Experimental investigation of the diameter effect on heat transfer characteristics of supercritical CO₂ flows through horizontal microtubes

Experimental study on convective heat transfer through microtubes encounters many challenges such as measuring the wall temperature of the tube with a reasonable uncertainty, relatively high heat loss from the outer surface of the tube, and difficulty in measuring the fluid bulk temperature at the tube outlet. These issues become exacerbated with the flow at low Reynolds numbers. The present study presents a new method to measure the tube wall temperature with a small solder cast where a thermocouple is embedded. The net heat flux to flows, determined by the heat loss on the outer surface and the axial heat conduction, is integrated to obtain the local mean enthalpy of fluid and then the mean fluid temperature along the tube. Thus, convection heat transfer coefficients are calculated without measuring the mean temperature at the outlet. The new method is successfully validated with laminar water flows through a horizontally configured microtube, 0.501 mm in inner diameter. All Nusselt numbers are found to lie between 4.36 and 4.36 + 10 percent. Then, the new method was applied to supercritical carbon dioxide flows through horizontal microtubes with inner diameters of 0.317, 0.509, and 0.847 mm at 8.0 MPa. The inlet Reynolds number was kept lower than 1100 in all tests to examine the supercritical carbon dioxide at low Reynolds numbers. A peak in the Nusselt number was observed when the fluid temperature is near the pseudo-critical temperature. The heat transfer enhancement at the pseudo-critical temperature was associated with three mechanisms such as high specific heat, a high temperature gradient near the wall, and buoyancy-induced secondary flow in the transverse direction. Finally, the results show that the magnitudes of the Nusselt number increase with an increasing diameter since the effect of buoyancy significantly rises with the diameter.