A nuclear reactor can't explode like an atomic bomb. But some experts feared "steam" explosions of lesser force might have serious consequences for nuclear plant safety.
Now recent experiments suggest that even this type of explosion is less likely than had been thought.
Theoritically a steam explosion could occur in an accident where the reactor core suddenly loses its cooling water. If all the backup cooling systems fail then the heat in the uranium fuel is great enough to melt the core. If the molten fuel then fell into water at the bottom of the reactor vessel, a steam explosion theoretically could result.
Is it probable, even possible, that a steam explosion could blow a hole in the heavy, concrete dome covering the reactor and so release large amounts of radioactive material into the environment?
Some nuclear experts say that steam explosions of such magnitude are impossible. But using conservative assumptions, the US Reactor Safety STudy (RSS) estimated that they represent 1 to 2 percent of the total risk in operating a nuclear reactor. Antinuclear forces have stressed the uncertainties involved in such estimates.
However, recent experiments conducted at Sandia National Laboratories in Albuquerque, supported by work being done in West Germany and Italy, indicate that the risk of such an explosion is 10 to 100 times less than was assumed in the RSS analysis.
"Generally, this reduces steam explosions from a marginal to an insignificant source of risk," maintains Rick Sherry, research program manager for the Nuclear Regulatory Commission (NRC). However, he adds that this conclusion applies only to Pressurized Water Reactors, one of two types operating in the US. Experiments applicable to the other type, General Electric's Boiling Water Reactor, are not yet completed, Mr. Sherry reports. "When they are I expect we will be phasing out research in this area," he adds.
The Sandia tests on which this conclusion is based involve dropping mixtures of molten metal and metal oxides which simulate the material in a reactor core into specially instrumented water chambers and measuring the results, explains Dr. Marshall Berman who supervised the experiments.
These ranged from small-scale tests, involving only ounces of molten material to those involving 50 pounds of melt. They used both a mixture of iron and aluminum oxide and "corium," the complex mixture of uranium, zirconium, oxygen, and stainless steel which experts feel would result if a reactor core melted.
The test results, coupled with considerable theoretical and analytical work, found that the chance of a steam explosion occurring after a core meltdown was five times greater than had previously been assumed but the likelihood that such an explosion could blow a hole in the concrete contain ment dome was far lower.