52

Scenario: Deadly virus wiped out 80 - 90% world population and it did it pretty quickly (in 2 months)

Although it is plausible that among these 10 - 20% of survivors will be people who actually know how to run nuclear power plant, it's safe to assume that they will have different tasks to solve now.

Now, I did read several apocalypse scenario questions here. And it seems that it is wide consensus, that should the nuclear power plants remain unattended for longer period of time, they will simply overheat and cause major damage to their surroundings.

I know that power plants are mainly run by computers. Also, the nuclear power plants have several backup power on and off-site to provide emergency power to help cool down.

So, In case of no natural disaster, will unattended nuclear power plant cause any damage to its surroundings?

Please try validate your claims by references.

White Fang
  • 780
  • 7
  • 13
PavelBot
  • 523
  • 1
  • 4
  • 9
  • 6
    Is there something stopping the nuclear operators from shutting down the plants before everyone dies? Two months is a long enough time for that. – Lacklub Mar 14 '16 at 12:34
  • They can shut it off. But I assume they will be unable to maintain it in safe state – PavelBot Mar 14 '16 at 12:35
  • I'd expect nuclear power plant to be at least as secure as trains in this domain : without human intervention every onece in a while, the shut down fairly quickly. – MakorDal Mar 14 '16 at 13:02
  • You may want to look into this Reddit thread asking the same question. – fi12 Mar 14 '16 at 13:43
  • Related question over on SciFi – JPhi1618 Mar 14 '16 at 14:00
  • note: Title and Last paragraph ask 2 different thinks. – Zaibis Mar 14 '16 at 14:18
  • Variations on this question appear quite frequently. I wonder whether we need to create an FAQ. Unless specifically designed to do so and given the correct fuel to do so, a fission power plant cannot detonate in a nuclear explosion. Unless you were interested in some sort of hybrid space drive/power plant, I just don't see this ever happening. – Jim2B Mar 14 '16 at 14:54
  • 6
    @Jim2B is exactly right. Nuclear fuel is made from low-enriched uranium, which, in normal operations in a power plant, is fissioning as fast as it can already. Atomic bombs require high-enriched uranium, (or more commonly, plutonium,) and even then they need extremely precise explosive triggers to set them off. This means it's quite impossible for a nuclear reactor to turn into a mushroom cloud. They can have other types of accidents, of course, but an atomic explosion is not one of them. – Mason Wheeler Mar 14 '16 at 17:35
  • 3
    It's never the fuel in the reactor that you need to worry about, it's the highly radioactive stuff in post-processing baths that will heat up and mess up your day... – Serban Tanasa Mar 14 '16 at 18:24
  • 9
    Thorium salt reactors literally can't melt down if designed and built properly, but the idea got mothballed after initial proof-of-concept by anti-nuke hysteria, ironically enough. Some reactors would probably leak due to exigencies... Whether those are caused by idiots trying to exploit them improperly, subtle defects, or disasters. That wouldn't do as much damage as you might think, though. Scale, you see. – The Nate Mar 14 '16 at 19:55
  • @Jim2B Well, 1986 we probably had a fizzle. Was that a bomb? It did explode. Was it a nuclear bomb? It was fueled by nuclear power. Reactor designs differ; some are not inherently stable, as demonstrated. When the first reactor in the US went critical, "Norman Hilberry stood ready with an axe to cut the scram line". The yield of a runaway event is not great, but sufficient to render a thousand square miles uninhabitable through fallout. – Peter - Reinstate Monica Mar 15 '16 at 11:52
  • 3
    @Lacklub: you can shut down the chain reaction, but you cannot shut down the ongoing decay of the fission products; without cooling, those will cause a meltdown which can start the chain reaction again. – Michael Borgwardt Mar 15 '16 at 13:03
  • @MichaelBorgwardt I agree. But I was trying to establish the scenario the question was asking. An operating reactor behaves differently than just decay heat. – Lacklub Mar 15 '16 at 13:06
  • 2
    @PeterA.Schneider, the explosion in 1986 (Chernobyl) was a steam explosion. The explosions in 79 (Three Mile Island) and 2011 (Fukashima) were $2 H_2 + O_2 \rightarrow 2 H_2O$ explosions. There was no nuclear detonation from any of these. The heat for the steam and $H_2O$ disassociation came from their respective reactors. Control rods (often not graphite these days) are used for controlling the speed of the reaction. If allowed to operate uncontrolled with no cooling, the fuel in any nuclear reactor's core will *never* explode from nuclear reactions, ever, under any conditions. – Jim2B Mar 15 '16 at 13:35
