The pressure boundary of a CANDU® fuel channel is composed of a cold-worked Zr-2.5Nb pressure tube, which has each end rolled into a stainless-steel end fitting. Heavy-water (D2O) coolant (250–310°C) flows over and through twelve or thirteen fuel bundles contained in each pressure tube. During operation, some deuterium generated by aqueous corrosion of the tube surface enters the metal. Additional deuterium also enters through the rolled joint between the tube and the end fitting.
Predictive models for deuterium ingress are required for fitness-for-service assessments for operating pressure tubes and for the development of new reactor designs. A predictive model for assessing the long-term oxidation of, and deuterium ingress into, the body of the pressure tubes has been developed from in-reactor tests of samples which had been pre-oxidized to obtain oxide thickness values representative of long-term behavior. Deuterium ingress is modeled based on a fraction (2–10 %) of the corrosion-freed deuterium entering the metal. The current version of the model contains relationships describing the oxidation rate as a function of oxide thickness, temperature, concentration of dissolved oxygen in the water, and fast neutron flux and fluence. It can successfully predict the observed deuterium-uptake history of pressure tubes in existing CANDU reactors. The model projects a slight increase in the rate of oxidation and deuterium ingress over time. This increase is much less than for Zircaloy-2, a material used in early CANDU units.
In parallel with model development, there are experimental programs involving detailed surface analysis of removed pressure tubes and irradiation tests focused on elucidating the mechanisms of oxidation and deuterium ingress. As the results of these programs become available, they will be incorporated into the predictive model. This presentation will focus on the model and recent results from the supporting experimental programs.