The First Nuclear Reactor on Earth

Eagerly, I pre-ordered Adam Higginbotham’s Midnight In Chernobyl [wc] which arrived last Friday. I’ll write a review of it in time but spoiler: it’s good. But he mentioned something, and jumped past it, and it immediately made me grab my ${internet_device} and start searching.

Here is what Higginbotham wrote:

The simplest form of nuclear reactor requires no equipment at all. If the right quantity of uranium 235 is gathered in the presence of a neutron moderator – water, for example, or graphite, which slows down the movement of the uranium neutrons so they can strike one another – a self-sustaining chain reaction will begin, releasing molecular energy as heat. The ideal combination of circumstances required for such an event – a criticality – has even aligned spontaneously in nature: in ancient subterranean deposits of uranium found in the African nation of Gabon, where groundwater acted as a moderator. There, self-sustaining chain reactions began underground two billion years ago, producing modest quantities of heat energy – an average of around 100 kilowatts, or enough to light a thousand lightbulbs – and continued intermittently for as long as a million years, until the available water was finally boiled away by the heat of fission.

I had never heard of this before! Well, it was pre-dinosaur so, if an underground nuclear reactor goes critical and there’s nobody around to see it, does it make any sound?

There’s much more detail in a 2005 article in Scientific American [sci] and Ethan Seigel has a typically Ethan, highly approachable, write-up. [es]

------ divider ------

The Scientific American article goes into how the researchers were able to determine what happened, based on the isotopes and their half-lives – it’s pretty cool! Nuclear forensics is another topic that interests me greatly. According to Rhodes and Stillman, it’s possible to sample the results of an explosion (or melt-down) and tell nearly everything about it based on the quantity and type of isotopes it contains. In an aside in Twilight of the Bombs, Rhodes mentions that the idea of “a nuclear weapon of unknown origin” is absurd – they can tell which centrifuge cascade refined what plutonium and what breeder reactor it came from as well as which process run. But try explaining something like that to one of your former friends who’s read and believed that nonsense about some unknown power doing an EMP airburst to throw the US back to the dark ages…


  1. Owlmirror says

    Also, the Oklo reactor made it into the list of creationist claims:

    Claim CF210:
    Radiometric dating assumes that radioisotope decay rates are constant, but this assumption is not supported. All processes in nature vary according to different factors, and we should not expect radioactivity to be different.
    [ . . . ]
    The Oklo reactor was the site of a natural nuclear reaction 1,800 million years ago. The fine structure constant affects neutron capture rates, which can be measured from the reactor’s products. These measurements show no detectable change in the fine structure constant and neutron capture for almost two billion years (Fujii et al. 2000; Shlyakhter 1976).
    [ . . . ]
    Fujii, Yasunori et al., 2000. The nuclear interaction at Oklo 2 billion years ago. Nuclear Physics B 573: 377-401.
    [ . . . ]
    Shlyakhter, A. I., 1976. Direct test of the constancy of fundamental nuclear constants. Nature 264: 340.
    [The pdf link in the original fails; this works]

  2. Jazzlet says

    This was talked about in one of the geology courses I took way back when, probably the university one, not the school one. It is cool!

  3. says

    I read the original article at the time, but don’t really remember anything other than that it was a uranium deposit where water flowed in & out of the (cracks? Fissures? Fossae?) in the deposit and that this ebb and flow affected the reaction.

    i don’t have enough physics to really understand what’s going on here, but it seems you’re saying that the normal process I think of as a chain reaction – an emitted neutron smashes into another nucleus which then emits more than one neutron, thus not only perpetuating but accelerating the reaction until some peak is reached where additional neutrons do not speed up the reaction because the area is already saturated with neutrons – doesn’t normally happen in pure U-235? Instead, an interaction with water or graphite has to slow an emitted neutron down enough for it to fracture a nearby U-235 nucleus?

    Again, I have no physics at this level. Think of me as barely past F=MA. Given that, is it possible to answer a couple questions?

