BISON is a nuclear fuel performance application built using the Multiphysics Object-Oriented Simulation Environment (MOOSE) finite element library. One of its major goals is to have a great amount of flexibility in how it is used, including in the types of fuel it can analyze, the geometry of the fuel being modeled, the modeling approach employed, and the dimensionality and size of the models. Fuel forms that can be modeled include standard light water reactor fuel, emerging light water reactor fuels, tri-structural isotropic fuel particles, and metallic fuels. BISON is a platform for research in nuclear fuel performance modeling while simultaneously serving as a tool for the analysis of nuclear fuel designs. Recent research in BISON includes techniques such as the extended finite element method for fuel cracking, exploration of high-burnup light water reactor fuel behavior, swelling behavior of metallic fuels, and central void formation in mixed-oxide fuel. BISON includes integrated documentation for each of its capabilities, follows rigorous software quality assurance procedures, and has a growing set of rigorous verification and validation tests.

No abstract

The work described in this report was performed under funding from the Nuclear Energy Advanced Modeling and Simulation (NEAMS) program. This report is issued in satisfaction of the level 2 milestone M2MS-18IN0201012 on Release Bison with Initial Metallic Fuel Performance Capability. Metallic fuel development began in Bison a few years after light water reactor development began (∼ 2009). The initial development was funded by the Advanced Fuels Campaign. With recent interest from industry, the Nuclear Regulatory Commission, and the Versatile Test Reactor program, the Nuclear Energy Modeling Simulation Program has increased funding for metallic fuel development and validation in FY18. The material models with the most mature development are binary uranium-zirconium and ternary uranium-plutonium-zirconium alloys for fuel and HT9 for cladding. The first material models were incorporated from open literature sources, which include mechanical and thermal material properties that are functions of temperature, porosity, and zirconium concentration, swelling, fission gas release, zirconium diffusion, creep, and sodium coolant channel boundary conditions. These material models have been utilized to simulate ERB-II fuel pins and the results compared to measurements. The EBR-II experiment measurements mostly consist of cladding strain and zirconium redistribution. As such, comparisons to EBR-II measurements are currently limited to these two figures of merit. Simulations and comparisons to experiment measurements have also been done for transient testing of EBR-II fuel pins in TREAT, which included temperature measurements. Preliminary comparisons between Bison calculations and experiment measurements are favorable. These comparisons have also highlighted the importance of the fuel swelling and fission gas release models on cladding strain. Also, Bison metallic fuel simulations are currently being run in support of ATR experiments and VTR exploratory design calculations. Further development and evaluation of individual material models and fuel system models are planned for FY19.

The restart of the Transient Reactor Test facility (TREAT) will once again provide the capability for rapid transient testing of fuel concepts. Under the auspices of the U.S. Department of Energy's Office of Material Management and Minimization, research is underway to assess the feasibility of converting the current high-enriched uranium (HEU) fuel in TREAT to low-enriched uranium (LEU) fuel. The LEU concept retains the fuel process that results in micrometer-sized UO 2 fuel grains dispersed in a graphite moderator matrix. The LEU fuel design includes more 238 U, which fundamentally changes the feedback mechanisms in the fuel. To explore the effects of conversion on a pulse transient, a simplified semi-infinite TREAT fuel element model of both the HEU and proposed LEU configurations was simulated using MAMMOTH with the requisite multiscale and multiphysics coupling. The developed method incorporates fission energy deposition at microscale locations from the Mesoscale Atomistic Glue Program for Integrated Execution (Magpie), heterogeneity effects from the microscale model in the form of time lag, and independent feedback temperature sources from the microscale fuel grain model and surrounding moderator. The fuel grain size was varied along with temperature feedback sources to explore the feedback mechanisms. Significant differences between fuel and graphite temperatures were found to develop for transients with large energy depositions, for large fuel grains, and for fission-fragment irradiated graphite. These differences in temperature do not influence the feedback for HEU fuel but have a significant effect on LEU fuel. The difference between HEU and LEU fuel is caused by the fuel temperature feedback coefficient for LEU fuel that is roughly 20% of the graphite temperature feedback coefficient. The immediate equilibrium assumption is invalid for LEU fuel in certain TREAT operating regimes. As the conversion of TREAT to LEU fuel aims to conserve HEU capabilities, MAMMOTH simulations of the LEU model explore the effects of matching the same period, peak power density, and deposited energy of the HEU model. The same pulse shape was not achievable due to the feedback mechanism changes.

No abstract