On the modeling of gas-solid fluidization: Which physics are most important to capture

Recent studies to predict biomass fluidization hydrodynamics motivated a new study to reassess how to model gas-solid characteristics that capture the same physics as that measured in experiments. An Eulerian-Eulerian multifluid model was used to simulate and analyze gas-solid hydrodynamic behavior of the fluidized beds. The relations for the pressure drop measured at fluidization were used to correct for the bed mass by either adjusting the initial solids packing fraction or initial bed height, two parameters that must be specified in a CFD model. Simulations using sand as the bed medium were compared with experiments and it was found that adjusting the bulk density, or in other words, the initial solids volume packing, correctly predicted the pressure drop measured experimentally, but significantly under-predicted the minimum fluidization velocity. By adjusting the initial bed height to correct for the mass, both the pressure drop and minimum fluidization velocity were successfully predicted. Ground walnut shell and ground corncob were used as biomass media and simulations were performed for two reactor bed diameters by simply adjusting the initial bed height to match the measured pressure drop. All of the simulations correctly predicted the pressure drop curves of the experimental data. Further examination of the simulations and experimental data for walnut shell confirmed that adjusting the bed height was the best approach to model fluidization without artificially altering the physics and retaining the known characteristics of the bed material.