Nuclear energy 101: Inside the "black box" of power plants


This morning, I got an email from a BoingBoing reader, who is one of the many people worried about the damaged nuclear reactors at Fukushima, Japan. In one sentence, he managed to get right to heart of a big problem lurking behind the headlines today: "The extent of my knowledge on nuclear power plants is pretty much limited to what I've seen on The Simpsons".

For the vast majority of people, nuclear power is a black box technology. Radioactive stuff goes in. Electricity (and nuclear waste) comes out. Somewhere in there, we're aware that explosions and meltdowns can happen. Ninety-nine percent of the time, that set of information is enough to get by on. But, then, an emergency like this happens and, suddenly, keeping up-to-date on the news feels like you've walked in on the middle of a movie. Nobody pauses to catch you up on all the stuff you missed.

As I write this, it's still not clear how bad, or how big, the problems at the Fukushima Daiichi power plant will be. I don't know enough to speculate on that. I'm not sure anyone does. But I can give you a clearer picture of what's inside the black box. That way, whatever happens at Fukushima, you'll understand why it's happening, and what it means.

At a basic level, nuclear energy isn't all that different from fossil fuel energy. The process of generating electricity at a nuclear power plant is really all about making heat, just as it is at a coal-fired plant. Heat turns water to steam, steam moves turbines in the electric generator. The only difference is where the heat comes from--to get it, you can light coal on fire, or you can create a controlled nuclear fission reaction.

A fission reaction is a lot like a table filled with Jenga games, each stack of blocks standing close to another stack. Pull out the right block, and one Jenga stack will fall. As it does, it collapses into the surrounding stacks. As those stacks tumble, they crash into others. Nuclear fission works the same way--one unstable atom breaks apart, throwing off pieces of itself, which crash into nearby atoms and cause those to break apart, too.

Every time one of those atoms breaks apart, it releases a little heat. Multiply by millions of atoms, and you have enough heat to turn water into steam*.

In a Boiling Water Reactor, like the ones at Fukushima, water is pumped through the core—the central point where the actual fission reactions happen. Along the way, fission-produced heat boils the water, and the steam rises up and is captured to do the work of turning turbines.

In the Core

The core is the part that really matters today.

In the core of a nuclear reactor, you'll find fuel rods—tubes filled with elements whose atoms are unstable and prone to breaking apart and starting the Jenga-style chain reaction.

Usually, the elements used are Uranium-238 or Uranium-235. They're refined and processed into little black pellets, about the size of your thumbnail, which are poured by the thousands, into long metal tubes. Bunches of tubes--each taller than a basketball player--are grouped together into square frames. These tall, skinny columns are the fuel assemblies.

The fission reactions that happen are all about proximity. In a fuel rod, lots of uranium atoms can crash into each other as they break apart. Pack the fuel rod into an assembly, and lots more atoms can affect one another—which means the reactions can release more energy. Put several fuel assemblies into the core of a nuclear reactor, and the amount of energy released gets even higher.

Proximity is also what makes the difference between a nuclear bomb, and the controlled fission reaction in a power plant. In the bomb, the reactions happen—and the energy is released—very quickly. In the power plant, that process is slowed down by control rods. These work like putting a piece of cardboard between two Jenga towers. The first tower falls, but it hits a barrier instead of the next tower. Of all the atoms that could be split, only a few are allowed to actually do it. And, instead of an explosion, you end up with a manageable amount of heat energy, which can be used to boil water.

In Case of Emergency

Now that you understand what's going on inside a nuclear reactor, you get a good idea of what happened at Fukushima. Like the other nuclear power plants in Japan, Fukushima Daiichi got a message from the country's earthquake warning system, and shut down in advance of the quake. Basically, that means that control rods—"big metal gizmos", as Charles Forsberg, executive director of the MIT Nuclear Fuel Cycle Project, described them to me—were inserted into the fuel assemblies, cutting the fuel rods off from one another. But, because you aren't completely separating all the uranium atoms from one another, shutting down the core isn't the same thing as flipping an "off" switch.

When a reactor core is shut down, its energy output drops not to zero, but about 6% of its normal output, Forsberg told me. The reactions grind to a halt over the next few days, as the falling Jenga towers run out of other towers they can actually hit. In the meantime, atoms keep breaking apart, releasing both heat and fast-moving particles that can penetrate human skin and damage our cells. Because of this, every nuclear reactor has ways of getting rid of the heat, and blocking those fast-moving radioactive particles.

When the reactor at Fukushima shut down, it should have been kept cool by water pumped through the core. But, because the tsunami damaged the diesel-powered generators that pumped the water, the core kept heating up. If that sounds like a design flaw, you're right. The Fukushima reactors were built in the early 1970s. In modern nuclear reactor designs, pumps aren't necessary to move water through the core in an emergency shut down. Instead, the water moves via gravity.

But, in this case, no pumps meant no water movement. So the core got hotter, which boiled off some of the water. The boiling caused pressure in the core to increase. To protect the core, and prevent a bigger problem, authorities had to vent some of that steam into the atmosphere, which means venting some of the radioactive particles along with them.

This is also probably tied into the explosion that happened, according to MIT's Charles Forsberg.

"There's zirconium in the fuel rods. When you overheat the reactor core, the first thing that happens is that the zirconium begins to react with steam or water and forms zirconium oxide and hydrogen," he says. "You get a mixture of steam and hydrogen. When you release steam into a secondary building [to decrease pressure in the core], the steam condenses and leaves behind just the hydrogen. Then all you need is an ignition source and you can get a hydrogen burn. That's what happened at Three Mile Island. I don't know if that's what happened in Japan, but it's likely to be the source of that explosion."

The good news is that the explosion seems to have happened outside the core. In that case, it's completely reasonable that an explosion could happen without releasing lots of radioactive material. That's because nuclear power plants come in layers, like an onion.

The core is contained within a building that has solid concrete walls, 3-to-6 feet thick. It's meant to withstand collision with an airplane. It's also meant to withstand an explosion from inside. But that bunker sits inside something a lot flimsier—a building more akin to a metal shed. It's the shed that exploded today at Fukushima. Because radiation levels didn't rise after the explosion, we can be pretty certain that the bunker is still intact.

