On the uses of classical or improved heat transfer correlations for the predictions of convective thermal performances of water-Al2O3 nanofluids

This paper reports calculations aimed at predicting the convective heat transfer enhancement evaluated in terms of the ratio of the nanofluid and base fluid heat transfer coefficients. Either usual single-phase heat transfer correlations or improved correlations involving slip velocity due to inertia, Brownian and thermophoretic diffusions are considered. In this regard, eight classical flow configurations, both for laminar and turbulent forced or natural convection, are first examined. In each case, five to six commonly used models for the effective thermal conductivity and viscosity are compared for water-Al2O3 nanofluids, with nanoparticle volume fractions ranging from 1% to 5%. In order to recover anomalous convective heat transfer enhancements, we examine mainly the influences of simultaneous augmentations of thermal conductivity and dynamical viscosity. The opposite influences of these two fluid properties on the heat transfer coefficient are clearly established. In the second part of the Result section, we attempt to give a critical interpretation of the previous results by using new published improved correlations derived with the objective to show that the usual correlations are not relevant for predicting nanofluid convective heat transfer. However, most of the present results reveal that the increase in the particle volume fraction tends to deteriorate the heat transfer in comparison with the base fluid.