Turbulent Rayleigh–Bénard convection in a supercritical [formula omitted]-based binary mixture with cross-diffusion effects

Abstract

CO2-based binary mixtures are promising in energy engineering to improve efficiencies of thermodynamic cycles where supercritical CO2 prevails. Indeed, under supercritical pressure and transcritical temperature conditions, special thermophysical properties of binary mixtures make the heat transfer process accompanied by enhanced cross-diffusion effects, i.e., the Soret effect and the Dufour effect. Their influences on heat transfer have not been well understood yet. This paper numerically investigates turbulent Rayleigh–Bénard convection in CO2-C2H6 binary mixture. Calculations are performed based on a spectral element method solver considering various temperature differences and pressures. Results indicate that the flow field presents a double-vortex structure. The thickness of the boundary layer is negatively correlated with the flow intensity. Under a positive separation ratio, the concentration gradient induced by cross-diffusion effects enhances buoyancy, which in turn reinforces the natural convection and heat transfer. The enhancement becomes stronger as the critical pressure is approached. Turbulent fluctuations are also strengthened by cross-diffusion effects, except for temperature fluctuations. The unusual reduction in temperature fluctuations is attributed to the increased size of the thermal plume caused by the Dufour effect.