Akihide Hidaka

Phenomenological studies on melt-coolant interactions in the ALPHA program

Abstract Two series of experiments to investigate melt-coolant interactions have been performed as part of the ALPHA program at JAERI. In the melt drop steam explosion experiments, melt simulating a molten core was dropped into a pool of water. Volume fractions of the melt, water and steam in the mixing region prior to the occurrence of spontaneous steam explosions were quantified. Other characteristics of melt-coolant interactions were evaluated for settling velocity of the melt in water, propagation and expansion velocities, energy conversion ratio and debris size distribution. It was found that the probability of the occurrence of spontaneous steam explosions could be reduced by using a melt dispersion device. Measurement of void fraction in the mixing region clearly showed that the melt dispersion device enhanced steam generation. However, one experiment indicated that the use of the dispersion device could possibly result in a more energetic steam explosion. It was found that the mixing of non-condensable gas in the steam phase of the mixing region during melt dispersion played an important role for the suppression of the spontaneous steam explosion. Knowledge of the parametric effects of melt mass, ambient pressure and water temperature was extended. In the melt coolability experiments, water was poured onto the melt to investigate melt-coolant interactions in a stratified geometry where water overlies on a melt layer. Melt eruptions which could induce an explosive interaction were observed when the subcooled water was poured through a pipe nozzle. The eruption was not observed when the water was near the saturation temperature or supplied through a spray nozzle. The explosive interaction in the stratified geometry was found to be much smaller in magnitude than the steam explosion in the melt drop configuration.

Enhancement of Cesium Release from Irradiated Fuel at Temperature above 2,800 K

2. Results and Discussion The total fractional releases during the two tests were estimated from changes of the off-line measured γ -ray spectra just before and after the tests because the energy resolution of on-line γ -ray measurement was insufficient due to influence of electromagnetic waves from high frequency induction coil. The fractional releases of 134 Cs, 137 Cs and 106 Ru at the end of tests evaluated by using several γ -ray peaks of isotopes are summarized in Table 1. In the evaluation, no release of Eu was assumed because the on-line measurement showed almost no change of γ -ray intensities of Eu during the tests. Moreover, one of radionuclides, that is, Eu had to be chosen as standard for comparison of the γ -ray spectra between before and after the test because the off-line γ -ray measurements were performed using different systems. The fractional releases of Cs were evaluated by averaging those obtained at the peaks of gamma energy to be 99.8% and 85% in VEGA-3 and -1, respectively. The main reason for larger fractional release of VEGA-3 than that of VEGA-1 is considered that the fuel of VEGA-3 experienced higher maximum temperature for relatively longer duration than those of VEGA-1. Figure 2 shows the fractional release histories of Cs estimated from the on-line γ -ray measurement and temperatures measured in the VEGA-1 and -3 tests. The curves were plotted so that the first and second plateaus of VEGA3 might be superposed on the second and third plateaus of VEGA-1. In order to quantify the differences of release behavior in VEGA-3 and -1 tests due to maximum temperature and its duration, the release rate coefficients 11) used commonly as a parameter of radionuclide release models were evaluated and compared with each other. The release rate coefficient is essentially equal to a fractional release rate per minute to the current inventory. The Cs release rate coefficients every minute calculated from the two tests results are shown in Fig. 3. The fitting curves of the release rate coefficients in VEGA-1 and -3 mostly follow an Arrhenius form excepting at the three temperature plateaus and in particular high temperature portion above 2,800 K. The regression equations of release rate coefficient, k (min −1 ), calculated by the least

Influence of Pressure on Cesium Release from Irradiated Fuel at Temperatures up to 2,773 K

The radionuclides release from fuel is a primary issue for evaluation of the source term in a severe accident of a light water reactor. After the TMI-2 accident, numbers of experiments have been conducted in this domain of research in the world.1–3) However, information is still insufficient for the precise source term evaluation, in particular such conditions of very high temperature and pressure. In order to obtain the data at such conditions, a systematic research program, VEGA (Verification Experiments of radionuclides Gas/Aerosol release) 4,5) was initiated at Japan Atomic Energy Research Institute (JAERI). The VEGA program aims in particular to clarify the effects of pressure and temperature on the radionuclides release. This article describes the effects of system pressure on the cesium release obtained in the first two experiments, VEGA-1 and VEGA-2, which were conducted at the same maximum temperature of 2,773 K and different system pressures.

Experimental and Analytical Study on the Behavior of Cesium Iodide Aerosol/Vapor Deposition onto Inner Surface of Pipe Wall under Severe Accident Conditions

The WAVE experiments have been performed at JAERI to investigate the CsI deposition onto the inner surface of pipe wall under typical severe accident conditions. It was shown that relatively large amount of CsI was deposited at the upstream floor of the pipe and that larger amount of CsI was deposited on the ceiling than the floor at the downstream. Analyses of the experiments have also been conducted with the three-dimensional thermohydraulic code, SPRAC, and the radionuclide transport analysis code, ART. The experimental results were well reproduced with ART by using peripherally subdivided pipe cross section and associated representative thermohydraulic information from SPRAC prediction. It was clarified through the present experiment and analyses that major deposition mechanisms for the chemical form of CsI are thermophoresis and condensation. Accordingly, the coupling of the FP behavior and the detailed thermohydraulic analyses was found to be essential in order to accurately predict the CsI depositi...

Vapor Condensation and Thermophoretic Aerosol Deposition of Cesium Iodide in Horizontal Thermal Gradient Pipes

The aerosol deposition test series is being performed to investigate the deposition of FP vapor and aerosol onto the inner surface of reactor coolant piping during a severe accident of a light water reactor. Vapor and aerosol of CsI as an FP simulant was introduced into the horizontal test section pipes. No substantial decomposition of CsI was identified to occur both in the high temperature inert and superheated steam environments. The comparison between the results of the aerosol deposition test series and the thermo-fluiddynamic analysis with WINDFLOW implied that a profile of the CsI deposition due to vapor condensation and thermophoretic aerosol deposition was influenced by local thermo-fluiddynamic conditions. Deposition velocities were evaluated for CsI vapor and aerosol based on the deposition characteristics of CsI and the thermo-fluiddynamic analysis.

Experimental and analytical study on aerosol behavior in WIND project

The tests on fission product (FP) behavior in piping under severe accidents are being conducted in the wide range piping integrity demonstration (WIND) project at JAERI to investigate the piping integrity which may be threatened by decay heat from deposited FPs. In order to obtain the background information for future WIND experiment and to confirm analytical capabilities of the FP aerosol analysis codes, ART and VICTORIA, the FP behavior in safety relief valve (SRV) line of BWR during TQUX sequence was analyzed. The analyses showed that the mechanisms that control the FP deposition and transport agreed well between the two codes. However, the differences in models such as diffusiophoresis or turbulence, the treatment of chemical forms and aerosol mass distribution could affect the deposition in piping and, consequently, on the source terms. The WIND experimental analyses were also conducted with a three-dimensional fluiddynamic WINDFLOW, ART and an interface module to appropriately couple the fluiddynamics and FP behavior analyses. The analyses showed that the major deposition mechanism for cesium iodide (CsI) is thermophoresis which depends on the thermal gradient in gas. Accordingly, the coupling analyses were found to be essential to accurately predict the CsI deposition in piping, to which little attention has been paid in the previous studies.