    1) What is the force that slows the neutrons as they travel through water?
    2) Why don’t faster neutrons blow apart U-235 nuclei?
    3) I assuming this must be an effect related to quantum-something (chromodynamics? mechanics?)?
    4) Without the special (probably quantum) effect, the intuition from observing macro-scale interactions would be that a projectile has to have sufficient speed (really, energy) to shatter the target, but that additional speed doesn’t prevent shattering the target. If anything it shatters the target more effectively (or into more pieces, if you call that “effective”). So… at the quantum level is it that the mere fact of having too much energy prevents certain interactions?
    5) Does this have anything to do with “quantum tunneling” which also requires a minimum energy threshold?
    6) So is it at least somewhat accurate to say that adding energy to a neutron is required to boost it above a threshold necessary to shatter a U-235 nucleus, and that adding more does not prevent a normal chain reaction up to some other threshold energy that then prevents the collision/interaction?
    7) So is it at least somewhat accurate to say that there’s a “goldilocks energy” for these reactions – if the energy is too low, the neutron can’t add enough energy to cause the fission of another nucleus, leading to fizzle, but if the energy is too high, the neutron somehow fails to interact with other nearby nuclei?

    and, finally,
    8) Is the CIA going to waterboard me for asking these questions?

  4. rq says

    Oh wow, this is interesting! Chernobyl has always had a weird fascination for me, due to that time-honoured tradition of humans liking dates that have personal significance – in this case, it happened on or about my birthday (I’ve read various narratives and sometimes there’s a small disparity in dates). Both books sound fascinating (ha, all books do!). But that natural nuclear reaction? Awesome to learn about. This is why I believe divine intervention is a boring, cheap way out – when things like this happen randomly on their own, it’s just so much cooler.

  5. Reginald Selkirk says

    RE Crip Dyke, Right Reverend Feminist FuckToy of Death & Her Handmaiden #4:

    Slow and fast neutrons

    Yes, the energy (speed) of the neutrons is key. If you want an analogy, think of absorbance and fluorescence of light. An incoming photon has a certain energy to give. An absorbent or fluorescent material has a certain energy it is capable of absorbing. The fluorescent material has a certain energy of light it is capable of re-emitting.

  6. says

    The detailed analysis of how and why the RBMK at Chernobyl blew up is really fascinating. It’s interesting that so much of the control of these incredible powers is manual. And reading about the effects of the disaster simply makes me gawp in disbelief: 100-foot diameter concrete blocks 7 feet thick flying into the air like poker chips…

    I have been to Chernobyl and I had some fascinating and scary experiences there. I wrote about them on my old website before I started blogging here. I’ve always been fascinated by the idea of having a reaction that’s on the verge of an explosion – and keeping it there.

    So, to Crip Dyke’s question: the chain reaction takes place nearly instantly in a weapon, but in a reactor there is a mediator that slows it down by absorbing neutrons and keeping the reaction going without it blowing apart, or cooling down. The specifications of the reactions are unimaginable – temperatures as hot as 6000C in melted zirconium/uranium puddles – they just eat right through anything.

    8) Is the CIA going to waterboard me for asking these questions?

    There’s not much secret about this stuff – it’s all “merely” extreme edge-case engineering. As Feynman said “the secret is that it works at all.” A lot of details are published. Coincidentally my copy has not arrived yet but there is an old US government publication that is basically “roadmap for building a nuclear reactor.”

  7. cvoinescu says

    Crip Dyke @ #4:
    I am not a physicist. This is my understanding:

    The way neutron moderators work is mostly classical physics. It’s both the oxygen and the hydrogen nuclei in water that slow neutrons down, but hydrogen does most of the work, because its mass is the same as that of the neutron. Imagine shooting a very fast billiard ball on a large table with widely spaced, slowly moving balls on it. If the fast ball hits a ball of about the same size head-on, it transfers practically all of its energy to it and remains nearly stationary. If it’s more of a glancing impact, it transfers only some of the energy, but it is slowed down quite a bit, on average. After a dozen collisions, it’ll be quite slow, while some of the initially stationary balls will have accelerated. In time, they’ll collide in turn, and so on, so the energy will disperse. Now, if the fast ball hits a bowling ball, it just bounces off with almost all of its energy remaining, and the bowling ball is affected only slightly. Assuming perfectly elastic collisions, which is the case with particles but not with billiard and bowling balls, it’ll take hundreds of collisions for the fast ball to slow down significantly.

    The actual picture is a bit more complicated, because the apparent “diameter” of the balls (nuclei), i.e., the probability of a collision with the neutron, is not a classical property, and doesn’t vary in the way you’d expect with the mass of the nucleus. Also, some of the balls can sometimes absorb your fast ball — it sticks instead of bouncing. Regular water would be a perfect moderator if it wasn’t for this. Heavy water already has extra neutrons in its hydrogen nuclei, so it is much less likely to absorb them. This makes it attractive as moderator in designs that need really slow neutrons (which would need to survive twenty collisions or so).