  • Someone sufficiently stupid could make fuel with that potential but there's no need to and no one (to my knowledge) ever has. – Jim2B Mar 15 '16 at 13:36
  • 1
    @Jim2B How can you be so sure? I know wikipedia is not a scientific source; but there seems to be a bit too much released energy (10t TNT equivalent) and too much radioactive Xenon for a pure hydrogen explosion, and some quick googling comes up with more sites talking about it, so it's not a lone lunatic wikipedia author. Of course "explosion" needs a definition; but it got hot, it went through the roof; that's good enough for me. 10 t TNT is the yield of the smallest tactical nuclear US weapon. – Peter - Reinstate Monica Mar 15 '16 at 13:49
  • To put the scale of nuclear accidents in perspective... More people died of e coli contaminated spinach in Germany during 2011 (50+), than have directly died from all nuclear reactors for the entire history of the world (<40 - with about 32 from Chernobyl, a reactor type so unsafe no non-Communist country would have built it). But I guess the keyword in irrational fears is "irrational". – Jim2B Mar 15 '16 at 14:05
  • @Jim2B True: It's very likely that in the OP's scenario most survivors would not die from exploding, burning or simply leaking nuclear power plants. (But instead from germs in the drinking water and lack of antibiotics against all kinds of trivial infections, etc.) But the number of fatalities due to Chernobyl is hotly contested and not trivially determined. The thing is that even the very slightest (statistically insignificant) rise in cancer probability results in an enormous number of victims. Yes, they fade against the backdrop of general mortality, but so do the E.Coli cases. – Peter - Reinstate Monica Mar 15 '16 at 14:13
  • 1
    Two months is a long time, but the scenario is not that it somehow becomes common knowledge that most of the population will die, and they have two months to wrap up their affairs. In order to shut down plants deliberately, the operators have to realize that they need to shut them down. If there is enough uncertainty about the course of the epidemic, or enough people are in denial about it, I could see some plants remaining active past the time when human intervention was possible, leaving only automated shutdown and other fail-safes. – David K Mar 15 '16 at 14:31
  • The error in your question is this: "they will simply overheat and cause major damage to their surroundings". As both Fukushima and Chernobyl showed, the surroundings are not damaged. They just make people not want to be there any more. They are made "inhabitable" and "not economically usable", but not damaged to any large extent. The trope that Radiation Is Lava... http://tvtropes.org/pmwiki/pmwiki.php/Main/ArtisticLicenseNuclearPhysics ...has very little bearing on reality until you reach insane doses, like several Gray or above. – MichaelK Mar 15 '16 at 18:52
  • You know that it's disturbing when someone asks the same question multiple times about a deadly virus: http://worldbuilding.stackexchange.com/questions/28836/can-average-joe-reboot-the-nuclear-power-plant?rq=1 The question I now have is, how deadly is the virus you have? – jfa Mar 15 '16 at 21:51
  • @PeterA.Schneider Don't oversell TNT, it's energy content actually isn't all that awe inspiring. Those ten tons of TNT roughly correspond to about 5 times your car's tank, or 20 tons of water vapourizing into steam. It was suprisingly hard to find the figure for how much energy is released in the H+O->water reaction, but if my math is correct, those 10 tons of TNT should correspond to about 2.5 tons of water being formed. There's plenty more water in your typical nuclear power plant. A comparable BWR get about 2 tons of water per second, and the reactor itself contains a lot more water. – Luaan Mar 16 '16 at 10:30
  • @Luaan Interesting -- true, H has a high specific energy, about 30 times that of TNT. Only a few hundred kg H are needed for a 10t TNT equivalent explosion. But sources seem to agree that it may have happened; I'd still contend that a fizzle "explosion" is principally not impossible. I think we agree that it will not blow up like a strategic nuclear weapon, if that was the actual question. – Peter - Reinstate Monica Mar 16 '16 at 11:02
  • Also, the most dangerous radiation sources left unattended in a plague like yours will be the nuclear weapows cores (plutonium flakes and that flake is soluble in water) and tomographers/x-rays radiation sources, like the one involved in the Goiania Disaster ( https://en.wikipedia.org/wiki/Goi%C3%A2nia_accident ). – Geronimo Mar 22 '19 at 12:57

