In reactor safety research, there has been a continued drive to improve the understanding of severe accidents in order to prevent significant releases of radioactivity under severe accident conditions. The work has been focussed on different levels corresponding to the defence-in-depth principle discussed earlier.

The first objective during a core melt accident is to maintain vessel integrity following attack by molten corium or debris from the higher up reactor core internals. The corium may or may not be in contact with water and even in the latter case may not coolable. Various research programmes have been carried out including investigation of early and late phase melting phenomena under different accident conditions and the energetics of corium/water interaction studies of heat transfer-related mechanisms from the debris to the vessel. Others include thermal-hydraulic cooling of debris beds, investigations of the structural response of the vessel and examination of the effectiveness of cooling of the vessel with water from the outside, etc. Theoretical programmes of work supported by experimental programmes have been carried out. The combined programmes have considered scaling effects and also how to extrapolate results obtained from simulant materials to reactor materials.

If the vessel is breached, then molten debris will be released in the cavity beneath the reactor. It is therefore important to understand the physical processes of the potential release of melt from the vessel, how it will spread over the concrete floor or how it might be impeded by other retention structures (in some of the newer designs). It is likely that there will also be interaction of corium with water from discharge of emergency core cooling systems (ECCS) and the need to quantify the load on the containment from any resulting steam explosions.

The EC programmes have covered experiments and theoretical studies on the thermochemistry of molten corium interactions with structures. They have included projects to determine the production of hydrogen and other non-condensable gases, e. g. carbon monoxide to establish the threat to containment from gas combustion. Another facet was to consider the retention of fission products in the melt with respect to the ‘Source Term’ (see below). Work items covered vessel failure and corium release modes, corium spreading effects and the consequent impact on direct containment heating. It also covered the interactions of corium with water and structures and generic studies on retention devices (e. g. core catchers).

An important aspect of severe accident research has been to quantify the Source Term. This is defined to be the quantity, timing and physical form of the radiological and chemical species release to the environment. It is dependent on the type of accident. The important inputs for determining the Source Term are fission product release from the fuel, and the transport in the primary circuit and the containment. Also important are the suspension, resuspension and condensation/revaporisation mechanisms within the reactor circuit and containment. Accident mitigation devices such as sprays and other measures have an important mitigation effect (Table 8.5). There are important large-scale integral tests and supporting separate effects tests to provide data to validate computer codes for analysis.

The ultimate Source Term to the environment depends on whether the containment is breached. There are different threats in the short term and long term.

Assuming the containment holds the Source Term will depend on the leak tightness of the containment. The short-term threat arises from corium/steam explosion, hydrogen

combustion, more particularly detonation, direct containment heating, secondary effects of missiles and containment isolation failure. The long-term threat comes from the build up of heat (and therefore pressure) due to failure of removal of decay heat or by failure of isolation devices, i. e. material failure. The containment strength is very design dependent and different mitigation systems will be feasible for different designs.

Within the EC framework programme, there have been generic experiments and theoretical studies on hydrogen combustion (deflagration and detonation), thermal — hydraulics, stratification and natural convection. Also included in the programme has been investigation of dynamic concrete behaviour at high impact velocity and studies of leakage of steam and aerosols through cracks and penetrations. These have been in conjunction with the identification of mitigation measures.