The Oklo Reactor — The Nuclear Power Plant That Nature Built Two Billion Years Ago

The ore from Gabon had the wrong isotope ratio. It was a tiny difference. It meant a nuclear reactor had been running here two billion years ago. All images generated by author unless stated

In May 1972, a physicist named Francis Perrin was examining a uranium ore sample at the Pierrelatte nuclear fuel processing plant in southern France. The sample had come from a mine in the Oklo region of Gabon in West Africa.

Something was wrong with it.

Uranium ore, wherever it is found on Earth, has a consistent isotopic composition. It contains approximately 99.3 percent uranium-238 and 0.7 percent uranium-235. This ratio is the same in ore from Australia, from Canada, from Kazakhstan and from uranium meteorites that have never been part of any planetary crust. It is a fixed feature of matter, set by the conditions of the nuclear reactions in the stellar remnants that produced the elements that make up the solar system. It does not change. Except at Oklo, it did.

The Gabon sample contained only 0.717 percent uranium-235, a tiny deviation, but in a number that never deviates. Perrin calculated what was missing and found that the deficit was large enough to account for the fissile material in several nuclear weapons of Hiroshima scale. The first question was straightforward: had someone stolen weapons grade material? Careful checks ruled out human tampering. The second question was harder. There was only one other explanation.

The uranium had already been burned, in a natural fission reactor, two billion years ago.

On 25 September 1972, the French Atomic Energy Commission announced their conclusion. Self-sustaining nuclear chain reactions had occurred on Earth approximately two billion years before the announcement. Nature had built a nuclear reactor more than two thousand million years before Enrico Fermi’s team achieved the first artificial one in Chicago in 1942.

Established: what the Oklo reactors were and how they worked

The Oklo discovery has been exhaustively investigated over the past five decades. Sixteen natural fission reactors have been found in three different ore deposits at the Oklo mine in Gabon, at Oklo, at Okelobondo and at Bangombe, all within 20 km of one another. Two reactors, designated RZ2 and RZ10, have been studied in particular detail through mass spectrometric analysis of borehole samples. The science is not contested. The Oklo phenomenon is as well established as any finding in nuclear physics.

Why it could happen then, and not now. The key is the ratio of uranium-235 to uranium-238. Today, uranium-235 comprises 0.7 percent of natural uranium, far too low a concentration to sustain an unmoderated natural chain reaction. But uranium-235 decays faster than uranium-238. Calculating backward, two billion years ago, the natural concentration of uranium-235 in uranium ore was approximately three percent, comparable to the enriched uranium used in modern light water reactors. At that concentration, under the right conditions, natural fission becomes physically possible.

The right conditions. Three requirements had to be met simultaneously at Oklo. First, a sufficiently high concentration of uranium ore. The Oklo deposit was exceptionally rich. Second, the absence of neutron absorbing materials that would have quenched the reaction before it began. The local geology was unusually pure. Third, and most critically: a neutron moderator. Free neutrons released by fission travel too fast to trigger further fissions efficiently; they need to be slowed. At Oklo, groundwater filled this role. Like in a man made light water nuclear reactor, the water acted as a moderator, absorbing the neutrons and controlling the chain reaction.

The self-regulation mechanism. The Oklo reactors did not simply run until the fuel was exhausted. They regulated themselves, cycling on and off over hundreds of thousands of years. Physicist Alex Meshik studied the trapped xenon isotopes within the ore samples and found the signature of this self-regulation: when the water moderator boiled off due to the heat of the reaction, the chain reaction slowed and eventually stopped. When the rock cooled and groundwater returned, the reaction resumed. The xenon isotope record shows this cycle repeating across the operating lifetime of the reactors, with individual cycles estimated at approximately 30 minutes on and several hours off.

The total energy output over the reactors’ approximately 200,000-year operating period is estimated at 100,000 megawatt-years. The fission products, the radioactive waste, remained remarkably confined within the ore body. Over two billion years, most of the fission products migrated less than 10 metres from their point of origin. This finding has significant implications for modern nuclear waste management and has been studied by nuclear engineers as evidence that geological containment of radioactive waste is achievable on very long timescales.

Water moderated the neutrons. When it boiled off from the heat of fission, the reaction slowed. When the rock cooled and the water returned, it started again. This cycle ran for approximately 200,000 years.

What the Oklo reactors prove about nuclear physics

The physical conditions at Oklo two billion years ago were different from today in one important way, the uranium-235 concentration, and identical in every other measurable way. The Oklo reactors operated on the same nuclear physics that governs modern reactors: the same cross-sections, the same decay chains and the same reaction products.

This has a profound implication that goes beyond the discovery itself.

The laws of physics have not changed. The Oklo reactors provide the most precise direct measurement ever achieved of whether the fundamental physical constants have varied over cosmological time. The argument runs as follows.

