Effect of Magnetic Field and Impingement Jet on the Thermal Performance and Heat Transfer of Hybrid Nanofluids

In this paper, we focus on modeling the flow and heat transfer behavior of SiO2–CuO/water hybrid-nanofluid impingement jet used for CPU cooling, where this flow is subject to a magnetic field. For this purpose, a new geometry has been adopted that contributes to the processor’s cooling while controlling the dynamic field and making it stable. The assessments were performed using two-phase mixture model under laminar forced convection flow setting. The working liquid consists of SiO2 and CuO nanoparticles with a diameter of 20 nm dispersed in the base fluid. The flow field, heat transfer, thermal efficiency, loss pressure and entropy production were analyzed in terms of volumetric concentration, Hartmann number, and Reynolds number. The simulation approach was applied to compare previous research findings, and a considerable agreement was established. Results indicate that the use of outside magnetic forces aids in maintaining the working fluid’s stability. Boosting the Hartmann number to maximum values increases pressure drop and pumping power while lowering system efficiency by 5%, 5% and 19%, respectively. Compared to pure water, hybrid nanofluids yield to a considerable drop in mean CPU temperature up to 10 K. The hybrid nanofluid’s efficiency improves as the Reynolds number and nanoparticle volume fraction rise, where the improvement in the best conditions reaches up to 21% and 27%, respectively. Using the following nanoparticles: SiO2, CuO and SiO2–CuO improves the Nusselt number of the base fluid by 15%, 36% and 30%, respectively. While the pressure drop values increase by 5%, 17% and 11%. Regarding the entropy production, the results reveal that the total entropy values increase slowly with the volume fraction of the nanoparticles, and the maximum increase does not exceed 5% in the best case. On the other hand, the increase in the total entropy values reaches 50% when Ha = 20. Lastly, two correlations for the Nusselt number and the friction factor are suggested, with errors of no more than ±9% and ±7%, respectively.