Thermal oxidation and high temperature structural behavior of uranium carbide

npj | materials degradation Article Published in partnership with CSCP and USTB https://doi.org/10.1038/s41529-025-00732-1 Thermal oxidation and high temperature structural behavior of uranium carbide Check for updates 1234567890():,; 1234567890():,; Emma C. Kindall1, Natalie S. Yaw1, Malin C. Dixon Wilkins2, Juejing Liu2, Sam Karcher2, Bryn Merrill1, Rushi Gong3, Shun-Li Shang3, Zi-Kui Liu3, John S. McCloy1,2, Hongwu Xu4,5, Adrien J. Terricabras6, Scarlett Widgeon Paisner6, Arjen van Veelen6, Joshua T. White6 & Xiaofeng Guo1,2 Uranium monocarbide (UC) exhibits physiochemical characteristics well-suited for nuclear fuel applications in Generation IV reactors, but its high susceptibility to oxidation remains a major barrier to deployment. A detailed understanding of the U-C-O system, including UC thermal oxidation, crystal chemistry, and thermodynamic/kinetic properties, is essential to predict its behavior under normal and off-normal reactor conditions. In this work, in situ high temperature synchrotron X-ray diffraction was conducted under sealed and open-air conditions to characterize UC thermal expansion and oxidation behaviors. From the sealed experiment, the mean coefficient of thermal expansion of UC was determined to be 9.8 × 10−6 K−1 from room temperature to 970 K. Open-air experiments conducted from room temperature to 773 K revealed the oxidation sequence UC → UO2 → U3O8. Notably, a tetragonal U(C1-xOx)2 phase, absent from current thermodynamic predictions, was observed at 840 K, lower than previously considered, suggesting potential relevance for advanced reactor fuel applications. These findings reveal ambiguities in existing knowledge of the U-C-O system, emphasizing the need for continued investigation to facilitate the use of UC-based TRISO and other carbide fuels in emerging reactor designs. With increasing demand for removal of fossil fuels from energy grids, the push for more advanced reactors has surged, and with it the search for and development of compatible advanced and accident tolerant fuels has become paramount. Non-oxide ceramics, including uranium nitride (UN) and uranium carbide (UC) have emerged as promising candidates for various Generation IV reactor designs. UC is notably considered for gascooled fast reactors (GFRs)1. The main challenges of implementing uranium carbide fuel in conventional rod fuel geometries are its high radiationinduced swelling, limited irradiation data at high burnups, in addition to rapid oxidation in the event of excess temperatures and air ingress, which has limited its use in commercial (e.g., Gen III) reactors2. However, uranium carbide boasts a number of advantages over traditional oxide fuels including higher fissile density, higher thermal conductivity, which are important characteristics for improving accident tolerance3,4, and better structural compatibility with other components in nuclear fuel matrices, like proposed claddings materials SiC and ZrC2. Additionally, UC phases incorporated with UO2, referred to as uranium oxycarbide (UCO) ceramics, are critical components in the robust accident tolerant TRi-structural ISOtropic (TRISO) fuel particle proposed for use in very high temperature gas-cooled reactors (VHTRs), molten salt cooled reactors (MSRs), and various small modular reactor designs, further motivating research into properties of the U-C systems2,5. For fuel qualification, candidate systems must be well characterized to enable accurate prediction of fuel behavior under normal, transient, and accidental conditions, including uncertainty. Thermal expansion is an important property for predicting matrix compatibility between components within fuel kernels and for optimizing cladding design6. Similarly, understanding the oxidation process, to which carbides are highly susceptible, is a critical aspect of evaluating uses and planning controls for fuel applications. For example, incorporating uranium carbides or other carbide species, as oxygen “getters”, into TRISO fuel kernels provides a more readily oxidizable substitute for carbon in the inner graphite layer, thereby reducing buildup of gaseous CO species that reduce fuel efficiency and compromise integrity7,8. Additionally, at the back end of the nuclear fuel cycle, pristine UC may have to be oxidized prior to final disposal in a geological repository due to its pyrophoric and oxidative behavior, whereas TRISO fuels are 1 Department of Chemistry, Washington State University, Pullman, WA, USA. 2School of Mechanical and Materials Engineering, Washington State University, Pullman, WA, USA. 3Department of Materials Science and Engineering, The Pennsylvania State University, University Park, PA, USA. 4Earth and Environmental Sciences Division, Los Alamos National Laboratory, Los Alamos, NM, USA. 5School of Molecular Sciences and Center for Materials of the Universe, Arizona State e-mail: University, Tempe, AZ, USA. 6Materials Science and Technology Division, Los Alamos National Laboratory, Los Alamos, NM, USA. npj Materials Degradation | (2026)10:19 1 Article https://doi.org/10.1038/s41529-025-00732-1 considered safe for direct disposal9,10. A rigorous understanding of UC’s thermal oxidation process and further mapping of the U-C-O system are critical for developing strategies to mitigate risks in both planned and potential alterations. UC thermal expansion has been previously investigated, primarily in the 1960s and early 1970s, with most experiments using dilatometry, which as a bulk technique is sensitive to factors such as microstructure and packing density2,11–15. This presents an opportunity to validate this critical material property on the atomistic scale using a different methodology, for improved fuel design and containment considerations. Although UC thermal oxidation has been more often studied than thermal expansion, the specifics of this process remain elusive. Reported oxidation intermediates include graphite, U2C3, UC2, and CO gaseous species10,16–21, but the fate of uranium during oxidation is generally thought to follow one of the following two simplified pathways: (1) UC → UO2 → U3O8 or (2) UC → UO3 → U3O8. In the second hypothesized reaction pathway, the intermediate UO3 phase was identified based on thermogravimetric analysis (TGA) without in situ phase identification19,21–23. In the discussion of this work, we will clarify the identity of “UO3” is only a mass equivalent. While mass balance methodology of starting materials and final products provides useful generalizations about the overall process, intermediate phases cannot be exclusively identified and confirmed without in situ monitoring. Recent efforts using environmental scanning electron microscopy (SEM) were undertaken in this pursuit, but this characterization focused on morphological changes instead of phase identification; rather the SEM study clearly shows the kinetic factors at work with high temperature oxidation of UC17. In this study, we used in situ synchr (...truncated)