10 Answers10

48

No, probably.

Here's why:

Nuclear power plants will be unlikely to go into a meltdown scenario in the event of operators' absence. After several days, most will shut themselves down if they have not received maintenance. However it is plausible that a lack of operators combined with some hitherto undiagnosed problem with the cooling cycle or systems could begin a series of events that lead to a meltdown.

Nuclear power plants are already some of the most failure-redundant systems we have. Such events as mass strikes, earthquakes, power surges are all planned for as a matter of course. A properly-designed nuclear plant would be much less likely to explode without human contact than some other things in cities such as

  • Gas works
  • Coal/Gas power plants
  • Sewage treatment centers
  • Oil refineries

Even if there is a runaway heating without humans present, there are several redundant cooling systems that can replace each other. Computers can dump the control rods if a large meltdown starts to occur, and even if the core burns though the container, it will be caught in a 'core-catcher'—a structure designed to stop radiation from escaping in the event of an accident.

However, in the unlikely case that damage does occur, what can we expect? Well. A nuclear reactor will not go off like an atomic bomb, because the fuel is not in a pressure container. The most likely scenario is that a runaway reaction would cause the fuel to melt through the bottom of its container like a thermite charge, and drop onto the floor slowly sizzling away down into the concrete below. large fires would be set in the immediate vicinity by the intense heat, and localised explosions would throw radioactive debris around, which could be moved several hundred kilometers by the winds to affect a long but thin area with radioactivity. However, this would mostly be unnoticeable apart from in the nearest few km.

Source(s)

Community
  • 1
Rugnir
  • 899
  • 8
  • 12
  • 23
    Decent answer, but see Fukushima. While the circumstances differ (natural disaster vs. lack of manpower), part of the problem was that much of the redundancy was defeated by concurrent failures with local infrastructure (plausible in a manpower scenario as well). Also, the effects from just that one incident have been felt globally (certainly far more than "the nearest few km"!), so you might want to consider revising the impact portion of your answer. Finally, atomic bombs are certainly not pressure containers, either! :-) – type_outcast Mar 14 '16 at 12:12
  • 6
    In terms of atomic bomb being a pressure container the idea is to keep the fuel together long enough to get a sizable explosion rather than just a small explosion that throws the fuel away in all directions.

    As a rebuttal to your statement of distance of effects, I would ask do you have any mention of severe life-threatening radiation more than a few km from an accident? Even in fields around fukishima there was few effects other than higher than usual radiation levels in milk, but in a pinch that milk wouldn't kill you, just make you slightly more vulnerable to cancer.

     – Rugnir Mar 14 '16 at 12:34
  • Also, I would add that the bulk of my answer is explaining that because of multiple redundant systems, there is little likelihood of any one plant going up. There may be a situation whereby several systems were being replaced and/or were damaged but certainly not a large amount of accidents would happen concurrently all over the world in this situation. – Rugnir Mar 14 '16 at 12:36
  • 5
    As far as the 'atomic bomb' thing goes, it's not just pressure - weapons-grade fuel is much, much more pure than the stuff used in nuclear power plants, and an atomic bomb is actually a very carefully designed piece of equipment that will fail to explode if it has even a little damage - even weapons-grade fuel does not explode easily. Put simply, a nuclear power plant will not explode like a bomb, ever. The worst thing it can do is leak into the groundwater (which is what happened at Fukushima). – IndigoFenix Mar 14 '16 at 12:45
  • 3
    @IndigoFenix A nuclear reactor can explode a little like a bomb. Not a nuclear bomb, sure, but still an explosion. One (nuclear) reason that it will explode is a prompt criticality accident, and there can be a variety of different explosion. Steam and hydrogen explosions for reactors with water in them. Sodium reactors will naturally combust whenever coolant leaks. Some reactors use oils, which burn with oxygen. These can make pressure build, and can either look like or turn into an explosion. – Lacklub Mar 14 '16 at 12:59
  • No, a nuclear bomb is what you get by compressing a mass of fissile material into a small enough volume to initiate a super-critical nuclear chain reaction, period. "Pressure containers" have absolutely nothing to do with nuclear bomb designs whatsoever! – type_outcast Mar 14 '16 at 13:23
  • 15
    @Lacklub makes an important distinction as well; whenever you hear of explosions associated with nuclear reactors, it's always in reference to comparatively minor, conventional (non-nuclear) explosions caused by secondary fires, steam pressure, etc. Your answer is correct in that a nuclear reactor is never going to blow up like a nuclear bomb. – type_outcast Mar 14 '16 at 13:34
  • @type_outcast It's used in the inertial thermonuclear (fusion) bomb. So I wouldn't say "absolutely nothing to do" :) Of course, we're not talking about the kind of pressure containers you use to cook meat - it's usually another explosion that provides the pressure. – Luaan Mar 14 '16 at 13:35
  • @type_outcast Fukishima was different because there were still people there. Such an extraordinarily failsafed system would certainly have a watchdog that, upon realizing there are no longer people in the control room, would enter a "safe mode," or at least as safe as nuclear reactors get. Of course, if the zombie outbreak hits and then a Fukishima sized earthquake hits in the next few days, maybe the watchdog wont be tripped. – Cort Ammon Mar 14 '16 at 14:15
  • @Luaan Fusion bombs don't use pressure containers. In case you were actually talking about inertial confinement fusion, ICF isn't a weapons technology, and it doesn't use pressure containers either. Anyway, I realize you were just trying to be silly (and that's fine by me!) :-) – type_outcast Mar 14 '16 at 14:17
  • 1
    @CortAmmon Agreed! Many things different about Fukushima vs. an under-manned facility. I couldn't find a precedent for an entire reactor staff all calling in sick on the same day. :-) My main point there was simply with a highly redundant design, if the staff and related infrastructure all start to fail (because those people called in sick, too), you've already lost much redundancy, so with ~500 reactors operating worldwide, it's not too much of a stretch to expect some serious problems. – type_outcast Mar 14 '16 at 14:28
  • 3
    @type_outcast I would argue that a large part of the Fukushima disaster was caused by the infrastructure rather than infrastructure failings. Things like intentionally dumping coolant water into the sea, etc. Similarly, the Chernobyl accident was largely caused by human intervention and a Benny Hill style progression of coincidences and bad decisions. Arguably, nuclear reactor accidents would be safer with less humans around. – Jake Mar 14 '16 at 16:27
  • 8
    @type_outcast: Incorrect to say that effects from Fukushima have been felt globally, at least if you discount mass hysteria produced by news media, which of course wouldn't exist in an apocalypse scenario. Even with Chernobyl, beyond those few kilometers it's hard to tease out effects from background data without sophisticated statistical analysis. – jamesqf Mar 14 '16 at 18:27
  • 2
    The only flaw in this answer I see is the presumption that the control systems will last. Of course, they will for some time, but not always and probably not as long as the reactor needs them to. That said, reactors that can divide the fuel into two sub critical masses don't need much control to remain safe as long as they actually do this. There cannot be a meltdown without the energy to produce it. – The Nate Mar 14 '16 at 20:04
  • 1
    @type_outcast, what happened at Fukushima was that the reactors had two emergency cooling systems: a gravity-feed system and a pumped system. When the earthquake hit, the reactors automatically shut down and both cooling systems activated. This cooled the core faster than was optimal, so the gravity-feed system was manually shut down. Then the tsunami hit, disabling both the pumped system and the valves for re-activating the gravity-feed system. – Mark Mar 14 '16 at 21:04
  • I know very well what happened at Fukushima. Where I went wrong was trying to use a specific example to help illustrate a general point, in providing feedback. But I chose my example poorly. Mea culpa. – type_outcast Mar 15 '16 at 01:53
  • @TheNate The presumption is not that the control systems have to last a long time; there is only fuel stored for at most a few days. After that, the plant will shut itself down. – Rugnir Mar 15 '16 at 12:11
  • 2
    @TheNate, you and Rugnir are both wrong. First of all, fuel in the reactor can be there for 18-36 months. Second of all, the decay heat will still provide problems after the reactor is sub-critical. The reactor has to behave like long-term waste storage. – Lacklub Mar 15 '16 at 12:29
  • The piles remain radioactive and dangerous until they stop being active; while heat is a consequence, it's mostly a negligible consequence absent a nuclear cascade. "Sub critical" means precisely that. Yes, you'd have to treat the system as long term storage. – The Nate Mar 15 '16 at 16:24
  • Do any of you guys have any suggested edits to the answer to improve its information? (with sources preferably) – Rugnir Mar 15 '16 at 17:44
  • 1
    The core catcher article states "Thus, in early 2011, the two reactors of the Chinese Tianwan Nuclear Power Plant are the only working nuclear reactors with core catchers." So virtually all plants in the world currently lack them. – ceejayoz Mar 15 '16 at 20:11
  • 1
    AFAIK "core-catchers" are, as of now, science fiction - i.e. not yet in production. Your typical atomic reactor is at least 30 years old and probably past it's original design time. – Martin Schröder Mar 16 '16 at 00:03
  • "Fukushima is different" -- so, natural disasters won't happen in unmanned plants? – syck Feb 16 '23 at 18:54
31

