The RSM approach to develop a new correlation for density of metal-oxide aqueous nanofluids

Density is one of the basic thermo-physical properties of a fluid that should be analysed before conducting performance analysis of the fluid. In the present study, the influence of design parameters on the density of nanofluids was investigated using response surface methodology (RSM). The quadratic model generated by RSM is used to predict the performance parameters, i.e. temperature and mass concentration with reasonably good accuracy. Improved empirical correlations were proposed based on the experimental data for density prediction of the metal oxide nanofluids. Experimentally measured densities of two different nanofluids at the nanoparticle mass fraction of up to 0.2% and the temperature range of 20°C-40°C were examined. The enhancement of densities compared to the density of base fluid at 20°C and 40°C, are about 0.217% for 0.2% fraction of ZnO nanoparticles. In addition, the densities of SiO₂ nanofluids are improved 0.117% compared to the density of distilled water. Finally the RSM outputs were compared with the results obtained from the experimental data. It was observed that the optimal RSM model is accurate and the absolute maximum deviations measured values from the predicted densities of ZnO and SiO₂ nanofluids are 0.008% and 0.014% respectively. (paper)