DEM simulation of the mixing equilibrium in fluidized beds of two solids differing in density

The distribution of solids in multicomponent fluidized beds strongly influences process performance, as undesired inhomogeneities usually result from a balance between the mixing and segregation tendencies. In the present work, a coupled DEM-CFD computational code is utilized to investigate on the equilibrium degree of mixing of two-component beds of particles with equal size and different density as function of the operating velocity and density ratio. Model validation is carried out first, by means of a direct comparison of simulated and experimental results on the mixing index and concentration profiles along the bed height, for a mixture of steel shots and glass ballotini of 433 μm average diameter. Since pseudo-2D geometry is used in simulations, the average voidage and the minimum fluidization conditions are very different from those of the experiments. Therefore, appropriate scaling of the velocity is necessary to compare the two systems. The DEM-CFD model is then utilized extensively to analyze the behaviour of the mixture at different velocities and varying component density ratio. A comprehensive data set is obtained for a dense-over-light density ratio ranging from 1 to 3 and a velocity ratio (actual over minimum fluidization velocity of the denser component) in the range 1 to 2. These results are re-examined within a unique framework by proposing a new model capable to interpret data while being mathematically and physically consistent with real observations. Inversion of this model provides a mixing map for the fluidized system whose behaviour is characterized for any combination of the two properties in the ranges investigated.

Graphical abstractA DEM-CFD approach is utilized to investigate the equilibrium degree of mixing of two-component beds of particles with equal size and different density as function of the gas velocity and density ratio. After validation, a comprehensive simulation data set is analyzed. The results are utilized also to propose a new model equation. Thus, a mixing map for the fluidized system is obtained.Figure optionsDownload full-size imageDownload as PowerPoint slide