Experimental and numerical study of convection heat transfer of CO 2 at supercritical pressures in vertical porous tubes

Abstract

Convection heat transfer of CO 2 at supercritical pressures in vertical sintered porous tubes with particle diameters of 0.1–0.12 mm and 0.2–0.28 mm was investigated experimentally and numerically. The study investigated the influence of the inlet fluid temperature, mass flow rate, pressure, particle diameter, heat flux, flow direction and buoyancy on convection heat transfer in porous tubes. The results show that the inlet temperature, pressure, mass flow rate, particle diameter and heat flux all strongly influence the convection heat transfer at supercritical pressures. When the inlet temperature is much larger than the pseudocritical temperature, T pc, the local heat transfer coefficients in porous tubes are much less than when the inlet temperature is less than T pc. For T 0 < T pc and wall temperatures not much larger than T pc, the local heat transfer coefficients have a maximum for both upward flow and downward flow along the porous tube when the fluid bulk temperatures are near T pc. Buoyancy caused the different variations in the local heat transfer coefficients along the porous tube for upward and downward flows. The results also show that the heat transfer coefficients increase as the particle diameter decrease. The numerical simulations were performed using the local thermal equilibrium model with consideration of the effects of variable porosity, thermal dispersion and area-of-contact stagnant effective thermal conductivity. The experimental and numerical results for the friction factor of CO 2 at supercritical pressures flowing in sintered porous tubes at constant temperature (without heating) corresponded very well with the known correlation. However, the predicted results for the friction factor of CO 2 at supercritical pressures flowing in the heated sintered porous tube differ from those measured in the experiments both for upward and downward flows. The calculated wall temperatures corresponded well with the measured wall temperatures.