Fluidization characteristics of a mixture of gasifier solid residues, switchgrass and inert material

Effective fluidization of materials present in the reactor bed is critical for optimizing reaction conditions in a fluidized-bed gasifier. An improper fluidization leads to inefficient conversion due to many reasons such as low heat and mass transfers, ineffective gas–solid phase reactions, and uneven reactor temperature in autothermal gasification. The objective of this study was to investigate effect of reactor bed composition, i.e. a mixture of gasifier solid residues (GSR), switchgrass, and inert material, on fluidization using a 0.25 m i.d. transparent column. In this cold-flow study, the amount of inert material, i.e. silica sand, in the bed was held at 20 kg. The switchgrass in the mixture ranged from 0.17 to 5% of the sand quantity while the GSR ranged from 5 to 35% of the switchgrass. The particle geometric sizes by mass of sand, GSR and switchgrass were 348 ± 1.6 μm, 80 ± 2.6 μm, and 10.3 ± 1.7 mm, respectively. For all conditions, with an increase in gas superficial velocity, i.e. ratio of volumetric gas flow and bed cross-sectional area, the pressure drop across the bed increased reaching a maximum level at the minimum fluidization condition. Results showed that when the bed consisted of only GSR and sand, with an increase in the GSR from 5% to 35%, the gas superficial velocity at minimum fluidization condition, called minimum fluidization velocity (Umf), decreased significantly (p < 0.05); however, corresponding bed pressure drop (dPmf) remained constant. When the bed consisted of GSR, switchgrass and sand, there were significant effects (p < 0.001) of GSR, switchgrass and their interaction (GSR*Switchgrass) on Umf and dPmf. Fluidization improved with an increase in GSR up to 35% in the mixture. Overall, both Umf and dPmf increased with an increase in levels of GSR (5 to 35%) and switchgrass (0.17 to 3%) in the mixture. Fluidization characteristics were found to be strongly dependent upon mixtureʹs effective properties, which were determined using properties of all mixture components. Correlations available in literature were used to predict Umf using effective properties of tertiary mixture with GSR, switchgrass and sand. Prediction of Umf from all selected correlations did not match well with the experimental data for the entire range of tertiary mixture compositions. Fluidization of bed materials sustained up to 3% level of switchgrass. However, segregation of bed materials and in-bed channelization caused ineffective fluidization at 5% level of switchgrass in the mixture.