How To Win This

This is a serious emergency, but there are some good reasons to be hopeful. According to World Nuclear News and Reuters, there were seven reactors in Fukushima that were affected by the earthquake. Of those, four have access to outside power to run their water pumps and are stable. Three lost their power. Out of those three, one has steady levels of water. Only two have decreasing water levels. But, in recent hours, workers have started pumping in seawater to one of those. Hopefully, both can be stabilized. But it's hard to say right now.

And then what happens? Remember, this is really just an emergency shutdown gone awry. The control rods are still in place. The Jenga columns are still separated. So, over time, the fission reactions will still slow down and stop. As they do, heat levels will drop, and so will levels of radiation.

Really, what we have here is a waiting game. The goal is to keep the reactors stabilized long enough that the shutdown can completely shut down.

For more information—and details I might have missed—I recommend checking out a recent BBC article, and an interview Skepchick blogger Evelyn did with her father, a nuclear engineer.

*If you want to understand thy physics of nuclear energy in more detail, I'd recommend reading Marcus Chown's The Matchbox That Ate a 40 Ton Truck

Image: Japan Ministry of Land, Infrastructure and Transport. AirPhoto


  1. Ok this may sound like besides the point but I saw a show on TV (I think you guys at BB recommended it) about what would happen if all humans disappeared. Being a fan of post-apocalyptic fiction I watched and was shocked that according to them all modern nuclear reactors would have melt-downs if left unattended. I sort of assumed they would have like a fail safe switch or something that made the water pump in and the safety rods to go into action and in some way permamently contain the reactor and fuel rods.

    (I have actually been worried about this. Not psychoticly worried, I dont believe all humans will disappear or anything but I’ve actually have had nightmares of being the only one left having to rush around the world switching off power plants before they blow up – so info about this would be great :))

    1. Not true, thankfully. Like Maggie said, these are a 1970’s design. Chernobyl in 1986 made everyone fundamentally change their plant design so that in the event of a failure, or a problem, the system cools itself. Since then, Reactors are made to always run on the ‘cool’ side of their reactions and the human intervention is to keep them ticking over. This is very important to consider when talking about say, new nuclear power plants being built – for example nobody would but a boiling water reactor these days.

      Like Maggie said the coolant is ideally gravity-fed rather than pumped, so if anything goes wrong the coolant still flows down and it all works. From what I have heard, the company involved with this plant had (compared to the rest of Japan) an interesting safety record, but the important point is that the reactor is contained. Most of the attempts earlier were to salvage the plant as a working facility. I think they have clearly shifted to making sure that it cools down and then writing off the facility, in which case their only concern is that it doesn’t breach the 6 foot dome – and that’s not likely even now. The containment dome is the real heavy engineering here, and whilst I empathise with a lot of people’s deep concerns the most I can say is ‘trust me’ when I say it’s looking good. I only wish the same could be said for the rest of Japan.

    2. What does it matter if they all melt down when all humans disappear? We’re all gone! Earth will survive, it’s our species that won’t.

    3. @osmo Never assume anything. Also with particular caution against assuming that any significant level of intelligence has been applied to any particular complex problem involving corporations and government of any flavour.

  2. Hi Maggie,

    Just a clarification:
    You write “Proximity is also what makes the difference between a nuclear bomb, and the controlled fission reaction in a power plant.” This is true in a sense, but may be misleading. No matter how close nuclear fuel rods get to one another there will never be a _nuclear explosion_. Uranium rods used in reactors are either natural uranium, or enriched to about 3% fissile U_235. For a bomb, you’d need far more enriched uranium (~90%).

    So the worry isn’t about a nuclear explosion–like Hiroshima–but about a conventional explosion–like hydrogen igniting–that brings radioactive isotopes with it.


  3. Hey Maggie, this is kinda weird. I was thinking of posting a comment on an earlier article asking if anybody had a good link to a rundown on the basics of nuclear energy, because I never understood how it could be so dangerous if it was essentially just a case of creating steam. I read Purple America last year and got a sense of the scale of the cooling mechanisms, but was hoping for something basic-but-not-too-basic. You rock.

  4. Do they use the term ‘The Brazil Syndrome’ as it looks like the antipode (opposite side of world) is just off the coast of Brazil?

  5. I wasn’t very worried in the first place but nonetheless it was still informative and reassuring to read through your article.

  6. Very good “layman” explanation, Maggie. It’s not often I see one that’s simplified without being overly-distilled.

    One thing that bears adding is the importance of the water, as it’s not just a coolant. It is integral to the reaction. The neutrons released in U-235 fission are too energetic to sustain a chain reaction. An analogy would be comparing firing a bullet through a piece of glass vs. throwing a rock at it (and yes, analogies stink, but that’s as close as I can get without breaking out the nuclear physics vocabulary). The water is full of hydrogen, which slows down the neutrons to a level more likely to produce fission. Without it, the reaction slows to a level sustained by natural fission, giving off a small amount of heat that varies with the level of enrichment. It’s this residual heat that must be removed, else one winds up with a damaged core or potentially a pile of molten slag in the bottom of the reactor vessel.

    In either case, there’s not really any danger of an explosion, since the reaction will only run at the normal rate of decay once the geometry of the fuel cells is destroyed.

    A more reasonable worry would be that the integrity of the cooling system may be compromised; leading to release of any “stuff” in the coolant into the environment. Fission reactions produce some rather nasty short half-life elements that we’d all like to see remain safely contained in the core.

  7. Minor correction. Once the rods are in, the fission reactions almost immediately drop to zero. But there are radioactive fission products still in the core. It takes days for these to decay to longer life isotopes, and that radioactive decay generates a lot of heat. It’s a substantial part of the normal heat generation within a nuclear reactor. There is nothing that can be done to stop this heat generation. You can only keep it cool and wait.

  8. There is also direct radiolysis of water as a source of hydrogen in a BWR. So even if it was a hydrogen/oxygen explosion at Fukujima and not just explosive release of steam from some secondary system, we can’t use that as evidence that the core did actually melt. Most likely we’ll have to wait for quite a while until we have a conclusive report about what really went on in the last couple of hours.

    Though it looks like
    1) The situation is now under control, and will further improve over the next couple of days.
    2) The accident released a significant amount of radioactive isotopes, but not enough for widespread contamination.
    3) The vast majority of the radioactive isotopes remain contained in the reactor core.

