Search

The SPECTRA code has been developed as an integrated safety analysis tool for comprehensive analyses of in- and ex-vessel phenomena during a severe accident (SA) in a sodium-cooled fast reactor (SFR). The individual modules, such as sodium fire and sodium-concrete reaction, have been implemented into SPECTRA to simulate specific phenomena in an accident of SFRs. However, SPECTRA has no capability for the overlapped event involving sodium fire and sodium-concrete reaction in the same compartment because the sodium pool and the floor concrete are modeled in each module independently. The capability of SPECTRA is enhanced by integrating the analyses of sodium pool fire and concrete ablation for overlapped events of the ex-vessel phenomena. A new shared module for the sodium pool and floor concrete region, named “PFC module” was introduced in the previous study. The sodium fire module is connected to the PFC module in this study. The integrated model is validated through the benchmark analysis of the F7-1 pool fire experiment. The calculation results of the temperature and combustion rate show good agreement with the experimental result. A demonstration analysis is also conducted for an overlapped event of the ex-vessel phenomena including both concrete ablation and sodium pool fire. SPECTRA can simulate a reasonable heat and mass transfer behavior during the overlapped event.

The effectiveness evaluation of safety measures against severe accident is necessary for restart of experimental sodium-cooled fast reactor Joyo in Japan. These safety measures correspond to those in defense-in-depth (DiD) level 4. In the previous study, an internal event level-1 probabilistic risk assessment (PRA) at power was performed to calculate frequencies of the accident sequences of failure of safety measures in DiD levels 1 to 3, to identify dominant accident sequence groups, and to identify dominant accident sequence for selecting important accident sequences in each accident sequence group, which are needed for implementing the effectiveness evaluation of safety measures in DiD level 4. Based on this, the present study implemented internal event level-1 PRA at power to show quantitatively reduction of those occurrence frequency by the safety measure in the DiD level 4. As a result, the frequency of each accident sequence group decreased significantly, and the total frequency of the accident sequence groups decreased to about 1×10-6 /reactor-year, which is about 1/1000 times the one estimated in the previous study. The protected loss of heat sink was the largest contributor in all the accident groups, and a dominant accident sequence in each accident group was also identified in this study.

This study aims to understand the time-dependent change in the occurrence rate of leak from steam generator (SG) tubes in sodium-cooled fast reactors (SFRs). The target SFRs in the present paper are Phenix in France and BN-600 in Russia. By reviewing publicly available literature that show data from the SFRs, we have investigated the numbers of tube-to-tubeplate welds and tube-to-tube welds, heat transfer areas of tube base metal, operating hours of SGs, dates when SG tube leak occurred, locations of leak, and corrective actions taken after tube leak events, such as replacement of the module, in which a leak occurred. Based on these, we have estimated the time to leak and quantitatively analyzed the time-dependent change of the occurrence rates of SG tube leak for each of the above-mentioned parts by hazard plotting method. The results show the time-dependent change in the rates of both Phenix and BN-600. For Phenix, the rate of leak at tube-to-tube welds shows a significant increase after 40,000 hours. This can be caused by thermal stress repeatedly exerted on the materials. In addition, the decrease in the rate was found, which is probably thanks to improved welding and SG operating conditions. For BN-600, it seems that in many cases, the probable cause of the leak was initial defects that developed to failure during the early stage of reactor operation, and that no special countermeasure was taken in the later stages. Therefore, it would be natural to assume that the rate simply decreased over time.

