Answers are grounded in this paper and cite their sources.
This study investigates the effect of fines content and the selected fractionation strategy on the accuracy of simulating a bubbling fluidized bed reactor using a two-fluid model. A cylindrical reactor (50 cm in height, 2.2 cm in diameter) was considered, filled with polydisperse alumina particles (size 26–105 µm, Geldart Group B, density 2400 kg/m3). Eight values of superficial air velocity were examined in the range from 0.019 to 0.1213 m/s. The continuous particle size distribution was approximated by a discrete set of one to five solid phases using the Sauter mean diameter for each group. Two fractionation methods were compared: Method 1 used the full particle size distribution, whereas in Method 2 fines that did not satisfy the terminal velocity criterion were excluded prior to discretization. The obtained results were compared with experiments on the average bed height. It is shown that increasing the number of discrete groups improves agreement with experimental data at low gas velocities; however, at high gas velocities, this effect is non-monotonic and depends on the emergence of discrete groups consisting solely of fine particles. Method 2 provides significantly better agreement at high gas velocities, with the relative deviation reduced to 0.42–2.38%. This represents the best result for Method 2 at high gas velocities, rather than a universal metric applicable to all groups. The obtained results indicate that, for simulating a bubbling fluidized bed, the use of two to four discrete groups is sufficient, depending on the gas velocity, terminal velocity, and particle size range.
No takes yet. Share an insight, caveat, or question.
Sergei Solovev (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: