Sub-stoichiometric Zirconium Carbide (ZrC) is a viable candidate for use in nuclear environments as it has shown desirable thermo-physical properties and an ability to tolerate radiation damage. These thermophysical properties, in some cases, are sensitive to stoichiometry making ZrC a customisable and versatile material whose properties can be optimised by varying the carbon content. Whilst the majority of studies have focused on producing and measuring different properties of ZrC, very few studies have focused on accurately measuring the carbon content for a specified fabrication route resulting in a scatter in the data of some physical properties (for example the lattice parameter). Accurately determining the carbon content is key to correctly referencing the thermo-physical properties.

To study variations in the carbon content and the influence of sintering temperature during fabrication, sub-stoichiometric zirconium carbide pellets of nominal stoichiometries ranging from C/Zr = 0.5 to 1.00 were prepared by reactive hot-pressing precursor powders of zirconium hydride and graphite in varying ratios at sintering temperatures of 1500, 1700 and 2000˚C. A novel method using 13C nuclear magnetic resonance (13C NMR) was used to identify carbon present within and dissociated from the ZrC structure, allowing the true stoichiometry of the ZrC phase to be determined.

Combustion carbon analysis was used to determine the total carbon content of the samples. These samples showed lower carbon contents than the nominal values reported in the literature. Solid-state, static 13C NMR spectroscopy of the samples revealed the presence of carbon bonded in the ZrC structure, amorphous carbon, and graphitic carbon. Results presented here show that the standard methods of obtaining the carbon content such as carbon analysis and nominal methods in overestimate the ZrC bonded carbon in the samples. The NMR corrected carbon content for the lowest nominal carbon content sample was found to have C/Zr = 0.50 which is lower than the stable compositional limit for ZrC according to the phase diagram (C/Zr = 0.55). This result was in good agreement with theoretically stable carbon contents predicted by recent density functional theory calculations in the literature. For the samples sintered at 2000˚C, the main ZrC carbon resonance was observed to systematically shift to higher frequency (ppm) for lower carbon content samples.

SEM on samples sintered at 2000˚C revealed the presence of microscale inter- and intra-granular dark structures throughout the sample. These were observed to become increasingly connected as the carbon content of the sample was reduced. An increase in the grain size was also observed for samples of lower carbon content.

The layered Nb4AlC3 MAX phase was also investigated due to its previously reported ability to accommodate a large number of carbon vacancies and its high yield. This sample was revealed to be deficient in carbon by 36.6% with respect to the nominal carbon content but did not exhibit exsolved sp2 phases. Two shoulders observed in the main resonance indicated that multiple carbon environments exist within the material. The intensities and the location of these resonances were observed to best fit a vacancy avoidance model.

Irradiations, with 5 MeV protons to 0.015 and 0.031 dpa, on samples sintered at 2000 ˚C produced a rearrangement of the ZrC bonded carbon within the sample. This was predominantly observed through systematic changes in the 13C NMR line shape with increasing irradiation dose. Comparison of the pristine and irradiated nominal C/Zr = 1.00 and 0.60 NMR spectra clearly showed the emergence and growth of a shoulder to the main peak at higher frequencies (more positive ppm). The intensity of this shoulder increased with increasing carbon content. Although no significant change in the centre of gravity was observed in the sample due to irradiation, the systematic evolution of this shouldering peak at higher ppm indicated that vacancies might have been created in the ZrC structure due to irradiation. This was evidenced by the NMR spectra of pristine samples of successively decreasing carbon contents which exhibited similar systematic line shape evolution. The possible creation of vacancies and the exsolution of carbon from within the ZrC structure was further evidenced by an increase in the signal strength of the Raman spectra in the dark regions of the sample with increasing irradiation dose, which implied that the quantity of carbon within this region increased. Irradiations were also observed to increase the lattice parameters of the samples for a given stoichiometry which increased with successive irradiation doses. This result indicated that the swelling of the lattice parameter increased with increasing irradiation dose. Irradiations to 2.4 dpa using 158 MeV Xe confirmed the trends observed in the Raman and SEM investigations on the lower dose proton irradiated samples. Increasing the irradiation dose to 2.4 dpa with (158 MeV Xe) resulted in an increase in the lattice parameter as well as a decrease in the average grain size with the exception of the nominal C/Zr = 0.60 sample. These results indicated that the irradiation damage continued to accumulate for samples irradiated to higher doses.

The rock salt structure of ZrC was retained for all samples after irradiation demonstrating that ZrC remains stable under the irradiation conditions used in this study. However, it was revealed that a redistribution of carbon environments occur within the sample during irradiation with exsolution to unincorporated sp2 carbon environments consistent with graphene sheets in various states of disorder. The exsolution of carbon and the production of lower carbon content ZrC appears to be the radiation response of the material under heavy (Xe) and light (1H) ion irradiation.