Abstract
Grain growth has been investigated in high purity uranium dioxide and in uranium dioxide containing a number of additives. High purity UO[2]F[2] powder was chosen as the source material for the fine UO[2] powder needed to produce the specimens for this work. Ammonium diuranate was precipitated from a solution of UO[2]F[2] using ammonia. Using a solution coating technique, quantities of ammonium diuranate were doped with various additives: Cr, Ni, La, Mo, and Al. The ADU was reduced to UO[2] by heating for two hours in dry hydrogen at 500°C. The resulting powder had a surface area of 18-20 m[2]/g. The UO[2] powder was cold pressed into pellets weighing 0.5g each and these were sintered at 1400°C in wet hydrogen. The water content of the hydrogen was not sufficient to cause deviation from stoichiometry. Completely pore free specimens were obtained. The surfaces of the specimens were then ground and polished flat suitable for microscopic examination because thermal etching was relied on to reveal grain boundaries. Grain growth experiments were performed at temperatures between 1400 and 1800°C, the majority however, were at 1500, 1600 or 1700°C. A multianneal technique on two specimens provided all the data for each experiment. Grain size measurements were made on photomicrographs of the specimen surfaces. Spectrographic analysis was carried out on the specimens at the end of the runs. These revealed that excessive evaporation of the dopant compounds had occurred in most cases resulting in lower dopant concentrations than were intended and also in contamination of each batch of UO[2] with dopants used for the previous experiments. The experimental results showed that the behaviour of the undoped UO[2] was controlled by the residual impurities of which iron and silicon were the most abundant. The molybdenum doped material showed results typical of a material containing inert second phase particles. The lanthanum doped materials behaviour indicated that solute drag was occurring. The activation energy for grain growth in this material was 40 kcal/mole. The Cr, Ni, and Al doped specimens all showed different grain growth exponents at different temperatures indicating that the rate controlling process changed with temperature. Exaggerated grain growth occurred in Cr and La doped specimens. It was concluded that impurity effects on grain growth are dependent on the concentration, the mobility and the energy of interaction of the impurity with the grain boundaries.