My credentials are limited to a fundamental knowledge of Transition Metal Chemistry, some Thermodynamics, and Electrical Engineering. Estimates are that it will take 40 years to decommission these reactors. Efforts are underway to stop the leakage of contaminated water to the ground and the ocean nearby. Mainly, an array of coolant pipes extending 100ft down and a mile in circumference around the plant are under construction to create an ice-dam to contain further radioactive water leakage. If the ice dam works to contain further leakage, then the slower more pragmatic approach they have in mind is probably the best way to go. However, if this plan fails then it might be worth considering a roll of the dice on an alternative approach with at least one of these reactors. The following idea has inherent risks that may not be worthwhile if leakage of contaminated water can be substantially slowed.
Depending on what kind of shape the containment vessel is in, a slurry of fine Lead particles, Borosilicates, and Pumice could be injected into the pressure vessel. the water is slowly drawn off so that eventually the entire containment vessel is filled with this mixture. communication of the molten material outside of the pressure vessel is an open question, but certainly steam must not be allowed to accumulate. Without water coolant, the core will super heat and melt the lead. The melting point of Lead is low around 327.5°C, 1,132°C for Uranium, 1342°C for Pumice, and a 820°C softening point for Pyrex(reference Borosilicate). The Lead will boil vigorously creating abrasive action of the Pumice and Borosilicate against the exposed melting Uranium. This convective action of the boiling Lead will also mix the neutron absorbing Borosilicate with the Uranium as it is assimilated into the mixture. Provided that the abrasive action does not create other problems, as the Uranium is assimilated its concentration is lowered to the point that temperature eventually drops. We finally end up with a large block of hot solid that, although still radioactive, no longer needs to be cooled with water. The final state can be entombed in concrete or cut up and reprocessed.
Potential problems are the initial plume of steam and possibly aerosolized Lead and other materials escaping to the atmosphere as the reactor is allowed to heat up. The longer the process takes the more potential for contaminated material to escape into the air. There is also the scale involved and the cost of material to consider. The weight of the molten material might be too much for the support structure. Probably the biggest consideration is if the idea doesn't work it could create a bigger mess than we have now.
I am presenting this idea raw without formal evaluation in the hope that it might be helpful in the effort to stop radioactive material from leaking into the ocean sooner rather than later. The devil is obviously in the details.
Tue Dec 17, 2013 at 1:44 AM PT: This process will be even easier if the corium has leaked out. No need to draw off the water. Just pump in the particularized Lead(possibly a lead salt solution) and Borosilicate( or even borax) and let the convective action of the boiling water and lead do the work to make a diluted amalgam of the Uranium with the Lead and Boron that eventually slows the chain reaction. Considering the greater density of Uranium the process might be extremely slow. The whole concept hinges on the kinetics of molten material underwater which may require some research..