The Japan Atomic Energy Agency has been developing “Advanced Reactor Knowledge- and AI-aided Design Integration Approach through the whole plant lifecycle (ARKADIA)” to offer the best solutions for challenges in the design and operation of nuclear plants. A part of ARKADIA for design study, which included design optimization of components, is named as ARKADIA-Design. In the development of ARKADIA-Design, we have been developing a process to automatically optimize design parameters of structural components subjected to various kinds of loads, including thermal transients. In this paper, we propose a simplified procedure to estimate the failure probability of components subjected to thermal transients for design optimization. An objective function of this optimization is defined on the basis of failure probability of the components, because failure probability can be commonly used as an indicator of component integrity for various mechanisms, and it helps future introductions of a risk-informed performance-based approach to component design. To enable the necessary number of estimations for design optimization with practical calculation time, we aimed to reduce the number of analyses required for one estimation. For this purpose, we adopted the first-order second-moment (FOSM) method as the estimation method for failure probability in the process of optimization. An orthogonal table in the experiment design method is utilized to define the conditions of the analyses for evaluation of the mean and variance of thermal transient stress, which are used as inputs in the FOSM method. The superposition of ramp responses is also utilized to evaluate the time history of thermal transient stress instead of finite element analysis. The proposed procedure was applied in a demonstration study to optimize the thickness of a cylindrical vessel subjected to thermal transients derived from shutdown. We confirmed that the procedure can evaluate the failure probability depending on the cylinder thickness with practical calculation time.

To develop rationalized maintenance plans for nuclear power plants, the characteristics of each plant must be considered. For sodium-cooled fast reactor (SFR) plants, constraints on inspections exist due to the specialty that equipment retaining sodium must be handled, which is one of the important points that must be considered in maintenance rationalization. In this study, we propose a maintenance optimization scheme, which is a design support tool, using risk information to develop a maintenance strategy based on the system based code (SBC) concept. The SBC concept intends to provide a theoretical procedure to optimize the reliability of structure, system and components (SSCs) by administrating every related engineering requirements throughout the life of the SSCs from design to decommissioning. “The American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel Code, Code Case, N-875, Alternative In-service Inspection Requirements for Liquid-Metal Reactor Passive Components” was developed based on the SBC concept. The basic procedure of the Code Case has also been implemented in “The ASME Boiler and Pressure Vessel Code, Section XI, Division 2,” which provides requirements for Reliability and Integrity Management programs for nuclear power plants, including advanced reactors. The purpose of this study is to establish detailed procedures for the maintenance optimization scheme based on the procedure in Code Case N-875. Furthermore, a quantitative trial evaluation of the core support structure of the next SFR under development in Japan is also performed using the maintenance optimization scheme.

For nuclear power plants, probabilistic risk assessment (PRA) should be performed not only against earthquake and tsunami, which are critical events especially in Japan, but also other external hazards such as strong wind. The aim of the present study is to develop a practical PRA methodology for sodium-cooled fast reactors (SFRs) against strong wind, paying attention to the final heat sink, ambient air, that removes decay heat under accident conditions. First, this study used Gumbel distributions to estimate hazard curves of the strong wind based on weather data recorded in Japan. Second, it identified important structures, systems and components (SSCs) for decay heat removal, and developed an event tree that results in core damage, focusing on the impacts of missiles (e.g., steel pipes) caused by strong wind. It also identified missiles that can reach SSCs at elevated places, and calculated the fragility of the SSCs against the missiles as a product of two probabilities. One is a probability of the missiles that would enter an inlet or outlet of the decay heat removal system, and another is a probability of failure caused by missile impacts. Finally, it quantified conditional decay heat removal failure probabilities by introducing the fragilities into the event tree. The core damage frequency (CDF) was estimated at about 5 × 10-10/y. The dominant sequence is that strong wind causes offsite power loss and missiles, the missiles penetrate the diesel fuel tank, cause a fire, and the fire increases air temperature around the reactor building where air cooler inlets of decay heat removal systems are installed, leads to loss of power for the diesel generator for forced circulation cooling, resulting in loss of decay heat removal. Through the above, this study has developed the practical PRA methodology for SFRs against strong wind.