Short-term, it's probably safer if they don't shutdown. When the generator no longer produces electric power, the cooling of the residual nuclear reactions depends on power coming from the grid - if the grid is down, the cooling fails and when the backup generators run out of fuel, you may get a steam explosion. This doesn't necessarily mean a release of radiation, and it definitely doesn't mean a nuclear explosion (that's just popular pseudoscience - nuclear power plants simply can't "go nuke").

However, in your scenario, most people are already dead. That means that even for those places that will get affected by a radiation release, the remaining people are going to have plenty of space to live on. Don't forget that the more area you "poison", the lower the radioactivity at any given place - worst case scenario, the site of the plant itself and its nearby surroundings might be dangerous, but most land already has plenty of natural sources of radioactivity that are stronger. Just avoid possibly contaminated water (which may be tricky, since you need a lot of water to run a power plant, so they tend to be close to big-ish rivers), and you'll be mostly fine. Life is actually quite resistant to radioactivity - you may get an overall increase in occurence of cancer and similar issues for a while, but nothing too major. I mean, we're talking relatively to 80-90% people outright dying basically overnight - you should really concern yourself with all the dead bodies everywhere, rather than a bit of nuclear fallout, maybe.

But let me stress again that there's not going to be a nuclear explosion. There will be very little to no direct damage to the surroundings. All we're talking about here is a (potential) release of radiation and radioactive substances. It's actually very hard to make a nuclear bomb - and nuclear power plant designers have pretty much the opposite design goals.

Luaan
  • 4,035
  • 1
  • 18
  • 19
13

The short of it, YES.

Both the radiation and poisonousness of the materials encased in these facilities will long survive their containers and especially the cooling systems.

Today maintenance is a big cost factor sometimes requiring these facilities to shut-down for a while. If you dive into incidence records even so there are many many small problems per year.

