Yue Ma1 Chenwe He2 Philippe Garcia1 Jacques Léchelle1 Audrey Miard1 Pierre Desgardin2 Barthe Marie-France2

1, CEA/DEN/DEC, St-Paul-Lez Durance, , France
2, Centre National de la Recherche Scientifique (CNRS), Orléans, , France

Uranium dioxide and hyper-stoichiometric oxide have been extensively studied over many years, but still a great deal remains to be understood in relation crystal imperfections and their formation and migration mechanisms. This is because UO2+x is a complex material in which anion, cation and electronic disorder changes with temperature, oxygen activity and impurity content. There are a number of complicating factors such as clustering and screening of charged defects that make this material difficult to study experimentally or using theoretical methods. The study of uranium vacancies is particularly challenging as these defects are generally present at low concentrations and in different charge states.
Recently, it has been shown by combining first principles approaches and experimentation carried out on charged particle irradiated UO2, that Positron Annihilation Spectroscopy (PAS) was indeed sensitive to the presence of uranium vacancy containing defects. This, in addition to the fact that the method is a priori sensitive to charge, makes PAS a candidate as a spectroscopic technique that could further shed light on the presence and nature of thermally induced uranium vacancies.
Now, it is reasonably well established that at a given temperature, the uranium vacancy concentration is an increasing function of non-stoichiometry. It is further surmised that the charge of this type of defect, while being negative when the material is close to stoichiometric composition, will increase with increasing deviation from stoichiometry. In order to verify these points, we recently determined the positron annihilation characteristics of dense sintered samples that had first been subjected to controlled oxidation annealing. Doppler broadening and lifetime spectroscopy are shown to be extremely complementary techniques. Both are sensitive to the presence of negative or neutral vacancies and the former is more sensitive to the chemical environment of the positron when it annihilates whereas the latter allows distinguishing trapping rates at different vacancy defects. The results obtained are discussed at length in the light of previous studies and but also with regard to their implications upon material property changes. The method we have used is shown to be extremely promising for understanding and possibly quantifying the complex changes that occur during oxidation of UO2 which involve the concomitant increase in oxygen interstitials and uranium vacancies.