Natural Convection Immersion Cooling of the Cylinders in Nanofluids: Developing a New Nusselt Number Correlation

Abstract

This study focuses on exploring the natural convection heat transfer within water-alumina nanofluids and its practical applications in immersion cooling. The aim is to improve the efficiency of this cooling method by utilizing fluids with improved thermal properties. The selected geometry is a vertical cylinder, which is of great significance in engineering applications and academic research. The used nanofluid consists of \({{\text{Al}}}_{2}{{\text{O}}}_{3}-{\text{water}}\) nanofluid, with varying volume fractions (0.1, 0.2, and 0.5%). Experimental assessments were carried out using a steady-state methodology. Numerical simulations employ the single-phase and two-phase mixture approaches. The results of this research reveal the impressive accuracy of the mixture method in simulating external natural convection when compared to concurrently obtained experimental data. Furthermore, in the present paper, the single-phase method has yielded results deemed acceptable and closely aligned with the outcomes from the two-phase method. This alignment was achieved through the utilization of appropriate relationships, ensuring the accurate estimation of thermophysical properties. Notably, it becomes evident that established correlations for the Nusselt number correlation of natural convection designed for conventional fluids and the mere incorporation of the thermophysical properties of nanofluids are insufficient for the accurate prediction of the Nusselt number for nanofluids. In response to this challenge, a novel Nusselt correlation is introduced, comprehensively considering the thermophysical properties of \({{\text{Al}}}_{2}{{\text{O}}}_{3}-{\text{water}}\) nanofluids, nanoparticle transport mechanisms, geometric attributes of the studied object, and nanoparticle volume fraction. Comparative assessments with previous correlations emphasize the enhanced predictive accuracy of the proposed innovative correlation.