Left unattended either in a 'safe' state or not will make for a poisoned area later on always. I expect few real blow-ups. I expect all of them to poison both the immediate environment and pretty far downwind and downstream while the containment structure degrades much much faster than the important nuclear degradation times, both for power plants and weapon systems.

Not to forget an entire ubiquitous industry carrying poisonous corrosive substances in truly BIG containers.

Examples? Cooling system problems caused both Three Mile Island and Chernobyl. Achilles heel, as heat will keep coming and to stop it from building you need the facility in working order. Fukushima is another example of the bad stuff surviving where the structure goes down. (Luckily) no examples of long unattended facilities. Yet.

Better carry a good, up-to-date map.

References:
https://en.wikipedia.org/wiki/List_of_nuclear_power_accidents_by_country
https://en.wikipedia.org/wiki/Three_Mile_Island_accident
http://www.nei.org/master-document-folder/backgrounders/fact-sheets/chernobyl-accident-and-its-consequences
https://en.wikipedia.org/wiki/Fukushima_Daiichi_nuclear_disaster
https://en.wikipedia.org/wiki/Kyshtym_disaster

http://www.forbes.com/sites/realspin/2014/02/27/the-dangerous-degradation-of-the-u-s-nuclear-arsenal/#3cc993be2104

Bookeater
  • 5,361
  • 2
  • 17
  • 38
  • It's tricky, yeah. There's a lot of different things that can happen - everything might be contained (especially for certain designs), or there might be a steam explosion with or without a release of radioactivity. Some designs will also release pressure (and heat) long before a catastrophic failure, so leaking radioactivity to the surrounding environment as slowly as possible. Depending on what happens, everything might be contained on premises, or within a small area around the plant, or leak into water sources, or a more global (and less intense) release as in Chernobyl. – Luaan Mar 14 '16 at 13:33
  • 15
    I down-voted this because — I am sorry to say — this is essentially a rehash of anti-nuclear fear mongering and does not reflect the scenarios that OP mentioned.

    Fukushima and TMI where sudden, unexpected losses of power where they failed to keep the core covered during the critical first week. That caused a meltdown which caused the fuel elements to fail.

    OP said - specifically - that this downfall will take place over at least two months and leave as much as 20% of the population. That is plenty of time and people to safe the plants and get them into so called "Cold Shutdown".

     – MichaelK Mar 14 '16 at 18:54
  • 2
    You can look at Wikipedia's page on nuclear meltdowns for more information as well. I'm confident some reactors would go into meltdown globally, but I think it's a bit of a stretch to say they would always contaminate the environment. Today's reactors are engineered pretty well to meltdown "safely". Even in the event of the entire reactor crew being the first plague victims, the odds of major local devastation aren't that high. – MichaelS Mar 15 '16 at 02:56
  • "poisonousness"... Try "toxicity." – jpmc26 Mar 15 '16 at 04:44
  • 11
    Very incorrect on the "poisonousness". Wildlife around Chernobyl is in reasonably good health. Radiation levels in the Fukushima exclusion zone are lower than natural background radiation in many areas of the world. Three Mile Island produced no additional risk of cancer to anyone in the area, to the limits of measurement. As far as poisonous substances go, you need to be very much more concerned about oil in all its forms, or any chemical plant dealing with chlorine. – Graham Mar 15 '16 at 14:08
  • Thanks for the toxicity tip. The commentator's comments are truly bracing. I would SO love to believe they are right...However, as Richard Feynman once said, "Nature cannot be fooled". – Bookeater May 05 '16 at 10:25
3

Short-term will be fine if the automatic shutdown systems work. The very long-term is more of a problem. The reactors will have the usual pre-reprocessing mix of fuel and toxic waste products, wrapped in zirconium tubes, inside a steel vessel filled with coolant (usually water), in a steel and concrete structure.

All those materials are durable, but they are not forever. The weather always wins in the end, possibly hundreds of years in the future. And when it does things will start to leak slowly into the water table.

