A new drag correlation for monodisperse suspensions using immersed boundary method simulations

Quantitative description of gas-solid momentum transfer is of fundamental importance to predict the behavior of suspensions of sedimenting or fluidized particles. It is usually characterized by the dependence of normalized average drag force F_d on mean Reynolds number (Re) and solids volume fraction (φ). In this work we report detailed direct numerical simulation (DNS) results of interphase momentum transfer in flows past fixed assemblies of monodisperse spheres using an iterative immersed boundary method (IBM). A methodology has been applied for all the IBM simulations, which was validated to obtain highly accurate results of drag force and flow velocity field at relatively low computational cost. Simulations were performed for Re ranging from 50 to 1000 and φ varying from 0.1 to 0.6. A new drag correlation is finally obtained with an average standard deviation of only 4.0% from the underlying IBM simulation data. In comparison to some correlations reported in literature, this new correlation gives more reliable predictions of the drag force for high Re and dense packings. It is considered to be so far the best possible expression of the drag law for gas-solid flows. In addition, this new drag correlation can be used in large-scale models, in which the interphase momentum exchange is not directly resolved, thus finally enhancing the predictive capability of computational fluid dynamics (CFD) simulations for gas-solid flows.