Engineering Papers⌕ Search

Engineering topics

Andersson, David

Publications and source records attributed to Andersson, David.

50 records · Page 3

Multi-Scale Modeling of Swelling in Accident-Tolerant U 3 Si 2 Fuel

U 3 Si 2 is a promising candidate for use as an accident-tolerant fuel for light water reactors. A multi-scale computational approach was used to calculate swelling in pellet-form U 3 Si 2 fuel. Swelling was assumed to be equal to the volume fraction of fission gas bubbles in the fuel, and the evolution of bubble volume fraction was determined from phase-field simulations. To parameterize the phase-field model, density-functional theory and molecular dynamics simulations were performed. To enable molecular dynamics simulations, a new interatomic potential for the U-Si system was developed based on the modified embedded atom method. A new phase-field model based on a grand-potential functional was also developed. The model is applicable to both intergranular and intragranular bubbles. To calculate the volume fraction of bubbles, the microstructure was decomposed into regions consisting of only intragranular or intergranular bubbles based on a truncated octahedral grain structure, and growth of the bubbles in the two regions was simulated separately. The total swelling was then calculated based on a a weighted average of bubble volume fraction in the two regions. Total swelling was of the same order of magnitude, but larger than that predicted by the existing empirical swelling model used in BISON and a rate-theory based model. Limitations of the present approach and suggestions for improvement are presented.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Gaseous swelling of U 3 Si 2 during steady-state LWR operation: A rate theory investigation

Rate theory simulations of fission gas behavior in U 3 Si 2 are reported for light water reactor (LWR) steady-state operation scenarios. We developed a model of U 3 Si 2 and implemented into the GRASS-SST code based on available research reactor post-irradiation examination (PIE) data, and density functional theory (DFT) calculations of key material properties. Simplified peripheral models were also introduced to capture the fuel-cladding interaction. The simulations identified three regimes of U 3 Si 2 swelling behavior between 390 K and 1190 K. Under typical steady-state LWR operating conditions where U 3 Si 2 temperature is expected to be below 1000 K, intragranular bubbles are dominant and fission gas is retained in those bubbles. The consequent gaseous swelling is low and associated degradation in the fuel thermal conductivity is also limited. Those predictions of U 3 Si 2 performance during steady-state operations in LWRs suggest that this fuel material is an appropriate LWR candidate fuel material. Fission gas behavior models established based on this work are being coupled to the thermo-mechanical simulation of the fuel behavior using the BISON fuel performance multi-dimensional finite element code.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

A Modified Embedded-Atom Method Interatomic Potential for U-Si

Uranium-silicide (U-Si) fuels are being pursued as a possible accident tolerant fuel (ATF). This uranium alloy fuel benefits from higher thermal conductivity and higher fissile density compared to UO{sub 2}. In order to perform engineering scale nuclear fuel performance simulations, the material properties of the fuel must be known. Currently, the experimental data available for U-Si fuels is rather limited. Thus, multiscale modeling efforts are underway to address this gap in knowledge. In this study, a semiempirical modified Embedded Atom Method (MEAM) potential is presented for the description of the U-Si system. The potential accurately describes not only the primary phase of interest (U{sub 3}Si{sub 2}), but also a variety of U-Si phases across the composition spectrum.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Mechanistic materials modeling for nuclear fuel performance

Fuel performance codes are critical tools for the design, certification, and safety analysis of nuclear reactors. However, their ability to predict fuel behavior under abnormal conditions is severely limited by their considerable reliance on empirical materials models correlated to burn-up (a measure of the number of fission events that have occurred, but not a unique measure of the history of the material). In this paper, we propose a different paradigm for fuel performance codes to employ mechanistic materials models that are based on the current state of the evolving microstructure rather than burn-up. In this approach, a series of state variables are stored at material points and define the current state of the microstructure. The evolution of these state variables is defined by mechanistic models that are functions of fuel conditions and other state variables. The material properties of the fuel and cladding are determined from microstructure/property relationships that are functions of the state variables and the current fuel conditions. Multiscale modeling and simulation is being used in conjunction with experimental data to inform the development of these models. Finally, this mechanistic, microstructure-based approach has the potential to provide a more predictive fuel performance capability, but will require a team of researchers to complete the required development and to validate the approach.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Meso-Scale Fuel Performance Modeling using the MARMOT Code