  9. Watching CNN now, they are saying that one of the Fukushima reactors may be melting down now.

    1. Maybe so, but “containment” is the key word.

      Tragically, the Japanese people have had a somewhat greater experience of radiation and its effects than others.

      i truly and sincerely hope that that experience may now, in some way, help them to deal effectively with these events.

    2. Looks like they elected to use the sea to cool the Fukushima nuclear power plant:

      Japanese officials took the extraordinary step on Saturday of flooding a crippled nuclear reactor with seawater in a last-ditch effort to avoid a nuclear meltdown…


      Although officials said the leaks did not pose a major health risk, they also told the International Atomic Energy Agency that they were making preparations to distribute iodine, which helps protect the thyroid gland from radiation exposure, to people living near Daiichi and a second nuclear plant that suffered damage in the quake, called Daini, [emphasis mine] about 10 miles away.

      Oh dear.

  10. And to put that in some more perspective…an atomic bomb would be like placing all the Jenga towers right next to each other and then hitting them with a wrecking ball. (Or starting their falls with a brick of C4.)

    That would be impressive to say the least.

    Then you step up to a hydrogen bomb, which would replace the normal Jenga tower with one made of bricks of pure nitroglycerin. (Not that a hydrogen bomb is substantially different than a standard atomic bomb, but there is a different set of reactions that take place and increase your “power” output.)

  11. To Osmo, Actually unattended nuclear power plants, assuming they don’t get flooded like the Japanese reactor, will eventually just shut themselves down. A couple of things happen.

    First, as the water heats up, it becomes less dense. This means that fewer neutrons will slow down (the water is called a neutron moderator). If the neutrons don’t slow down, they wont cause the fissions to occur. Instead they fly right through/buy Uranium atoms and other fissionable material (Uranium 238 is fissile, meaning that once it captures a neutron and goes through a beta decay process, becomes plutonium 239 which is fissionable and will split and give off energy when split with a moderated (slowed down) neutron.)

    This is referred to as a “negative temperature coefficient” all commercial nuclear power plants are designed with a negative temp coeff (as temp goes up, power goes down)

    Another self limiting effect is the build up of Neutron “poisons” When a Uranium-235 (or PU-239) atom fissions, it gives off fission products and energy (heat). The fission products are other elements such as Xenon or Iodine (smaller atoms) The sum of their weights and the mass equivalent of the energy (e=mc^2) equal the total weight of the U/PU atom). These fission products do not fission when struck with a neutron, instead they absorb neutrons – hence the term neutron poison.

    Also, under normal conditions, when a reactor shuts down, the control rods (made of Boron – another Neutron poison) either fall into or are injected into the reactor core (if the BWR in Japan is like the GE designed BWRs in the US, the control rods are driven in from the bottom of the core).

    The core meltdown, while possible, under normal shutdown conditions is unlikely because cooling is provided to the reactor core to carry away the decay heat (heat generated from the decay of radioactive isotopes and any stray fission reactions.)

    BS/MS Nuclear Engineering (non Practicing, now I install solar systems so the only nuclear reaction I care about is 93 million miles away – and that’s fusion not fission)

  12. You say “Usually, the elements used are Uranium-238 or Uranium-235.” I think that this probably needs some clarification: Since U-238 is not fissile, that means that it can’t be used by itself as nuclear fuel, right? U-235 and Pu-239 are the most common nuclear fuels.

    1. Anon #23:

      More like an array of tubes.

      Anon #24: thye’re probably hard-wired to seismograph central, and know when they feel it.
      From what I’ve read, they immediately scrammed the 5 or 6 separate reactors (which I take to mean, fully inserted all control rods,thereby shutting down the fission reaction), but then the power supply for the coolant systems ( first on unit 1 and now unit 3 as well) failed or partially failed – from what I read, the result of the tsunami knocking out the diesel generators generating the power for the coolant system.

      So they have the reactors shut down and within containment, but dissipating the residual heat those reactors are continuing to emit is giving them problems, due to their difficulty in getting a enough coolant to flow.

      The key is to keep that heat draining away as it slowly subsides,to keep it from reaching a high enough temperature to “breach containment”, which temprature is around 3000 degrees C (iirc).

      Last I read, the thing was at about 750 degrees c.

      Oh wait here’s the source:

      “As of Friday afternoon, additional backup generators were en route to the plant, and unit 1’s coolant system was running temporarily on a battery. Japanese regulators stated that pressure in the reactor had risen to 1.5 times normal levels. At 750 degrees, an engineer familiar with the BWR design told the Los Angeles Times Friday, the temperature is well below the 2,200-degree design limit for preventing cladding failure.”


  13. “Like the other nuclear power plants in Japan, Fukushima Daiichi got a message from the country’s earthquake warning system, and shut down in advance of the quake.”

    Request for clarification: did they get warning before the _quake_, or before the _tsunami_? And if they did hear that the quake was coming, how much warning did they have, and how?

    1. Seismic waves travel much slower than light, so we can send warning of a quake after it starts, but before the damaging tremors arrive. The power plants are a bit over 100km away from the epicenter, so I wouldn’t be surprised if there were several tens of seconds of warning.

  14. Great explanation, but there’s one thing that is still unclear to me. You note that when the control rods are inserted, “you aren’t completely separating all the uranium atoms from one another” and the fuel rods continue to generate heat. This makes sense, but if uranium pellets packed together in a single fuel rod can generate dangerous levels of heat, how are the pellets transported to the reactor in the first place?

    1. The fuel before it has been in the core does not have fission fragments in it, it is not producing heat nor is itt very radioactive. The explanation here is very much simplified, it is actually the fission fragments, not the uranium which is producing heat and radiation.

    2. Believe it or not, the fuel assemblies are not very radioactive when they are first produced. In a new plant, when we do our initial fuel load out, we typically have to introduce a neutron source into the reactor in order to get the chain reaction going.

      As stated earlier, without a moderator, the reaction in the fuel cannot occur. In most commercial reactors, the moderator is water (the hydrogen atoms have approx. the same mass as nuetrons, so they are very efficient at slowing down, or ‘thermalizing,’ the neutrons produced from fission. Typically, when we store the rods in water (both before and after they have been in the core,” it is in borated water. boron is a nuclear poison, in that it absorbs neutrons so that they are not available to cause fission.

