A New Modelling of the Kinetics of Uranium Dioxide Oxidation in Air
- PDF / 112,397 Bytes
- 6 Pages / 612 x 792 pts (letter) Page_size
- 62 Downloads / 211 Views
$1HZ0RGHOOLQJRIWKH.LQHWLFVRI8UDQLXP'LR[LGH2[LGDWLRQLQ$LU L. Desgranges (1), C. Poinssot (2) Commissariat à l’Energie Atomique, Nuclear Energy Division, (1) Department of Fuel Studies, CEA CADARACHE, F-13107 St Paul Lez Durance, France (2) Department of Physics and Chemistry, CEA SACLAY, F-91191 Gif-sur-Yvette, France $%675$&7 A finite difference based modelling of the oxidation of uranium oxide is presented. This modelling involves only oxygen diffusion from the atmosphere into the solid and then inside the solid described with Fick‘s law. Some calculations performed with this modelling evidenced that the shape of the weight gain curves observed experimentally could be reproduced. This modelling is compared to the formula previously used to interpret the thermograms. ,1752'8&7,21 The oxidation of nuclear fuel is a relevant mechanism for the description of the evolution of spent fuel rod in (interim) storage incidental (accidental) conditions. The oxidation state of uranium can indeed go from +4 to +6 due to the air oxidation of UO2 in U3O8, which dramatically modifies its properties, regarding leaching for example. Moreover the swelling of U3O8 compared to UO2 induces strains on the cladding which can ruin the fuel rod and make it unusable for further handling in the case of interim storage. Up to now, the oxidation was described using a semi-empirical method based on experimental results from which activation energy is derived which allows the extrapolation of the kinetic laws [1]. However, in (interim) storage conditions, the prediction of the oxidation kinetics is difficult for low temperature (lower than 200°C) because the time scale of this kinetics lasts longer than several years which make experimental data difficult to obtain. As a consequence the validity of the extrapolation is not fully warranted. In order to improve this description we present in this paper a modelling of the oxidation of uranium oxide based on a finite difference approach. The aim of this modelling is to rely on physical laws at the lowest level and not on relationships which needs an interpretation of the experimental data. In the following the modelling is described, some examples of calculation are presented and the results are discussed in comparison with the existing empirical description. 02'(//,1* *HRPHWU\ The geometry of the grain is described by a sphere divided in a set of N concentric shells with the same thickness e. The purpose of the modelling is to determine, at each calculation step, the crystalline phase and the oxygen concentration in each shell from which the weight gain is deduced as the sum of the oxygen fluxes from the atmosphere. Thanks to the chosen geometry, the modelling is reduced to a one-dimension problem. In the modelling, three crystalline phases (UO2, U3O7 and U3O8) and five interfaces (UO2atmosphere, U3O7-atmosphere, U3O8 atmosphere, UO2-U3O7 and U3O7-U3O8) are taken into consideration. As it is done in [1], U3O7 is the generic name for several crystalline phases U4O9, αU3O7, βU3O7, whic
Data Loading...
