Numerical analysis of MHD combined convection for enhanced CPU cooling in NEPCM-filled a trapezoidal cavity

Abstract

This study investigates the cooling of a central processing unit (CPU) using a nano-encapsulated phase change material (NEPCM)-water mixture in a trapezoidal cavity with rotating cylinders and baffles. A numerical model based on the finite element method (FEM) is employed to solve the governing equations. The system is subjected to a sinusoidal temperature profile from the CPU and a constant magnetic field. Key parameters examined include Reynolds number (Re: 10–100), Richardson number (Ri: 0.1–10), Hartmann number (Ha: 5–80), NEPCM volume fraction (ϕ: 0.015–0.035), Lewis number (Le: 0.1–10), buoyancy ratio (Nz: 1–5), NEPCM fusion temperature (θ_f: 0.1–0.9), and Stefan number (Ste: 0.1–0.9). Results show that increasing Re and Ri significantly enhances heat and mass transfer, with the average Nusselt number (Nu_av) increasing by up to 80.5 % and average Sherwood number (Sh_av) by up to 147.9 %. The magnetic field suppresses convection, reducing Nu_av by 12.7 % and Sh_av by 39.5 % as Ha increases. Increasing ϕ improves heat transfer (Nu_av up by 32.5 %) with minimal effect on mass transfer. Le strongly influences mass transfer, with Sh_av increasing by 284.6 % as Le increases. The NEPCM fusion temperature exhibits a non-monotonic effect on Nu_av, with an optimal value at θ_f = 0.5. In conclusion, the study reveals complex interactions between parameters, with Re, Ri, and Le having the most significant impacts on system performance. These findings provide valuable insights for optimizing CPU cooling systems using NEPCM-water mixtures and magnetohydrodynamic (MHD) effects.