  15. The crux of the nuclear power safety issue is that the only way plant operators can purchase liability insurance coming close to the amount of damage that may result from operation is that governments “socializes” the liability by artificially limiting it. Refer to
    The fact no commerical insurer will fully underwrite an asset is the clearest business case against commerical nuclear power.

    1. I wonder how many other industries/projects you could say the same thing about? The First sky scraper? The first Farris wheel? Steam powered boats and trains or cars? Everything is new, scary and risky until it’s old boring and everyday….

  16. The Jenga analogy is both good and bad. It captures the idea of atoms breaking apart and causing others to do the same. But the atoms themselves never hit each other. Uranium never crashes into uranium.

    I’m going to build on a classic example that might explain things in a bit more detail for people who are curious.

    Imagine each atom is a mousetrap. All of them are set, glued to the floor, and have a pingpong ball or two sitting on each one. In a room without walls. U-235 is like a very sensitive trap, U-238 isn’t sensitive at all (trigger is glued?). The pingpong ball is a neutron (a part of the atom). Each trap is powerful enough that when it snaps shut, it breaks in half.

    A bomb is very high in U-235. Toss in a pingpong ball and you get a massive chain reaction. Balls (neutrons) fly everywhere, traps are smashing apart. Lots of energy.

    A reactor is relatively low in U-235. Toss in a pingpong ball and it might hit a trap, sending out more pingpong balls. But because they are separated, a pingpong ball might just fly out of the room. A chain reactions can happen, but slower. Reactors are designed to reach a sort of steady state. Lots of activity creating lots of heat, but that’s carried away by coolant (water, sodium, helium are commonly used)

    Moderators are like walls, bouncing balls (or neutrons) back in the room. Without moderators, most of the balls leave the room without setting anything off. A reactor won’t run without them. Lots of choices for moderators. Graphite is great, but burns (which is what happened at Chernobyl). They also control the speed (aka temperature) of the neutrons, which in turn controls all sorts of things such as how much plutonium is produced. So moderators are very carefully designed.

    Control rods are designed to be “sticky” and catch neutrons. Sort of like fly strips on the ceiling could catch pingpong balls. Move the fly strips closer, and it catches balls, slowing down the reaction. Boron (like in plain ol’ Borax) is a favorite. Sometimes it’s even mixed in the water. But for all intents and purposes, these sticky control rods are the brakes for the whole process.

    Actually, I could go on and on, but am already making this more complicated than most people care to deal with.

    Oh, as a side note to someone else’s comment, control rods slowly change into atoms or isotopes that don’t absorb neutrons. So if not maintained, eventually a reactor core will heat enough to melt down.

  17. My knowledge about EVERYthing pretty much boils down to what I learned on The Simpsons…

    And would I have been able to turn on my TV without nuclear power plants propping up the grid? No way dude.

  18. “Radioactive stuff goes in. Electricity (and nuclear waste) comes out.”

    Sums it up doesn’t it?

    If there was even a backup system of using wind, water or solar power this situation would have been easier. It is madness to have nuclear power plants situated on unstable tectonic plate lines. When is the world going to grow up. We have everything we need in the sun and the ocean and wind to provide all our energy needs and it’s time to develop it.

    1. The question I have when someone makes this argument (“We have everything we need in the sun and the ocean and wind”) is always: what are the long-term climate effects of these technologies?

      If we built enough solar panels to satisfy all our energy needs, we’d be taking a huge amount of heat out of the atmosphere. How does that change the climate?

      If we build enough wind turbines to satisfy our energy needs, we affect the speed of the wind. How does that change the climate?

      Nobody knew when we started burning fossil fuels that it would lead to global climate change. Does anyone know what the long-term large-scale effects of any of the other solutions would be? I’ll admit that I’ve not done a lot of terribly hard searching, but it’s a question that I’ve never seen answered.

      1. “If we built enough solar panels to satisfy all our energy needs, we’d be taking a huge amount of heat out of the atmosphere. How does that change the climate?”
        No, we’re not; we’re actually adding heat to the atmosphere. Solar cells reflect less heat out into space than just ground; ground isn’t that dark! (in most places) Ditto solar thermal, though it’s not as obvious. So the overall effect is still warming, but less so.

        “If we build enough wind turbines to satisfy our energy needs, we affect the speed of the wind. How does that change the climate?”
        That’s actually a harder question, because it depends where the turbines are and where they’re taking power out of the atmosphere (plans for stratospheric or jet stream turbines will have a much different effect than the regular ones) ; there has been some computer modeling but it’s safe to say we’re not exactly sure.

        There’s a maximum limit to the amount of heat we can add to the environment, be it from burning fossil fuels, solar, nuclear (fission or fusion) or geothermal energy* without producing a noticeable warming effect. This puts a fundamental limit on human industrial activity on this planet.
        The obvious answer is to take industrial activity off the planet, if you ask me.

        *most power-producing geothermal applications (as opposed to just heat pumps for HVAC) increase the rate at which heat is leaving the planet and entering the atmosphere, so it counts too.

      2. Your point as to wind power troubles me as well.

        My own pet theory for the long term salvation of mankind’s energy needs is not nuclear, but tidal, power generation- but the moving sea is one of the most punishing environments for equipment possible.

        As o tho this matter, still think it’s more of a delay or glitch in shutdown procedures due to not tsunami-proofing the diesel generators on-site
        rather than anything fundamentally flawed in the design of the reactors proper.

        And i note (hopefully) that the radiation releases so far have been negligible.

        May they continue to be so!

        Go engineers!

  19. My ‘observation’ of the communications of the actual nuclear technicians in the Fuji reactor suggest to me that there is a big fact gap between what the Japanese govt & press are telling you and what the reality is -AT EVERY STAGE OF THE DISASTER.
    While less cautious readers may be happy dying of radiation poisoning in front of their nuclear power station powered tv….
    The reactor was out of control within a hour of the quake. The only question was just whether it was catastrophic or indeed a new word should be invented. The one the techs were using was ‘meltdown’. The techs knew the jig was up when the borate failed to help. Using sea water to cool the reactor was last chance desperation. It failed.
    The new open ‘air cooling’ is working?
    May as well listen to Homer Simpson.
    Why not leave my comments up and see just who is right. Time will tell.
    Cursory knowledge of how this fairly simple reactor works tells it all. I have yet to hear how no power & cooling improves an already cooked reactor.

