The Moderator

Field Note: neutron moderation — the art of making violence gentle enough to be useful.

You cannot split uranium with a neutron traveling at fifteen thousand kilometers per second. The nucleus simply does not care. It is like throwing a baseball at a wall and expecting it to crack. The neutron bounces off, or worse, passes straight through — uranium-235 presents a cross-section so small at high energies that the neutron has better luck finding the needle in the cosmic haystack.

But slow the neutron down, and everything changes.

At thermal energies — roughly 0.025 electron volts, corresponding to a neutron moving at about 2,200 meters per second, or roughly Mach 6 if you prefer to think in human terms — the uranium-235 nucleus becomes an irresistible target. The probability of fission, the cross-section, jumps from something like one barn (one times ten to the negative twenty-four square centimeters) to over five hundred barns. Five hundred times more likely to catch the neutron. Five hundred times more likely to split.

This is the miracle of moderation. It turns an indifference into an obsession.

How moderation works

The physics is elegant in its simplicity. A fast neutron collides with a light atomic nucleus. Momentum transfers. The neutron loses energy. Repeat. Repeat. Repeat.

The key insight is that the moderator atoms must be light. A neutron colliding with a heavy nucleus — say, lead — is like a ping-pong ball hitting a bowling ball. The ping-pong ball bounces back with almost all its energy intact. The bowling ball doesn't move. Not much useful slowing happens.

But a neutron colliding with a light nucleus — hydrogen, carbon, deuterium — is like two ping-pong balls colliding. Energy transfers back and forth. The neutron loses a significant fraction of its energy with each collision. Lighter is better.

Hydrogen is the lightest element. Ordinary water — H₂O, the stuff you drink — is therefore the most efficient moderator per collision. And it is abundant, cheap, and convenient. This is why the most common reactor design in the world, the light water reactor, uses ordinary water as both moderator and coolant. Two functions, one substance. Elegant engineering.

But water has a problem: it also absorbs neutrons. Not much — hydrogen's absorption cross-section is small — but enough that light water reactors need enriched uranium. The natural concentration of uranium-235 (about 0.7 percent) is insufficient when water is stealing some of the slowed neutrons before they can cause fission. You need more than 3 percent U-235. You enrich it. That is expensive. That is political. That is why the light water reactor design, for all its elegance, carries the burden of the enrichment industry.

Alternatives

Heavy water (D₂O) uses deuterium — hydrogen with a neutron added — as the moderator. Deuterium is just as effective at slowing neutrons (nearly, anyway — the extra mass makes it slightly less efficient per collision) but absorbs dramatically fewer neutrons. The result is that heavy water reactors can run on natural uranium. No enrichment needed. CANDU reactors in Canada use this design. It is elegant in a different way. The penalty is cost: heavy water is expensive to produce. You have to separate deuterium from ordinary water through enormous distillation columns or electrochemical processes that consume vast quantities of electricity.

Graphite is another option. Carbon atoms are heavier than hydrogen but light enough to moderate effectively. Graphite was the moderator of choice for the first nuclear reactors, including the Chicago Pile-1 experiment that launched the nuclear age in 1942. The British Magnox reactors and the Russian RBMK reactors (the type at Chernobyl) used graphite. Graphite doesn't absorb many neutrons, and it doesn't need to be contained under high pressure like water does. The trade-off is that graphite is flammable. Graphite fires are not hypothetical. They are among the worst things that can happen in a reactor.

The moderator temperature coefficient

Here is where the physics gets interesting, and where safety is baked into the equations rather than bolted on as an afterthought.

When a moderator heats up, it generally becomes less dense. Water expands. Graphite expands slightly. The moderation becomes less efficient because there are fewer moderator atoms per unit volume for the neutrons to collide with. Fast neutrons escape the core without being slowed. The reaction slows down.

This is called a negative moderator temperature coefficient. It is, in the nuclear engineering lexicon, a very good thing. It means the reactor is self-stabilizing. If the reactor gets too hot, the moderator stops moderating as well, the reaction slows, the heat generation drops, and the reactor cools back down. The physics itself provides the safety.

Not all reactors have this property. Some designs — particularly those using fast neutrons without a moderator at all — have positive coefficients. In those designs, the reactor gets more reactive as it gets hotter. This is fundamentally unstable and requires active, engineered systems to prevent runaway. It is why the RBMK design, with its positive void coefficient (water turning to steam in the core made the reaction go faster instead of slower), is now rightly regarded as one of the most dangerous reactor designs ever built.

A good moderator makes a good reactor. A bad moderator makes a reactor that fights its own operators.