Iranian Journal of Chemical Engineering (IJChE)

Abstract The impact of bed loading on minimum spouting velocity (ums) of polydispersed TiO2 particles was studied in a conical fluidized bed. The experiments were performed at different bed loadings according to Gaussian and narrow-cut particle size distribution (PSD). The bed consisted of simple-agglomerates in size range of 30-90 µm belonging to Geldarts’ group A classification. The effect of PSD and interparticle force (IPF) on the predicted ums and hysteresis in the pressure profiles were studied through a combination of computational fluid dynamics and discrete element method (CFD-DEM). The experimental data showed that the choice of bed with Gaussian PSD-type led to more accurately predicting ums than the narrow-cut particle PSD. The impact of IPF on the expected ums became more critical than the PSD type because of an increase in bed loadings. The lowest deviations the results were obtained in the low bed loadings, which is confirmed the accuracy of simulation results. The simultaneous effects of PSD-type and IPF led to a change in the fluidization behavior of the bed. The bed with narrow-cut PSD has a hydrodynamic behavior similar to spouting and slugging regimes, while the fluidization quality of the bed improves by fine particles.

Abstract > The hydrodynamic characteristics of dense conical fluidized bed were investigated experimentally and numerically. Experimental studies have been carried out in a bed containing TiO2 particles belonging to A/C boundary of Geldart's classification with a wide particle size distribution. Pressure measurements and an optical fiber technique allowed determining the effect ofhigh bed particles loading on the minimum fluidization velocity, local solid volume fraction and solid velocity. Two-fluid model approach with three different drag models and boundary conditions (BCs) consisting ofno-slip, partialslip and free-slip BC is presented for the numerical predictions. In this paper, we show the Gidaspow drag function with k-Â turbulent model by applying the partial-slip BC can improve the numerical results at high particle loading.