Abstract
Gas fluidized beds with bi-dispersed particles of different sizes, densities, and shapes are encountered in many industrial processes spanning the manufacture of energy, material and resources to greenhouse emissions mitigation and renewable resources utilization. The design and operation of such systems heavily rely on the prediction of minimum fluidization velocity of the mixture (Umf(mixture)) to avoid particle segregation and maximize mixing. Many correlations exist for predicting Umf(mixture); however, they exhibit limitations in accuracy and/or applicability. The limitations of these correlations are often expressed in terms of particles properties (e.g. effective size and density), flow conditions, and solids composition. This study has critically assessed highly cited correlations, delivered a comparative analysis of their applicability and predictability against a large pool of experimental data, and deepened the understanding of parameters that affect the prediction of Umf(mixture) using different correlations.
[Display omitted]
•All collected data were correlated by the mixture's Archimedes and Reynolds numbers.•Correlations used Umf and Wt% of each particle had the highest prediction accuracy.•Applicability of correlations used Umf and Wt% of each particle can be generalized.•Overestimations were observed by correlations developed involving biomass particles.