A direct reactor auxiliary cooling system (DRACS) under natural circulation (NC) conditions with a dipped-type direct heat exchanger (D-DHX) in the upper plenum of a reactor vessel (RV) has been investigated for enhancing the safety of sodium-cooled fast reactors. Studies of the past have revealed that core-plenum interactions, which consists of penetration of the coolant from D-DHXs into the subassemblies and the narrow gap between them (IWF: inter-wrapper flow), and the heat transfer through a wrapper tube among subassemblies (radial heat transfer), occurred and increased core cooling performance during the DRACS operation. Therefore, a multidimensional thermal-hydraulics analysis model in the RV using a computational fluid dynamics (CFD) code (RV-CFD model) was developed to evaluate the core cooling performance. For the design study, the RV-CFD model must simulate reasonable calculation costs while maintaining accuracy. In this study, the subchannel analysis method using the CFD code for fuel subassemblies (subchannel CFD model) was applied to the RV-CFD model. In the subchannel CFD model, the porous media approach was used to consider local geometry in the fuel subassembly, and the effective heat conductivity coefficients in a diffusion term of the energy equation were set to fit the actual radial thermal diffusion between subchannels. Two numerical simulations were compared to the experimental data obtained from the sodium experimental apparatus PLANDTL-1. In the first case, the focus was only the radial heat transfer without the D-DHX operation. In another case with the D-DHX operation, the IWF noticeably occurred, and the focus was on the core-plenum thermal interaction. The calculated sodium temperature in the core correlated well with the experimental results. The RV-CFD with subchannel CFD model was validated for core-plenum interactions during the DRACS with the D-DHX operation under NC conditions.

In order to improve the safety of nuclear power plants, it is necessary to make sure measures against their severe accidents. Especially, in the case of a sodium-cooled fast reactor, there is a possibility that molten core material would be discharged through control rod guide tubes into the inlet coolant plenums beneath the rector cores in the event of a core disruptive accident (CDA). It is important to ensure in-vessel retention that keeps and confines damaged core material in the reactor vessel even if the CDA occurs. In this study, effective cooling of the melt in coolant was confirmed by comparing the experiment and analysis. CDA scenario initiated by a unprotected loss of flow condition , which is a typical cause of core damage, is generally categorized into four phases according to the progression of core-disruptive status, which are the initiating, early-discharge, material-relocation and heat-removal phases for the latest design in Japan. During the material-relocation phase, the molten core material flows down mainly through the control rod guide tube and is discharged into the inlet coolant plenum below the bottom of the core. The discharged molten core material collides with the bottom plate of the inlet plenum. Clarification of the accumulation behavior of molten core material with such a collision on the bottom plate is important to reduce uncertainties in the safety assessment of CDA. In present study, in order to make clear behavior of core melt materials during the CDAs of sodium-cooled fast reactors, analysis was conducted using the SIMMER-III code for melt discharge simulation experiments by Imaizumi et al. in which low-melting-point alloy was discharged into a shallow water pool. As the result, temperature and pressure behaviors during the discharge almost coincided between the analysis and the experiment. Therefore, it can be concluded that the validity of the analysis cell system model was confirmed.

In a postulated severe accidental condition of sodium-cooled fast reactor (SFR), the eutectic reaction between boron carbide (B4C) as control rod element and stainless steel (SS) as control rod cladding or related structure materials may take place. Thus, the kinetic behavior of the B4C-SS eutectic reaction is one of the important phenomena to be considered when evaluating the core disruptive accidents in SFR. In this study, for the first step to obtain the fundamental information on kinetic feature of the B4C-SS eutectic reaction and compare the pervious findings, the thermal analysis using the pellet samples of B4C and Type 316L SS as a different experimental approach was performed up to 1773 K at different heating rates of 2.5-10 K min-1. The differential thermal analysis (DTA) endothermic peaks for the B4C-SS eutectic reaction appeared from 1483K to 1534K and systematically shifted to higher temperatures with increasing heating rate. Based on this kinetic feature, apparent activation energy and pre-exponential factor for the B4C-SS eutectic reaction were determined by Kissinger method. It was found that the kinetic parameters obtained by thermal analysis were comparable to the literature values of the thinning experiment at high temperatures. In addition, the microstructure and element distribution formed in the interdiffusion layer composed of the B4C-SS system were analyzed by the electron probe microanalyzer (EPMA), which can provide key validation data on elemental interdiffusion behavior in the early stage of the eutectic reaction for reflecting the reaction kinetic modeling.

