Experiments on Supercritical Heat Transfer to CO2 in Tubes & Annuli

by Yoon Y. Bae
Korea Atomic Energy Research Institute (KAERI)

Click Here To Access The Presentation

Heat transfer at a supercritical pressure has received continued research interest from scientists and engineers in various fields since the advent of a fossil fuel fired power plant, which operates at a condition above the critical point. Recently a heat exchanger with a micro channel introduced supercritical pressure carbon dioxide as a coolant to enhance the heat transfer efficiency by taking advantage of a dramatic change of the physical properties at around a pseudo-critical temperature. Since the resumption of research on a concept of a supercritical pressure water cooled reactor (SCWR) in the early 1990s, a supercritical heat transfer in a sub-channel, where the pressure is well above the critical pressure, has become an interesting research topic. Heat transfer at a supercritical pressure is totally different from that at a subcritical pressure due to the substantial variations of the physical properties of fluids at around a pseudo-critical temperature. The convective heat transfer correlations developed based on the test data obtained at a subcritical pressure are no longer valid since the nondimensional parameters based on the bulk temperature or the averaged temperature are not sufficient to interpret the dynamic and thermal behaviors at a supercritical pressure.

Experiments on heat transfer at supercritical pressures have been carried out by many researchers with different mediums such as water, carbon dioxide and Freon. Although the coolant in a SCWR is a water, carbon dioxide was used as a surrogate medium in our experiments in order to reduce experimental cost and lower safety risk since its properties show very similar behaviors at supercritical pressures when they are reduced by critical values except for the kinematic viscosity. It has been suggested that a heat transfer correlation obtained from an experiment with carbon dioxide may be directly applied to a calculation of a heat transfer to water without any correction, since an inevitable error arising from the fluid-to-fluid scaling effect between water and carbon dioxide is not as significant as expected.

A series of experiments using various test sections including tubes and annuli have been carried out at KAERI using a test facility named as SPHINX. The tested geometries include three different size tubes (4.4, 6.32 and 9.0 mm) and two types of annulus channels (concentric 8.0 x 10 mm and eccentric 9.5 x 12.5 mm). The effect of heat flux, mass flux, channel hydraulic diameter and pressure was investigated. The concentric annulus channel corresponds to the 4.4 mm ID tube and the difference between them was examined also. The test results were compared with the predictions from the existing heat transfer correlation in order to review the known correlation critically. Upon realizing a need for a unified correlation, which is valid for a deterioration range as well as a normal range, a new correlation was suggested based on our experiments although a further improvement is still needed.