Tamotsu Kudo

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.

Experimental study on in-vessel debris coolability in ALPHA program

Abstract In-vessel debris coolability experiments were performed in ALPHA program at JAERI. Aluminum oxide (Al 2 O 3 ) produced by a thermite reaction was used as a debris simulant. Approximately 30 and 50 kg of Al 2 O 3 were poured into a pool of nearly saturated water at the ambient pressure of approximately 1.3 MPa formed in a lower head experimental vessel. The post-test visual observation and measurement using an ultrasonic technique indicated the formation of a thin porous layer at the vicinity of the surface of the solidified Al 2 O 3 and the interfacial gap between the solidified Al 2 O 3 and the lower head experimental vessel wall. Thermal transient characteristics on the lower head experimental vessel wall observed in the experiments implied that the interfacial gap and the thin porous layer in the solidified Al 2 O 3 layer acted as a thermal resistance during the initial heat-up stage, and water subsequently penetrated into the interfacial gap to effectively cool the lower head experimental vessel wall. The maximum heat flux removed from the experimental vessel was ranged from approximately 190 to 360 kW m −2 while the temperature of the vessel wall decreased rapidly.

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.

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.

Small-scale component experiments of the penetration leak characterization test in the ALPHA program

A small-scale penetration leak characterization test has been performed as a part of the ALPHA program at Japan Atomic Energy Research Institute (JAERI). Two series of experiments were performed using test sections which simulate relevant parts of an EPA (Electrical Penetration Assembly) used in Japanese PWR containments. One of the test sections simulates an alumina module and the other includes the silicone resin portion of the EPA. The test section was heated in a leak test vessel which simulated thermal-hydraulic conditions inside and outside of the containment in a severe accident. From the experimental results, it was concluded that although the silicone resin may melt at high temperature, the alumina module will remain intact under severe accident conditions. The EPA as a whole is estimated to maintain leak-tightness during a severe accident. It was found in the experiments that heat conduction along the metal portion of the test section had a strong influence on the melt progression of the resin. It was also found that the measured strain of the alumina module was predominantly caused by the elevated temperature. Therefore, the thermal load will be more of a threat to the EPAs integrity rather than the pressure load.

Revaporization of a CsI aerosol in a horizontal straight pipe in a severe accident condition

Abstract Aerosol revaporization in piping is being investigated in the WIND project at the Japan Atomic Energy Research Institute. The objectives of this study are to characterize the aerosol revaporization from piping surfaces under various thermal-hydraulic conditions and to obtain insights applicable to the validation of analytical models. Cesium iodide aerosol was introduced into the test section with a carrier gas. After quantifying the deposited mass of cesium and iodine, the test section was reheated to realize the revaporization. The revaporized materials were deposited onto another test section with an axial temperature gradient located downstream. Two runs (WAV1 and WAV2) were conducted. In WAV2, the influence of metaboric acid was examined. Most of the deposited cesium and iodine in the test section was revaporized and transported downstream. In WAV2, deposition density of cesium was much larger than that of iodine. It was supposed that a part of the cesium iodide that was deposited in the upstream test section reacted with boric oxide to form cesium metaborate.

Effect of Microstructure on Failure Behavior of Light Water Reactor Coolant Piping under Severe Accident Conditions

In a severe accident of light water reactors, the reactor coolant system (RCS) piping might be subjected to thermal loads caused by the decay heat of the deposited fission products and the heat transfer from the hot gases, with an internal pressure in some accident sequences. Tests on the RCS piping failure were performed along with high temperature tensile and creep rupture tests including metallography to investigate the failure behavior. The prediction of the 0.2% proof stress by Arrhenius equation is in good agreement with the measured stress above 800°C for served RCS piping materials. The modified Nortons Law for the short term creep rupture model agrees with the experimental values between 800 and 1,150°C for type 316 stainless steel. The microstructural change was discussed with the effect of the very rapid formation and resolution of the precipitation on the strength at high temperature. The result of the piping failure tests which simulated the severe accident conditions, i.e., in short-term at...