During normal operation in nuclear reactors, the nuclear fuel cladding corrodes as a result of exposure to high temperature cooling water. During this process, hydrogen can enter the zirconium-alloy of the fuel cladding, and under proper conditions, precipitate as brittle hydride platelets which can severely impact cladding ductility and fracture toughness. Hydrogen tends to migrate to and precipitate at colder spots. Because high local hydride concentrations increase the risk of cladding failure, it is important to predict the local hydrogen distribution. Hydrogen transport depends on different phenomena. Even though migration can only occur when the hydrogen is in solid solution, the cladding temperatures during operating condition allow a portion of the hydrogen to be in solid solution. Therefore, as hydrogen is picked up during the corrosion reaction between the cladding and the coolant, it can migrate following Ficks law and the Soret effect. Once the local hydrogen content reaches the terminal solid solubility for precipitation, the hydrogen will start precipitating as zirconium hydride. Previous work in our laboratory implemented a model that describes these different phenomena, into the 3D fuel performance code BISON. A first attempt to benchmark this model has been made in this study by comparing the results given by BISON to the hydrogen distribution measured in a nuclear fuel rod, which underwent a five cycle exposure at the Gravelines nuclear power plant. This was feasible because of the very detailed information about that reactor and fuel pin were available from the thesis of J.-H. Zhang in 1992. The benchmark performed with BISON showed very good agreement between the calculations experimental observation.The calculation above used a value of the precipitation kinetics parameter 2 based on a fit that ignored discrepancies in the early part of the precipitation process. The calculation was revised according to a new model. The new model, which assumes an initial dependence of the precipitation rate on (C_ss-TSS_p )^2 , provides a better fit for this date and we believe a more precise value of 2, which varies mildly with temperature.

We used a range of advanced microscopy techniques to study the microstructure, nanoscale chemistry, and porosity in zirconium alloys at different stages of oxidation. Samples from both autoclave and in-reactor conditions were available, including ZIRLOTM, Zr-1.0Nb, and Zr-2.5Nb samples with different heat treatments. Scanning transmission electron microscopy (STEM), transmission Kikuchi diffraction (TKD), and automated crystal orientation mapping with TEM were used to study the grain structure and phase distribution. Significant differences in grain morphology were observed between samples oxidized in the autoclave and in-reactor, with shorter, less well-aligned monoclinic grains and more tetragonal grains in the neutron-irradiated samples. A combination of energy-dispersive X-ray mapping in STEM and atom probe tomography analysis of second-phase particles (SPPs) can reveal the main and minor element distributions respectively. Neutron irradiation seems to have little effect on promoting fast oxidation or dissolution of ?-niobium precipitates but encourages the dissolution of iron from Laves-phase precipitates. An electron energy-loss spectroscopy (EELS) analysis of the oxidation state of niobium in ?-niobium SPPs in the oxide revealed the fully oxidized Nb5+ state in SPPs deep into the oxide but Nb2+ in crystalline SPPs near the metal-oxide interface. EELS analysis and automated crystal orientation mapping with TEM revealed Widmanstatten-type suboxide layers in some samples with the hexagonal ZrO structure predicted by ab initio modeling. The combined thickness of the ZrO suboxide and oxygen-saturated layers at the metal-oxide interface correlated well to the instantaneous oxidation rate, suggesting that this oxygen-rich zone is part of the protective oxide that is rate limiting in the transport processes involved in oxidation. Porosity in the oxide had a major influence on the overall rate of oxidation, and there was more porosity in the rapidly oxidizing annealed Zr-1.0Nb alloy than in either the recrystallized alloy or the similar alloy exposed to neutron irradiation.

Because hydrogen ingress into zirconium cladding can cause embrittlement and limit cladding lifetime, hydrogen pickup during corrosion is a critical life-limiting degradation mechanism for nuclear fuel. However, mechanistic knowledge of the oxidation and hydrogen pickup mechanisms is still lacking. In an effort to develop such knowledge, we conducted a comprehensive study that included detailed experiments combined with oxidation modeling. We review this set of results conducted on zirconium alloys herein and articulate them into a unified corrosion theoretical framework. First, the hydrogen pickup fraction (fH) was accurately measured for a specific set of alloys specially designed to determine the effects of alloying elements, microstructure, and corrosion kinetics on fH. We observed that fH was not constant and increased until the kinetic transition and decreased at the transition. fH depended on the alloy and was lower for niobium-containing alloys. These results led us to hypothesize that hydrogen pickup during corrosion results from the need to balance the charge during the corrosion reaction such that fH decreases when the rate of electron transport through the protective oxide increases. To assess this hypothesis, two experiments were performed: (1) micro-X-ray absorption near-edge spectroscopy (?-XANES) to investigate the evolution of the oxidation state of alloying elements when incorporated in the growing oxide and (2) in situ electrochemical impedance spectroscopy (EIS) to measure oxide resistivity as a function of exposure time on different alloys. With the use of these results, we developed an analytical zirconium alloy corrosion model based on the coupling of oxygen vacancies and electron currents. Both modeling and EIS results show that as the oxide electric conductivity decreases the fH increases. These new results support the general hypothesis of charge balance. The model quantitatively and qualitatively predicts the differences observed in oxidation kinetics and hydrogen pickup fraction between different alloys.

A review is presented of work performed in our group over the years in the areas of radiation damage, corrosion, hydrogen pickup, hydriding, and the mechanical behavior of zirconium alloy nuclear fuel cladding with the goal of developing a greater mechanistic understanding of cladding performance in service.