Now that's scary! :)
I've been doing a little research on zirconium. My thought, that ziconium (a tranistion metal), released from melting ZrO, would "mix" with the uranium/plutonium and slow reaction.
"Nuclear applications Consuming about 1% of the Zr supply, zirconium is used for cladding nuclear reactor fuels.[15] For this purpose, it is mainly used in the form of zircaloys. The benefits of Zr alloys is their low neutron-capture cross-section and good resistance to corrosion.[5][6] The development of efficient methods for the separation of zirconium from hafnium was required for this applications.
One disadvantage of zirconium alloys is their reactivity toward water at high temperatures leading to the formation of hydrogen gas and to the accelerated degradation of the fuel rod cladding:
Zr + 2 H2O → ZrO2 + 2 H2 This exothermic reaction is very slow below 100 °C but rapid at higher temperatures. Most metals undergo similar reactions. The redox reaction is relevant to the instability of fuel assemblies at high temperatures,[27] This reaction was responsible for a small hydrogen explosion first observed inside the reactor building of Three Mile Island accidented nuclear power plant in 1979, but then, the containment building was not damaged. The same reaction occurred in the reactors 1, 2 and 3 of the Fukushima I Nuclear Power Plant (Japan) and in the spent fuel pool of reactor 4 after the reactors cooling was interrupted by the earthquake and tsunami disaster of March 11, 2011 leading to the Fukushima I nuclear accidents. After venting of hydrogen in the maintenance hall of these three reactors, the explosive mixture of hydrogen with air oxygen detonated, severely damaging the installations and at least one of the containment buildings. To avoid explosion, the direct venting of hydrogen to the open atmosphere would have been a preferred design option. Now, to prevent the risk of explosion in many pressurized water reactor (PWR) containment buildings, a catalyst-based recombinator is installed to rapidly convert hydrogen and oxygen into water at room temperature before explosivity limit is reached."
Fuel cladding interactions:
The study of the nuclear fuel cycle includes the study of the behaviour of nuclear materials both under normal conditions and under accident conditions. For example, there has been much work on how uranium dioxide based fuel interacts with the zirconium alloy tubing used to cover it. During use, the fuel swells due to thermal expansion and then starts to react with the surface of the zirconium alloy, forming a new layer which contains both fuel and zirconium (from the cladding). Then, on the fuel side of this mixed layer, there is a layer of fuel which has a higher caesium to uranium ratio than most of the fuel. This is because xenon isotopes are formed as fission products that diffuse out of the lattice of the fuel into voids such as the narrow gap between the fuel and the cladding. After diffusing into these voids, it decays to caesium isotopes. Because of the thermal gradient which exists in the fuel during use, the volatile fission products tend to be driven from the centre of the pellet to the rim area.[16] Below is a graph of the temperature of uranium metal, uranium nitride and uranium dioxide as a function of distance from the centre of a 20 mm diameter pellet with a rim temperature of 200 oC. The uranium dioxide (because of its poor thermal conductivity) will overheat at the centre of the pellet, while the other more thermally conductive forms of uranium remain below their melting points.
Release of radioactivity from fuel during normal use and accidents:
The IAEA assume that under normal operation the coolant of a water cooled reactor will contain some radioactivity[18] but during a reactor accident the coolant radioactivity level may rise. The IAEA state that under a series of different conditions different amounts of the core inventory can be released from the fuel, the four conditions the IAEA consider are normal operation, a spike in coolant activity due to a sudden shutdown/loss of pressure (core remains covered with water), a cladding failure resulting in the release of the activity in the fuel/cladding gap (this could be due to the fuel being uncovered by the loss of water for 15–30 minutes where the cladding reached a temperature of 650-1250 oC) or a melting of the core (the fuel will have to be uncovered for at least 30 minutes, and the cladding would reach a temperature in excess of 1650 oC).[19]
Based upon the assumption that a PWR contains 300 tons of water, and that the activity of the fuel of a 1 GWe reactor is as the IAEA predict,[20] then the coolant activity after an accident such as the three mile island accident (where a core is uncovered and then recovered with water) can be predicted (I don't think this would hold true for Fukushima).