Search

The D/W (Drywell) and S/C (Suppression Chamber) pressure data of Fukushima-Daiichi Nuclear Power Plant Unit 3 was analyzed in depth. This analysis provided valuable information related to the accident progression behavior on one hand, and gave a hint for understanding of the debris-to-coolant heat transfer when fuel debris relocated to the pedestal on the other hand. In this unit, the D/W and S/C pressure increased and decreased cyclically with a relationship, which seems to have been dependent on the composition of vapor and non-condensable gases in the S/C cover gas region. Based on this characteristic, the vapor pressure in the S/C cover gas region was evaluated for two pressure decrease cycles during and after the expected debris relocation to the pedestal respectively. This evaluation allowed an understanding that the S/C vapor pressure increased due to the heat transfer from the debris relocated to the pedestal. With this understanding, several possible S/C vapor pressure histories were assumed. Based on this parametric study it was suggested that some part of the S/C water was heated up within about 75 min when the main part of the fuel debris would have been relocating to the pedestal.

The onsite investigations of Fukushima Daiichi Nuclear Power Plant (1F) Units-2 and 3 indicate possibilities of multiple breaches of the Reactor Pressure Vessels (RPVs). In the meantime, some analytical works indicate possibilities that the core materials of 1F Units-2 and 3 were once relocated to the lower plenum of the RPVs and cooled (quenched) before the water inventory boiled-off (dry out) and the once-cooled debris re-melted. This study utilizes Lagrangian-based Moving Particle Semi-implicit (MPS) method to investigate such complex solid-liquid multiphase re-melting and the debris-vessel wall interactions to obtain new insight on 1F Units-2 and 3 RPV failure modes to help the future debris retrieval from these reactors. Two major modifications/developments have been carried out based on the previously developed MPS method. Namely, stabilization of rigid-body contact model and the further improvement of the speedup algorithm to enable large and long scale debris-bed re-melting analyses of the real plant scale. The numerical accuracy of the pressure boundary condition at fluid-wall boundary has also been improved. Sensitivity analyses have been carried out with the developed new MPS method. The results indicate the possibility that the lateral part of the RPV may have been subject to not only convective heat transfer of metallic melt pool, but also by conductive heat transfer by oxidic debris conglomerate. The results also indicate that following the initial vessel breach, discharge of metallic melt induces relocation of oxidic debris conglomerates, leading to concentration of the heat source around the central (bottom) part of the lower head. However, large uncertainties associated with 1F are acknowledged and further model validations are necessary before drawing further insights.

Estimation of the final debris distribution at the Fukushima Daiichi Nuclear Power Station (1F) is inevitable for a safe and effective decommissioning. It is necessary to clarify possible failure modes of the reactor pressure vessel (RPV), which is influenced by the thermal status of slumped debris that highly depends on the in-vessel accident progression. In this study, the accident analysis of 1F Unit 2 (1F2) was conducted using the RELAP/SCDAPSIM code, in order to understand better the in-vessel accident progression. One of the unsolved issues of 1F2 is the mechanism of three pressure peaks measured through late Mar. 14 to early Mar. 15, 2011. Despite various analyses, its mechanism is not clearly understood. Comparing the results of previous boiling water reactor (BWR) core degradation experiments and that of 1F2 numerical analysis, it can be estimated that most relocated metallic materials had solidified at the core bottom at the onset of first pressure peak. It is likely that the pressure increase occurred due to the evaporation of injected water reaching the heated core plate structures. Between the first and second pressure peaks, the water is assumed to have been injected continuously and the water level was likely to have recovered to BAF at the initiation of the second pressure peak. Probable slumping of a certain amount of molten materials initiated the second pressure peak and the subsequent gradual pressure increase continued possibly due to massive reaction between coolant and remaining Zircaloy in the core. Assuming the closure of the safety relief valve (SRV) at 0:00 on Mar. 15, the third pressure peak was well reproduced in the analysis. Although the total amount of the slumped material was evaluated, large deviation exists among the cases and uncertainty is still large.