  20. #33: a ‘switch’ implies that fusion exists. It doesn’t and billions of dollars and thousands of scientists are working on it.

    #34: The earthquake epicentre was in the middle of the ocean, and the power plant is right on the coast. To detect the earthquake, and then send a light signal, you would need to have some sort of underwater underground seismometer, wirelessly transmitting data to the shore (cables can break). Even if it was instantaneous (it won’t be) it would only give you a few second’s warning of the shockwave. Hardly enough to close down 6 nuclear reactors.

  21. i’ve just spent some time looking at more of the images coming out of japan – it is now less than 48 hours after the quake.

    I hope they get these reactors cooled down asap, but really, as important as this nuclear accident is, this is yet just a sideshow to this immense immense tragedy.

    man oh man
    poor japan

  22. To clarify for Anon, it is the earthquake that triggers the shutdown, but then it is the earthquake that is the fastest travelling effect..

    My point is these reactors are shutting down pretty much *whilst* the earthquake hits them – they other ones around the country have got a minute or more, you are right but it is very time-sensitive. It’s no surprise the affected plants are the ones closest to the epicentre, basically.

  23. One of the interesting thing is with the cooling water. Normally ultra pure water is used in a closed system with a heat exchanger with sea water to cool the core. Not only does the sea water get radioactive, but also the contaminants in the form of biologicals will tend to cause blockages over time. With a Nuke this is not really an issue as you are wanting to not release any water coming in.

  24. First, I don’t know if this has been verified, and IANANE (I am not a nuclear engineer) but I had read that a safety trainer with knowledge of the design of this reactor said that the explosion had almost certainly occurred outside of the core, and that the most likely scenario was that an excess of hydrogen was released to the turbines causing the explosion. Anyone have any more info on that?

    Second, anyone care to comment on the wisdom of placing a reactor on the shoreline of a country that experiences regular tsunamis? I understand that the prevailing thought at the time that the reactor was built was that most tsunami were considered most likely to occur in the south of the country. Still, even setting aside the emergency fallback of pumping sea water in as the cooling agent which may have been part of the placement consideration, it does seem awfully risky. Pumping seawater wouldn’t have been necessary if the diesel generator hadn’t been swamped and the battery fallbacks exceeded their limits… Thoughts?

    1. Moose, I have been having the same thoughts about the plant siting, especially when I saw the photos of the plant location. I live on the coast of North Carolina and I cannot imagine locating a nuclear power plant on a coastline, especially since Japan’s coastlines are subject to tsunamis. There is dumb and then there is just plain stupid!

      1. Anon #116:

        The Indians did not have untoward problems when the 2004 Christmas tsunami washed against the Tamil Nadu coast, site of their Madras nuclear power plant.

        It depends on the standard to which it was engineered.
        Turned out to be not enough, so where did their estimates go wrong?

        The forensic engineers will have their opportunity to study this. I hope they learn all there is to learn from this disaster, whatever those lessons may be.

  25. This is really an interesting write-up. But, I’m lost in this part: If steam is used to turn turbines of electric generators, why does it need to be vented out to atmosphere when the pressure inside the core becomes too much. Can’t it be channeled to turbines just like in normal operation time? Anyone care to explain?

    1. To increase the rate of cooling/heat dissipation, the transport away of excess energy, thus helping to speed a reduction in pressure, by that reduction of temperature…I think.

      1. Thanks for explanation, but I still don’t get it. Why can the steam be released to turbines to reduce pressure?Why atmosphere?

        1. Ah i see Noodles has beat me to the draw.

          Too much pressure, too soon for turbines not engineered for the speed of release.

          It appears that the periodic controlled release of steam is one way to deal with overheating yet contained nuclear reactor cores afflicted by compromised cooling systems.

        2. Look at this diagram from the BBC. The steam is part of a closed loop – the water is exposed to the fuel rods, heated to steam, powers the turbines, and is then condensed back into water to go round again. In normal operation, this water never leaves the reactor (it couldn’t, because it contains radioactive isotopes, which is why the steam of this water is bad).

          When the coolant system fails, there’s no way to condense the steam, but the water keeps boiling off, so pressure inside the sealed system rises. Eventually it’s going to rupture something and cause serious problems, so the steam is vented out into the reactor building – which is more of a bunker with 6-foot walls. But of course that too can only hold so much pressure, so eventually you have to let some of the steam out. That’s why it ended up in the atmosphere.

        3. Bulone:

          It really doesn’t matter where one bleeds steam off, in this situation. All one is doing is removing heat. In normal operations, the steam, after the enthalpy is exhausted, is condensed back into feedwater and fed back into the reactor. This is necessary because of the strict chemistry requirements, among other things. However, a lot of equipment and systems have to be in operation for this. In an emergency, you ust vent steam off and try to pump water in only to rpovide a heat sink.

    2. They need to keep the water level in the reactor vessel above the nuclear material. Normally the steam is condensed into liquid water an recirculated back into the reactor vessel. They apparently can’t pump cool water back into the reactor vessel to keep the water level high, so they don’t let steam out into the turbines.
      /not a nuclear engineer

  26. One thing not being taken into account is the ongoing seismic activity that could create further damage, this is my concern. Time is everything in regard to what will happen.

  27. Bulone – I expect the pressures then were far too high for the turbines to spin and not tear themself apart – which would then make the coolant leak everywhere. Rather then, they kept the coolant a liquid and sealed inside, but at high pressure whilst they hope to cool it enough that the pressure drops – it clearly hasn’t and they needed to vent the pressure in the pipes before the burst. I would guess that something burst or some hydrogen evolved out of the coolant, and blew out a load of debris in the video. The structure is still there though, so it wasn’t very forceful and the core is built like a bunker so that’ll be fine.

    Moose: Nuclear plants have to be built near large bodies of water to use in their cooling systems, typically the ocean because lakes are closed. The west coast of Japan isn’t doing much, and there are mountains in the way, so it makes at least some sense to put the plants where they are.

