Validation and application of a new reflooding model for the Sampson Code

Reflooding of a heated core is a complex phenomenon that can result in renewed temperature rise with localized melt formation, additional release of fission products, and in an increased hydrogen production. It has been studied during the last decades among other core degradation phenomena in experiments such as CORA, QUENCH, Phebus SFD and Phebus FP, but in the last years it got some relevance due to Fukushima Daiichi accident. Those experiments have been the base for the validation of severe accident codes, such as MELCOR, MAAP, ATHLET-CD, Scdap and SAMPSON, for core degradation scenarios including those with core reflooding. Unfortunately, the accuracy of the computational analyses during the reflooding transient is limited due to the stochastic phenomena involved. In the case of SAMPSON a detailed analysis to evaluate its capabilities is performed using CORA-17 experiment as a validation scenario. This analysis will provide insight about how the code behaves under core reflooding conditions as well as its flaws. The main conclusions of the validation shown that the code by default, not only underestimated the amount of hydrogen released during the quench phase, but also predicted a more immediate response and aggressive reaction. Based on previous studies, the code has been modified to consider the possibility of a molten pool formation in the lower part of the core as well as the crack formation and growth in the Zr clad due to the thermal shock incrementing the amount of H2 in about 79%. The importance of this validation is to obtain high level of accuracy in order to use SAMPSON for Fukushima Daiichi Unit 2 accident analysis during which a reflooding scenario it is theorized to occur.