Brannerite ceramic and glass-ceramic composite materials for immobilisation of damaged and degraded nuclear fuels

Future disposal of high actinide content nuclear wastes that are unsuitable for further reprocessing requires robust, durable wasteform materials. Brannerite, nominally UTi2O6, is a naturally occurring mineral that can retain the majority of its uranium inventory despite metamictisation and extensive weathering. Due to this, synthetic brannerites have been proposed as candidate materials for immobilisation of high actinide wastes, either as a ceramic or as the ceramic phase in a glass-ceramic composite. This work has investigated the underlying chemistry of a range of brannerite structured materials to examine their suitability for this application.
Initial investigations examined the formation of UTi2O6 from the component oxides in a glass-ceramic composite for the first time. A suite of samples with variations in UO2:TiO2 molar ratio, and processing time and temperature was produced and characterised by XRD, SEM, and U L3 edge XANES. Materials with stoichiometric amounts of UO2 and TiO2 contained a small fraction of UO2 in the glass matrix, but additions of hyperstoichiometric amounts of TiO2 resulted in high quality UTi2O6 glass-ceramic composites. It was observed that some grains of UTi2O6 were encapsulating small regions of UO2, suggesting that brannerite formation occurs mainly around particles of UO2, with diffusion kinetics playing a key role.
After the success of the initial investigation, a further suite of UTi2O6 glass-ceramic composites was produced, with varying glass:ceramic ratios. The compatibility of these materials with hot isostatic pressing was also examined. Materials containing up to 90% UTi2O6 by weight were successfully synthesised in a one step process at 1200 °C, showing the flexibility of this glass-ceramic system. Compared to formation of a pure ceramic brannerite, the addition of only 10% glass by weight allowed for a reduction in reaction time and temperature compared to pure ceramic brannerites (1200 °C for 6 hours, compared to 1300 °C for 24 hours or more). Hot isostatically pressed UTi2O6 glass-ceramic composites were also produced, and the impact of HIP temperature and time examined.
The impact of the glass phase composition (in the system Na2Al2-xBxSi6O16) on the formation of UTi2O6 was also examined. Unlike the formation of zirconolite, the glass composition did not have a strong effect on the formation of brannerite, with all materials forming glass-ceramic composites with UTi2O6 as the major crystalline phase. This difference was ascribed to the lack of easily formed U/Ti silicate phases in these brannerite glass-ceramic composite systems, where analogous zirconolite systems have been reported to form sphere/titanate and zircon ceramic phases when the glass phase has a relatively higher silica activity.
The effect of Ce oxidation state and O vacancy fraction as a function of temperature and atmosphere on CeTi2O6 stability was also examined, utilising thermogravimetric techniques, XRD, and Ce L3 edge XANES. It was shown that the temperature dependent Ce3+/Ce4+ redox couple is active in CeTi2O6, and the maximum stable Ce3+ content determined to be between 13.1% and 15.7% Ce3+. The thermodynamics of CeTi2O6 formation were also discussed, and it was proposed that the reduction of a small fraction of Ce4+ to Ce3+ at temperature must occur for the formation of CeTi2O6 to be energetically favourable. The results of this investigation demonstrated that Ce is a poor surrogate for Pu in some systems, particularly in thermal regimes where the Ce3+/Ce4+ redox couple is active.
Two investigations into the crystal chemistry of actinide brannerites were performed. The first examined the formation of ATi2O6 and A0.5B0.5Ti2O6 (A and B = U, Th, Ce) ceramic phases in glass-ceramic composites in two different process atmospheres, air and Ar. The key factor controlling the formation of brannerite was identified: the availability of an overall A-site charge of 4+. This was most directly observed in the samples batched as U0.5Ce0.5Ti2O6: when processed in air a Ce3,4+/U5+ brannerite was formed; under Ar, a Ce3+/U4,5+ brannerite. Although the addition of glass allows for complete formation of UTi2O6, the kinetics of ThTi2O6 formation were slower, with larger fractions of ThO2 and TiO2 observed in the final phase assemblages compared to UO2 and TiO2 in UTi2O6 glass-ceramic composites produced by the same heat treatment.
The second investigation examined the solid solubility of Al3+ in UTi2O6, and was the first systematic investigation into the use of a high fraction, lower oxidation state Ti-site dopant to stabilise the formation of U5+ brannerite. A near single phase material was produced in the material batched as UTiAlO6, and the structure examined by time-of-flight neutron diffraction. The impact of cation size on the brannerite structure was examined, with changes in average U oxidation state and Al3+ content causing shrinkage of the brannerite unit cell. The use of a Ti-site dopant to stabilise the brannerite structure when formed in air was successfully demonstrated, allowing for charge-balancing of U5+ whilst retaining the high U content that makes brannerites particularly attractive as wasteforms for high actinide content wastes.