Numerical simulation of turbulent particle-laden flows with significant fluid to particle inertia ratio

Turbulent liquid-particle flow represents a particle-laden flow regime in which the microscale fluid inertia influence on the particle fluctuating motion and consequently the fluid-particle interaction is significant. The present work examines the predictive capability of a two-phase flow CFD model that is based on the kinetic theory of granular flow in simulating dilute-phase turbulent liquid-particle flows. The model predictive capability is evaluated at both the mean and fluctuating velocity levels, where the impacts of employing different drag correlations and turbulence closure models to describe the fluid-particle interactions are examined. The results suggest that the present model is capable of producing reasonably good predictions for both phases, though not yet quantitatively accurate, provided that appropriate drag correlation and the turbulence closure model are selected. In addition, the model predictive capability is also assessed for a gas-particle flow regime in which the gas to particle inertia ratio is not insignificant. For this purpose, gas-particle flow experiments involving low inertia particles are conducted using laser Doppler velocimetry technique. In this gas-particle flow regime, the results indicate that the present model can accurately predict the gas-phase turbulence though its predictive capability for the granular temperature is still lacking particularly near the pipe wall.