    That the system hinged on the diesel generator failing is the most concerning part of all this. It absolutely should not have failed and since it did, the core should then have some well planned system for cooling without it! This is basic risk management. There will be some hard questions asked of the company after this.

    1. Yeah, they kept the generators at ground level….and the tsunami was 25 feet tall.

      A lesson learned.
      For all seaside reactors, too.

    2. Yeah, I suppose that question of the diesel generators is what I can’t wrap my brain around. I understand the need to have access to a large body of water, but if doing so represents such a clear danger to tsunami, wouldn’t you compensate for that with diesel and battery generators positioned in such a manner as to weather such an eventuality? Hindsight being what it is, this seems like an error of Gaussian proportions…

  28. >> In the bomb, the reactions happen—and the energy is released—very quickly. In the power plant, that process is slowed down by control rods.

    In a power plant, the reactor is just amplifying delayed neutrons.
    Delayed neutrons: Some fission products give off neutrons seconds or minutes after fragging off. The reactor keeps itself going by generating new delayed neutrons as it fissions. But absorbing some, over seconds-timescales, you can control it.

    In a bomb, the chain reaction is over in a microsecond, you couldn’t possibly control it with mechanical devices.

    See The Los Alamos Primer, and look up delayed neutrons and reactor physics. Its subtle but the difference between a bonfire and a detonation.

  29. Nice explanation, after the explosion I made a layman’s guess that it was a hydrogen bubble that blew, although I did not consider that there was a mechanism to vent the gas away from the core into a secondary area, other than outside. Must admit when I saw those steam clouds billowing I thought they were dealing with a core exposed to the atmosphere and boiling off its remaining water. Scary stuff, but not Chernobyl scary yet.

  30. I was anticipating this post, Maggie. Thank you. :-)

    I have a fair bit of confidence in many nations’ nuclear plants to contain their messes, even in the event of a partial meltdown; This is by design. Sciencey folks should review the horrible design and operational failures which allowed Chernobyl to happen, and subsequently, contaminate much of the Ukraine, Belarus, and even parts of Western Europe.

    Humans have harnessed many powerful and dangerous forces, not the least of which is nuclear energy. The important thing is to treat these forces with the respect they require. If we don’t, the price is indeed a high one.

  31. @gravitysrainbow

    No. All modern reactors come with automatic shutdown mechanisms when there is a problem. But they are pretty good at running themselves, so what would more likely happen is that they would operate until they ran out of fuel and shut down when there was no one around to refuel (like a car running out of gas, it would just putter to a stop).

    Modern reactors also are designed such that they will shut down in the event of loss of coolant (albeit this can take some time when you’re at the GW+ scale), and have a maximum achievable power. This is guaranteed by the laws of physics in what is called a “negative temperature coefficient” if you want to look up the details.

  32. Good post, Maggie — please keep ’em coming.

    I hope our Japanese brothers (and sisters) come out of this okay.

    I will also say, that I hope we install plenty more fission power plants in the US until we’re able to create useful fusion reactors. The Westinghouse AP10k should be an appropriate power plant for us until that time. (I could only find the AP1k listed on Wikipedia, designed in Pittsburgh: )

  33. finally an article that explains the situation and does not intend to scare everyone shitless, like those in international media.

    thanks, hopefully japan resolves this in a way that causes the least amount of troubles to the people of japan and its neighboring countries.

  34. Still not a clear picture. The article says “Usually, the elements used are Uranium-238 or Uranium-235. They’re refined and processed into little black pellets, about the size of your thumbnail, which are poured by the thousands, into long metal tubes. Bunches of tubes–each taller than a basketball player–are grouped together into square frames. These tall, skinny columns are the fuel assemblies.”

    My question is: how hot does a single tube get all by itself once it is filled?

    My confusion is: First, I assume that a single tube does not get very hot, at least not boiling hot. Certainly, a single tube would not get so hot as to start melting the tube…right? So, if these tubes in the core are separated by control rods then why is there any threat of melting?


  35. Just a quick clarification. The heat/steam from the reactor does not normally drive the turbine directly. The heat is transferred in a device called a steam generator (a large heat exchanger) to a different loop of water/steam. This isolates the radio active water in the core from the non-active steam driving the turbine.

    1. I don’t claim to be an expert but isn’t that extra loop with the steam generator the key difference between a BWR (which this is) and a PWR? You can see the difference in the side-by-side diagrams here.

    2. This makes sense (isolation and less contamination of turbine/condenser equipment). But how would the water level drop within the reaction pressure vessel if it is a closed system? Wouldn’t this imply there is a leak?

    3. The reactor is the steam generator in a BWR. Boiling water reactors don’t use steam generators; pressurized water reactors do.

  36. Still a lot of trouble, danger and expense to go through to boil water. We can put a man on the moon. Maybe we should use the power they use. Solar. No meltdowns, no liability, no waste storage problem, no spent fuel rods, no weapons transfer, no oil wars, not even necessary to boil water. Only power.

    1. That’s a strawman argument if I ever saw one. Yes, satellites use solar power, but that’s only a small part of what is used for energy in space. Space shuttles use tremendous amounts of liquid hydrogen and oxygen as propellant and most satellites use chemical propellants. Solar power is sufficient for powering the electronics, but it doesn’t replace conventional power generation for the larger tasks. [] Spacecraft also lack the shackles of friction and gravity that makes things so difficult here on Earth, which allows them to use very low energy means of adjusting their trajectory once in orbit.

      Solar power is a wonderful thing, but it’s certainly not a direct replacement for the fossil and nuclear fuels that we use today. Even just powering a building with solar power takes a large area of solar panels. We’re constantly improving the efficiency and cost of obtaining solar energy, but it’s still a long way from replacing our current methods of generating electricity. Not that we shouldn’t keep working on it — it’s probably our best shot at getting power without too many harmful effects — but it’s unreasonable to expect that we can switch away from our current fuel sources to solar power for most purposes with what’s available today.

  37. Here is a link to the most recent information on the worst of the six Japanese reactors with crippled cooling systems. Despite an apparent hydrogen gas explosion, it appears that they have avoided a complete breach in the reactor containment walls (both concrete and steel walls are probably intact – though the concrete wall may be damaged). They have been pumping sea water into the reactor building, a highly corrosive last ditch effort to cool the plant which officially ends its effective life as an energy producing unit. Let’s hope they continue to succeed in preventing container breach here and with the other crippled plants.

