The oxidation of nuclear fuel has been studied for over 40 years and some engineering criteria has been established to evaluate the evolution of a defective UOX fuel rod in contact with air in order to assist the designer of an (interim) storage facility. Within the French context, MOX fuel also has to be taken into consideration. In the PRECCI program, some oxidation tests were conducted on UOX and MOX samples, non-irradiated and irradiated, to compare and to analyse the difference between these two types of fuel.
The weight gain curves show a different behaviour depending on irradiation: non-irradiated MOX is less oxidised than UOX, but irradiated MOX oxidises faster than UOX. This difference is explained by the fast oxidation of irradiated UO2 grain boundaries, because of which irradiated UO2 ceramics behave as a powder from an oxidation point of view. The weight gain curves for irradiated samples were treated to determine the oxidation rate constant k, which was found consistent with the values reported in the literature. Yet our results tend to prove that k depends on burn-up.
The existence of an oxidised layer in the UO2 grains was confirmed with the examination of samples taken during the oxidation test. Optical observation revealed a new phase on the grain edge which was proved to be oxygen enriched by EPMA. Although the oxidation of plutonium enriched clusters was proved with EPMA, it was not possible to characterise it with optical observations.
In order to better understand the relationship between oxidation kinetics determined by weight gain measurements and actual fuel rod degradation, an attempt was made to estimate the gas fluxes in a defective fuel rod. The gas propagation inside the fuel rod was estimated at room temperature by interpreting fuel rod piercing data. The gas flux inside a fuel rod was then compared to the oxygen consumption by oxidation and to the gas supply through the defect. It was concluded that oxygen is readily consumed by oxidation and does not have the ability to propagate in the fuel rod. A stop and go degradation process for the evolution of a defective fuel rod is thus proposed based on this assumption.