    For your 2-7, it’s helpful to think of it not as the neutron cracking the nucleus, but rather as the neutron needing to stick to the nucleus long enough for the nucleus to crack by itself. If the neutron is too fast, it simply bounces off. You’re trying to win the big plush toy by throwing the ball so that it stays in the bucket, not to break the bucket by throwing the ball fast enough.

    In reality, some (very few) nuclei will break even with fast neutrons, but most bounce, so you need loads of collisions with fast neutrons to keep the chain reaction going. Your fuel needs to be tightly packed to give the neutrons the opportunity to hit hundreds of fuel nuclei before being absorbed or lost to the outside of the reactor. With slow neutrons, the fuel can be spaced widely, such as in the leisurely arrangement of the CANDU reactors. This makes engineering easier (for one, widely spaced fuel is easier to keep cool). At the other end of the spectrum, bombs have no neutron moderation at all. They just need a solid ball of “fuel” of a minimum size (a critical mass) to exist for a brief time. This can be made smaller by surrounding it with a neutron reflector (e.g. beryllium). The secret sauce is in “assembling” this ball before it melts itself down — a job for precisely calibrated conventional high explosives, as unlikely as that sounds.

    8) Probably not.

  8. DonDueed says

    It’s also important to note that the fuel in a power reactor is very different from that in a nuclear weapon.

    All natural uranium is a mixture of two isotopes. Over 99% is U238, which is not fissile (except under very extreme conditions). The rest is U235, the explodey stuff. To make a bomb you have to separate and concentrate the U235 to 80% or higher, which these days is mostly done in the centrifuges that you may remember we accused Saddam of trying to produce.

    Power reactors use somewhat enriched uranium, around 7% U235 or so. They cannot explode like a nuclear bomb — even Chernobyl didn’t, but even so a large reactor can produce a very massive power surge, and that’s what happened at Chernobyl. That disaster was entirely due to human error (a whole shitstorm of them, actually), not any breakdown of the reactor itself; but its design was an inherently dangerous one compared to those used in the US.

    It is also possible to make a reactor out of natural uranium, but that requires pretty massive amounts of U and a moderator. The US did this using graphite moderators to cook up the plutonium for the Fat Man style bombs (and their descendants). That’s also what happened in the very unusual case of the Gabon reactor.

    By the way, the quoted article above is full of technical errors. “Molecular energy” — no. Just no. I hope it’s a result of poor translation or something. If not, I wouldn’t trust a word it says, at least where technical details are concerned.

  9. lochaber says

    I’ve heard of this previously. Unfortunately, it was from someone who produced it as “proof” that aliens previously inhabited earth, and built the pyramids and banged apes or some such nonsense.
    (I did look up the event/details, and it’s fascinating, I just wish I had the sense at the time to stop talking to that person…)

  10. voyager says

    Fascinating stuff. I think I understand the basics so I hope I don’t get too lost once I’m down this rabbit hole.

  11. cvoinescu says

    One of the articles says that the energy spectrum of neutrinos released by uranium fission was thought to be a signature of a technological civilization (no known natural processes emit the same kind of neutrinos). They say that the discovery of the Oklo natural reactors invalidates that.

    I argue that that’s not the case, because our reactors are continuous and Oklo worked on a 2-hour on-off cycle. Also, not sure if this is generally the case or specific to the circumstances at Oklo, but it had to wait a couple billion years from the formation of Earth for groundwater to become oxygenated, before it could dissolve and concentrate the uranium deposits to the point where the reaction could start. While not a sign of intelligent life, free oxygen, required for an Oklo-type reactor, is a sign of life nonetheless.

  12. says

    Eagerly, I pre-ordered Adam Higginbotham’s Midnight In Chernobyl [wc] which arrived last Friday. I’ll write a review of it in time but spoiler: it’s good.

    Today I finished reading this book. I agree with your assessment: it’s good. Thanks for the excellent book recommendation!

    There could be lots of things that particularly caught my attention in this book, so I’ll only mention a few.

    One of the underlying themes in this book was people who were in charge and knew about the dangers of radiation intentionally subjecting their workers/citizens to it. This trend was pretty much all over the place. And it was sickening. The example that particularly caught my attention was the radium girl story, because I’m an artist, and I cannot avoid cringing about the sheer stupidity of telling people to lick paintbrushes. You just never lick paintbrushes, you shape their tips with a wet paper towel instead. And, no, that’s not more time consuming than licking the brush, because an artist can learn to perform this simple motion quickly. Historically, artists used various very toxic pigments that are no longer commonly used (copper acetoarsenite, various lead-based pigments, mercuric sulfide). Even with the most toxic pigments being heavily restricted and hard to obtain for the modern artist, it’s still an extremely stupid idea to lick paintbrushes. Especially if one uses cadmium or cobalt pigments (as I do). You just don’t lick paintbrushes, that’s stupid.

