used, segregation occurs due to a tendency of
agglomeration, and a larger increase of flow rate is observed. This results
from the effect of cohesion on axial motion especially when the scale of mixing
process is small. Our method can be used to provide specific guidance of scale-up
of different powder mixing cases.
Reference