The severe accident at the TMI-2 reactor stimulated research to understand how a severe beyond-design accident progresses, how to mitigate its consequences, and how to terminate it. A central issue is the behaviour of the reactor core during an unprotected severe accident: when and how the core loses its original geometry, what configurations are formed, how much hydrogen is generated by the steam oxidation of core materials and how the rate of oxidation is affected by changes in geometry, which influence of core degradation exists on the release of fission products, and by what processes solid and liquid core materials are either evaporated or transported to the lower plenum of the pressure vessel. Degraded core processes are a key factor in the progression of an accident: they provide the initial conditions for in- and ex-vessel phenomena, represent direct threats to the containment and determine the source term. The knowledge of in-vessel melt relocation processes is also important with respect to cooling recovery actions and RPV failure analysis.

The objective of the project "Core Degradation" is to improve the existing code systems and to perform a series of separate-effects tests in order to develop a more accurate model of the time- and temperature-dependent core degradation process. In particular it is planned to develop failure criteria of ZrO2 layers and to compute the relocation behaviour of solid and liquid materials (e.g. candling, slumping) and its influence on the debris bed formation.
Completion of the remaining CORA tests and of the analysis of the PHEBUS SFD tests, independent peer reviews by the USNRC of its major modelling codes SCDAP/RELAP5 and MELCOR, performance of the first two International Standard Problems (ISP-28 and ISP-31) in the melt progression area, and an increased pace generally in the areas of code development, assessment and application to power plants.

The heat-up of phase of severe accidents is sufficiently well understood and modelled. There are however still significant deficiencies in the modelling following the onset of core degradation, in particular concerning fuel cladding chemical interactions and oxide shell failure mechanisms, which can dominate uncertainties in the predication of accident progression in commercial reactors. The lack of understanding of quench processes of degraded cores has been underlined in ISP-31. In these cases, further separate-effects tests and mechanistic model development are recommended.
Work programme

The modelling and validation efforts are based mainly on a European data bank provided by a series of integral (in-pile) PHEBUS-SFD tests at CEA Cadarache and integral CORA (out-of-pile) tests at KfK, which are considered as complementary. The main effort of this project is devoted to the experimental validation and the benchmarking of models able to predict a wide range of thermomechanical and chemical processes during the core degradation phase. One of the key phenomena is the oxidation behaviour of the core materials by steam and the resulting H2 generation before, during and after core material relocation. Specific quenching tests, simulating cooling recovery under faulty ECCS or accident management conditions, are foreseen at KfK with emphasis on the behaviour of ZrO2 layers on the zircaloy cladding and on materials interactions with stainless steel, Inconel and absorber materials (Ag-In-Cd and B4C). Calculations using ICARE-2 at CEA-IPSN/Cadarache and MELCOR or SCDAP/RELAP-5 at AEA/Winfrith are performed with regard to early and late phase phenomena, like oxidation of degraded core and molten-fuel/materials interactions. Complex melt chemistry aspects, including the modelling of low temperature eutectics formation, fission product release and debris bed behaviour, are examined by ENEA-ERG/Rome in connection with the ICARE-2 code developments. The University of Pisa focuses on the validation of H2 generation models due to solid and molten zircaloy oxidation against a series of PBF, LOFT and CORA tests. Calculations with KESS are performed at IKE/Stuttgart, in connection with ATHLET-CD (GRS) and ESTER (JRC) predictions, to better understand cladding embrittlement and core thermo-hydraulics effects on the quenching behaviour (oxidation energy). Radioactive heat transfer effects and late phase phenomena including debris bed formation are investigated by the University of Bochum.