Fast Reactors, Slow Progress
GNEP is a very long-term vision; most of the initial $250 million would be spent just to study how the new technologies might work and what they would cost. But its proponents’ thinking is that we need a very long-term vision. The Energy Department predicts that 1,000 nuclear power plants will be running worldwide by midcentury, up from 441 today. And the existing uranium supply, GNEP advocates argue, won’t feed that many reactors.
The size of the uranium supply is in fact unknown, because uranium went through a long period of depressed prices, and not many people have been looking for it lately. According to industry sources, about 3 million tons are known to exist, but another 12 million tons or so may be out there. (An MIT study in 2003 predicted that enough uranium was still available to build 1,000 reactors and run them for 40 years.) To the extent that we may need to stretch this resource, however, GNEP offers a way – at least on paper – to recover vast amounts of additional energy from it.
Existing reactors generate energy through a chain reaction that begins when a free neutron hits an atom of U-235, an isotope of uranium, and splits its nucleus. The split atom throws off two or three neutrons; usually, one splits another U-235 atom, and others are absorbed by atoms of another uranium isotope, U-238, to form plutonium-239 and other transuranic elements (those beyond uranium in the periodic table). These transuranics, along with fission products such as cesium isotopes, are among the components of nuclear waste.
The trouble is, U-235 is a relatively rare isotope; natural uranium consists of about one part U-235 to 142 parts U-238, which is not as easily split. Uranium used for reactors is enriched so that U-235 occurs at a concentration of one part in 20. GNEP would use uranium more efficiently by burning transuranics from spent fuel, after they are separated from the other by-products through reprocessing. It could also exploit some of the U-238. The key would be to develop a new generation of reactors, called “fast reactors.”
Reactors that are cooled by water, as almost all reactors are today, slow the neutrons considerably after they’re released by the chain reaction. But the reactors proposed by GNEP would not; they would use a different material, probably molten metal, to carry off the heat. (Unfortunately, the preferred metal for this purpose – sodium – burns on contact with water or air.) Like a billiard ball shot by a more powerful cue, the neutrons would pack a bigger punch – enough to split some of the U-238 as well as the transuranic isotopes.