This is why waste reprocessing plans include treatments like "vitrification": enclosing it in glass. Because it remains potentially dangerous for tens of thousands of years. The design of the Waste Isolation Pilot Plant is interesting here, especially their PDF on how to design warning signs to last a hundred centuries: http://www.wipp.energy.gov/picsprog/articles/wipp%20exhibit%20message%20to%2012,000%20a_d.htm

pjc50
  • 4,900
  • 13
  • 26
  • Agreed. We can quibble about how long it will take, but ultimately weathering (chemical and physical) will eat through all the protective systems and expose the core to the environment. It may be that the really "hot stuff" elements will have decayed by then but it will emit a higher level of radiation than background. You can handle the pure metal (Pu or U) without special precautions but Plutonium Oxide (the chemical) is highly toxic in small doses. Then again, very few reactors will have Pu in them (unless they are breeder reactors making the stuff). – Jim2B Mar 14 '16 at 17:48
  • No... no no no no... the trope that plutonium is very toxic is nonsense. It falls on the same level as lead. Botulin — you know the stuff people inject into their faces as Boox? — is more than 1000 times as toxic. – MichaelK Mar 14 '16 at 18:47
  • 5
    I down-voted this as well because that claim "It gets into the water-table and then it gets everywhere" is pretty much anti-nuclear nonsense. The time it will take for the plants to deteriorate that much is measured in centuries, if not millennia. That means the main contamination concerns - Cs-137 and Sr-90 - will have decayed to insignificance. What you are left with then are the actinides.... uranium, americium and plutonium. They have the fun chemical qualities of 1) being next to unsoluable in water and 2) having a fantastic affinity for dirt and rock. They will stay on the site. – MichaelK Mar 14 '16 at 19:06
  • Eventually, the layers of concrete surrounding the core will erode. Eventually that shielding, now expised, will weather and errode. Eventually the waste will cease to be radioactive. Eventually, the wind will grind down the mountains we know. Eventually, new mountains will rise. Eventually the sun will die out. How long do these events take, though? You need to check yourself in your relative time scales. – The Nate Mar 14 '16 at 19:45
  • 3
    I have checked myself into the relative time scales in this issue. Maybe you should too. The three major contamination concerns are I-131, Cs-137 and Sr-90. I-131 is gone in much less than 10 years. I can show you the math for it if you like. Cs-137 and Sr-90 both have a half-life of ~30 years. Which means that in 300 years the inventory is down to 1/1000 of the original. And then there is the matter of actually dispersing them. The actinides are so low in intensity and so immobile in the environment that they are of little concern in this apocalyptic scenario. Other things matter more. – MichaelK Mar 14 '16 at 20:26
  • Ah. Heh. That was directed at the "tens of thousands of years" and the dangers of exposing the cores.( I.e. I was directing the comment to the answer) That same sort of weathering applies to the hills and mountains and clearly takes it's sweet time...which actually supports your point, there. – The Nate Mar 15 '16 at 16:35
3

In short: no.

EDIT 2: And the reason for this is how you phrase the question: "should the nuclear power plants remain unattended for longer period of time, they will simply overheat and cause major damage to their surroundings".

That(!) — the boldfaced part — does not happen, even in the very unlikely case of meltdowns. Not even in the worst case scenario of Chernobyl did we see that. Chernobyl is currently an unintentional wildlife preserve. Nature and wildlife are doing just fine, save for one part, known as the "Red Forest" where the fallout was so heavy it actually killed the vegetation. That however is a shining exception to the rule. The rest of the area — and Fukushima even more so — remain undamaged.

Then we can start to ponder what "damaged" actually means. "Damaged" as in "destroyed", "disfigured", "dysfunctional" or "disturbed"... no, that does not happen. But "damaged" as in "uninhabitable" or "economically unusable", that is different matter since humans are quick to abandon such areas.

