1. PF Insights is off to a great start! Fresh and interesting articles on all things science and math. Here: PF Insights

Nuclear Reactor

  1. This may not be an appropriate place to ask this, and if that is the case please chastise me and cast me out. Given that the members of this forum seem to be very educated and intelligent, especially in regards to these fields I wanted to ask a question. What would be the absolute worst case nuclear reactor meltdown? What sorts of effects could one expect and at what types of distances? And bear in mind, I really am asking about the worst it could possibly get, as in, no one there to SCRAM the control rods, no fire crews responding, that sort of disaster scenario.
  2. jcsd
  3. jimgraber

    jimgraber 190
    Gold Member

    Think Chernobyl times ten, but not Hiroshima-Nagasaki. As I understand it, a small explosion is possible, but not kilotons. The big tragedy is a large escape of radioactive by products. You would probably get better answers if this were moved to the Nuclear Engineering forum.
    Jim Graber
  4. vanesch

    vanesch 6,236
    Staff Emeritus
    Science Advisor
    Gold Member

    The absolute worst case was Chernobyl. I don't know why another poster said "times ten" because I don't see what could be worse: a serious part of the core material (and other stuff, as the Chernobyl reactor was also a military irradiation facility) thrown high into the atmosphere by a huge fire enhanced by a still working reactor for more than 36 hours, at the moment of shutdown (that means, with the highest possible amount of fission products).

    But beware: that's not a *meltdown*. It was a blowing up of the reactor by sudden overheating. Also, a nuclear reactor can *never* undergo a nuclear explosion (as in an atomic bomb). It is physically impossible. You can at most have strong heat production which will make the confinement give up under heat and pressure. At Chernobyl, it was the water that got heated too much in the tubes that blew up the reactor structure.

    A *meltdown* is a much less severe accident, and it doesn't imply necessarily any release of radioactivity to the environment. A meltdown happens when the fuel elements are not cooled enough, so that they reach their melting point - inside the reactor. For instance, a meltdown can happen when there is a loss of coolant, or in a PWR, a loss of pressure, so that the cooling water starts boiling off. This is what happened (by wrong manipulations by the control crew) in the 3-miles island accident: the rods containing the fuel can get so hot as to melt. This can happen even with the reactor shut off (in fact, with a loss of coolant, normally the reactor shuts itself off immediately), but because of the heat of radioactive decay. We are talking here about much lower power productions than the 30 GW that blew up the Chernobyl reactor, but nevertheless they can melt the fuel. The thing to do is to restore the cooling, and the incident is closed (but the reactor is damaged). If the cooling cannot be restored, then things can get worse: the heat can end up damaging the reactor vessel, and as such, the fuel can escape the reactor vessel and arrive in the confinement building. If one still doesn't cool this, the confinement building can end up damaged by the hot fuel, and some fuel can escape to the outside (but not in a plume in the atmosphere!). When spread out enough in the neighbourhood, the fuel will cool down and that's it. This is very heavy, but local, contamination.

    Again, serious as this is, this is nothing comparable to Chernobyl. In Chernobyl, everything was really put together to make the accident as bad as possible. That's why I cannot imagine anything worse.

    In fact, in modern reactor design (such as the EPR), a meltdown is considered as a possible accident that should be handled. Hence, there is a cooled "core catcher" underneath the reactor vessel which stops the accident right there, without any release to the outside.

    You have to know that western reactors are all so designed (contrary to the Chernobyl type reactors, which are therefor considered unsafe - and this was well-known, even before the accident) that they *naturally* shut down by physical mechanisms when things go wrong seriously. There's no action required nor from the control crew, nor from any automatic system: the physics of the reactor is such, that the reactor stops when it gets too hot or when it looses a serious part of its coolant.
    This is why even a worst-case accident with a western power plant can never be as severe as the Chernobyl accident (where the reactor was unstoppable because of bad design - they stopped it finally by dumping boron on the carcass with helicopters).

    So the only driving force for an accident that remains is decay heat, which is far less than the reactor power.
  5. So in a scenario where there was aboslutely no one available to man the plant, unexpectedly, (think neutron bomb or the rapture) the plant would still shut down safely by default?
  6. vanesch

    vanesch 6,236
    Staff Emeritus
    Science Advisor
    Gold Member

    Yes, it would shut down safely. However, and this relates to the discussion about the cooling pools, current power plants are not designed to be left alone for extended periods.
    The culprit is the needed electricity for the cooling pumps, which are in case of blackout generated by emergency generators (usually redundant diesel engines). If they run out of fuel (after days or weeks), there can be a problem with the cooling (of the residual heat of radioactive decay). Now, I think (but I'm not sure) that at least the primary water circuit of a modern PWR is designed such that pure passive convection is sufficient to bring the decay heat to the steam generators without the primary pumps activated. However, there needs to be secondary water cooling to evacuate the heat from the steam generators - I don't think things have been designed completely passively there (although that is in principle possible but cumbersome).

    That said, a nuclear power plant can also work "independently" loose from the main grid, at low power, and provide itself with electricity - the reactor is now active, but at something like 10% of its nominal power. In this mode, a power plant can run a very long time "all by itself". However, I think it needs manual control in this mode.
    Last edited: Apr 1, 2008

  7. There are just too many variables to have a reasonable answer to that. Plants require people to run them (top up gasses, maintain chemistry etc) but safety systems are designed to put the plant in a safe state when certain parameters are out of norm, so it would self scram.

    Assuming a normal scram, I think the plant would be just fine for quite a long while due to the engineered design features. But since there are lots (understatement) of people that work at nukes, it's just not something that is very probable.
  8. jimgraber

    jimgraber 190
    Gold Member

    A point that has been implied but not mentioned specifically is that it depends on the design of the particular plant. In general, older designs are less safe and newer ones more safe. So called third and fourth generation designs are intended to "self shut down" in essentially the way you mention. In fact, some experts who know more than I do claim that even an incident as bad as Chernobyl is impossible or highly unlikely for second generation designs such as most of those now operating in western countries.

    Of course, you can think of worse scenarios, like dropping a real nuke on a reactor, or just a deliberately dirty bomb...
    Best, Jim Graber
  9. I ask these questions because I am a writer. In al my work I try to strive for as much realism as possible, within the confines of the fantastic situations I create. The particular tale which gives rise to this question centers around a zombie plague and the attempts to contain and correct it. Basically, I am just looking for the worst case scenario so I can make sure to not go overboard in my depiction of a reactor event.
  10. vanesch

    vanesch 6,236
    Staff Emeritus
    Science Advisor
    Gold Member

    How about the zombies eating the fuel rods and then vomiting it all over the place ? :wink:
  11. A zombie plague hits the plant around 4 in the morning most nights, but it seems to be mostly giddy zombies that don't make much sense when talking.
  12. mheslep

    mheslep 3,451
    Gold Member

    I had always thought a melt down that continued through the floor of containment building and down until the mass hit ground water explosively sending steam into the atmosphere was understood to be the absolute worst scenario. Pebble reactors, for instance, eliminate that possibility.
    Edit: I see after reading more above that this was only possible w/ the earlier reactor designs.
    Last edited: Apr 3, 2008
  13. vanesch

    vanesch 6,236
    Staff Emeritus
    Science Advisor
    Gold Member

    I guess that such a scenario is only possible with a reactor that doesn't have passive safety (meaning, continues to be critical even after loss of coolant), because I have a a hard time believing that the power of the radioactive decay by itself is sufficient to do such a thing. Now, this is just my gut feeling, but it is my impression that this is again one of those fear mongering scenarios entirely made up (I mean "melting itself a way through the floor"). Even the Chernobyl reactor didn't do that, and that one continued to function still quite a while after the accident.

