Abstract: DEM–CFD simulations of a gas–solid fluidized bed are presented and compared to experimental results of the small scale challenge problem (SSCP) from the National Energy Technology Laboratory. It has been well recognized that the pressure drop and the solid velocity profile across the bed may be over-predicted with the widely used constitutive relations for gas–solid flows in the Eulerian two-fluid frame work. An attempt is made to address this problem using energy and force balances across a bubbling fluidized bed to attenuate the drag coefficient while preserving the correct velocity profile. This is done with the DEM model for particulate phase and the new drag law is applied to the Lagrangian–Eulerian model for particulate flows. We show that a drag model used in simulating gas–solid fluidized bed should account for the energy dissipation in both a particle-rich and particle-lean (bubbles and slugs) regions of the fluidized bed. The new drag model proposed in this work has taken into account the additional energy dissipation from forming heterogeneous structures and performs better than the standard closure models used for bubbling fluidized bed simulations in the selected cases we have examined against NETL data. Such approaches to closure model building enforce the macroscopic conservation principles while attempting to still accommodate local variations in drag due to cluster formation mechanisms.