A mechanism believed responsible for the explosive vapor growth observed in Armstrong’s experiment with LMFBR materials (liquid sodium injected into molten UO2) is discussed. Basically, liquid sodium globules can be entrained and wet the liquid UO2 surfaces. The lack of nucleation sites in the liquid-liquid-like system results in the overheating of the liquid sodium until homogeneous nucleation occurs. When the superheat limit is reached, vaporization is rapid enough to produce shock waves.The validity of the proposed entrainment-wetting-superheat mechanism to explain the observed UO2-Na explosions has been demonstrated by above surface injection of liquid Freon-11 (normal boiling point 23.8°C) into hot water (70 to 90°C). This liquid-liquid system is believed to simulate Armstrong’s UO2-Na system, since in the latter, the transition from film boiling to nucleate boiling based upon Henry’s film boiling correlation will take place at a temperature well above the melting temperature of UO2.Based upon this mechanism, large-scale coherent UO2-Na vapor explosions would appear impossible in a reactor environment. In a real reactor system, fission gases and fragments, as well as solid materials and gas bubbles entrained in the liquid sodium, would be present and would promote boiling prior to reaching the threshold for homogeneous nucleation in sodium, therefore resulting in mild interactions of the type observed in the TREAT in-pile experiments.