The heat capacity and enthalpy of condensed UO 2: Critical review and assessment

Abstract

Having established the role of the heat capacity, C p ( T), of condensed UO 2 in various FBR accident scenarios, e.g. HCDA and PAHR, and having noted the unsatisfactory state of present knowledge concerning this basic thermophysical property of the fuel, all existing enthalpy and heat capacity data are collated and assessed, and certain recommendations made. The conventional method of obtaining C p ( T) by analytical differentiation of some adopted fit to this enthalpy data is then critically examined. The attendant problems are illustrated both for solid UO 2, where the contribution to C p ( T) from the weak, sigmoidal, enthalpy structure (which is just discernible in the data of Hein and Flagella) is missed and for molten UO 2, where not even the direction of the trend of C p ( T) with T can be definitively established, resulting, upon extrapolation to 5000 K, in C p values which can differ by as much as 60 J mol −1K −1. Some recent progress towards a more acceptable, “model-independent” approach, known as quasi-local linear regression (QLLR), is then reviewed and applied to enthalpy data of UO 2 on both sides of its melting point, T m . In the case of solid UO 2, a pronounced heat capacity peak, extending over about 100 K and centred on 2610 K., is revealed, whose magnitude and location is very similar to that found in other fluorite structured materials near 0.8 T m wherein it indicates a (Bredig) transition to a state characterised by giant ionic conductivities. Whilst it is impossible to establish any definite T-dependence for the C p (QLLR) values in molten UO 2, the tendency to slightly decrease appears to marginally outweigh the converse, in qualitative accord with the dependence advocated by Hoch and Vernardakis. In the post-transitional region T t < T< T m the opposite holds, as is necessary for consistency between the independently established T-dependences of the thermal conductivity and diffusivity, which requires that C p ( T) increases with T faster than the density decreases. Attention is then drawn to some interesting comparisons which exist between the behaviour of UO 2 and other (non-actinide) fluorites near their melting points, which suggest the existence, in UO 2 of a significant degree of (i) cation disorder in the post-transitional region, 0.8 T m < T < T m and (ii) electronic disorder in the melt. The review concludes with an extended, retrospective overview of the present situation regarding the heat capacity of condensed UO 2, and identifies some specific experimental goals in connection with the current experimental programme of the Joint Research Centre, Karlsruhe, Fed. Rep. Germany.