Nuclear scientists at the Pierrelatte uranium enrichment plant in south-east France noticed a strange deficit in the amount of uranium-235 they were processing in June 1972. That’s a serious problem in a uranium enrichment plant where every gram of fissionable material has to be carefully accounted for.
The problem lay in the ratio of uranium isotopes in their samples. Natural uranium contains three isotopes, always in the same ratios: uranium-238 (99.2744 per cent), uranium-235 (0.7202 per cent) and uranium-234 (0.0054 per cent).
The problem was with the uranium-235 of which there was only 0.600 per cent.
Physicists soon traced the anomaly to the supply of uranium ore from Gabon in West Africa, which contained far less uranium-235 than the ore from anywhere else on the planet, a problem that caused some consternation among nuclear scientists.
So France’s top nuclear scientists began an investigation and, in the process, made one of the more remarkable discoveries in recent history.
This kind of depleted uranium is only found inside nuclear reactors, which burn uranium-235.
That set off a hunt for a reactor that could have produced this stuff.
On 25 September 1972, they announced that the depleted uranium had come from Gabon where nuclear scientists had discovered a 2 billion year-old nuclear reactor at the site of the Oklo uranium mines near a town called Franceville. This was a naturally occurring deposit of uranium where the concentration of uranium-235 had been high enough to trigger a self-sustaining nuclear reaction.
Today, say Edward Davis at Kuwait University and a couple of pals who review the scientific history of the discovery at Oklo, one of the most extraordinary natural phenomena on the planet.
Since its discovery, the Oklo reactor has been a significant driver of important research in nuclear physics. In particular, physicists have used it to study how buried nuclear waste might spread through the environment. And since the reactor began operating some 2 billion years ago, they’ve also used it to study how the universe’s fundamental constants may have changed during that time.
But the first puzzle that physicists had to deal with in 1972 was how a naturally-occurring reactor could work at all. Nuclear scientists well know that reactors do not work with natural uranium because the level of uranium-235 is too low at only 0.7202 per cent. Instead, the uranium-235 has to be enriched so that it is about 3.5 per cent of the total. So how did so much end up at Oklo?
The answer to this puzzle is that uranium-235 has a shorter half-life than other uranium isotopes and so would have been present in much higher quantities in the Earth’s distant past. When the Solar System was created, for example, about 17 per cent of uranium would have been the 235 isotope. That percentage has fallen steadily since then. When the ore in Gabon was laid down some 2 billion years ago, the concentration of uranium-235 would have been about 4 per cent, more than enough for a self-sustaining nuclear reaction.
The idea is that when a neutrons hits an atom of uranium-235, the atom splits producing two smaller nuclei and several neutrons. These neutrons go on to split other atoms in an ongoing chain reaction.
However, the liberated neutrons are high-energy particles that tend to fly away rapidly. So nuclear reactors usually contain a moderating material that slows down the neutrons so that they can interact with other uranium atoms.
It turns out that water is a reasonable neutron moderator. So an important component of this natural reactor was the presence of water seeping through the uranium ore. And this had an interesting impact on the way the reactors operated.
Nuclear scientists believe that the Oklo reactors operated in pulses. As water flowed into the rock, it moderated the neutrons, allowing a chain reaction to occur. But this increased the temperature of the rock, boiling the water into steam which escaped. When this happened, the neutrons were no longer able to interact with and split uranium nuclei and the chain reaction stopped. The rock then cooled allowing water to flow back in.
So the Oklo reactors operated in pulses. Today, nuclear scientists have calculated that the chain reaction probably lasted for 30 minutes and then switched off for about 2.5 hours, a pulsing process that continued for about 300,000 years.
While they were on, the reactors were powerful devices.
“The reactors likely operated under conditions similar to present day pressurized water reactor systems, with pressures about 150 atmospheres and temperatures of about 300 degrees C,” say Davis and co.
French nuclear scientists carried out a detailed survey of the Oklo site, discovering not just one reactor zone but up to 17 of them over an area of several tens of square kilometres. Some of these were close to the surface and so had been influenced by weathering processes while others were at depths of up to 400 metres and were more or less pristine.
In addition to the depleted uranium-235, these zones contained numerous fission fragments such as isotopes of zirconium, yttrium, neodymium and cerium. The unusual ratios of these isotopes was an important indicator of what had gone on there almost 2 billion years earlier.
The presence of these fission by-products immediately piqued the interest of nuclear scientists, particularly in the US. Perhaps the biggest challenge facing the nuclear industry is to find a way to deal with the highly radioactive waste that reactors produce. One idea is to bury it but that raises the question of what would happen to this waste over the millions of years during which it remains toxic.
The Oklo reactors were a natural test of this question. So US scientist, in particular, began a program to measure the way in which different fission products migrated away from the reactor zones.
“One of the most important, and surprising, early ﬁndings was that uranium and most of the rare earth elements did not experience signiﬁcant mobilization in the past two billion years,” say Davis and co.
“Because the wastes were contained successfully in Oklo, it appears not unrealistic to hope that long-term disposal in specially selected and engineered geological repositories can be successful.”
This evidence has since become one of the main arguments in favour of nuclear waste repositories such as the one planned at Yucca Mountain in Nevada.
Oklo has also become the focus of physicists studying the possibility that the universe’s fundamental constants may have changed over time. The reason that Oklo may be able to help is that it stopped operating over 1.5 billion years ago. So the nuclear processes that occurred at that time must’ve been governed by the fundamental constants as they were then.
In particular, physicists are interested in the fine structure constant which determines the strength of the electromagnetic interaction. This in turn determines the way neutrons are absorbed in chain reactions and consequently the yields of different fission products.
The focus of most research has been on the amount of samarium-149 produced by these natural reactors. The data places bounds on how much the constant may have changed in the past. The consensus is that the data is consistent with the fine structure constant being actually constant although it doesn’t rule out tiny changes. Davis and co point out that the Oklo data can also constrain changes in other constants, such as the ratio of light quark masses to the proton mass. To date, this work is consistent with these constants being constant.
The Oklo story ends with a damp squib. After a period of intense interest in the early 1970s, mining continued at Oklo and eventually all the natural reactors were mined out. The one exception was a shallow reactor zone at a place called Bangombé, some 30 kilometres from Oklo, although this has largely been washed out by ground water.
So these zones have been largely lost to science. That’s a shame. It also means that nuclear scientists are unlikely to get better data on natural nuclear reactors using the advanced techniques than those of available in the 1970s.
Cover Image, Black Rocks | Axel Rouvin