Our Nuclear Future: Facts, Dangers and Opportunities

by Edward Teller; Albert L. Latter

One of the first acts of the Atomic Energy Commission was to establish a Committee for Reactor Safeguards. With the passing of years this committee had to take on more heavy responsibilities. At first it had to operate under secrecy. With the wider and more public use of reactors the safety considerations are becoming more available to the public. The question of safe operation of a machine cannot be separated from a thorough understanding of the working of the machine. We cannot attempt to give an adequate description of a reactor or of the safety rules. A few general statements have to suffice.

A working reactor is full of neutrons. In a small fraction of a second these neutrons produce fission and a new generation of neutrons comes into being. In slow reactors which contain lots of light elements like hydrogen or carbon, the neutrons move with speeds little greater than that of sound and a generation may last as long as a millisecond (one thousandth of a second). In fast reactors which contain almost exclusively heavier elements like uranium or iron, neutrons move with a great speed which is about three per cent of the speed of light. In this case one generation replaces another in less than a microsecond (one millionth of a second).

Fortunately not all the neutrons get reproduced so rapidly. Some fissions produce delayed neutrons which are emitted usually with a delay of several seconds. In a steadily working reactor each generation should have the same number of neutrons as the previous one. If each succeeding generation has even a slight surplus, the reactor will become hot and may explode in a small fraction of a second. The main reason why safe operation is possible is the fact that fast multiplication can occur only if each generation becomes more populous even when one does not count the delayed neutrons. A slightly overactive reactor is easily governed, but there comes a point when the dormant dragon begins to stir. This happens when there are enough neutrons produced so that multiplication can occur without waiting for the delayed neutrons. At that point a well behaved dragon will perform a harmless action. For instance it may blow a fuse. But a vicious dragon will spit radioactive fire.

It is not easy to predict whether the dragon will be always well behaved. But with careful analysis one can make such a prediction. For instance one must look into the question of whether the reactor is stable. If it gets hotter, does this make the reactor proceed even faster so that the rate of heating increases and the reactor runs away? In a stable reactor excess heat should tend to stop the energy production and thus the reactor cools and returns to its normal operating temperature.

But too great a stability may also be dangerous. Heating may be overcompensated by the cooling mechanism; after the reactor has become too cold it may then heat up too fast and overshoot again. We must guard not only against a simple run-away, but also against increasing oscillations.

In many reactors unusual chemical compounds are used. A reactor accident may start with nothing worse than an ordinary chemical reaction between strange compounds under strange conditions. But if this chemical reaction destroys the reactor sufficiently to allow some fission products to escape, then such a chemical accident can be as bad as one of nuclear origin.

In the interior of the reactor materials are exposed to unusually strong radiation. Under this effect some materials can change their chemical properties so that what has been inert as a construction material may become dangerous during the operation of the reactor.

Perhaps the most important single item is the arrangement of mechanical controls. The reactor is adjusted by a system of sheets or rods made of a material which absorbs neutrons. This arrangement must be so constructed that the control rods can be withdrawn only at a very slow rate. But it must be possible to put them back quite fast. Any danger signal should shove the absorbers in at maximum speed. The technical expression is “scram.”

The main point, however, is that all the dangers and safety devices can be studied and after careful study a nuclear accident can be avoided. Some reactors are now so thoroughly understood that they can be safely used for training of future nuclear engineers. Other reactors which are more powerful or less well studied have to be used more carefully. Some reactors should be, and are being, enclosed in gas-tight containers. If an explosion occurs the fission products will be harmlessly confined inside the container. Of course, one must be quite sure that the reactor is of such a type that it cannot produce an explosion great enough to burst the container and what is even more important one should be quite sure that the container is closed except when the reactor is shut down and completely safe. Often it may be best to build the reactor underground.

The safety of a reactor, of course, depends to a great extent on the use to which the reactor is put. In general a power station is less likely to give trouble than a moving power source. It is not probable that nuclear locomotives will ever be safe. In nuclear ships more room is available and more room permits more safety measures. But even so the safety of nuclear motors in ships will have to be considered particularly carefully because ships will have accidents in harbors.