Deuterium migration in nuclear graphite: Consequences for the behavior of tritium in CO2-cooled reactors and for the decontamination of irradiated graphite waste
Résumé
Abstract In this paper, we aim at understanding tritium behavior in the graphite moderator of French CO2-cooled nuclear fission reactors (called UNGG for “Uranium Naturel-Graphite-Gaz”) to get information on its distribution and inventory in the irradiated graphite waste after their dismantling. These findings should be useful both to improve waste treatment processes and to foresee tritium behavior during reactor decommissioning and waste disposal operations. The purpose of the present work is to elucidate the effects of temperature on the behavior of tritium during reactor operation. Furthermore, it aims at exploring options of thermal decontamination. For both purposes, annealing experiments were carried out in inert atmosphere as well as in thermal conditions as close as possible to those encountered in UNGG reactors and in view of a potential decontamination in humid gas. D+ ions were implanted into virgin nuclear graphite in order to simulate tritium displaced from its original structural site through recoil during reactor operation. The effect of thermal treatments on the mobility of the implanted deuterium was then investigated at temperatures ranging from 200 to 1200 °C, in inert atmosphere (vacuum or argon), in a gas simulating the UNGG coolant gas (mainly CO2) or in humid nitrogen. Deuterium was analyzed by Nuclear Reaction Analysis (NRA) both at millimetric and micrometric scales. We have identified three main stages for the deuterium release. The first one corresponds to deuterium permeation through graphite open pores. The second and third ones are controlled by the progressive detrapping of deuterium located at different trapping sites and its successive migration through the crystallites and along crystallites and coke grains edges. Extrapolating the thermal behavior of deuterium to tritium, the results show that the release becomes significant above the maximum UNGG reactor temperature of 500 °C and should be lower than 30 % of the total amount produced over a reactor operating time corresponding to about 11 effective full-power years. Moreover, it would mainly concern the tritium located close to the free surfaces. Furthermore, the total extraction of the remaining tritium in graphite waste should be more efficient in dry inert gas than in humid gas, but would require temperatures higher than 1300 °C for the total removal of the most deeply located deuterium.