Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Exploding Reactors? Does that even happen?

  1. Jul 24, 2011 #1
    The media is always seeming to portray it like http://tvtropes.org/pmwiki/pmwiki.php/Main/NuclearPhysicsGoof" [Broken] waiting to go off. Now I've done a fair amount of research into nuclear power and it's readily apparent to me that this is not the case.

    I was wondering what, if any circumstances would cause a major explosion in a nuclear reactor. Is there any set of circumstances that would cause an actual nuclear bomb type explosion in the reactor? What reactors would be most prone to disasters? Are there statistics on the safety of reactors verses conventional power?
    Last edited by a moderator: May 5, 2017
  2. jcsd
  3. Jul 24, 2011 #2


    User Avatar
    Staff Emeritus
    Science Advisor

    Responsible media don't portray that.

    There really would be an explosion, but more of a pressure increase due to a pipe or vessel rupture. The most plausible accident would be a "Loss of Coolant Accident" (LOCA) or a "Reactivity-Initiated Accident" (RIA). We also consider loss of power, which may lead to a LOCA if the pressure in the primary circuit becomes too high.

    Commercial reactors are designed to preclude an explosion like that of an atomic (fission only) bomb. One cannot inject the necessary reactivity into a commercial reactor.

    Now there have been some reactors that did undergo power excursions that resulted in some type of explosion. One was the SL-1 reactor, and another was a test reactor for spacecraft propulsion, Kiwi, and the test was Kiwi-TNT.

    http://www.cddc.vt.edu/host/atomic/images/ktntb.jpg [Broken]

    But the Kiwi-TNT wasn't really an explosion like an atomic bomb, but more of a core disruptive accident. The core got above melting and just let loose.
    Compare the last picture at bottom of page with the nuclear weapons explosions.
    http://www.cddc.vt.edu/host/atomic/testpix/index.html [Broken]
    Last edited by a moderator: May 5, 2017
  4. Jul 24, 2011 #3

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    It's a shame the bomb came first.

    Had the bomb been developed from a power reactor project, public opinion would be instead "What? use uranium for a BOMB? What a waste of perfectly good uranium - you can make electricity out of that stuff ! "

    the analogy i like to use is this:

    "Compare a campfire to a stick of dynamite. Both are chemical energy but they proceed at fundamentally different rates.
    Dynamite has a lot of potential energy in that triple nitrogen bond that firewood lacks.
    You can make dynamite burn slowly, just light it with a match, but to make it explode you must compress and physically shock it so it all burns at once.. I defy you to make a stick of firewood explode.

    A reactor and a bomb are both fission energy, proceeding at fundamentally different rates. The rate is determined by geometry and you just can't pack the parts of a power reactor close enough together to make a bomb."

    You could make a steam explosion. That can happen with any kind of boiler. Just google "boiler explosion pictures".
    old jim
    Last edited: Jul 24, 2011
  5. Jul 24, 2011 #4
    So in conclusion, we're talking about blasts orders of magnitude less potent than even a poorly built fission bomb(IE: a fizzle)? 1/2 - 5 tons tnt range?
  6. Jul 24, 2011 #5


    User Avatar
    Gold Member

    One of the phrases I remember from my nuclear days was; "http://en.wikipedia.org/wiki/Prompt_critical" [Broken]".

    This was not a good thing as I recall.

    Googling the phrase today yields another reactor explosion which did not happen until I was out of the industry(We were well versed in what happened at SL1 & TMI):

    Gads. It's been so long, I don't even know what that means anymore.

    Time for math!

    200 MeV / fission * 5e18 fissions = 1e21 MeV
    3600 joule = watt hours
    1 MeV = 1.6e-13 joules
    reactor output = 160,217,646 joules =44505 watt hours
    3600 seconds per hour
    0.5 fraction of a second(guessing!)
    and the answer is: 320 megawatts


    That reactor was rated at 70 megawatts. The fraction of the second it took for the reactor to reach maximum power was probably less than half a second.

    Astro would have a better grasp of the time frame probably:

    Last edited by a moderator: May 5, 2017
  7. Jul 24, 2011 #6


    User Avatar
    Gold Member

    My math must have been wrong for that Russian sub:

    0.038 tons of tnt = 5e18 fissions
    2000 lbs per ton
    and the answer is: 76 lbs of TNT????

    hmm.... I've never played with TNT, so I really can't say.
  8. Jul 25, 2011 #7

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    your math looks okay to me...

    160 megajoules would be 160 megawatt-seconds, which in a half second is your 320 megawatts (320 average - probably peaked many times that)

    A megajoule is ~948 BTU's,
    so 160 of them should boil somewhere around 150 pounds of water, making a very respectable steam explosion.

    your initial statement was right, orders of magnitude less .
    Observe most of the pieces of that submarine landed within a few hundred yards. A genuine atomic-bomb like explosion would have destroyed the harbor, i'd think.
    A power reactor just won't stay together long enough to reach the power levels of atomic weapons.