A sodium-cooled fast reactor has been designed to attain a high burn-up core in commercialized fast reactor cycle systems. The sodium-cooled fast reactor adopts a wire spacer between fuel pins. The wire spacer performs functions of securing the coolant channel and mixing between subchannels. In high burn-up fuel subassemblies, the fuel pin deformation due to swelling and thermal bowing may decrease the local flow velocity in the subassembly and influence the heat removal capability. Therefore, understanding the flow field in a wire-wrapped pin bundle is important. This study performed particle image velocimetry (PIV) measurements using a wire-wrapped three-pin bundle water model to grasp the flow field in the subchannel under conditions, including the laminar to turbulent regions. The PIV results confirmed that the normalized flow velocity near the wrapping wire in the low Re number condition was relatively decreased compared to that in the high Re number condition. Meanwhile, in the region away from the wrapping wire, the maximum flow velocity was increased by decreasing the Re number. Accordingly, the PIV measurements using the three-pin bundle geometry without the wrapping wire were also conducted to understand the effect of the wrapping wires on the flow field in the subchannel. The results confirmed that the mixing due to the wrapping wire occurred, even in the laminar condition. These experimental results are useful not only for understanding the pin bundle thermal hydraulics, but also for the code validation.

Many sensitivity calculations are necessary to evaluate the effect of the uncertainties of Core Disruptive Accident (CDA) scenarios on debris bed coolability for Sodium-cooled Fast Reactors (SFR). This paper describes a calculation model and a technique for high-speed calculations of debris bed coolability. Firstly, the detailed coolability calculation model taking into account phase change and vapor-liquid advection for a boiling zone in the debris bed is derived as a single equation. The vapor-liquid advection is based on the momentum equation including inertial, viscous, capillary and gravitational effects. In addition, the streamlined calculation which ignores inertial and gravitational effects for the boiling zone is proposed. Secondly, both the detailed and streamlined calculation models are validated through the analysis of the Sandia ACRR-D10 in-pile experiments. Then, the effects of the ignoring on debris bed coolability are evaluated through sensitivity calculations. Thirdly, the speed-up technique for the streamlined calculation model is proposed which solves the boiling zone by the combination of transient and steady state calculations and allows using coarse mesh size and time step. Owing to the technique, the calculation time becomes about 1/50. The application of the speed-up technique to the streamlined calculation model is confirmed under typical SFR conditions. Finally, it is demonstrated that the streamlined calculation model and the speed-up technique enable us to perform many sensitivity calculations in a realistic time. The model and technique developed in this study is practical to evaluate the effect of the uncertainties of CDA scenarios on debris bed coolability for SFR.

Understanding the effect of uncertainties of Core Disruptive Accident (CDA) scenarios on debris bed coolability on a core catcher is required for decision making on design options to mitigate a CDA consequence. For the understanding, a huge number of calculations are required but are extremely difficult to perform because a huge number of calculations require much calculation time to solve non-steady equations in the coolability calculation model. Thus, we applied Artificial Neural Network (ANN), which is one of models for machine learning, to debris bed coolability calculations. The application of ANN is expected to exponentially improve the calculation speed of debris bed coolability because ANN provides results from experimental rules learned through training without solving non-steady equations. The application is in three steps. Firstly, we created many data for training ANN and validating the trained ANN through coolability calculations parameterizing main dominant inputs (particle diameter of debris bed, porosity of debris bed, etc.) by using Latin hypercube sampling. Secondly, ANN was trained and validated with the created data. The accuracy rate of the results by the ANN to the validation data exceeded 99%. In addition, the calculation time using ANN was micro seconds order. Finally, through demonstration calculations, it was confirmed that we can easily understand the effect of uncertainties of CDA scenarios on debris bed coolability owing to results visualization based on a huge number of parametric calculations using ANN. Thus, the application of ANN to debris bed coolability calculations should contribute to the decision making on design options to mitigate a CDA consequence.

