Investigations are performed to study the mixing between immiscible liquids in a pool configuration due to an upward gas flow. A water-R113 system is used in the bubbly/churn-turbulent regimes to determine the effects of the unagitated pool depth on layer mixing. The superficial gas velocity at which full mixing is attained is observed to increase with the pool depth, although it is concluded that this is a weak dependency. Mixing in the churn-turbulent regime is studied with Wood’s metal-water and Wood’s metal-silicone fluid (100 cS) as pairs of fluids. Additional past mixing data from six other fluids are also included in the data base. A criterion is proposed to determine if two liquids will entrain in bubbly or churn-turbulent flow. Correlations are derived that, for a set of given conditions, allow prediction of the mixing state (mixed or segregated) of a system. Because of the indirect method of measuring the mixed layer thickness, pool void fraction experiments are also performed. For the case of water and R113, the effect of unagitated pool depth on the void fraction is studied. The drift flux two-phase flow model is used for the analysis of the void fraction experiments, and correlations are proposed to predict the void fraction for both the bubbly and churn-turbulent flow regimes. These correlations take into account the physical properties of the liquid, the unagitated pool depth, and the superficial gas velocity. After comparison with independent data, it is concluded that they are suitable for molten core/concrete interaction modeling in the absence of solid crusts.