    .... my two cents. 1.5 cents euro...
  9. Jul 25, 2011 #8
    What's with Chernobyl? The reactor suffered a power excursion mighty enough to shatter the whole reactor.
    That's still not even close to a small, tactical nuke. But the reactor jumped to 30 GW within seconds. So there was an uncontrolled, nuclear power spike. Not nearly close to a nuke. But the same physics, at least in my eyes.
  10. Jul 25, 2011 #9


    User Avatar
    Staff Emeritus
    Science Advisor

    That was a steam explosion, and it was a narrow pulse. One has to look at the energy, i.e., power integrated over time, rather than the peak power. Also, there was large volume of water which more or less flashed to steam, such that the pressure exceeded design limits and the pressure vessel and piping ruptured.


    An estimate of the Chernobyl energy release is about 10t TNT. (I need to verify that, because I've seen conflicting estimates).
    Ref: http://www.springerlink.com/content/d71710g0012116x4/
  11. Jul 25, 2011 #10

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    you're correct, at least to my unsophisticated understanding.

    Recall that western power reactors shut themselves down when the water in them starts to boil. Reason is the water molecules separate so become less effective at slowing down neutrons to the range they'll cause fission.

    That effect is called "Moderator Void Coefficient" and it's analogous to rarified air starving a gasoline engine.

    Western power reactors are designed to have a "Moderator Void Coefficient" that is negative, negative meaning that boiling shuts down the reaction.
    A positive void coefficient would make boiling accelerate the reaction , analogous to engaging the supercharger on a gas engine.

    That Russian reactor was a scaled up WW2 design with a lot of graphite in the core. That design has a "Moderator Void Coefficient" that can be strongly positive, ie once it starts to boil it makes even more power further aggravating the boiling. They got a runaway. Steam blew it apart, hot fuel ignited the graphite(think charcoal) and the fuel itself was so hot a lot of it shed its clad and scattered, probably some of it vaporized.

    The published "Moderator Void Coefficient" for Chernobyl is +$4, but $ is an obsolete unit. What +$4 means is means boiling all the water in the core adds 4X enough reactivity to take it prompt critical - so it may well have got there.
    EDIT: In fairness to the designers one must admit there are protections built into the design. The fellows had bypassed some of these to run an "experiment" ... The article linked by Astro speaks to that.

    And as mentioned a few posts ago - it can't stay together long enough to reach power levels of a weapon.

    Still I'm glad they never finished that same design plant on South coast of Cuba.

    old jim
    Last edited: Jul 25, 2011
  12. Jul 25, 2011 #11
    Yeah, but cause to the steam explosion was thermal energy emitted by the reactor. There wouldn't have been a steam explosion without the massive power spike.
    The "nuclear bomb type energy excursion", if I may call it so, didn't do much damage to the reactor. But it caused the steam explosion and THAT killed the whole thing.
    If a nuclear fission bomb goes off, a chain reaction is fissuring the nuclear fuel, plutonium or uranium. If those isotopes are fissured, energy in form of heat and neutron rays is released.
    And there's so much energy in such a short amount of time that a fireball is formed - the angry, red sun we've come to known as a nuclear explosion's heart.
    My point now is that physically the same happened in the Chernobyl reactor - uncontrolled chain reaction, uranium is fissured, energy in form of heat and neutron rays is released. Only difference - compared to a real nuke it was just a tiny amount, so no nuke fireball. But it still was enough to cause a steam explosion.
  13. Jul 25, 2011 #12


    User Avatar
    Staff Emeritus
    Science Advisor

    Yes, it was an explosion. However, the reactor was placed in a dangerous configuration, outside of design space, it had an inherent positive moderator coefficient, and the safety systems were over-ridden. That is quite different from an LWR design, and the practices would be illegal outside of the USSR. The folks I know in the industry wouldn't allow such an event.

    The reactor didn't explode like an atomic bomb, and while the yield appears to be less than an atomic bomb (to be verified), it was sufficient to destroy the core and reactor.

    There is also the matter of a lack of containment.
  14. Aug 13, 2011 #13


    User Avatar
    Science Advisor
    Dearly Missed


    Quite correct the reactor didn't explode like an atomic bomb. As you stated in one of your previous posts, the explosion at Chernobyl produced a yield of 10 tons TNT equivalent.

    The very first atomic bombs had yields of about 20 kilotons, or about 2,000 times greater.

    So the difference between Chernobyl and even the relatively low yield first atomic bombs,
    is a factor of 2,000.

    So all this calling the Chernobyl or Fukushima explosions as "atomic bomb-like" is really a bunch of nonsense. It's done by people who don't know the relative magnitudes of the two types of explosions.

  15. Aug 13, 2011 #14


    User Avatar
    Gold Member

    Steam explosions are very dangerous even in conventional boilers. In the pulp and paper industry, one of the most feared failures is the introduction of water or diluted black liquor onto the molten smelt bed of a Kraft recovery boiler. When you have a boiler 10-15 stories high operating at 600-900 psi, such an explosion could put your pulp mill out of commission for years, with the potential for significant loss of life for operators in the building.

