(479b) Rheology of Powder Mixtures in a Planetay Blender

Rheology of powder mixtures in a planetary blender

L. Legoix, C. Gatumel, H. Berthiaux

Université de Toulouse, centre RAPSODEE, Ecole des Mines d’Albi-Carmaux, campus Jarlard, 81013 Albi Cedex 09, France

leonard.legoix@mines-albi.fr / cendrine.gatumel@mines-albi.fr / henri.berthiaux@mines-albi.fr

Powder mixing is a key process in different industrial fields: pharmaceutical, agro-food, civil engineering, metallurgy...  This step has to be done differently depending on the cohesiveness of the powder. In this study we will try to understand the rheological behavior of different powder mixtures. A four bladed planetary mixer Triaxe® seems to be a good way to blend free flowing and cohesive powders thanks to the variety of the trajectories and blade speeds. The blade motion is composed of two rotations: one around the blades axis called ‘rotation’ and one around the vertical axis of the spherical tank called ‘gyration’. So the Triaxe blender can involve all of the three flow mechanism (convection, shearing and diffusion) essentials to get a good mix even if the powder is cohesive.

Free flowing semolina (312μm), cohesive lactose (61μm) and fine lactose (26μm) are processed in this pilot scale (48L) blender. Eleven mixtures are made of semolina and lactose, with a mass fraction of lactose increasing from 0 to 1 with a 0.1 pitch. Other eleven are made of semolina and fine lactose to investigate the impact of particle size for this kind of mixtures.

During each experiment, thirty kilograms are stirred in the Triaxe® mixer. It’s blade tip speeds ranges between 0 and 2.5m/s. The torque of each engine is measured as a function of the rotating speeds, allowing us to calculate different coefficients a and b of the relation between a modified power number N_pM and a modified Froude number F_rM: N_pM = a . F_rM^b. These numbers are useful to classify powder rheological behavior and to to perform the scale up for different blenders in industry. Other parameters like filling ratio are also studied.

At the higher blade speeds, results show that the mixer experiences lower power consumption for cohesive powders, probably due to an aeration of the granular media and/or the creation of shear plans. Concerning lactose-semolina mixtures, a rheological behavior change, from free flowing to lactose-like, is observed for a mass fraction in lactose around 0.5, due to a lack of contact between semolina’s grains surrounded by lactose particles. A similar transition is observed in fine lactose-semolina mixtures for a lower fraction, close to 0.3.

We therefore find a dimensionless correlation to understand the rheology of mixtures of two different ingredients, one being cohesive and the other one free flowing, depending on the composition. As a perspective to this work, we propose to study flow regimes, the rheology and the mixing kinetics in a lab-scale similar transparent mixer, and eventually propose a stochastic model of convective powder mixing.