  38. “Proximity is also what makes the difference between a nuclear bomb, and the controlled fission reaction in a power plant. In the bomb, the reactions happen—and the energy is released—very quickly. In the power plant, that process is slowed down by control rods.”

    Maggie, as much as I appreciate your efforts to explain the situation calmly. and rationally, all the more following Rachel Maddow’s moment of blissful ignorance with the UCS flack, you fail pretty bad at meeting your own stated purpose of the article with this quote here. These words suggest the control rods are only thing keeping a reactor from blowing up like a bomb (a common fear) and you fail to mention here that the unenriched fuel used in reactors precludes this outcome entirely. This is not a minor detail.

  39. Ok So, apparently, the Japanese are using the american fast breeder system and not the Candu system.

  40. If you get the feeling that public officials attempt to keep people in the dark when a nuclear event is unfolding, you are probably right.

    The first step for people to become informed about these issues is that many people should own a geiger counters. Most people in the world can afford them. Mine cost around $300usd, a few decades ago. If most of the citizens of Japan near the failed nuclear plant had geiger counters, then they would have a better individual and collective idea what was going on.

    If people do not own gieger counters, then authorities will maintain and control the hood of ignorance and manipulation over the people.

  41. Thanks for the great post. It’s getting some proper information out there. About the design as it relates to tsunami; the backup power systems were designed to withstand a ~2-3 meter tsunami and the reactor was designed to withstand a ~7.5 tremblor. As we know, those parameters were greatly exceeded.

  42. What sort of disaster would befall the earth and its people if a wind turbine or a solar panel melted down…..?

    1. “What sort of disaster would befall the earth and its people if a wind turbine or a solar panel melted down…..?”

      Probably nuclear war, but it would already be going on. That’s the only scenario I can come up with.

  43. Thank You for explaining this, I feel a little better for the folks in Japan, but still worry that if there is an explosion, where will all the “toxic” particles go? Do they go into the atmosphere and follow the weather paterns, or do they drop to the ground within a certain radius of release? Just wondering what other countries should be worried about, or if this is even a legitimate concern. Again Thanks for the explaination, it’s made it a little more understandable…

  44. Maybe I am simplifying this waaay too much, but it seems to me to be a simple design to have a self sustaining cooling mechanism. First; maintain a volume of water well above the height of the core. Second; as the shutdown is initiated, a simple valving system would water to flow (via that little gravity issue here on the planet) into the core at a controlled rate-equal to the rate of evaporation. Third; the subsiquent steam plumbed through a series of coils (i.e. a moonshine still) and the resulting condensation be re-introduced along with the gravity fed water. It seems to me, that with an adequate volume of water on stand by at all times, this wouldn’t really be an issue. I’m sure that it’s not that easy; but maybe it really is. Anyone with a better knowledge of how this stuff works please comment. An open forum is often the mother of invention…

  45. There are a couple of mildly disturbing things here:
    – They had some window between the generator failure and meltdown. I would have been scrambling to bring up the cooling system back. Fly in the equipment – wire up to the adjacent generators – something. It seems the reaction time, leadership, ability to improvise were all pretty bad.
    – The containment vessel still has a bunch of holes in it, coolant, venting etc.
    – Has the explosion damaged the release valves? Blown the whole thing open?
    – What is stopping us from having an explosion inside the containment vessel and having “stuff” blown out through the holes?
    – Once they pump the seawater in how do they let it out? Steam? How much radioactive steam is going to be released until the thing cools down?
    – Seems like a terrible design…

    Any thoughts?

  46. This is a well written basic introduction to some very complex stuff. What is most refreshing is Maggie’s writing like an intelligent person who respects the intelligence of her readers. There was no ooh and ahh and hystrionics over mind-boggling big numbers. Janet Maslen recently, in the NYT, reviewed Brian Greene’s latest orgasm over string theory, and Maslin gushed, blushed and sputtered over Greene’s mind-boggling speculations, and generally made an ass of herself. Not all non-physicists are as gee-golly stupid as Maslin and those write like her.

    Anyway, good work, I enjoyed it and learned a lot. Thanks

  47. This may seem like a rather simple question from a naive perspective, but, can anyone explain why they don’t just use liquid nitrogen to cool the core in an emergency?

    1. The liquid Nitrogen, while having an EXCELLENT delta-T, would require (extra!) energy to cool down before putting in the reactor coolant loop. Also, the temperature extreme would make most pipes very brittle, and is NOT a good idea from an engineering standpoint. Better to use the cheap, plentiful water that is easily handled, easily cooled, and easy to purify. (You have to use distilled/deionized water to keep corrosion to a minimum).

      Nice idea, though. Not practical, however.

  48. Agree with the anon post above–if they did put in the control rods—why is there even a possibility of a “melt down”? Yes, as Maggie’s article stated, you can get pressure, explosions, but for a melt down to occur, wouldn’t the control rods have to be removed?

    OK, exactly what is a melt down anyway?

  49. That was the shortest, clearest explanation I could even imagine. Thanks for the info, but more than that, thanks for putting my mind at ease. Now at least I’ll know how to interpret the stories that follow.

  50. I am glad they use the “basketbll-player-scale”, because it’s a universally recognised measure when talking about fuel rods.

  51. Great article. As others have said, the only improvement worth making is explaining that a nuclear fuel rod cannot explode like a bomb, even without any control rods.

    @Daneyul: A meltdown is literally when the core material (fuel rods and such) become so hot that they melt. At that point, it becomes very difficult to keep the reaction materials safely contained in the core. If molten reaction materials escape from the core, then the outer containment (the bit that’s exploded in Japan) is all that’s stopping radioactive crap from getting into the environment.

    A meltdown will also mean that the reactor is beyond repair, though this is probably the least of one’s concerns at the time!

    [I am not a nuclear engineer, this is not nuclear advice, atomic numbers may go up or down, your home may be at risk if you fail to look after your cooling system or any reactors secured upon it]

  52. FWIW those core enclosures look to be built with explosion relief panels, meaning the corrugated sheet siding panels are designed to break away in an explosion to relieve stress on the underlying structural frame. So the steel skeleton you see left is not damaged by the force of expanding gas in an explosion because the fasteners holding the siding in place are designed to break under stress allowing the siding panels to blow away.

