Fracture toughness and fracture surface energy of sintered uranium dioxide fuel pellets

Abstract

The fracture properties of sintered uranium dioxide fuel pellets for thermal reactors play an important role in the formation of stress-induced cracks and fission-gas swelling during the reactor operation. The fracture toughness or critical stress intensity factor, K~c, and the fracture surface energy, 7~, (energy required to create unit area of fractured surface) are two parameters which can be used for predicting the mechanical behaviour of the fuel pellets. The reported values of fracture surface energy, 7s, for uranium dioxide pellets show a large variation (0.4 to 8 J m-Z) [1, 2]. An attempt was made to study the variation of Kit and 7s in sintered uranium dioxide pellets in the density (4) range of 9.86 to 10.41 g cm 3 using Vickers indentation technique, now commonly used for determination of fracture parameters of brittle and ceramic materials [3-6]. Half-penny shaped cracks are developed in ceramics and the crack length is related to the fracture parameters (K~c and 7s). A new parameter called fracture modulus Ef = (Klc):/7s was also calculated from the experimental data. Uranium dioxide pellets of different densities (90 to 95% theoretical density) were made from a single batch of green powder (ammonium diuranate route) by cold compaction and sintering at an average temperature of 1680°C for 5h. The geometric density of the pellets was measured as specified by ASTM standard C-766-79 for standard uranium dioxide pellets. The height of the pellets varied from 14 to 16mm and the average diameter of the pellets was 12.35 mm. Vickers hardness indentation was made on metallographically polished specimens using a universal hardness tester (Eseway, Type: DVRB-M) and the hardness was determined. A load of 294N was applied and the radial crack lengths in the four directions were measured using an optical microscope at x 400 immediately after the indentation. The ratio of the crack length (C) to half diagonal of indentation (a) made on the samples was greater than 2. Fig. la presents a photomicrograph of a specimen showing the cracks when viewed in an optical microscope. Fig. l b shows the crack tip ending when viewed through a scanning electron microscope. A minimum of four indentations were made on each of the samples and the average crack length (C) of each indentation and the average of hardness values were used for the calculation. Indentations resulting in chipping were discarded and fresh indentations were made. It was confirmed that there was no change in crack length as a function of time after the indentation was made. The overall average crack-length (C) and the volume fraction porosity (p) in the pellets were plotted (Fig. 2). The data were fitted to a straight line C(#m) = 1073 to 3990p, withacoefficientofregression r = 0.96. The fracture toughness, KIo, was calculated from the average crack length and the hardness using the equation Kjc = f(E/H)~/2(p/c3/2) (1)