Modelling of thermo-mechanical and irradiation behavior of metallic and oxide fuels for sodium fast reactors

Abstract

A robust and reliable code to model the irradiation behavior of metal and oxide fuels in sodium cooled fast reactors is developed. Modeling capability was enhanced by adopting a non-empirical mechanistic approach to the extent possible, so that to increase the ability to extrapolate the existing database with a reasonable accuracy. Computational models to analyze in-reactor behavior of U-Zr and U-Pu-Zr metallic alloy fuel pins and U0 2-PuO 2 mixed oxide fuel pins have been developed and implemented into a new code, the Fuel Engineering and Structural analysis Tool (FEAST). FEAST has several modules working in coupled form with an explicit numerical algorithm. These modules describe (1) Fission Gas Release and Swelling, (2) Fuel Chemistry and Restructuring, (3) Temperature Distribution, (4) Fuel Clad Chemical Interaction, (5) Fuel and Clad Mechanical Analysis and (6) Transient Creep- Fracture Model for the clad. Given the fuel pin geometry, composition and irradiation history, FEAST can analyze fuel and cladding thermo-mechanical behavior at both steady state and design-basis accident scenarios. FEAST was written in FORTRAN-90 program language. The FEAST-METAL code mechanical analysis module implements the old Argonne National Laboratory (ANL)'s LIFE code algorithm.

(cont.) Fission gas release and swelling are modeled with the Korean GRSIS algorithm, which is based on detailed tracking of the fission gas bubbles within the metal fuel. Migration of the fuel constituents is modeled by means of the thermo-transport theory. Fuel Clad Chemical Interaction (FCCI) models were developed for steady-state and transient situations, based on precipitation kinetics. A transient creep fracture model for the clad, based on the constrained diffusional cavity growth model, was adopted. FEAST-METAL has been benchmarked against available EBR-II database for (steady state) and furnace tests (transients). The results show that the code is able to predict important phenomena such as cladding strain, fission gas release, clad wastage, clad failure time and axial fuel slug deformation, satisfactorily. A similar code for oxide fuels, FEAST-OXIDE, was also developed. It adopts the OGRES model to describe fission gas release and swelling. However, the original OGRES model has been extended to include the effects of Joint Oxide Gain (JOG) formation on fission gas release and swelling. The fuel chemistry model includes diffusion models for radial actinide migration, cesium axial and radial migration, formation of the JOG, and variation of the oxygen to metal ratio. Fuel restructuring is also modeled, and includes the effects of porosity migration, irradiation-induced fuel densification and grain growth.

(cont.) The FEAST-OXIDE predictions has been compared to the available FFTF, EBR-II and JOYO databases, and the agreement between the code and data was found to be satisfactory. Both metal and oxide versions of FEAST are rather superior compared to many other fuel codes in the literature. Comparing metal and oxide versions, FEAST-OXIDE has a more sophisticated fission gas release and swelling model, which is based on vacancy flow. In addition, modeling of the chemistry module of the oxide fuel requires a much more detailed analysis to estimate its impact on the thermo-mechanical behavior with a reasonable accuracy. Finally, the melting of the oxide fuel and its effect on the thermo-mechanical performance have been modeled in case of transient scenarios.