A probabilistic safety assessment (Level-2 PSA) methodology was developed for comprehensive risk evaluation of sodium-cooled fast reactors. As part of this development, in this paper, phenomenological event trees were developed as well as technical database to quantify the probability of event sequences in the Level-2 PSA, focusing on the initiating and transition phases of unprotected events. Typical and important accident categories were selected: unprotected loss of flow (ULOF), unprotected transient overpower (UTOP) and unprotected loss of heat sink (ULOHS). Based on the state-of-the-art knowledge, the headings of these event trees were selected so that dominant factors in accident consequences can be represented appropriately. For each of the headings, available information for the probability quantification were reviewed and integrated as the technical database for the Level-2 PSA. It was clarified that the headings of the ULOF category, for which experimental database and evaluation models have been reasonably established, can be commonly applied to certain part of the different accident categories except for some specific points, which were identified in this study. For the ULOHS category, an additional event tree is necessary before the core disruption providing various boundary conditions for the initiating phase. For the transition phase, dominant factors were also identified through parametric analyses. In the Japan sodium-cooled fast reactor, an inner duct is introduced into a fuel subassembly for enhancing molten fuel discharge from disrupted core in the transition phase. The parametric study showed that the analytical case without the fuel discharge through the inner duct resulted in an occurrence of recriticality regardless of the fuel discharge through control-rod guide tubes. This suggests that the fuel discharge through the inner duct is essential to avoid severe recriticality in the transition phase.

For the transition phase analysis of core disruptive accidents, the development of a three-dimensional reactor safety analysis code, SIMMER-IV, has been carried out based on the technology of the two-dimensional SIMMER-III code. The world first application of SIMMER-IV to a small-sized sodium-cooled fast reactor has also been attempted to clarify event progression in the early stage of the transition phase. This SIMMER-IV calculation is compared to the two-dimensional case calculated by SIMMER-III, neglecting the presence of control rod guide tubes. The present analysis with the three-dimensional representation suggests that the conventional scenario leading to rather early high-mobility fuel pool formation is unrealistic and the degraded core tends to keep low mobility in the early stage of transition phase.

Experimental verification of a reactor safety analysis code, SIMMER-III, was undertaken for transient behaviors of large-scale bubbles with condensation. The present study aimed to verify the code for numerical simulations of relatively short-time-scale multi-phase, multi-component hydraulic problems. Among these, vaporization and condensation, or simultaneous heat and mass transfer, play important roles. In this study, a series of transient bubble behavior experiments dedicated to condensation phenomena with noncondensable gases was carried out. In the experiments, a pressurized mixture of noncondensable gas and steam was discharged as a large-scale single bubble into a cylindrical pool filled with stagnant subcooled water. The concentration of noncondensable gas was taken as an experimental parameter as was the species of noncondensable gas. The characteristics of transient behavior of large-scale bubbles with condensation observed in the experiments were estimated through experimental analyses using SIMMER-III. In the experiments with steam condensation, dispersion of the gas mixture discharged into the liquid pool was accompanied by vapor condensation at the bubble surface. SIMMER-III simulations suggested that the noncondensable gas had a less inhibiting effect on the condensation of large-scale bubbles. This is a different characteristic to that of the quasi-steady condensation of small-scale bubbles observed in our previous experiments.

The possibility of severe recriticality could be excluded if the molten core materials are discharged from reactor core in the early stage of core disruptive accident (CDA). Based on this idea, several design measures for future commercial liquid metal-cooled fast breeder reactors (LMFBRs) have been proposed to enhance the molten fuel discharge from core in order to prevent formation of the core-wide molten pool with high mobility. One promising concept in these design candidates is modified-FAIDUS (Fuel subassembly with Inner DUct Structure). The event progression in unprotected loss of flow (ULOF) accident in a sodium-cooled large scale FBR with modified-FAIDUS was analyzed to assess the effectual performance of modified-FAIDUS in preventing severe recriticality using the SAS4A and SIMMER-III codes. Two parametric cases were performed covering the uncertainty of duct wall failure mechanism, one with stable fuel crust and another with unstable crust condition. The calculation showed that the final amount of discharged fuel from core in both cases was more than 20% of initial core inventory. The degraded core after fuel discharge is composed of the mixture of solidified fuel, swollen fuel chunks and molten steel, of which low mobility prevents massive fuel motion. The reactor power lowered to decay heat level and the reactivity lowered around −20 $, thus, the possibility of severe recriticality was eliminated.

