Impact of B{sub 4}C on the stratification of molten steel and BWR-type core material under IVR conditions

As part of the validation of the In-Vessel Melt Retention (IVR) strategy for its KERENA BWR, AREVA NP has performed a quantitative assessment of the potential impact of thermochemical phenomena. This was motivated by the fact that several of these phenomena, namely the formation of a dense metallic phase, have the potential to lead to a strong increase in local heat fluxes, with the risk of early IVR failure. In this context, experiments performed in the MASCA project with PWR-type corium melts were repeated using a typical BWR core melt: characterized by a lower U/Zr-ratio and higher contents of Zr, steel, and boron carbide (B{sub 4}C). Applying an improved 'cold crucible' induction heating technique, two series of test were performed, first without B{sub 4}C and then with a B{sub 4}C content of about 1.4 wt%. These two values bound the uncertainty with respect to the incorporation of B{sub 4}C into the molten pool. The target of the tests was to localize the point of equal densities between the dense metallic phase (Fe, U, Zr and O) and the residual oxidic phase in thermo-chemical equilibrium. An interesting result of these experiments was that, different from earlier MASCA tests with PWR-type coriummore » that showed an only insignificant impact of B{sub 4}C on metal density, the new experiments reveal a strong corresponding effect, which can over-compensates the density increase caused by U-migration into the metallic melt. This deviating result is attributed, first, to the higher Zr-fraction in the BWR-type core melt, and second, to the higher content of B{sub 4}C in the BWR-type melt in comparison to the MASCA tests (B{sub 4}C-content <0.5 wt%). Based on the obtained results it is predicted that - under certain conditions -B{sub 4}C can completely prevent the formation of a dense metallic phase, independent of the amount of molten steel in the melt. The paper gives an overview of the performed experiments and their main results and provides theoretical models to explaining the observed strong reduction in metallic phase density in presence of B{sub 4}C Their predictions are then compared to the experimental data. (authors)« less