Correctly predicting the mechanical behavior of the zirconium fuel cladding during a Loss-of-Coolant Accident (LOCA) is critical for nuclear safety analysis as the fuel rod needs to maintain its coolable geometry throughout the high-temperature transient.
A Loss-of-Coolant Accident (LOCA) is a designed accidental scenario in which the fuel cladding is subjected to a rapid temperature transient from 360 °C to 1200 °C, because of the loss of coolant. The high temperature regime lasts a few hundred seconds and is followed by a quench to room temperature. At high temperature, the oxidation kinetics is greatly increased compared to operating conditions. Furthermore zirconium alloys undergo a phase change around 815 °C from an HCP α phase to a BCC β phase. In addition, oxygen diffusion in the metal ahead of the oxide creates an oxygen-stabilized α phase layer. After the quench, the oxide and the oxygen-stabilized α phase layers are brittle, so the remaining cladding ductility is only due to the prior-β phase and is highly dependent on oxygen and hydrogen content. It is thus critical to correctly assess the oxygen and hydrogen concentration profiles in the fuel rod at the end of the LOCA sequence to evaluate the mechanical integrity of the fuel cladding.
So far, nuclear fuel performance codes such as BISON only use rather conservative empirical models which only provides oxide thickness and weight gain. The physically-based model that is being developed, called the Coupled Current Charge Compensation (C4) model, is meant to provide oxide, oxygen-stabilized α phase and prior-β phase layer growth kinetics, as well as oxygen and hydrogen concentration profiles. Both operating temperature (360 °C) and high temperature (1200 °C) isotherms as well as temperature transients such as LOCA can be correctly modeled for different zirconium alloys with this single model. It is based on a finite element modeling of charged species (oxygen vacancies, electrons and hydrogen) transport through the oxide layer, and oxygen diffusion inside the metal.
Preliminary results show that the C4 model accuracy is comparable to most of the empirical model and is in good agreement with experimental data.
The model is currently being validated using multiple criteria. In addition to metallographic analysis of corroded samples, Electron Probe Microanalysis (EPMA) and Glow Discharge Optical Emission Spectroscopy (GDOES) are used to obtain experimental concentration profile of oxygen and hydrogen.
The objective for this model is to be implemented in the nuclear fuel performance code BISON. Extended Finite Element Method (XFEM) will be used to model the moving interfaces (oxide/oxygen-stabilized α phase and oxygen-stabilized α phase/prior-β phase).
Related publications
- L. Borrel and A. Couet, “Modeling Corrosion Kinetics of Zirconium Alloys in Loss-of-Coolant Accident (LOCA),” in Proceedings of the 18th International Conference on Environmental Degradation of Materials in Nuclear Power Systems – Water Reactors: Volume 2, Portland, OR, 2017, pp. 553-565: Springer International Publishing.