Due to the improvement of heat transfer by nanofluids, an understanding of the interactions between nanoparticles and the base fluid is essential for simulation. The relative or slip velocity between nanoparticles and the base fluid is one of the main factors in choosing the multiphase mixture model approach. In this paper, a new slip velocity is proposed and used to compare the simulation result to the experimental results of natural convective flow in a cavity filled with an alumina nanofluid. Therefore, the ANSYS-Fluent 15.0 software is employed and the new slip velocity is applied as a user-defined function. The new slip velocity is a result of the combination of Brownian and thermophoretic diffusions, lift, buoyancy and centrifugal forces, virtual mass, pressure gradient, Van der Waals attraction and electric double layer repulsion forces. The comparison between these forces and induced drag force will provide the corresponding slip velocity. The simulation results were in good agreement with the flow pattern and heat transfer features of the experimental studies in the literature. It was found that thermophoretic and electrostatic slip mechanisms should essentially be considered in simulations, as well as buoyancy force. The major effects of electrostatic slip velocity are mainly seen in concentration higher than 1 vol%, while thermophoresis could not be ignored in any concentration. Therefore, the implemented slip velocity reveals some critical aspects of nanoparticle and base fluid interactions compared to an algebraic velocity. (C) 2016 Elsevier B.V. All rights reserved.