Thermal conductivity of nanofluids: A review of the existing correlations and a scaled semi-empirical equation

Efficiency of energy systems can be improved in different ways. One of these consists in adopting heat transfer fluids with better thermo-physical properties, e.g. thermal conductivity and dynamic viscosity, which can improve convective heat transfer coefficient. Being mixtures of high-conductive nanoparticles and common base fluids, nanofluids have the potential to increase the efficiency of a large number of energy systems. Thus, great importance should be dedicated to a correct and reliable estimation of the thermophysical properties of these fluids. In this work, the thermal conductivity of 11 nanofluids, for a total of 239 experimental points, was analyzed in detail: 7 nanofluids have water as base fluid (Ag, Al2O3, CuO, Fe2O3, SiO2, TiO2, ZnO), while 4 nanofluids have ethylene glycol (Al2O3, CuO, SiC and SnO2). All nanofluids’ thermal conductivity data derive from experimental measurements available in literature, carried out with samples characterized by stable preparation methods. The thermal conductivity of the nanofluids was estimated with 13 well-known correlations, in order to verify their accuracy. A new semi-empirical, scaled equation for predicting the thermal conductivity of nanofluids was also proposed. The equation requires the use of six parameters (volume fraction, temperature, base fluid critical temperature, nanoparticle diameter, nanoparticle thermal conductivity, base fluid thermal conductivity) and shows small deviations respect to the experimental data, having an average absolute relative deviation of 2.60%. This value was found to be the lowest among the other studied correlations. The study also highlights some issues and limitations that the research field related to nanofluids should overcome.