  53. “They had some window between the generator failure and meltdown. I would have been scrambling to bring up the cooling system back. Fly in the equipment – wire up to the adjacent generators – something. It seems the reaction time, leadership, ability to improvise were all pretty bad.”

    From what I’ve read, they did do that, but the electrical cables on the generator trucks they brought in were incompatible with the electrical system of the plant.

  54. Liz writes a prettty good technical article, but unless the Japanese operators tell us, there is no way to tell if the control blades inserted (partially or fully) for a controlled shutdown (the earthquake may have damaged the blades prior to insertion, but this is a low probability, and not all of the blades have to be inserted to achieve shutdown). That aside, think of the fuel as a big pile of charcoal in your bar-b-q: they don’t cool down very fast in the air, but squirt them with a firehose and they will cool quickly. It just so happens that without water, the fuel rods are so hot as to melt the (zirconium) metal jacket around the stack of fuel pellets, allowing radioactive gases and the fuel to escape. The presence of high levels of Cesium outside the plants indicates this has happened. The important thing to do now is to continue to use water to remove the decay heat and if necessary, due to the failure of the control rods to insert, add boron as well.
    – I am a Nuclear Engineer.

    1. Anon #103: is there, or have there been, any efforts towards developing mineralogical blends, or ceramic materials, or metallic alloys, which have radiation-absorbent abilities, not otherwise found in nature?

      1. Perhaps the nature or even existence of such materials would be considered a military secret.

  55. Anon #104: that would be nice, but the problem with ceramics is with the heat expansion properties are bad and ceramic is ‘brittle’. It does make a good barrier for mixing in nuclear waste though. The radiation absorption of different elements is different for each fission product – or said simply, what may absorb/block issues for uranium may not work for cesium, plutonium, so on, and there are a lot of products produced by the fission process. Add cost on top of that, and you see why its so hard and costly to contain and remove and store radioactive waste. Mankind would rather put this problem off to future generations then deal with it now.

  56. The Steam from these reactors in Japan directly turn the turbine and the steam turbine is in a radiation area the same as the reactor. Hence the term BWR GE Boiling Water Reactor.
    If you want to talk about steam generators then we are talking about Westinghouse PWR which is a Pressurized Water Reactor and the steam generator is a water to water heatexchanger which keeps the radiation inside the reactor building and away from the steam turbine. In testing a PWR before operating they perform a natural circulation test which insures it will circulate water in the primary reactor loop with the pumps off and no power. A PWR also has a very large Borated water tank under very high pressure to flood the primary loop in the event of primary loop pressure loss. This borated water kills the reaction rapidly and is referred to as a safety injection type of trip.

  57. About Anon #107
    People here in California who are worried about our 2 nuc plants, Diablo Canyon and San Onofre, should know that they are both the PWR type described above.
    Even though they are 1970’s technology, they were designed to the state of the art at the time and updated several times since then… the San Onofre plant, which I worked on the construction of in the 70’s, has backup cooling water in a pond up on the cliff above, with strong gravity-feed… and huge diesel generators backing that up…
    Pretty dang safe… and earlier descriptions of the containment vessels being “3 to 3 feet” of concrete…?? San Onofre’s are 14 feet thick..
    They poured concrete day and night for 5 years to build them!

  58. #37 – Though fusion appears to be perpetually fifty years in the future, there is an excellent reactor design that can be built right now, the Liquid Thorium Salt Reactor. Far superior to uranium plants in every way. Easily shut down. No weapons-grade byproducts. Widely available fuel. A hundredth of one percent of the hot waste.
    The only reason it wasn’t the design chosen in the 1950’s is that it doesn’t provide bomb fuel.

  59. ” In that case, it’s completely reasonable that an explosion could happen without releasing lots of radioactive material.”

    Take a good look at the pictures. The reactor buildings have been seriously damaged, not just the structure on top. The spent rods are stored in pools on top of those buildings. Once that water drains off those spent rods will heat up catch fire and spread highly radioactive ash everywhere. There is every indication that this is already happening. The radiation levels have spiked dramatically and they are saying it my be impossible to approach the buildings to make emergency repairs.

  60. It is difficult to get a handle, let alone an informed understanding and position, on what is happening in Japan in relation to its nuclear facilities – damaged by the earthquake last Friday. Every day brings news of a new issue or some risk coming to the fore. The Japanese haven’t been all that forthcoming – and sadly their past record of keeping the public informed is very poor.

  61. “When you release steam into a secondary building [to decrease pressure in the core], the steam condenses and leaves behind just the hydrogen. Then all you need is an ignition source and you can get a hydrogen burn. That’s what happened at Three Mile Island.”

    I think he meant that’s what happened at Chernobyl, since there was no explosion at Three Mile Island. But hydrogen vapors exploding is why most people believe the Chernobyl plant’s reactor containment exploded.

  62. I agree with Skywatcher (#37).

    Thorium is a superior nuclear fuel and has several important advantages over uranium:

    –Thorium powered nuclear reactors are more efficient and produce less than 1% of the waste of today’s uranium nuclear reactors.
    –Thorium reactors are safer, less expensive, smaller and can be configured to eliminate the possibility of melt downs or accidents.
    –Thorium does not produce plutonium and thus, could effectively eliminate further weapons production in volatile regions and reduce proliferation on a global scale, thus ending stalemate arguments over dubious nuclear programs such as exist in Iran and North Korea.
    –Proprietary thorium technology, capable of safely and efficiently dismantling nuclear stockpiles and eliminating spent uranium, now exists.

    The modern concept of the Liquid-Fluoride Thorium Reactor (LFTR) uses uranium and thorium dissolved in fluoride salts of lithium and beryllium. These salts are chemically stable, impervious to radiation damage, and non-corrosive to the vessels that contain them. Because of their ability to tolerate heavy radiation, excellent temperature properties, minimal fuel loading requirements (i.e., ease of continual refueling) and other inherent factors, LFTR cores can be made much smaller than a typical light water reactor (LWR- Lightbridge – LTBR). In fact, liquid salt reactors, and LFTRs specifically, are listed as an unfunded part of the U.S. Department of Energy’s Generation-4 Nuclear Solution Plan.

Comments are closed.