Left: Test facility for qualification of melt spreading tests. Photo: Argonne National Laboratory (ANL).
Right: Example of corium melt pouring for a melt spreading preparatory test. Photo: Argonne National Laboratory (ANL).

The NEA Reduction of Severe Accident Uncertainties (ROSAU) project aimed to reduce knowledge gaps and uncertainties associated with severe accident progression and mitigation. Following the Fukushima Daiichi accident, systematic re-evaluations of assumed accident initiators were carried out worldwide. These re-evaluations examined plant responses to severe accident conditions and operator actions for coping with and terminating severe accident progression. Based on these analyses, gaps were identified in two areas of severe accident progression where additional knowledge was needed and uncertainties could be reduced. These areas were: 1) the spreading of core melt in the containment cavity after the reactor pressure vessel rupture and the effect of metal content in the melt on molten core-concrete interaction; and 2) in-vessel and ex-vessel core melt and debris coolability. The status of technical knowledge in these two areas, focusing on ex-vessel configurations, had been thoroughly reviewed in State-of-the-Art Report on Molten Corium Concrete Interaction and Ex-Vessel Molten Core Coolability and in Status Report on Ex-Vessel Steam Explosion.

The ROSAU Project, supported by 15 partners in 8 countries, comprised 10 tests in the above two areas. Experiments were conducted at the Argonne National Laboratory (ANL) in the United States, under the co-ordination of the US Nuclear Regulatory Commission, and include five large underwater melt spreading tests (MST) with up to 300 kg of molten prototypic material in a newly designed facility and five smaller melt-core-concrete interaction (MCCI) and coolability tests (DCAM) with up to 80 kg of molten prototypic material. Tests were conducted at temperatures up to 2 500°C, with different metal and concrete contents in the melts, with different cooling water flow rates and sub-cooling. In addition to these experiments and with the support of project partners, the ANL carried out an analytical activity to refine and validate models and codes for each test category, so as to form the technical basis needed to extrapolate experiment findings to plant conditions.

The project started in June 2019 and was completed at the end of 2024. Five DCAM tests and five MST tests were completed successfully and documented. Small-scale melt spreading experiments, contributed in-kind to the project, were also completed and documented at the KTH Royal Institute of Technology, Sweden, providing additional data on melts spreading with and without water in the cavity. In addition, the ANL provided project members with new versions of the CORQUENCH and MELTSPREAD codes, where models have been significantly enhanced based on the ROSAU project tests results. These codes were used by the project partners to simulate the project results.

A new four-year phase of the project, ROSAU 2, has been proposed by the ANL to investigate in- and ex-vessel melt coolability to close issues with conventional fuels and to check models’ applicability to ATFs. New data are expected to inform severe accident management for the current fleet as well as new water-based SMRs. Discussions of the proposal have advanced and there are plans to start this new project phase at the end of 2025 or in early 2026.

ROSAU Members' area (password protected | reminder)