Prediction of gas–solid bed hydrodynamics using an improved drag correlation for nonspherical particles

Abstract

Gas–solid beds are ubiquitous in industrial and energy production applications. Examples include fluidized beds, which are used in many systems such as in integrated gasification combined cycle power plants or in chemical looping systems. These examples and others involve complicated interactions between each phase of reactants in the system. The motivation of this work stems from the need for a better understanding of bed hydrodynamics in existing energy systems; results from this work can be used directly in software such as Fluent to more accurately predict flow behaviors of gas and solid phases. The experimental data are collected from two setups including an optically accessible drag measurement facility that was used to obtain the drag coefficient at various particle Reynolds numbers and a lab-scale gas–solid packed bed which was used to validate the computational correlation through pressure drop measurements across the packed bed. Results showed that the new correlation predicted drag coefficients as accurate as 10% and deviated by up to 15% for particles with sphericities less than 0.9. This is a significant improvement compared to existing correlations, which can deviate as much as 50% for the same range of tested values. Similar findings are observed when the drag correlation is implemented into Fluent. It was found that the model predicted pressure drop in a particle bed with nonspherical particles with an error as low as 5% and as high as 28% near the fluidization velocity.