Fully natural circulation decay heat removal systems (DHRSs) are adopted for sodium fast reactors, which is a passive safety feature without any electrical pumps. It is needed to grasp the thermal-hydraulic phenomena in the reactor vessel and evaluate the coolability of the core under the natural circulation not only for the normal operating condition but also for severe accident conditions. In this paper, the numerical results of the preliminary analysis for the sodium experimental condition with the PLANDTL-2, in which the core and the upper plenum with a dipped-type direct heat exchanger (DHX) were modeled, are discussed to establish an appropriate numerical models for the reactor core including the gap region among the subassemblies and the DHX. The transient analysis simulating the reactor scram reveals that the 3-dimensional large scale flow structure is developed through the gaps in the whole of the core area during the reactor scram. The steady-state analysis coinciding Richardson number between the PLANDTL-2 and the reactor operation condition reveals that the hot spot and cold spot appear depending on the location of the DHX, which is caused by the complex thermal-hydraulic phenomena driven by the natural circulation. From these preliminary analyses, the characteristics of the thermal-hydraulics behavior in the PLANDTL-2 to be focused are extracted.

The loss-of-reactor-level (LORL) where the coolant circulation path is lost is one of the important accident types of loss-of-heat-removal-system (LOHRS) in loop-type sodium-cooled fast reactors (SFRs). Due to the non-negligible possibility obtained by probabilistic risk assessment (PRA), pipe failures and succeeding sodium leakage in two points both occurred in primary heat transport system (PHTS) was assumed in this study, unlike the conventional studies assuming single pipe failures. The sodium level in reactor vessel (RV) is affected by many elements such as leakage position, guard vessel volume, operational state of pumps, and countermeasures to maintain sodium level which are sodium pump-up into RV and siphon-breaking of the pipe between RV and pump. Thus, a calculation program was developed in this study to evaluate and discuss the effectiveness of the countermeasures and safety margins for the loss of coolant circulation path. In addition, the crack size was discussed and evaluated realistically, and t 2 (t : pipe thickness) was obtained for a sufficiently conservative value, instead of Dt/4 (D : pipe diameter) that was assumed in the conventional studies. Time interval between two leakages was also given by PRA, considering failure rates of the pipes and components. Representative sequences and leakage positions where the sodium level can decline below emergency sodium level (EsL) were chosen, and the sodium level transient in RV was calculated where the crack size of the second leakage was set t 2. The calculations were also conducted where the larger crack size, Dt/4, was set for both the first and second leakage, in order to investigate additional requirements to maintain the RV sodium level. The evaluation results clarified that the coolant circulation loop can be maintained even after the second leakage in PHTS, taking into account the effects by the countermeasures.

This paper describes the development of a probabilistic risk assessment (PRA) methodology against a combination hazard of strong wind and rainfall. In this combination hazard PRA, a hazard curve is evaluated in terms of maximum instantaneous wind speed, hourly rainfall, and rainfall duration. A scenario analysis has provided event sequences resulting from the combination hazard of strong wind and rainfall. The typical event sequence was characterized by the function loss of auxiliary cooling system, of which heat transfer tubes could crack due to cycle fatigue caused by cyclic contacts with rain droplets. This cycle fatigue crack could occur if rain droplets enter into the air cooler of the system following the cooler's roof failure due to strong-wind-generated missile impact. This event sequence has been incorporated into an event tree which addresses component failure caused by the combination hazard. As a result, a core damage frequency has been estimated to be about 10-6/year in total by multiplying discrete hazard frequencies by conditional decay heat removal failure probabilities. The dominant sequence is the manual operation failure of an air cooler damper following the failure of external fuel tank due to the missile impact. The dominant hazard is the maximum instantaneous wind speed of 20−40 m/s, the hourly rainfall of 20−40 mm/h, and the rainfall duration of 0−10 h.