The in-pile behavior of light-water-reactor (LWR) fuels is governed by the physics from various time and spatial scales. Accordingly, multi-scale modeling and simulation are needed to accurately predict the fuel performance at the engineering scale. To improve the traditional fuel performance modeling, the Department of Energy (DOE) Nuclear Energy Advanced Modeling and Simulation (NEAMS) program targets at developing multi-scale, multiphysics modeling tools for in-pile fuel behavior prediction under various operating or accident scenarios. In traditional fuel performance modeling, the history of fuel is represented by a single parameter, fuel burn-up. Burn-up measures the energy that has been extracted from fuels. However, it is not an adequate measure of fuel history as it ignores the flux effect, which is critical for the irradiation-induced microstructural evolution in fuels. Moreover, the traditional fuel performance models are usually fitted to experimental data. Their capabilities of being extrapolated to new operating conditions and being extended to new types of fuels are strongly limited by the empirical or semi-empirical nature. To improve the burn-up based models, a microstructure based fuel approach, the MOOSE-BISON-MARMOT (MBM) tool-kit, is being developed under the fuel product line (FPL) of the NEAMS program for predictive fuel performance modeling. In the MBM approach, the evolution of fuel microstructure and changes in fuel properties are described by the MARMOT code. The transient fuel properties obtained from MARMOT are then fed into the BISON code for engineering scale fuel performance modeling. Both BISON and MARMOT are developed within the MOOSE framework, which was developed at Idaho National Laboratory. In this summary, the development and the capabilities of the MARMOT code will be introduced. In the past few years, several materials models have been developed including: (1) a grain growth model that describes the grain size evolution as a function of time, temperature and fuel porosity, considering the effects of impurity and porosity dragging and anisotropic grain boundary energy, (2) a fission gas release model that describes the fission gas diffusion in grain interior, fission gas bubble formation in grain interior and at grain boundaries, and fission gas release to the fuel-cladding gap, (3) a fracture model that describes the fracture initiation and propagation in grain interior and along grain boundaries, and (4) a thermal conductivity model that describes the change in thermal conductivity due to grain growth, gas bubble formation, fracture and radiation damage. Several other models are also being developed in MARMOT for hydride formation in Zr-based cladding, radiation damage and grain size subdivision. In most cases, these materials models are coupled with each other. For instance, to accurate predict the transient thermal conductivity, information on grain size, fission gas density, porosity and fracture is required. Further development is on going to improve all these models. These models are developed primarily for UO{sub 2} fuels used in LWRs. They can be extended to other fuels with possibly new development. All models in MARMOT will be rigorously verified and validated before being applied for fuel performance modeling. A version of MARMOT was recently released in September 2015 with an assessment report. (authors)

36 MATERIALS SCIENCE↗

Atomistic Simulations in Support of UO{sub 2} Fuel Performance Modeling

Large fission gas atoms such as Xe have very low solubility in the UO{sub 2} fuel matrix, which drives precipitation and growth of intra-granular fission gas bubbles, segregation of gas atoms to grain boundaries and evolution of intergranular bubbles. The inter-granular bubbles develop percolation networks leading to release of the fission gas. Fission gas bubbles induce fuel swelling and the fuel thermal conductivity is reduced by both fission gas bubbles and dispersed gas atoms. When fission gas is released from the fuel to the plenum, the pressure on the clad walls increases and the gap thermal conductivity decreases. All of these detrimentally affect fuel performance. The thermal conductivity of UO{sub 2} is low across a wide temperature range compared to iso-structural materials such as ThO{sub 2}, which is caused by the coupling between phonons carrying heat and uranium spins or magnetism providing a scattering mechanism with very low mean free path. For in-pile conditions the thermal conductivity is further reduced due to, for example, radiation damage, generation of fission products and fission gases and fuel microstructure evolution. This results in a complex burn-up dependent thermal conductivity. Thermal conductivity and fission gas evolution are two of the most important properties determining UO{sub 2} nuclear fuel performance. Both properties have significant uncertainty. In order to reduce this uncertainty, new models are being developed based on multi-scale simulations that attempt to improve the description of material properties and also couple them to the microstructure state of the material. In this paper we highlight atomistic calculations and simulations used to develop meso-scale models of the evolution of fission gases and thermal conductivity in UO{sub 2}. Literature examples of the application of these results and the corresponding models to fuel performance simulations will also be given. (author)

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

First-Principles Study of Surface Properties of U{sub 3}Si{sub 2} for Accident Tolerant Fuel

Due to the high thermal conductivity and uranium density compared to UO{sub 2} fuel used in current Light Water Reactors (LWRs), U-Si fuels, such as the U{sub 3}Si{sub 2} and U{sub 3}Si{sub 5} compounds, are currently under investigation for the development of accident tolerant fuels (ATF) in response to the Fukushima Daiichi accident in Japan. In order to perform meaningful mesoscale and macroscale fuel performance simulations, fundamental material properties of nuclear fuels should be known. One example is the simulations of fission gas bubble evolution and gas bubble-induced swelling in nuclear fuels using mesoscale methods, such as phase field and rate theory. To properly describe the gas bubble behavior in the fuel, it is crucial to know the surface energies of the fuel. Unfortunately, there is no study of the surface properties of U{sub 3}Si{sub 2} from both experiment and theory. To this end, we systematically investigate the surface properties of U{sub 3}Si{sub 2} using density functional theory (DFT) based first-principles calculations. Of several low index facets, we studied the (001) and (100) surfaces of tetragonal U{sub 3}Si{sub 2} with all the possible terminations. Since all the studied surfaces are not stoichiometric, we also investigated the effect of elemental chemical potentials on the formation of surface. We systematically investigate the surface properties of U{sub 3}Si{sub 2} using density functional theory calculations. Of several low-index facets, the (001) and (100) surfaces show similar averaged surface energies calculated using even number of layers. At the high Si chemical potential, the Si-terminated U{sub 3}Si{sub 2}(100) and U{sub 3}Si{sub 2}(001) surfaces show lower energies than those of the U-terminated surfaces, with U,Si-terminated U{sub 3}Si{sub 2}(001) as the most stable surface. The surface energies of all non-stoichiometric U{sub 3}Si{sub 2} are linearly dependent on the Si chemical potential. The surface energies of U-terminated surfaces decrease as Si potential decreases. The U{sub 2}-terminated U{sub 3}Si{sub 2}(100) surface becomes that the most stable surface at U-rich side. Based on the current results, an averaged surface energy of U{sub 3}Si{sub 2}, i.e., 1.45 J/m{sup 2}, is recommended for the description of the rounded facets of gas bubbles in U{sub 3}Si{sub 2} by mesoscale simulations. (authors)