    The training requirements of BLRBAC were a driving force for many of my best consulting contracts. Don't adhere to their guidelines? No insurance for you.

    http://blrbac.org/wp/ [Broken]

    Not to go off-topic, but the heat-source is secondary to steam explosions in pressure vessels. I'd rather work in a nuke plant than work in a Kraft boiler-house with spotty operational standards.
    Last edited by a moderator: May 5, 2017
  16. Aug 13, 2011 #15


    User Avatar
    Science Advisor
    Dearly Missed


    It's the same old paradox; the more inherently dangerous something is, the safer we make it in practice.

    Think how much more inherently dangerous airline flight is where you are 6 miles high in the air traveling at 600 mph as opposed to being on the ground going only 60 mph, as in a car.

    So which kills more people yearly, airliner crashes or automobile crashes? Automobile crashes by a wide margin.

    Likewise, even with Chernobyl, Fukushima, and the non-event (at least in terms of public safety ) at Three Mile Island; nuclear power is safer than practically any other technology.

  17. Aug 13, 2011 #16

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    ""I'd rather work in a nuke plant than work in a Kraft boiler-house with spotty operational standards.""

    Senior year in college i got to tour a deep shaft coal mine with a power plant right at mine mouth.
    Decided then and there i'd ratrher work thirty years in a Nuke plant than thirty days in a coal mine.
    That's exactly what i did.

    If you ever get a chance to see movie "October Sky" - do. There's no question they took a camera down into a real coal mine.
    But even that scene doesn't capture it.
    That's awesomely grueling, dangerous work and miners have my deepest respect.

    old jim
  18. Aug 14, 2011 #17
    As Astronuc pointed out, it's the amount of energy released into the fuel, rather than the absolute maximum power that matters. The rupture enthalpies for fuel pellets have been experimentally verified, and typically lie around 1000 J/g UO2, which is the acceptance criterion for reactivity initiated accidents, Beyond that, the power transfer rate from fuel pellets to the surrounding coolant would be rapid, as the pellet shatters and the contact surface suddenly increases (as happened in Chernobyl). Prompt criticality in itself won't be a problem, as long as the physical feedback mechanisms limit the pulse width so that the maximum enthalpy criterion will not be exceeded.

    Typical reactivity initiated accident (RIA) mechanisms in boiling water reactors are pressure transients and control rod drops. As an example of the design requirements for a BWR core, dropping of the most reactive control rod from the fully inserted to fully withdrawn position shall not lead to violation of the pellet burst enthalpy. In those cases I'm familiar with, the control rod drop has been a more limiting RIA case from fuel integrity point of view than the pressure transient.

    In a PWR, typical RIA mechanisms are a control rod ejection (caused by a leak in the control rod drive mechanism housing), boron dilution and possibly some transients leading to rapid cooldown of the primary circuit. Typically, ejection of the most reactive control rod from fully inserted to fully withdrawn position will make the core promptly critical, and the pellet burst criterion will limit the maximum reactivity worth that one is allowed build into a PWR control element.

    It is a core design / fuel design / reactor licensing task to make sure that even the most limiting of the aforementioned reactivity insertion mechanisms will not lead to too large power excursion that would in its turn create a mechanism for rapid pressure increase in the primary circuit. Fortunately, these analyses are rather straightforward to do.

    All that said, there are always questions of whether we certainly have mapped all possibilities to insert reactivity in a core. For example: is the sudden ejection of one control rod certainly the limiting RIA in a PWR, taking into account the damages found in the Davis-Besse pressure vessel head? It appears that it's the rate of ejection that matters most, and if the multiple ejections would have a few milliseconds between them, it would probably not make the situation any worse, but these are issues that in my opinion need consideration and further studies.

    In any case, a nuclear explosion in a reactor is certainly impossible.
  19. Aug 15, 2011 #18
    Considering the comparatively low yield of even a worse case reactor explosion to a nuclear bomb, why is it that we as a culture are so afraid of nuclear reactors. I hear far more complaint about the peaceful use of nuclear power than I do about nuclear weapons stockpiles, which are orders of magnitude more destructive.

    On a slightly different note, how clean/dirty from an ionizing radiation standpoint are reactor explosions as compared to a first generation nuclear device as used on Hiroshima and Nagasaki. How about modern nuclear weapons? Generally speaking, would you rather be downwind of a Chernobyl style accident or Modern nuclear test detonation?

    Lastly, is my understanding correct that the Chernobyl power-plant lacked the concrete containment vessel that TMI had? I'm fairly certain I heard somewhere that US plants were better built safety wise that their equivalents in the USSR. Is that extra containment what kept TMI from turning into a US Chernobyl?
  20. Aug 16, 2011 #19
    In this category Chernobyl "wins" hands down. Nuke releases at most several tens of kilograms of fission fragments, Chernobyl released several *tons* of them.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook