Abstract
This paper investigates the cooling performance of nanofluid (NF) mixed convection in a porous I-shaped electronic chip with an internal triangular hot block using Buongiorno's two-phase model. This type of cavity and hot block geometry has not been studied formerly. The NF was assumed to be a mixture of water and CuO nanoparticles (NP) up to 4% of volume concentration. As most published mathematical models for the thermal conductivity of NF give inaccurate predictions, a new predictive correlation for effective thermal conductivity was also developed with a high accuracy compared to the experimental data. The results showed that any increase in the NP volume concentration enhances the average Nusselt number (Nu¯) and the normalized entropy generation, and reduces the thermal performance of the cavity in all orientations of the hot block. The maximum enhancement in cooling performance was 17.75% and occurred in the right-oriented hot block in the sand-based porous cavity. Furthermore, adding the NP to the base fluid leads to a more capable cooling system and enhances the irreversibility of the process.