Influence of oxygen content on thermally activated Kr diffusion in UO₂₊ₓ

Eric Gilabert1

Denis Horlait1, Marie-Lyne Amany2, Marie France Barthe2, Gaëlle Carlot3, Philippe Garcia3

1CNRS/IN2P3 and University of Bordeaux, CENBG UMR 5797, Chemin du Solarium, 33175 Gradignan, France;
2CEMHTI, CNRS UPR3079, Université d′Orléans, F-45071 Orléans Cédex 2, France;
3 CEA Cadarache, DEN/DEC, St. Paul Lez Durance, France.

Fuel performance codes developed by CEA, are designed to predict the release of fission gases and nuclear fuel swelling under normal, incidental and accidental operating conditions. In order to make relevant models more predictive, research has been carried out for several years at the Department for Fuels Studies (DEC) at CEA Cadarache based on separate-effects studies combining experimental and modelling methods which describe the material at atomic, mesoscopic and macroscopic length and time scales. Model assessment or model parameter identification relies on data derived from microstructural or spectroscopic characterisation techniques, most notably TEM (transmission electron microscopy), PAS (positron annihilation spectroscopy), TDS (thermal desorption spectroscopy) and NRA (nuclear reaction analysis) [1,2].
The TDS technique and the experimental protocols developed on the AITNA/PIAGARA platform of CENBG enable the release of noble gases (He, Kr, Xe) to be monitored from ion implanted UO2+x samples. From the study of samples prepared in different conditions, and implanted at different ion energies and fluences as low as 1011 ions.cm-2, it became apparent that deviation from stoichiometry had a crucial influence on the release kinetics of noble gases. This has long been known from the literature and the study irradiated material.
The purpose of this talk is to present a newly developed a protocol to quantify with greater accuracy, the effect of deviation from stoichiometry on the diffusion kinetics of Xe and Kr. Modifications to the TDS device have therefore been made to control in situ the oxygen uptake in UO2±x samples. This was done by setting up an oxygen tank that allows controlled additions of O2 in our ultra-high vacuum TDS setup over a wide range of quantities (≤10-6 moles of O2). This study was carried out using a new laser heating setup purposefully designed for nuclear materials such as UO2 [3]. The most notable difference with our former resistance furnace is the very limited volume of material that requires heating; the only components that are heated are the sample itself and the sample holder. All other furnace components remain close to room temperature so that laser heating constitutes a safeguard against undesirable oxidation of surrounding components.
The first results of this study, carried out under the auspices of the H2020 INSPYRE project, are presented. A preliminary TDS release curve was obtained following annealing of a sample, implanted with 2 MeV Kr ions at a fluence of 1012 ions.cm-2, at 1300°C under vacuum. Increasing amounts of O2 were subsequently introduced into the heating system, whilst maintaining the sample temperature at 1300°C. We found that a small increase in x can induce a significant rise in Kr release rates .
We subsequently modelled the Kr release curves using Fick’s diffusion equation, whence Kr diffusion coefficients were determined upon each stepwise increase in O2 quantities. Assuming the oxygen atoms introduced in the furnace are entirely absorbed by the sample and redistributed homogeneously on a very short time scale in comparison to annealing time, we derive a power law dependence of the Kr diffusion coefficient upon oxygen non-stoichiometry: we note a twenty-fold increase in the diffusion coefficient as x rises from 10-7 to 10-4 . The power law dependence of the trace diffusion coefficient to the supposed deviation from the stoichiometry of the sample is used to discuss potential mediating defect and diffusion mechanisms. We also discuss our data in the light of Kr diffusion coefficients reported in the literature, derived using different experimental methods.

[1] C. Onofri, C. Sabathier, C. Baumier, C. Bachelet, D. Drouan, M. Gérardin, M. Legros, Extended defect change in UO2 during in situ TEM annealing, Acta Materialia. 196 (2020) 240–251.
[2] P. Garcia, E. Gilabert, G. Martin, G. Carlot, C. Sabathier, T. Sauvage, P. Desgardin, M.-F. Barthe, Helium behaviour in UO2 through low fluence ion implantation studies, Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms. 327 (2014) 113–116. https://doi.org/10.1016/j.nimb.2013.11.042.
[3] D. Horlait, D. Gosset, A. Jankowiak, V. Motte, N. Lochet, T. Sauvage, E. Gilabert, Experimental determination of intragranular helium diffusion rates in boron carbide (B4C), Journal of Nuclear Materials. 527 (2019) 151834.

Event Timeslots (1)

Thursday – 16th September 2021
-
Eric Gilabert