The production of specific fission products at Oklo, particularly samarium-149, a neutron absorbing isotope created in the fission chain, is exquisitely sensitive to the value of the fine structure constant, the dimensionless number that governs the strength of electromagnetic interactions and is one of the fundamental constants of nature. If the fine structure constant had been even slightly different two billion years ago, the neutron capture cross-section of samarium-149 would have been different and the abundance of samarium-149 in the Oklo ore would have been measurably different.

The samarium-149 abundance at Oklo has been measured precisely. Damour and Dyson, in a study conducted in the mid-1990s, concluded that the Oklo data on the absorption of neutrons by samarium-149 limits the change in the fine structure constant to less than 0.1 parts per million over the last 1.8 billion years. Subsequent analyses using improved reactor models have refined this to limits on the order of 10 to the minus 17 per year, effectively zero change, measured over two billion years.

This is not a small result. Some theories of dark energy predict that fundamental constants should vary over cosmological time. The Oklo data is one of the most powerful constraints against those theories. A natural nuclear reactor that ran two billion years ago has given physicists a measurement that no telescope or particle accelerator has been able to match.

The samarium-149 abundance at Oklo constrains the change in the fine structure constant to less than 0.1 parts per million over 1.8 billion years. Some theories of dark energy predict it should have changed. It didn’t.

Contested and open: what the Oklo reactors imply

The physics of Oklo is not contested. The broader implications are where the questions begin.

Could it happen elsewhere? The conditions that produced the Oklo reactors were specific: a particular uranium concentration, a particular geological chemistry, a particular groundwater regime and at a particular period in Earth’s history when uranium-235 concentrations were high enough for natural criticality. The question of whether similar conditions exist or existed elsewhere, on other planets, or in other periods of Earth’s history, is genuinely open. The current consensus is that natural nuclear reactors were probably possible only within a limited window of time: after enough uranium had been produced by stellar nucleosynthesis and concentrated in planetary crusts, but before the decay of uranium-235 reduced its concentration below the threshold for natural criticality. On Earth, that window ran from approximately three to one billion years ago. Whether any other worlds hit the same conditions is unknown.

The fine tuning question. The Oklo constraints on the fine structure constant feed into a broader philosophical and scientific debate about fine tuning: the observation that the fundamental constants of physics appear to be set to values that permit complex chemistry, nuclear stability and ultimately life. If the fine structure constant were significantly different, nuclear reactions would not produce the heavy elements that make planets and organisms possible. Oklo shows that the fine structure constant has not varied measurably over two billion years, which means whatever mechanism set it to its current value set it very stably. Whether this stability requires an explanation, and if so, what kind, is a question that sits at the intersection of physics, cosmology, and philosophy and is not resolvable by any current empirical method.

The nuclear waste implication. The Oklo fission products’ remarkable confinement over two billion years has been studied by nuclear engineers and geologists as a natural analogue for the long term behaviour of geological nuclear waste repositories. The finding that most fission products migrated less than ten metres from their origin over two billion years provides empirical evidence that the right geological conditions can contain radioactive material over timescales that dwarf any human institution. Whether this evidence translates directly to modern waste management decisions involves technical questions about geological specificity that Oklo cannot fully resolve.

What it means

The Oklo phenomenon sits at the intersection of several large questions. It demonstrates that nuclear fission is not a human invention but a natural process, one that occurs wherever the physical and geological conditions align. It provides the most precise long term measurement available of the stability of a fundamental constant of nature. And it offers an empirical precedent for the geological containment of radioactive waste that no human experiment could replicate.

None of these things was apparent from a routine uranium assay at a French processing plant in 1972. What was apparent was that a number that never changes had changed. The explanation that followed, that nature had been running a nuclear reactor underground in West Africa for 200,000 years, two billion years ago, before quietly shutting itself down when the fuel ran out, is one of the more extraordinary things the physical record of the Earth has revealed.

It is also entirely ordinary, in the deepest sense: it is what the laws of physics produce, given the right conditions and without any intervention. The same laws that Fermi’s team applied in Chicago in 1942 were operating in the Franceville Basin two billion years earlier. The physics was always there. Gabon just had the right rocks.

Sources: Francis Perrin, CEA announcement, 25 September 1972; Damour, T. and Dyson, F., The Oklo bound on the time variation of the fine structure constant revisited, Nuclear Physics B (1996); Meshik, A.P., The workings of an ancient nuclear reactor, Scientific American (2005); IAEA, Meet Oklo, the Earth’s Two-Billion-Year-Old Natural Nuclear Reactor (2025); Physics Review / Hachette Learning, Oklo: a natural nuclear reactor; Shlyakhter, A.I., Direct test of the constancy of fundamental nuclear constants, Nature (1976); Petrov et al., Oklo natural nuclear reactor and variation of fundamental constants, Physical Review C (2006).