For the prototype sodium-cooled fast reactor, MONJU, the mechanical energy and structural response under energetics caused by neutronic power excursion during Unprotected Loss of Flow (ULOF) accident were preliminarily evaluated. In the first licensing of MONJU, pressure-volume relation (P-V relation) was evaluated based on the maximum theoretical work energy possible for an expanding core. It was adopted in the structural response analysis of the reactor vessel as the input. The maximum theoretical work energy is called Fuel Vapor Work Potential (FVWP) in this paper. In the successive studies of the energetics, mechanical energy was evaluated with the code in which mechanistic modelling of core expansion was implemented and this might reduce the Actual Work Potential (AWP) by an order of magnitude below FVWP. In order to evaluate the realistic structural response of the reactor vessel using AWP, method to convert the AWP to the P-V relation is necessary. Therefore, we developed the method to obtain realistic P-V relation based on the AWP by tracing the surface of the expanding core, and then we evaluated the mechanical energy and structural response under energetics during ULOF accident in MONJU using the developed method. The AWP is evaluated to 3 MJ based on the result of the latest ULOF analysis in which FVWP was evaluated to 30MJ, and sodium slug does not impact on the lower surface of the shield plug and no residual strain of the reactor vessel is evaluated. When FVWP is assumed to be 500 MJ as a hypothetical condition covering the conservative energy production, corresponding AWP is evaluated to 33 MJ. In this case, sodium slug impacts on the lower surface of the shield plug and residual strain of the reactor vessel of 0.008% at the maximum is evaluated, however the integrity of the primary boundary is still maintained.

Under the cooperative effort of the Civil Nuclear Energy R&D Working Group within the framework of the U.S.-Japan bilateral, Argonne National Laboratory (ANL) and Japan Atomic Energy Agency (JAEA) have been performing benchmark study using the Japan Sodium-cooled Fast Reactor (JSFR) design with metal fuel. In this benchmark study, core characteristic parameters at the beginning of cycle were evaluated by the best estimate deterministic and stochastic methodologies of ANL and JAEA. The results obtained by both institutions show a good agreement with less than 200 pcm of discrepancy in the neutron multiplication factor, and less than 3% of discrepancy in the sodium void reactivity, Doppler reactivity, and control rod worth. The results by the stochastic and deterministic approaches were compared in each party to investigate impacts of the deterministic approximation and to understand potential variations in the results due to different calculation methodologies employed. From the detailed analysis of methodologies, it was found that the good agreement in the multiplication factor from the deterministic calculations comes from the cancellation of the differences in the methodology (0.4%) and nuclear data (0.6%). The different treatment in reflector cross section generation was estimated as the major cause of the discrepancy between the multiplication factors by the JAEA and ANL deterministic methodologies. Impacts of the nuclear data libraries were also investigated using a sensitivity analysis methodology. The differences in the inelastic scattering cross sections of U-238, ν values and fission cross sections of Pu-239 and μ-average of Na-23 are the major contributors to the difference in the multiplication factors.

In the last two decades, CFD codes have been widely used for the design studies of NPPs. Recently high-precision simulation models have been developed to evaluate complicated phenomena, e.g. fuel melting in severe accident. The authors also are developing a high-precision CFD code with an interface tracking method to simulate the gas entrainment (GE) phenomena in sodium-cooled fast reactors (SFRs), which might be caused by a highly-intensified free surface vortex. The GE in SFRs is characterized by an elongated interfacial dent along the vortex core and the bubble pinch-off at the tip of the dent. To simulate this complicated phenomenon, the authors’ CFD code has physics-basis algorithms which model accurately the interfacial dynamic behavior, the pressure jump condition at an interface and the surface tension. Several verification problems, e.g. the slotted-disk problem, have been already solved and the accuracy of each individual algorithm is confirmed. In this paper, a basic experiment of the GE is simulated to validate the developed CFD code. In the experiment, the entrained gas flow rate is measured by image processing with a high-speed video camera. The simulation result of the entrained flow rate shows comparable value to the experimental data, that is, our CFD code is considered applicable to the evaluation of the GE in SFRs.