    The other underlying theme was the sheer stupidity of people who routinely rushed and cut corners when building nuclear facilities. Words cannot even express how stupid it is—you just don’t cut corners on something that can kill millions of people if you are a sane person. This one, by the way, really pisses me off. Politicians on both sides of the Iron Curtain rushed their nuclear projects for their idiotic arms race. None of the citizens had consented to that. Hell, they weren’t even informed about the risks. Cutting corners on something like space exploration wouldn’t piss me off so much—if shit happened, only a few astronauts would die, and they were people who had made at least a somewhat informed decision to take a risk. But when politicians are risking my life without even telling me what they are doing, now that’s infuriating. Reading about how nuclear reactors were built in USSR and also other countries makes me wonder how comes that humanity still hasn’t destroyed itself. There must be lots of luck involved. Then again, we are about to destroy ourselves by producing CO2, so probably at the end it will make little difference how exactly we annihilate ourselves. Then again, I’m probably unfair to those engineers who hurried to build nuclear facilities quickly and ended up cutting corners for that—there’s hindsight bias for me, and also optimism bias for the people who rushed to meet their deadlines and hoped for the best.

    Another interesting aspect was the comparison between how countries on both sides of the Iron Curtain handled the same things. After reading about how those in charge of Pripyat delayed informing their citizens about radiation, I was happy to read that people overseeing the Forsmark nuclear power plant in Sweden informed locals to stay away from the reactor almost immediately after having detected the radiation that came from Chernobyl and initially assuming that there might be a problem at Forsmark. Propaganda, on the other hand, was awful on both sides. It was almost funny to read about how the Western press published rumors about what might have happened in Chernobyl—journalists wrote about an unconfirmed ridiculously high death toll and pretty much invented their “facts.” That was propaganda at its worst. For the first part of the book I was pissed off about the Soviet press with it’s bullshit, so it was sad to read that the Western version wasn’t any better.

    By the way, one more thing, in this book there was one story missing—the one that generally appears whenever Latvian press writes about Chernobyl liquidators. The book focused on “important” people, those who were in charge of things or did notable jobs. Of course, the book also mentions all those unwilling army conscripts who were used as a cannon fodder and ordered to shovel nuclear waste, especially on the roof, but their individual stories aren’t carefully examined or given much attention, and none of them is mentioned by name. The narrative and experiences of these people actually differs from those with the important jobs. By the way, Latvians who got sent to Chernobyl were mostly the nameless cannon fodder types, this is why Latvian press talks about them.

    Anyway, imagine there’s some person who is already married and has some children. He has an important job at Chernobyl where he manages things and orders others around. He has already started to build a successful career. If he survives the ordeal with at least semi decent health, he can actually use his Chernobyl experience as a selling point in his CV, he can use it to further his career. The book featured several people who experienced exactly this kind of fate and mentioned their later successful careers. Of course, for fairness sake, I have to mention that the book also talked about people who did “important jobs” and died afterwards or experienced mental health issues or drunk themselves to death.

    Still, the story that’s missing is that of the nameless army conscripts who were the cannon fodder, the living robots sent to the rooftop. Imagine a boy in his late teens who is sent to Chernobyl as a cannon fodder to shovel around radioactive debris. He is still so young that he has no career, no wife, no children. He returns back home from Chernobyl with his health ruined. His life is pretty much over before it even started. Making a career is pretty damn hard when your health sucks. Most women are reluctant to marry somebody who would highly likely give them sick babies. Those liquidators who did choose to have children after Chernobyl later had to deal with their children’s health problems on top of their own. Those were ruined lives full of misery. These people got no awards or recognition or fame, no book deals or options to travel around the world giving interviews. Instead they had to fight to obtain health care. (Countries, unwilling to pay for their medical expenses, routinely pretended that their illnesses had no relation to Chernobyl radiation.) Many of these people died of cancers after living in misery for some years. They also lived in poverty unable to work or have careers. I really felt like these kinds of stories were missing in the book. The author interviewed people with the important jobs and didn’t mention any personal stories of people who were the cannon fodder and living robots. In this book they remained nameless, only mentioned as a group, with no prominently featured personal stories, even no names were given, they were treated as a statistic.

Leave a Reply