The COLOSS project was a 3-year shared-cost action which started in February 2000. The work-programme performed by 19 partners was shaped around complementary activities aimed at improving severe accident (SA) codes. Unresolved risk-relevant issues regarding H2 production, melt generation and the source term were studied, through a large number of experiments such as a) dissolution of fresh and high burn-up UO2 and MOX by molten Zircaloy, b) simultaneous dissolution of UO2 and ZrO2, c) oxidation of U-O-Zr mixtures, d) oxidation of B4C material and e) degradation-oxidation of B4C control rods. In parallel, corresponding models were developed for SA computer codes. These codes were then used to apply experimental results in plant calculations and evaluate their consequences on key SA sequences in different plants involving B4C and in the TMI-2 accident.
Significant results have been produced :
- Main B4C oxidation issues have been largely resolved. The large database produced from SETs and large scale CODEX and QUENCH bundle tests showed that B4C oxidation is strongly dependent on thermal-hydraulic conditions. Productions of limited H2, CO, CO2, very low CH4 and large amounts of aerosol were found. The CO and CO2 production is sufficient to affect the volatile FP chemistry in the circuit, in particular the iodine speciation.
- Data obtained on oxidation of U-O-Zr mixtures showed faster kinetics than for pure Zr. These data confirm that mixtures are a significant source of H2 production during core degradation and, mainly, during core reflooding. This is a key insight for the modelling the H2-peak production during reflooding, which is presently inadequate in codes.
- Results on simultaneous UO2 and ZrO2 dissolution improved understanding of fuel rod liquefaction, nevertheless more data are needed on the induced clad failure and rod collapse.
- Data were produced on the burn-up effect on UO2 dissolution and, for the first time, on MOX fuel dissolution, showing enhanced kinetics and greater apparent dissolution than for fresh fuel. Burn-up is a key underlying phenomenon suspected to favour the early fuel rod collapse observed in some conditions 450 K below the UO2-ZrO2 eutectic point ~2870 K.
- Mechanistic and parametric models were produced on B4C oxidation, fuel rod dissolution and mixture oxidation enabling SA codes under development in the EU to be upgraded. This effort was continued after the project to take full account of the findings and focussing more on the coupling between phenomena suspected to be as important as the separate models.
- A large series of plant calculations was done, final ones run with codes improved by the project. Sensitivity studies and code-benchmarks enabled code uncertainties to be evaluated on the H2 production, B4C-reaction gases and corium behaviour. The implications for safety were identified as well as strengths and weaknesses of SA codes to predict core degradation.
Experimental and analytical results are expected to be consolidated in the future SARNET network (6th FP), in particular regarding experiments on dissolution of irradiated fuel and oxidation of melts under transient conditions. Further plant applications with sensitivity studies and code benchmarking are strongly recommended to re-valuate code uncertainties.