In a sodium-cooled fast reactor (SFR), liquid sodium is used as a heat transfer fluid because of its excellent heat transport capability. On the other hand, it has strong chemical reactivity with water vapor. One of the design basis accidents of the SFR is the water leakage into the liquid sodium flow by a breach of heat transfer tubes. This process ends up damages on the heat transport equipment in the SFR. Therefore, the study on sodium-water chemical reactions is of paramount importance for security reasons. This study aims to clarify the sodium-water reaction mechanisms using an elementary reaction analysis. A quasi one-dimensional flame model is applied to a sodium-water counter-flow reaction field. The analysis contains 25 elementary reactions, which consist of 17 H2-O2 and 8 Na-H2O reactions. Temperature and species concentrations in the counter-flow reaction field were measured using laser diagnostics such as LIF and CARS. The main reaction in the experimental conditions is Na+H2O→NaOH+H and OH is produced by H2O+H→H2+OH. It is demonstrated that the reaction model in this study well explains the structure of the sodium-water counter-flow diffusion flame.

This paper describes experimental analyses using SIMMER-III and -IV, which are respectively two- and three-dimensional multi-component multi-phase Eulerian fluid-dynamics codes, for the purpose of integral code validation. Two topics of key phenomena in core disruptive accidents of sodium-cooled fast reactors are presented in this paper: duct-wall failure and fuel discharge/relocation behavior. To analyze the duct-wall failure behavior, the SCARABEE BE+3 in-pile experiments were selected. The SIMMER-III calculation was in good agreement with the overall event progression; which was characterized by coolant boiling, clad melting, fuel failure, molten pool formation, duct-wall failure, etc.; observed in the experiment. The CAMEL C6 experiment investigated the fuel discharge and relocation behavior through a simulated control rod guide tube, which is important in evaluating the neutronic reactivity. SIMMER-IV well simulated fuel-coolant interaction, sodium voiding, fuel relocation behavior observed in the experiment. These experimental analyses indicated the validity of the SIMMER-III and -IV computer code for the duct wall failure and fuel discharge/relocation behavior.

Studies on the self-leveling behavior of debris bed are crucial in the assessment of core disruptive accident (CDA) that could occur in sodium-cooled fast reactors (SFR). To clarify the mechanisms underlying this behavior, several series of experiments were elaborately designed and conducted within a variety of conditions in recent years, under the close collaboration between Japan Atomic Energy Agency (JAEA) and Kyushu University (Japan). The current contribution, including knowledge from both experimental analyses and empirical model development, is focused on a recently developed comparatively larger-scale experimental facility using gas-injection to simulate the coolant boiling. Compared to the previous investigations, this facility can achieve a much wider range of gas velocities (presently up to a flow rate of around 300 L/min). Based on the experimental observation and quantitative data obtained, influence of various experimental parameters, including gas flow rate (～ 300 L/min), water depth (180 mm and 400 mm), bed volume (3 ～ 7 L), particle size (1 ～ 6 mm), particle density (beads of alumina, zirconia and stainless steel) along with particle shape (spherical and non-spherical) on the leveling is checked and compared. As for the empirical model development, aside from a base model which is restricted to predictions of spherical particles, the status of potential considerations on how to cover more realistic conditions (esp. debris beds formed with non-spherical particles), is also presented and discussed. This work, which gives a large palette of favorable data for a better understanding and an improved estimation of CDAs in SFRs, is expected to benefit future analyses and verifications of computer models developed in advanced fast reactor safety analysis codes.