Abstract
An original methodology for studying powder flow in a cylindrical convective blender has been developed. A free-flowing and a cohesive powder were studied, at a fixed stirring speed, in rolling regime. For both powders, three apparent flow mechanisms were evidenced: convection in the volume swept by the blades, diffusion/shearing between the agitated zone and the stagnant one, as well as in the stagnant zone itself, and avalanches at the powder bed surface between agitated and stagnant zones. After defining six zones in the blender, tracing experiments were carried out by placing appropriate tracers in different starting zones and sampling the whole bed at different stirring times, which lead to mixing kinetics of the powders into themselves. A Markov chains model of the blender allowed the quantification of the three mechanisms respective magnitude by fitting the experimental data. This simple model has a good agreement with the free-flowing powder data, but is not able to represent well the observations for the cohesive powder. Bed consolidation should probably be taken into account for this kind of powders and thus a linear Markov model is not sufficient.
Highlights
In spite of the recurring use of convective blenders in the pharmaceutical industry, their global modelling has not been carried out yet, when focusing on fine or cohesive powders behaviour
The objective of this work is to study the differences in flow behaviour between a free-flowing and a cohesive powder when agitated by straight blades in a cylindrical convective blender, both experimentally and numerically thanks to a systemic modelling based on Markov chains
The influence of powder cohesion in convective blending has been characterized through mixing kinetics experiments
Summary
In spite of the recurring use of convective blenders in the pharmaceutical industry, their global modelling has not been carried out yet, when focusing on fine or cohesive powders behaviour. When dealing with batch planetary blenders, which combine two circular movements (namely rotation of the blades and gyration of their axis), the task is yet more cumbersome. For these reasons, a systemic approach has been undertaken in previous works, based on rheological observations at pilot scale [1]. A systemic approach has been undertaken in previous works, based on rheological observations at pilot scale [1] This lead to the setup of a cylindrical convective blender (Fig. 1): this device allows carrying out rheological studies, flow observations and determination of mixing kinetics. It should be noted that the term “mixing” is used in the sense of self-mixing of the powder in this study: the tracer employed for the mixing experiments are supposed to have the same flow behaviour and particle size than the powders studied
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
