Weekend Open Thread – The World’s First Nuclear Reactors

Most TAH readers likely know enough about science to know that nuclear reactors operate via a chain reaction involving nuclear fission. And most probably think that they know where the first nuclear reactor on earth was located.

Well, if you’re thinking that it was Enrico Fermi’s graphite pile at the University of Chicago – think again. Because that wasn’t the earth’s first nuclear reactor.

You see, Fermi was a latecomer to the reactor game – by somewhere around 1.7 billion years.

Yeah, you read that correctly. And yes, I’ll explain.

But first, a bit of a very-high-level overview of nuclear reactors.

. . .

A nuclear reactor operates by creating the conditions that allow a controlled self-sustaining fission chain reaction. (A nuclear warhead uses the same general mechanism, but by design the reaction is uncontrolled; the chain reaction in that case runs away until explosive disassembly due to the energy produced by the reaction destroys the warhead and its surroundings.)

Among the naturally-occurring elements, only the element uranium has naturally-occurring isotopes that are fissile.* So virtually all nuclear reactors depend on creating a self-sustaining, controllable chain reaction involving uranium. For uranium, the naturally-occurring fissile isotope of uranium is uranium-235.

For a self-sustaining chain reaction to occur, a critical mass must exist. Having a critical mass depends on a number of factors; among those factors are (1) the concentration of fissile isotopes in the nuclear fuel; (2) the amount of the fuel present; (3) the geometry of the reacting mass; (4) the density of the reacting mass; the (5) temperature of the reacting material, and (6) what other elements are also present in or surrounding the reacting mass.

As noted above, the naturally-occurring fissile isotope of uranium is uranium-235. Today it constitutes approximately 0.72% of natural uranium. Further, it fissions preferentially with neutrons at thermal energies (most neutrons initially produced by nuclear fission have substantially higher energies than those found in the thermal energy range). Other naturally-occurring uranium isotopes generally don’t fission when they absorb thermal neutrons. So in most reactor designs, neutrons have to be slowed (moderated) so that a significant fraction have average energies in the thermal range for a chain reaction to occur.

Water is a common coolant used in nuclear reactors. Because normal hydrogen is a mild neutron absorber regular water can’t easily be used today (if it can be used at all) with natural uranium to produce a chain reaction. (In contrast, “heavy” water, made via oxidizing deuterium vice normal hydrogen, can be so used; deuterium doesn’t readily absorb neutrons.) As a consequence, uranium used in power reactors cooled by normal water thus is typically enriched to around 3% uranium-235 content. The fuel is also typically rich in uranium oxides.

All isotopes of uranium are radioactive – that is, they decay to lighter elements over time. However, uranium-235 decays much more rapidly than the more common isotope of uranium, uranium-238; its half-life of 704 million years is only somewhat over 1/6 that of the half-life of uranium-238 (4.47 billion years). As a result, around 1.7 billion years ago, uranium-235 was relatively more plentiful; it constituted somewhat over 3% of naturally-occurring uranium.

OK, enough background. Now, on to the main story.

. . .

In 1972, a nuclear reprocessing plant in France noticed something peculiar. A sample of uranium ore from a particular mining area in Gabon – in the vicinity of the town of Oklo, to be precise – had an isotopic assay more like depleted uranium than natural uranium. Specifically, it was only about 0.6% uranium-235. Natural uranium has about 20% more uranium-235: 0.72%.

This raised a huge “red flag” – because French regulations for processing/reprocessing fissile materials required all fissionable material to be strictly accounted for lest it be diverted in small amounts over time and used to construct an illicit nuclear weapon. The low uranium-235 concentration in the ore sample therefore triggered an investigation by the French government’s nuclear regulatory agency.

What they found was amazing. At 16 locations in the Oklo mining area, uranium ore – which like reactor fuel is high in uranium oxides – showed markedly lower concentrations of uranium-235 than did natural uranium from anywhere else in the world. One sample showed a uranium-235 concentration of only 0.44% – or barely 61% of the amount of uranium-235 that should have been there.

The area hadn’t been previously mined before the current mining operations started, and it also wasn’t a dumping ground for depleted uranium. Further investigation also discovered clear evidence of fission decay products in the same ore in concentrations indicating that nuclear chain reactions had indeed occurred within the ore body.

But how?

It seems that around 1.7 billion years ago – when natural uranium had a uranium-235 concentration of roughly 3.1%, about the same as is used today in water-cooled nuclear reactors – groundwater permeated the uranium ore in question. This allowed water to slow (moderate) the neutrons naturally present in uranium due to spontaneous fission to thermal energies. The ore was unusually pure; at 3.1% uranium-235, it had a high enough concentration of uranium oxides to serve as reactor fuel in a reactor cooled by normal water. And in 16 places, the physical geometry and concentrations of other chemical elements within and around the ore body was right to allow a chain reaction to occur.

But there was no evidence of any explosion or other gross deformation due to runaway chain reaction. Why?

It turns out that a natural control mechanism also existed. In those places where geometry was right, after a time a self-sustaining chain reaction did occur. This chain reaction persisted for a time, increasing in power – and heating the ore body to a few hundred degrees Kelvin. This in turn boiled away the local groundwater, stopping the reaction. After the ore body cooled, the groundwater again permeated it; the cycle then repeated itself. This repetition continued until the ore in those regions with favorable geometry was depleted enough in uranium-235 (through it being consumed via fission) that a self-sustaining chain reaction was no longer possible. About 1.7 billion years later, those same regions produced the ore samples with anomalously low uranium-235 concentrations found in 1972.

The heat/boil dry/cool down/refill cycle was thought to have taken about 3 hours. The cycles are estimated to have continued over a period of a few hundred thousand years, and to have produced power at an average of somewhat less than 100kW thermal. Around five tons of uranium-235 is believed to have been consumed by the process.

In short: 1.7 billion years or so ago, for a few hundred thousand years those 16 locations IVO what is today Oklo, Gabon, were naturally occurring nuclear reactors. To date, no others have been found anywhere on earth.

The Japanese-American physicist Paul Kazuo Kuroda had theorized in 1956 that such a natural reactor could have been possible in the geological past. The Oklo findings proved him correct 16 years later.

. . .

So there you have it. Fermi indeed built the first man-made nuclear reactor on our planet. But when it comes to nuclear reactors, it appears Mother Nature beat him to the punch by about 1.7 billion years. (smile)

OK, enough nuclear trivia for today. Enjoy the WOT, everyone – and the weekend.

—–

* Author’s Note: technically, there aren two other naturally-occurring fissile isotopes. Uranium-238 can absorb neutrons, forming (briefly) uranium-239; uranium-235 occasionally spontaneously fissions, producing neutrons. Uranium-239 decays in short order (through a few intervening steps) to produce plutonium-239, which is indeed fissile. Through a similar process, thorium-232 absorbs neutrons produced from spontaneous fission of thorium and (after a short radioactive decay chain) produces uranium-233. Uranium-233 is also fissile.

However, both of these other two fissile isotopes only exist naturally in negligible trace quantities (both can, however, be produced in quantity in nuclear reactors). So for all practical purposes, the only naturally-ocurring fissile isotope of consequence is uranium-235.

Category: Historical, It's science!, Open thread