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Multiscale Investigation of Thermal Transport in UO{sub 2} Fuel using Modeling and Simulation

The thermal conductivity of the UO{sub 2} fuel in light water reactors (LWRs) defines how efficiently heat is transported from the fuel to the coolant, and thus directly influences the reactor efficiency. Unfortunately, the same fission events that generate heat also cause radiation damage and fission product formation that decreases the thermal conductivity. This reduction in thermal conductivity reduces the safety of the reactor in addition to its efficiency, because excess heat generated during an accident cannot be transported from the fuel, resulting in high temperatures that can result in melting. Because thermal conductivity plays such an important role in reactor efficiency and safety, many experiments have been conducted to measure the thermal conductivity of UO{sub 2} and how it changes during reactor operation. Experiments on unirradiated fuel have resulted in a large set of data indicating how the bulk thermal conductivity changes as a function of temperature, which was used to develop a semiempirical model. Additional experiments have been conducted to determine how the thermal conductivity of UO{sub 2} changes as a function of burnup, leading to burnup dependent models fitted to the data. While these empirical models perform well within the conditions at which the data was collected, they cannot be extrapolated outside of those conditions. In addition, while burnup is a measure of the amount of fissioning that has occurred, it cannot account for the impact of the operating conditions on the thermal conductivity. While experimental data provides an empirical view on the value of the thermal conductivity in the fuel during reactor operation, a more mechanistic understanding of how temperature and burnup impact the transport of heat through the fuel would be extremely valuable. Such an understanding would allow for a more accurate prediction of fuel performance, but it could also be used to direct modifications to the fuel that would raise the thermal conductivity or reduce the impact of burnup. In this work we summarize recent efforts to use modeling and simulation at the atomistic and mesoscales to obtain a mechanistic understanding of thermal transport in UO{sub 2}. The thermal conductivity of LWR fuel is one of the most critical material parameters impacting the efficiency and safety of the reactor. Thus, a significant effort is being made to develop a mechanistic understanding of the impact of temperature and burnup on it. In this work, we have provided a short summary of modeling and simulation efforts being carried out at the atomistic and mesoscale to understand this critical property. These research efforts are providing a fundamental understanding of the thermal transport in unirradiated UO{sub 2}, as well as the impact of radiation damage and fission products on thermal transport. These efforts have been focused on UO{sub 2}, but they also are laying the framework for conducted similar simulations to investigate new accident tolerant fuel concepts. (authors)

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Mesoscale Benchmark Demonstration, Problem 1: Mesoscale Simulations of Intra-granular Fission Gas Bubbles in UO 2 under Post-irradiation Thermal Annealing

A study was conducted to evaluate the capabilities of different numerical methods used to represent microstructure behavior at the mesoscale for irradiated material using an idealized benchmark problem. The purpose of the mesoscale benchmark problem was to provide a common basis to assess several mesoscale methods with the objective of identifying the strengths and areas of improvement in the predictive modeling of microstructure evolution. In this work, mesoscale models (phase-field, Potts, and kinetic Monte Carlo) developed by PNNL, INL, SNL, and ORNL were used to calculate the evolution kinetics of intra-granular fission gas bubbles in UO 2 fuel under post-irradiation thermal annealing conditions. The benchmark problem was constructed to include important microstructural evolution mechanisms on the kinetics of intra-granular fission gas bubble behavior such as the atomic diffusion of Xe atoms, U vacancies, and O vacancies, the effect of vacancy capture and emission from defects, and the elastic interaction of non-equilibrium gas bubbles. An idealized set of assumptions was imposed on the benchmark problem to simplify the mechanisms considered. The capability and numerical efficiency of different models are compared against selected experimental and simulation results. These comparisons find that the phase-field methods, by the nature of the free energy formulation, are able to represent a larger subset of the mechanisms influencing the intra-granular bubble growth and coarsening mechanisms in the idealized benchmark problem as compared to the Potts and kinetic Monte Carlo methods. It is recognized that the mesoscale benchmark problem as formulated does not specifically highlight the strengths of the discrete particle modeling used in the Potts and kinetic Monte Carlo methods. Future efforts are recommended to construct increasingly more complex mesoscale benchmark problems to further verify and validate the predictive capabilities of the mesoscale modeling methods used in this study.

99 GENERAL AND MISCELLANEOUS↗