Investigation of causes of layer inversion and prediction of inversion velocity in liquid fluidizations of binary particle mixtures

• Published in: Powder technology 2019-01, Vol.342, p.418-432
• Main Authors: Abbaszadeh Molaei, E., Yu, A.B., Zhou, Z.Y.
• Format: Article
• Language: English
• Subjects: Activated carbon; Balance criterion; Beads; Binary mixtures; Biological properties; Bioreactors; CFD-DEM; Computer simulation; Drag; Drag force; Fluidization; Fluidizing; Glass beads; Inversion; Layer inversion; Liquid fluidization; Mathematical models; Mineral processing; Model accuracy; Particulates; Predictions; Pressure; Pressure gradient force; Simulation; Velocity
• ISSN: 0032-5910; 1873-328X
• DOI: 10.1016/j.powtec.2018.10.011

Liquid fluidization plays an important role in mineral processing such as solid classifiers and biological reactors. When particle properties differ in size, density or shape, segregation can occur. Typically, a segregation phenomenon so-called layer inversion in two-phase solid-liquid fluidization of binary mixtures has been reported in the literature. At low liquid velocities, the two species form distinct layers with the denser smaller particles at the bottom and the lighter larger particles at the top. At high liquid velocities, however, the two layers are inverted with smaller particles being at the top while larger ones at the bottom. In this work, glass beads (193 μm) and activated carbon (778 μm) are used in the simulation as those in Jean and Fan (1986), and the layer inversion is successfully reproduced by CFD-DEM. The underlying segregation mechanism is analyzed in terms of particle-fluid interaction forces. The results reveal that the layer inversion is caused by the relative change of particle-fluid interaction forces on particles. For large particles, the reduction of the pressure gradient force cannot be compensated by the increased drag force, resulting in the downward flow. Meanwhile, small particles tend to move upward because the increased drag force exceeds the decreased pressure gradient force. In addition, the effects of particle and liquid properties on layer inversion are examined, and different drag force models in predicting the layer inversion are also assessed. It indicates that Rong et al.'s drag model represents the layer inversion with the best accuracy. Finally, a model based on the force balance criterion is proposed to predict the inversion velocity showing a better accuracy. [Display omitted] •Layer inversion in liquid fluidization is successfully reproduced by CFD-DEM.•Drag force and pressure gradient force are responsible for the occurrence of layer inversion.•Rong et al. (2014)'s drag model shows better accuracy in the prediction of layer inversion.•Effects of particle and fluid properties on the inversion velocity are quantified.•A predictive model is proposed to determine the inversion velocity with a better accuracy.