Nuclear reactors can be made in many ways. Some look good on paper but turn out to be impractical in actual use. In its effort to develop low-cost nuclear power, the Atomic Energy Commission has long experimented at such places as Brookhaven National Laboratory on Long Island, with new liquid reactor fuels—a low-melting alloy of U-233 and bismuth, a solution of uranyl sulfate, and others. But AEC soon discovered that the program was leading only to prohibitively expensive means of obtaining competitive electrical energy, and last week it announced a shift in emphasis: funds for the Brookhaven liquid-fuel project and similar ones elsewhere have been largely diverted to AEC's Oak Ridge laboratory for development of a thermal breeder reactor that will make uranium 233 from thorium, the first natural raw material other than uranium to be used as a producer of peaceful atomic energy.
Thorium, on the AEC's back burner for at least five years, is more abundant in the earth's crust than uranium, but usable concentrations are limited. It occurs in monazite sand deposits throughout the world, notably in Brazil, India, South Africa, Ceylon, Madagascar, Indonesia, Malaya and Russia's Ilmen Mountains. In the U.S. it is present in the sand of East Coast beaches, is also found in Idaho and Wyoming.
Since it cannot sustain a chain reaction itself, natural thorium is not classified as a nuclear fuel but as a "fertile" material. The thorium may be placed in a reactor fueled with uranium 235. Neutrons produced by the fissioning U-235 are run through a moderator made of graphite or heavy water, thus slowed from 10,000 miles a second to a modest one mile a second. Fast neutrons would bounce off or pass through the thorium, but at this speed neutrons are moving slowly enough to be captured by thorium 232 atoms, turning them into unstable thorium 233. When thorium 233 decays, after a brief half life of 23 minutes, it becomes protactinium 233, which in turn decays after a half life of 27 days, becomes uranium 233, a full-fledged nuclear fuel as fissionable as the better known radioactive uranium isotope U-235.
The fissioning U-233 produces more than enough neutrons to maintain the original chain reaction. If these extra neutrons are captured by thorium, the reactor will produce more U-233 fuel than it can use. The AEC thinks this breeding factor is the key to cheap nuclear power. If a breeding reactor such as the one being planned for Oak Ridge were to start operation in 1959, it could be expected to produce enough material to fuel a duplicate of itself by 1984.