Proposal of Simplified Model of Radionuclide Release from Fuel under Severe Accident Conditions Considering Pressure Effect

The VEGA tests on radionuclides release from fuel under severe accident conditions showed that the Cs release rate at 1.0 MPa decreased by about 30% compared with that at 0.1 MPa. To explain this pressure effect, a numerical release model that considers the lattice diffusion in grains followed by the gaseous diffusion in open pores was developed. However, this model is not practical for the PSA analyses due to much computation time and therefore a simplified model called CORSOR-M with the release rate coefficient multiplied by (P≧1 atm) was derived from the numerical model. The multiplier comes from the pressure dependency of gaseous diffusion flux in pores at the pellet surface. The effect of pressure on source term was also estimated for a transient sequence at BWR with JAERIs THALES-2 code in which the simplified model was incorporated. Since the adequacy and applicability of CORSOR-M model were confirmed for the pressures up to 16 MPa through comparison with the VEGA tests and mechanistic models, it is proposed that the model be used for the source term analyses.

Radionuclide Release from Mixed-Oxide Fuel under High Temperature at Elevated Pressure and Influence on Source Terms

The radionuclide release from mixed-oxide fuel (MOX) under severe accident conditions was investigated in the VEGA program to provide the technical bases for safety evaluation including probabilistic safety assessment (PSA) for light water reactor (LWR) using MOX. The MOX specimens irradiated at Advanced Thermal Reactor (ATR) Fugen were heated up to 3,123K in helium at 0.1 and 1.0MPa. The release of volatile fission products (FP) was slightly enhanced below 1,623 K compared with that of UO2. The volatile FP release at elevated pressure was decreased as in the case with UO2. The total fractional release of Cs reached almost 100% while almost no release of low-volatile FP even after the fuel melting. The release rate of plutonium above 2,800 K increased rapidly although the amount was small. Since the existing models cannot predict this increase, an empirical model was prepared based on the data. The present study showed that there are no large differences in total fractional releases and inventories of important FP in PSA between UO2 and MOX. This suggests that the consequences of LWR using MOX are mostly equal to those using UO2 from a view point of risks.

Decrease of cesium release from irradiated UO2 fuel in helium atmosphere under elevated pressure of 1.0 MPa at temperature up to 2,773 K

In the case of severe accidents, the radionuclides release from fuel could mostly occur at high temperature under elevated pressure. The effect of temperature on the release has been clarified in many previous studies while the pressure influence has been scarcely investigated so far due to difficulty in the experimental operation. To investigate the effect of pressure on the release, two tests under the same conditions except for the system pressure were performed in the VEGA program at JAERI by heating up the irradiated UO2fuels up to 2,773 K in inert helium. The test results uniquely showed that the release rate of cesium for the temperatures below 2,773 K at 1.0 MPa could be suppressed by about 30% compared with that at 0.1 MPa. This article describes the outlines of the two tests and the observed effects of system pressure on cesium release as well as the results of various post-irradiation examinations. Moreover, the mechanisms and models that explain the pressure effect are discussed.

DEPOSITION OF CESIUM IODIDE PARTICLES IN BENDS AND SECTIONS OF VERTICAL PIPE UNDER SEVERE ACCIDENT CONDITIONS

Abstract A relatively small-scale aerosol deposition experiment (called WAVE) in a quartz glass pipe with a 90° bend followed by a short straight section was performed at Japan Atomic Energy Research Institute to investigate the effect of pipe orientation on the cesium iodide (CsI) aerosol deposition. In these basic configurations, the section after the bend was either horizontal, upward and downward, respectively. The upward case showed 5 and 10 times larger CsI mass deposition in the vertical section of pipe than the horizontal and downward configurations, respectively. The experiments were analyzed by coupling a three-dimensional fluid-dynamic and an aerosol behavior codes. The calculations were in reasonable agreement with the measured aerosol mass deposition except for the upward case because the principal CsI deposition mechanism is thermophoresis which depends on the thermal gradient in gas and the gradient was well predicted by the fluid-dynamic code. In order to better predict the deposited mass in vertical section of upward case, Nusselt number which is used for thermophoretic deposition calculation had to be reevaluated carefully by considering the effect of secondary free convection in the vertical pipe which flowed opposite to the main stream.