The WF (wall failure) test of the EAGLE program, in which ∼2 kg of uranium dioxide fuel-pins were melted by nuclear heating, was successfully conducted in the IGR (Impulse Graphite Reactor) of NNC/Kazakhstan. In this test, a 3 mm-thick stainless steel (SS) wall structure was placed between fuel pins and a 10 mm-thick sodium-filled channel (sodium gap). During the transient, fuel pins were heated, which led to the formation of a fuel-steel mixture pool. Under the transient nuclear heating condition, the SS wall was strongly heated by the molten pool, leading to wall failure. The time needed for fuel penetration into the sodium-filled gap was very short (less than 1 s after the pool formation). The result suggests that molten core materials formed in hypothetical LMFBR core disruptive accidents have a certain potential to destroy SS-wall boundaries early in the accident phase, thereby providing fuel escape paths from the core region. The early establishment of such fuel escape paths is regarded as a favorable characteristic in eliminating the possibility of severe re-criticality events. A preliminary interpretation on the WF test results is presented in this paper.

SAS-SFR (derived from SAS4A) is presently the most advanced computer code for simulation of the primary phase of the Core Disruptive Accident (CDA) of MOX-fueled Sodium-cooled Fast Reactors (SFR). In the past two decades, intensive model improvement works have been conducted for SAS-SFR utilizing the experimental data from the CABRI programs. The main target of the present work is to confirm validity of these improved models through a systematic and comprehensive set of test analyses to demonstrate that the improved models has a sufficient quality assurance level for applications to reactor conditions. In order to reach these objectives, an approach of PIRT (Phenomena Identification and Ranking Table) on a set of accident scenarios has been applied. Based on the fact that there have been a significant amount of validation studies for decades, development of the code validation matrix concentrated on key issues. Different accident scenarios have been chosen for the PIRT considering typical SFR accident transients that address a large range of phenomena. As the most important and typical Core Disruptive Accident scenarios leading to generalized core melting and to be addressed with SAS-SFR in the present study, ULOF (Unprotected Loss Of Flow), UTOP (Unprotected Transient OverPower) and ULOHS (Unprotected Loss Of Heat Sink) are selected. The PIRT process applied to a given accident scenario consists in an identification of the phenomena involved during the accident, the evaluation of the importance of the phenomena regarding to the evolution and consequences, and the evaluation of the status of knowledge based on the review of available experimental results. The identified phenomena involved in ULOF are explained as follows for the primary phase. Starting from initiating events, a loss of grid power leading to flow coast down without scram is assumed. The scenario up to coolant boiling is the main point within the first part of the ULOF phenomenological chart. Those elements related to reactivity feedback, such as heat up of coolant, fuel and various structures and their deformation due to the thermal transient are picked up. Depending on the time scale before boiling starts, primary, secondary and tertiary loop heat transfer including the DHR (Decay Heat Removal) system response is concerned since it defines the core inlet coolant temperature. Core inlet coolant temperature gives direct impact on the thermal condition of the core. It also affects reactivity through thermal expansion of the grid plate. In the second part of the ULOF phenomenological chart, elements such as coolant boiling, mechanical response of the fuel pin leading to cladding failure, FCI (Fuel-Coolant Interaction) and post-failure material relocation are picked up. This part of the chart is basically common to the ULOHS. Respective identified phenomena are to be simulated in the SAS-SFR code. To validate the function of the models in the code, ten high priority CABRI experiments are selected. Validation studies on these tests are underway. With the present study, important phenomena involved in ULOF, UTOP and ULOHS were identified and an evaluation matrix for the selected CABRI experiments was developed.