Failure characteristics of cladding tubes under RIA conditions

Abstract

In the frame of actions for improving the safety of its nuclear power plants, Electricité de France needs to build the mechanical criteria ensuring the clad integrity for several operating conditions. This paper presents analytical mechanical models used to derive failure criteria for reactivity insertion accidents (RIA) and interpretation of the CABRI REP-Na experimental tests. Building analytical criteria requires an experimental database. Mechanical tests performed on non-irradiated and irradiated cladding tubes have been provided from French and international programmes (PROMETRA, EPRI, …). These tests consist of tube burst and axial tension, and ring tension. Several strain biaxiality ratios are thus available: pure circumferential tension (from ring tension), pure axial tension (from tube axial tension), and plane strain conditions (from tube burst tests). Several strain rates, temperatures, irradiation conditions are also available. The major feature of our study has been to make it possible that these several thermomechanical conditions be representative of “standard” RIA loading conditions. To this aim, we have derived some biaxiality and strain rate corrections to be applied to the results of experimental tests, in such a way that they could be representative of RIA biaxiality conditions (which are assumed to be strain equibiaxiality), and also representative of RIA strain rate conditions (which are assumed to be 5 s −1). The corrections that we derive are based on the fracture properties of hydrided zirconium alloys (especially in terms of anisotropy), and also on an assumed form of the material constitutive equations. Each test of the “homogenized” database has thus been used to calculate a strain energy density, representative of its fracture (the strain energy density is defined as the integral of the stress times strain rate states, over the duration of the mechanical test). The SED values are plotted against the sample's oxide thickness, and a lower bound limit can be established, with respect to oxide thickness. In order to address the problem of representativeness of the laboratory database, an experimental set-up has been developed that aims at characterizing the failure behavior of cladding tubes under RIA conditions. The developed experimental set-up is based on electromagnetic forming. The development of the test and in particular of the die is delicate but leads to repeatable results with a controlled strain biaxiality ratio higher than those obtained through conventional tests such as ring tests or PSU ring tests. The use of electromagnetic forming process allows testing the specimen with very high strain rates. For the next test series other zircaloy alloys at the reception state and in hydrided conditions will be tested in order to look at hydrides influence on fracture strains. A first finite element simulation of the test was engaged. The simulation and the experimental results are in quite good agreement. In future, the consistency of the previously developed analytical mechanical criterion with the electromagnetic forming experimental results will be verified.