Sub-surface precipitations of UH3 have been modelled with Finite Element methods. The model includes a pre-stressed oxide layer, metal anisotropy, metal plasticity and a hyper elastic coherent hydride. The model was used to investigate UH3 precipitating at a variety of depths from 0 to 2um (spherical shape), with a variety of sizes from 0.08 to 0.8 um diameter (spherical shape) and with a variety of shapes from horizontal ellipsoid to vertical ellipsoid (depth 0.5 um). UH3 precipitation close to the surface was found to be energetically favourable as observed in experiments. Simulations on the shape of the precipitate found that the vertical ellipsoidal precipitates were found to be energetically favoured in contrast to what has been experimentally observed. In all cases the system could not accommodate the phase change by elastic deformation alone but by a combination of elastic and plastic deformation. When anisotropy is introduced into the metal matrix, the precipitate is surrounded by compressive and tensile regions. Tensile regions are found in the x-y plane adjacent to the precipitate and it is suggested that these regions are more likely to transform into further UH3 (through increased hydrogen diffusion, solubility or ease of phase change). Such precipitate development in the lateral direction would result in the experimentally observed horizontal ellipsoids. Multiple sub-surface precipitates were simulated (0.5 um depth) and it is suggested that compressive regions that develop between the precipitates could act as a barrier to coalescence. The surrounding stress regions and energetic factors suggest that there is a barrier to amalgamation for a distance of 1.5 um. Whereas for precipitates closer than 1.5 um there is an energetic benefit to coalescence. The transformation from sub-surface UH3 precipitates into growth centres (exposed UH3 on the surface) was examined by monitoring the oxide stress. This work also suggests that the transformation of sub-surface UH3 to growth centres could be retarded by an increased oxide thickness and the presence of a work hardened layer. To aid confidence in the model nano-indentation experiments were carried out on constrained UH3 surface films in the absence of air. Collected data shows a bulk modulus of 180 +/- GPa which is more in line with DFT calculated results as compared to Diamond Anvil Cell experimental work. The nano-indentation work represents the first time this type of data has been derived for UH3 in this way.

In this work a manufacturing method for UN, ZrN and (U,Zr)N pellets was established at the nuclear fuel laboratory at KTH Stockholm/Sweden, which consists of the production of nitride powders and their sintering into pellets by spark plasma sintering. The nitride powders were produced by the hydriding-nitriding route using pure metal as starting material. This synthesis was performed in a stream of the particular reaction gas. A synthesis control and monitoring system was developed, which can follow the reactions in real time by measuring the gas flow difference before and after the reaction chamber. With the help of this system the hydriding and nitriding reactions of uranium and zirconium were studied in detail. Fine nitride powders were obtained; however, the production of zirconium nitride involved one milling step of the brittle zirconium hydride. Additionally uranium and zirconium alloys with different zirconium contents were produced and synthesized to nitride powders. It was found that also the alloys could be reduced to fine powder, but only by cyclic hydriding-dehydriding. Pellets were sintered out of uranium nitrides, zirconium nitrides, mixed nitrides and alloy nitrides. These experiments showed that relative densities of more than 90% can easily be achieved for all those powders. Pellets sintered from mechanically mixed nitride powders were found to still consist of two separate nitride phases, while nitride produced from alloy was demonstrated to be a monophasic solid solution both as powder and as sintered pellets.