An ongoing Sandia National Laboratories’ (SNL) research study is evaluating a potential design of an injectable sacrificial material (SM) system that could contain and cool corium ejected from a reactor vessel lower head failure during a potential severe accident involving melting fuel at a commercial light water nuclear reactor (LWR). An injectable system could be installed at any existing LWR, without significant modification to the cavity or to the drywell pedestal region of the plant. The conceptual design under consideration is a passive system. The SM is being optimized to quickly cool the corium mixture while creating gas to form porosity in the solid, such that subsequent water flooding can penetrate the structure and provide additional cooling. The SM would form a barrier and limit corium-concrete interactions. This three-year project takes a joint experimental and computational approach. In this paper, we will first discuss the success of our small-scale experiments conducted on the interactions between the surrogate corium material (SCM) and SM, used to evaluate the injectable concept. A larger experimental study, currently underway, will further validate the injectable concept, with a focus on accurately measuring interactions. This paper details the modeling study and its progress, including modeling the experiments on a surrogate system and extending the model to bench-scale corium flow from validation experiments. The project’s modeling studies will use the SNL engineering code suite SIERRA Mechanics to understand the interaction of injectable SM and molten corium and predict corium spreading. Spreading is modeled using a level set method to track the front in conjunction with a pressure-stabilized finite element method on the fully three-dimensional mass, momentum, and energy conservation equations. Using this diffuse-interface method, the corium spreading front can be tracked and an appropriate pseudo-solidification viscosity models can be implemented to accurately model the corium spreading physics. Finally, an injectable SM delivery system is discussed along with its deployment to the six-common commercial LWR designs currently operating in the United States. At the end of this project, a simplified model based on SIERRA simulations will be developed for implementation into MELCOR, a severe reactor analysis code, developed at SNL for the U.S. Nuclear Regulatory Commission. This will allow us to demonstrate the ability of the injectable SM system to mitigate the ex-vessel corium spreading, provide containment and negate the release of radionuclides.
This paper describes the ongoing study of nuclear facility safety enhancement using Sandia National Laboratories’ (SNL) computer codes, supported by U.S. Department of Energy (DOE) Nuclear Safety Research and Development (NSR&D) Program. Continued DOE NSR&D support, since 2014 has allowed the use of the SNL engineering code suite (SIERRA Mechanics) to further substantiate data in the DOE Handbook published in 1994: DOE-HDBK-3010-94, “Airborne Release Fractions/Rates and Respirable Fractions for Nonreactor Nuclear Facilities.” The use of SIERRA codes allows for a better understanding of the mechanics, dynamics, chemistry and overall physics of airborne release scenarios. SIERRA codes provide insights into the contributing phenomena of source term releases from events such as liquid fires. The 1994 Handbook documents small-scaled, bench-top and limited experiments involving liquid fires, powder spills, pressurized releases, and mechanical insult-induced fragmentation scenarios. Data recorded from these scenarios has been substantiated using SIERRA solid mechanics and fluid mechanics codes. Data passing among multi-physics SIERRA codes predicted the contaminant release from a drum rupture due to fire even though there is no experimental data available. In the anticipated revision effort of the Handbook by DOE, these computational capabilities could enhance the data in a broader usage and could provide confidence in the safety analysis SIERRA codes can provide the initial source term to be used in the leak path factor (LPF) analyses, which predicts the ST release out of the facility. Typical LPF analysis is done using the MELCOR code, developed at SNL for the U.S. Nuclear Regulatory Commission. Widely used in nuclear reactor applications, MELCOR is a toolbox safety code in the DOE’s central registry for LPF applications. A recent LPF guidance study done by SNL indicated that MELCOR 2.1, along with updated guidance, should replace the obsolete MELCOR 1.8.5 guidance. This new guidance is significantly improved over the previous guidance, utilizing extensive MELCOR validation, including applicable reactor experiments and experiments described in the DOE-HDBK-3010-94 for LPF applications. The latest version of MELCOR should be included in DOE’s central registry, and should be used by safety analysts for LPF analyses.
Estimates of the source term from a U.S. Department of Energy (DOE) nuclear facility requires that the analysts know how to apply the simulation tools used, such as the MELCOR code, particularly for a complicated facility that may include an air ventilation system and other active systems that can influence the environmental pathway of the materials released. DOE has designated MELCOR 1.8.5, an unsupported version, as a DOE ToolBox code in its Central Registry, which includes a leak-path-factor guidance report written in 2004 that did not include experimental validation data. To continue to use this MELCOR version requires additional verification and validations, which may not be feasible from a project cost standpoint. Instead, the recent MELCOR should be used. Without any developer support and lack of experimental data validation, it is difficult to convince regulators that the calculated source term from the DOE facility is accurate and defensible. This research replaces the obsolete version in the 2004 DOE leak path factor guidance report by using MELCOR 2.1 (the latest version of MELCOR with continuing modeling development and user support) and by including applicable experimental data from the reactor safety arena and from applicable experimental data used in the DOE-HDBK-3010. This research provides best practice values used in MELCOR 2.1 specifically for the leak path determination. With these enhancements, the revised leak-path-guidance report should provide confidence to the DOE safety analyst who would be using MELCOR as a source-term determination tool for mitigated accident evaluations.
A sodium coolant accident analysis code is necessary to provide regulators with a means of performing confirmatory analyses for future sodium reactor licensing submissions. MELCOR and CONTAIN, which have been employed by the U.S. Nuclear Regulatory Commission for light water reactor licensing, have been traditionally used for Level 2 and Level 3 probabilistic analyses as well as containment design basis accident analysis. To meet future regulatory needs, new models are being added to the MELCOR code for simulation of sodium reactor designs by integrating the existing models developed for separate effects codes into the MELCOR architecture. Sodium properties and equations of state, such as from the SAS4A code, have previously been implemented into MELCOR to replace the water properties and equation of state. Additional specific sodium-related models to address design basis accidents are now being implemented into MELCOR from CONTAIN-LMR. Although the codes are very different in the code architecture, the feasibility fit is being investigated, and the models for the sodium spray fire and the sodium pool fire have been integrated into MELCOR. A new package called Sodium Chemistry (NAC) has been added to MELCOR to handle all sodium related chemistry models for sodium reactor safety applications. Although MELCOR code requires the ambient condition to be above the freezing point of the coolant (e.g., sodium or water), the high relative freezing point of sodium requires MELCOR to handle situations, particularly far from the primary circuit, where the ambient temperatures are usually at room temperature. Because only a single coolant can be modeled in a problem at a time, any presence of water in the problem would be treated as a trace material, an aerosol, in MELCOR. This paper addresses and describe the integration of the sodium models from CONTAIN-LMR, and the testing of the sodium chemistry models in the NAC package of MELCOR that handles sodium type reactor accidents, using available sodium experiments on spray fire and pool fire. In addition, we describe the anticipated sodium models to be completed in this year, such as the atmospheric chemistry model and sodiumconcrete interaction model. Code-to-code comparison between MELCOR and CONTAIN-LMR results, in addition to the experiment code validations, will be demonstrated in this year.
This report describes the progress of the CONTAIN - LMR sodium physics and chemistry models to be implemented in MELCOR 2.1. In the past three years , the implementation included the addition of sodium equations of state and sodium properties from two different sources. The first source is based on the previous work done by Idaho National Laboratory by modifying MELCOR to include liquid lithium equation of state as a working fluid to model the nuclear fusion safety research. The second source uses properties generated for the SIMMER code. The implemented modeling has been tested and results are reported in this document. In addition, the CONTAIN - LMR code was derived from an early version of the CONTAIN code, and many physical models that were developed since this early version of CONTAIN are not available in this early code version. Therefore, CONTAIN 2 has been updated with the sodium models in CONTAIN - LMR as CONTAIN2 - LMR, which may be used to provide code-to-code comparison with CONTAIN - LMR and MELCOR when the sodium chemistry models from CONTAIN - LMR have been completed. Both the spray fire and pool fire chemistry routines from CONTAIN - LMR have been integrated into MELCOR 2.1, and debugging and testing are in progress. Because MELCOR only models the equation of state for liquid and gas phases of the coolant, a modeling gap still exists when dealing with experiments or accident conditions that take place when the ambient temperature is below the freezing point of sodium. An alternative method is under investigation to overcome this gap . We are no longer working on the separate branch from the main branch of MELCOR 2.1 since the major modeling of MELCOR 2.1 has been completed. At the current stage, the newly implemented sodium chemistry models will be a part of the main MELCOR release version (MELCOR 2.2). This report will discuss the accomplishments and issues relating to the implementation. Also, we will report on the planned completion of all remaining tasks in the upcoming FY2017, including the atmospheric chemistry model and sodium - concrete interaction model implementation .
Safety basis analysts throughout the U.S. Department of Energy (DOE) complex rely heavily on the information provided in the DOE Handbook, DOE - HDBK - 3010, Airborne Release Fractions/Rates and Respirable Fractions for Nonreactor Nuclear Facilities, to determine radionuclide source terms. In calculating source terms, analysts tend to use the DOE Handbook's bounding values on airborne release fractions (ARFs) and respirable fractions (RFs) for various categories of insults (representing potential accident release categories). This is typically due to both time constraints and the avoidance of regulatory critique. Unfortunately, these bounding ARFs/RFs represent extremely conservative values. Moreover, they were derived from very limited small-scale bench/laboratory experiments and/or from engineered judgment. Thus, the basis for the data may not be representative of the actual unique accident conditions and configurations being evaluated. The goal of this research is to develop a more accurate and defensible method to determine bounding values for the DOE Handbook using state-of-art multi-physics-based computer codes. This enables us to better understand the fundamental physics and phenomena associated with the types of accidents in the handbook. In this year, this research included improvements of the high-fidelity codes to model particle resuspension and multi-component evaporation for fire scenarios. We also began to model ceramic fragmentation experiments, and to reanalyze the liquid fire and powder release experiments that were done last year. The results show that the added physics better describes the fragmentation phenomena. Thus, this work provides a low-cost method to establish physics-justified safety bounds by taking into account specific geometries and conditions that may not have been previously measured and/or are too costly to perform.