  14. Well. We have 2 reactors that have totally melted down in Japan.

    What I'd like to know is what the consequences of that is.........i.e. Now what?

    What will happen now that you have molten pools of radioactive material at the bottom of the containment vessels........with no carbon rods slowing down the reaction?

    PS. I think it's time for the cement trucks.....................and bury all that stuff........

  15. Worstcase in slow neutron reactor: have the fuel cook itself with decay heat in a pressure vessel that has water on outside but is missing the lid (or is vented very effectively). Do nothing. Let the fuel temperatures in the middle reach boiling, liberating so many various isotopes with various nasty behaviours (Sr-90 for one thing) that you'll stop thinking that Chernobyl was a worst case. This corium won't melt through the pot because of low thermal conductivity of corium and cooling on outside. For more lulz, as the last thing, let water level outside the vessel drop , so that the vessel melts and it falls into water, resulting in a steam explosion that spreads the **** all over the place, and turns a lot of molten corium into aerosol. For even more lulz, have it happen at 4 or more reactors with 4 spent fuel pools each storing 3x the core load, re-racked for max density, and let the spent fuel run dry and catch fire. For even more lulz, build a lot of powerplants on a coast all to same standards, and have several of them suffer this at once as you are hit by the tsunami.

    also: Chernobyl's criticality ceased when it explosively disassembled.

    Worstcase in fast neutron reactor (highly enriched fuel): steam explosion in one part of the core compressing rest of the core, resulting in low power nuclear kaboom that'd vaporize the core. Have it happen as close to the end of fuel cycle as possible.

    Face it, neither Chernobyl nor Fukushima are as bad as it can be. Chernobyl barely released anything besides noble gasses, caesium, and iodine. Just the stuff that easily leaves fuel. It wasn't burning very hot.
    Last edited: Apr 16, 2011
  16. Well.......the water table can't be that deep there.......when it hits it......will get a lot of radioactive steam coming up right?
  17. Yeah....people keep comparing this with Chernobyl........down the road aways, when the next one happens, they'll be comparing it with Fukushima.......lol.........z
  18. its not radioactive steam, its radioactive elements released by fuel that go along with steam, that you have to worry about. It's also aerosolization of the fuel. The water steam itself is not radioactive, unless it gone critical, in which case it is very short term radioactive.

    water table is like ground with gradually increasing water content. For best results, you need corium to drop, with some speed, right into the water. That did NOT happen in Chernobyl. 3 volunteers, true kamikaze, dived into highly radioactive water (on par with that water Tepco doesn't know what to do with), using scuba gear, to open the valve. Their light failed and they had to find valve by hand. 2 of them died later. All of the liquid nitrogen of the western soviet union was used also for preventing this worstcase. All in first week or two.

    The west, of course, loves to portray Chernobyl as worst case, even with utter FUD BS about it making some nasty isotopes for military (what isotopes? the problem everyone had was with iodine and caesium, just the two most volatile fission products except noble gasses). Truth is - it is nowhere near as bad as it could have been, thanks to very quick heroic actions of people who died to prevent the worst case.
  19. also, look at table of isotopes for Chernobyl here:
    look what fraction was NOT released. And would have been released if the molten fuel was thrown with an explosion, turning it into tiny droplets. Hell, there's no good assessment for health effect of inhaling such particle. What if it sits inside lung, and causes a tiny tiny constant radiation burn that is constantly being healed. A lot of cell divisions, of irradiated cells, plus the autopsy after death from cancer can reveal the fuel particle that caused it. No denying the cause of death with such. With the usual cancers, you can bs about hormesis, you can link to health studies performed in third world, you can play an advocate and try to apply 'innocent until proven guilty' to your defendant, the radiation. With fuel dust, you can't deny it so easily and it very obviously doesn't dilute, just like bullet in russian roulette doesn't dilute over the chambers of revolver.
    Last edited: Apr 16, 2011
Know someone interested in this topic? Share a link to this question via email, Google+, Twitter, or Facebook

Have something to add?