Microstructural-Level Fuel Performance Modeling of U-Mo Monolithic Fuel (original) (raw)

An interaction zone forms between U-10Mo and Zr, referred to as the UMo-Zr interaction zone, and grows during irradiation. At high burnups, fracture develops in the fuel plate, primarily in a region with high density of gas bubbles along the interface between the different sublayers in the interaction zone. This indicates potential degradation in fracture stress, either in the U-10Mo fuel or in the interaction zone, or both. The degradation in fracture properties induced by gas bubbles in U-10Mo fuel (21.8% in atomic percent) and in the UMo-Zr interaction zone have been studied. The change in fracture properties in bulk U-10Mo fuel is calculated using MD simulations. The change in fracture properties in the UMo-Zr interaction zone is simulated using the phase-field fracture method, with the model implemented in Idaho National Laboratory's (INL's) Multiphysics Object-Oriented Simulation Environment (MOOSE). U-10Mo was found to be very ductile under uniaxial tension by MD simulations adopting the bicrystal model. No fracture propagation along GBs was observed with up to 50% engineering strain. Considering that the extremely high-loading rate in MD simulations usually facilitates fracture, the ductile nature of U-10Mo indicated by the MD simulations seems to be convincing. Three plastic deformation modes were identified depending on the loading orientation with the selection of deformation mode consistent with the Schmid law. The absence of fracture propagation in U-10Mo has some interesting implications on the fracture observed in the UMo-Zr interaction zone at high burnups. The appearance of fracture may be caused by three possible reasons: (1) extremely high gas-bubble coverage or gas bubble density, (2) creep damage at stress levels lower than the yield stress, and (3) phase separation in the UMo-Zr interaction zone. It should be noted the above three mechanisms are not mutually exclusive; they may v in fact operate together, and the failure of the UMo fuel matrix may be caused by their compounding effect. These three effects will be the subject of future research. A phase field fracture model was designed and demonstrated for single-layer domain (e.g., bulk UMo) and for multiple-layer domain (e.g., UMo-Zr interaction zone) in MOOSE. The model is for brittle fracture without plastic deformation. The model is capable of modeling fracture initiation and propagation in domains with a distribution of gas bubbles in both 2D and 3D. In the future, the model will be further extended to include creep fracture, which is expected to be an important mechanism for failure of UMo fuel matrix. The model will be applied to study fracture in the interaction region with varying bubble size, density, and connectivity. A correlation between the fracture stress of the interaction zone and bubble morphology will be developed based on simulation results. The effects of phase separation, thickness of each layer, and fluctuation in layer thickness on fracture initiation and propagation will be studied as well. The Effect of Carbides on Mechanical Properties and Swelling of U-Mo Fuel Nonmetallic inclusions, such as carbides, are often found in the U-Mo alloy fuels, whether due to residual feedstock impurities or the formation during the manufacturing process. Carbide inclusions may affect the U-Mo fuel manufacturing process, the microstructure evolution, and fuel performance under irradiation. We investigated the effect of carbide inclusions on the mechanical properties of U-Mo fuel using different simulation approaches. Semi-empirical models and finite-element method modeling show that carbides have a minor impact on the mechanical properties of U-Mo fuel when the carbide inclusions have a typical volume fraction (0.5~1%) and average aspect ratio (1.5~2.5) as observed in experiments. Density functional theory calculations show that decreased Mo concentration of U-Mo alloy or increased hypo-stoichiometry in carbides can slightly decrease the mechanical strength of the UMo/ uranium carbide (UC) interface. Meanwhile, experiments show that the 235 U enrichment in carbides could be different from that in the U-Mo fuel matrix due to the incomplete mixing of depleted and highly enriched uranium feedstocks. Different enrichments in carbides and the fuel matrix present fission rate and temperature gradients in the fuel and potentially deleterious microstructural behavior during operation. By extending our previous work of fission rate effect on fuel swelling, we developed phase-field models to simulate the effect of variable 235 U enrichment in carbides on the gas bubble swelling in U-Mo fuel. Our simulation suggests that the 235 U enrichment and the volume fraction of UC inclusions have a minor impact on the gas bubble swelling if the targeted 235 U enrichment in the final U-Mo fuel can be achieved. Phase-field simulations also show that increased fission rate can result in accelerated gas bubble swelling enhanced by the increased 235 U enrichment in the U-Mo fuel. Radiation-Enhanced Diffusion in UMo Accurate prediction of fission-gas swelling requires accurate descriptions of the diffusion coefficients of relevant species in the fuel. Radiation-enhanced diffusion coefficients of U, Mo, and Xe in U-10Mo were calculated using rate-theory models and MD simulations. In addition, intrinsic diffusion of Xe was calculated using MD simulations. Utilizing the intrinsic diffusion, radiation-enhanced diffusion, and radiation-driven diffusion, the total diffusion of U, Mo, and Xe under irradiation was also determined in the temperature range between 300 and 1400 K. It was found that radiation-enhanced diffusion of U and Mo were dominant in the intermediate temperature range (450 to 650 K) at the evaluated fission rates, whereas the radiation-enhanced diffusion of Xe did not significantly contribute to the total diffusion of Xe under irradiation at any temperature range. The total diffusion coefficients of U, Mo, and Xe calculated in this work will be utilized as important parameters in mesoscale and engineering-scale nuclear-fuel models. vii Irradiation-Enhanced Creep A mesoscale model of irradiation-enhanced creep in polycrystalline UMo with a Zr layer has been developed. The model integrates a spatial-dependent cluster-dynamics model of radiation defect evolution, a phase-field model of non-equilibrium gas bubble evolution, and elastic-plastic deformation under a crystal-plasticity framework. The radiation defects including U and Mo interstitials, U and Mo vacancies, vacancy and interstitial clusters, and Xe fission-gas atoms are considered. The lattice mismatch among host atoms (U and Mo) and defects (interstitial, vacancy, and Xe atom) is described by a stressfree strain tensor. It enables one to consider stress-driven diffusion of solutes and vacancies. It is assumed the irradiation and stress-enhanced diffusion is one of dominant creep mechanisms in UMo. In the phasefield model of non-equilibrium gas bubble evolution, the Xe concentration inside gas bubbles is determined by the absorption of vacancies and Xe atoms. Therefore, the model can describe the transition between over-pressurized gas bubbles and voids which is determined by the local flux of vacancies and Xe atoms to gas bubbles. The thermodynamic and kinetic properties of radiation defects are described in a function of order parameters which presents different phases including UMo, gas bubble, and Zr cladding. Therefore, the model captures the evolving thermodynamic and kinetic properties with gas-bubble evolution. Plastic strain rate-based crystal plasticity is employed to describe the elastic-plastic deformation. It enables the analysis of the effect of anisotropic mechanical properties, such as grain orientation and individual slip system, on elastic-plastic deformation and creep. In summary, this is a physics-based model with a multiphysics coupling of radiation damage, gas bubble swelling, stress-driven diffusion creep, and elastic-plastic deformation. The model has been used to study the effect of gas bubble structures on elastic-plastic deformation, the effect of radiation conditions and thermodynamic, and kinetic properties of radiation defects on defect accumulation and gas bubble evolution. Atomistic modeling to support mesoscale creep models Irradiation creep models rely on the fundamental behavior of point defects in a stress field. How that applied stress field affects diffusion or equilibrium concentrations of defects will in turn affect the timeand stress-dependent evolution of the material system. How point defect properties vary as a function of applied pressure is largely unknown for U-Mo systems. It has been shown in Fe that application of pressure can significantly affect both the formation energy of defects and their generation under irradiation. In this work, we study how the application of hydrostatic tension and compression affects the formation energy and diffusion coefficient of interstitials and vacancies in U-Mo as a function of pressure, temperature, and composition. On average, the maximum applied pressure of 10 kbar produces a 6% increase in the interstitial formation energy and a 3% decrease in the vacancy formation energy. Under reasonable applied bulk pressures below the yield point (<100 MPa), negligible deviations in the defect formations are observed. Also, applied pressures should yield negligible variation on point defect diffusion at relevant temperatures and pressures. There are impacts of the applied pressure on defect formation and diffusion, and clear trends can be observed, but these effects are sufficiently small; even at large pressures, they likely can be neglected for practical purposes. However, in circumstances where the pressures may be quite large (e.g., in the area surrounding a highly pressurized nanometer-sized bubble) statistically significant changes in the local defect formation energy and diffusion coefficient could be...