The main areas for additional research on in-vessel related response, following a core melt event, relate to the timing and influence of reflood, both early and late (Krugmann, 2001). There are uncertainties in the late core degradation mechanisms and how these affect the pressure vessel failure mode. There have been experimental programmes at Sandia

National Laboratories, USA, the Paul Scherrer Institute (PSI), Switzerland and the Royal Institute of Technology (RIT), Sweden, addressing these issues. Particular EC projects in relation to vessel failure are the EC funded ARVI project and the FOREVER experiments at RIT. It has been shown that the pressure vessel failure mode impacts the integrity of the vessel supports, corium dispersal, missile generation and direct containment heating risk. In regard to outstanding issues, there is a research need to consider the hydrogen production rate in the event of delayed depressurisation as this impacts the hydrogen management control system. The composition and temperature of the gas discharge will depend on the response of the primary system discharge valves. At high pressure, the integrity of the SG tubes may also be an issue.

Activities currently in progress within the EC 5th Framework Programme include the following.

The core loss during a severe accident (COLOSS) (Adroguer et al., 2001) programme considers various issues concerning core degradation phenomenology. For both PWR and VVER rods, it includes the impact of UO2 and ZrO2 dissolution by molten Zircaloy on core geometry degradation. The objective is to examine the consequences on hydrogen production, melt generation and the source term. It also addresses how the burn-up effect affects the dissolution of UO2 and MOX fuel by molten Zircaloy for PWR rods.

The experimental programme considers how the oxidation of U-O-Zr mixtures contributes to the peak hydrogen production during core reflood. Separate effects tests are carried out using a number of different composition U-O-Zr alloys. The results show that the oxidation of mixtures contributes to significant hydrogen release during degraded core quench.

Several large-scale tests are included to examine the B4C effects, from absorber rods, on core degradation and melt progression. These include a large-scale VVER-1000 bundle test with a central B4C rod, carried out in AEKI, Hungary, and a similar test with a B4C rod carried out at FZK, Karlsruhe in Germany. Results show large escalation of oxidation and hydrogen during the final steam cooling phase, this phenomenon had not previously been observed.

The programme is also examining whether the oxidation of B4C rods can induce volatile organic iodine production.

Some of these issues have been examined in earlier EC 4th FP projects, CIT (Adroguer et al., 1999) and COBE (Shepherd et al., 1999), which, respectively, were concerned with core material interactions and quench effects during core degradation.

Within the NEA collaborative programme, the MASCA (NEA Annual Report, 2002) project has also investigated the consequences of core melt within a severe accident. Experiments are being carried out in the Kurchatov Institute in which prototypical corium compositions are used. The experiments address the uncertainties on heat load to the reactor vessel and, therefore, on the uncertainties of vessel failure.