Experimental study of choking flow of water at supercritical conditions

Future nuclear reactors will operate at a coolant pressure close to 25 MPa and at outlet temperatures ranging from 500°C to 625°C. As a result, the outlet flow enthalpy in future Supercritical Water-Cooled Reactors (SCWR) will be much higher than those of actual ones which can increase overall nuclear plant efficiencies up to 48%. However, under such flow conditions, the thermal-hydraulic behavior of supercritical water is not fully known, e.g., pressure drop, forced convection and heat transfer deterioration, critical and blowdown flow rate, etc. Up to now, only a very limited number of studies have been performed under supercritical conditions. Moreover, these studies are conducted at conditions that are not representative of future SCWRs. In addition, existing choked flow data have been collected from experiments at atmospheric discharge pressure conditions and in most cases by using working fluids different than water which constrain researchers to analyze the data correctly. In particular, the knowledge of critical (choked) discharge of supercritical fluids is mandatory to perform nuclear reactor safety analyses and to design key mechanical components (e.g., control and safety relief valves, etc.). Hence, an experimental supercritical water facility has been built at École Polytechnique de Montréal which allows researchers to perform choking flow experiments under supercritical conditions. The facility can also be used to carry out heat transfer and pressure drop experiments under supercritical conditions. In this thesis, we present the results obtained at this facility using a test section that contains a 1 mm inside diameter, 3.17 mm long orifice plate with sharp edges. Thus, 545 choking flow of water data points are obtained under supercritical conditions for flow pressures ranging from 22.1 MPa to 32.1 MPa, flow temperatures ranging from 50°C to 502°C and for discharge pressures from 0.1 MPa to 3.6 MPa. Obtained data are compared with the data given in the literature including those collected with fluids other than water. It is also important to mention that present models used to predict supercritical choking flows have been developed for fluids under subcritical conditions. Even though none of these models were developed to handle the expansion of supercritical fluids, we tested some of the models (Homogenous Equilibrium Model, Modified-Homogeneous Equilibrium Model and Bernoulli equation) under supercritical conditions and compared their predictions with our data and those of other researchers, available in the literature. In addition, a simple polytropic model is proposed to estimate the critical flow rate of water. It is found that the Modified Homogeneous Equilibrium Model is the most appropriate model to estimate the discharge flow rate of water under supercritical conditions. Results of the model comparison must help SCWR designer to choose safety devices correctly. As a common practice, the difference between the fluid temperatures with respect to the pseudocritical value (DTpc) is used to treat the data. To this aim, it must be mentioned that a new relationship is proposed to estimate the pseudo-critical temperature of water and carbon dioxide. In particular, for flow temperatures lower than pseudo-critical values, choking flow seems to occur within a very limited region. Close to the pseudo-critical temperature, our experiments provide data in a region where up to now, are very scarce. In general, an excellent agreement with experiments carried out by other researchers is obtained. It is observed that the mass flux decreases with increasing the flow temperature upstream of the orifice. In particular, the proposed experimental arrangement (i.e., use of two loops running in parallel) permitted us to determine flow conditions that trigger supercritical water choking flow. Furthermore, a small pressure gradient occurring upstream of the orifice is systematically measured. It is also observed that close to the pseudo-critical point, the heat transfer coefficient changes very rapidly which affects the difference between the inner tube surface and coolant temperatures. These fast variations combined with the corresponding change in fluid density make it very difficult to control and maintain flow conditions in the proximity of the critical point. The research work presented in this thesis has been the subject of two presentations at international conferences, a poster session and a publication in a scientific journal. (author)