Recent advances in understanding ruthenium behaviour under air-ingress conditions during a PWR severe accident

In a hypothetical severe accident in a Pressurised Water Reactor (PWR), Fission Products (FPs) can be released from the overheated nuclear fuel and partially transported by gases, composed of a mixture of superheated steam and hydrogen, to the reactor containment. Subsequent air ingress into a damaged reactor core may lead to enhanced fuel oxidation, affecting some FP release, especially that of ruthenium. Ruthenium is of particular interest because of its high radiotoxicity and its ability to form very volatile oxides. In the reactor containment, such volatile forms are very hazardous as they are much less efficiently trapped than particulate forms by emergency filtered venting. In the four and a half years of SARNET, collaborative research dedicated to the "ruthenium story" has been performed by several partners. This paper presents the main achievements over the whole project period. Starting from experimental observations showing that fuel could be extensively oxidised by air to, and that a significant fraction of ruthenium inventory can be released, rather satisfactory models have been developed. In addition, the effect of the air interaction with Zircaloy cladding, as well as with UO2 itself, has been studied. Experiments on the complex transformations of ruthenium oxides upon cooling through the reactor circuit have been performed. An unexpectedly large effect of temperature on the decomposition rate of gaseous ruthenium compounds has been found, as well as effects of the nature of circuit internal surfaces and other FP deposits. So it has been highlighted that various forms of ruthenium can reach the containment, but the most probable gaseous species under these conditions is ruthenium tetroxide. Preliminary analysis of ruthenium transport supports these conclusions. Experiments and analysis have also been launched on the radio-chemical reactions undergone by these ruthenium oxides in the reactor containment. Competing effects of gaseous decomposition to solid particles and re-volatilization from these ruthenium deposits have been demonstrated and modelled. The paper concludes by identifying the remaining work needed to achieve full resolution of the ruthenium source term issue. Recommendations are made for future research activities in the follow-up programme SARNET2. © 2009 Elsevier Ltd. All rights reserved.