However even with that definition you are going to have much bigger problems elsewhere. Chemical plants, refineries, oil wells, waste facilities and waste dumps, and — which is particularly alarming — dams of all sorts.... hydro power dams, tailings dams, coal ash dams and water regulation dams in particular. And the breakdown of clean water and sewage facilities is going to take a much higher toll on us.

And this is still while assuming that catastrophic failures do happen at under-staffed/abandoned nuclear power plants... a scenario which I am about to explain why it is not very likely at all.

The long answer: A nuclear power-plant can be shut down in seconds. Literally so. The issue then is residual decay and the heat that creates. And here is where it gets a bit curious and most people misunderstand.

Used nuclear fuel is not in a steady state where it remains at the same, let us call it "danger level", all the time and then - after a set time - click, it is suddenly turned off and stops being dangerous. Anti-nuclear campaigners aside, this is not how it works. Spent nuclear fuel starts off at insanely dangerous when you have just closed the reactor, to "worrisome" within a week, "handle with care" within a year, to "let's wrap this up and put it away" within 50 years... and then the rest of the time is pretty much just us being unnecessarily paranoid.

Spent nuclear fuel is like the embers of a recently extinguished campfire. At the beginning everything is crazy hot. But the hotter something is, the faster it cools off. So in the beginning, the activity of the fuel is intense. But the isotopes that cause this intense heat decay the fastest and therefore disappear quickly. The more time passes, only the less and less intense isotopes remain.

A quote from Blade Runner captures the essence of this:

"The light that burns twice as bright burns half as long"

The critical time where you need to keep actively cooling the fuel elements — in order that the residual heat will not melt them — is about one week. After that you need to keep them soaked but they will not boil away the water.

Within a month you can open up the reactor, take out the fuel elements and put them in the storage pool. Natural circulation is more than enough to keep the fuel elements nice and snug and safe to be around, as long as you top up the pool. Not because the water keeps them from melting but because water is an excellent radiation shield.

So with your scenario of 2 months until the apocalypse and 10-20% of the population remaning, nuclear power plants will have plenty of time to safe their fuel. If the worse comes to happen and some plants are left without the resources to safe them this thoroughly, they only need to keep the pumps going for a week, then leave it filled with water. The reactor pressure vessel and the containment will handle the rest.

In the long run we are then left with sites of spent fuel elements in pools, dry storage on the surface or in reactor vessels. Does that present radiation hazards?

Not really no. Sure there may be some local contamination from damaged fuel elements, but unless someone deliberately goes in there and starts lifting elements out of the pools and try to break them, the fuel cladding, the pools, the reactor vessels and containment buildings will keep the nasties — I-131, Cs-137 and Sr-90 in particular — inside. That is after all why they are there.

Sure... the "Irradiated Wasteland" trope is very popular and an effective plot generator. But if you are going for "reality check" here, then it will not happen with your scenario. If you desperately want to use it, the downfall will have to be much faster.

And — again — then your problems will be much larger elsewhere.

EDIT: In a pre-Fukushima scenario, then the disaster scenario where nuclear plants blow up left and right might have been slightly credible. Post-Fukushima however it becomes outright nonsense. Not only has Fukushima set the baseline for what a nuclear plant must be able to handle — a sudden and catastrophic loss of both cooling and emergency cooling — but other measures to mitigate the damage done by a meltdown have also been put into effect. Two of them are especially noteworthy...

  • Hydrogen re-combiners. Noted already during the Three Mile Island accident, accumulation of hydrogen in the containment is a major issue. That may end up exploding, as it did(!) during both TMI and Fukushima. The solution to this is to install passive hydrogen re-combiners. These are catalyzers in the ceiling that makes hydrogen combine with oxygen — in a non-explosive manner — and become water again.

  • Release filters/scrubbers. The other issue identified by TMI was the need to be able to vent containments in a controlled manner. This is actually rather simple to achieve, using scrubber pools and stone filters, which can absorb up to 99.9% of the substances of most concern.

Some counties had already after TMI started employing these countermeasures. Here is an example from Sweden, the Barsebäck Nuclear Power Plant. The cylindrical structure to the left of the two reactor blocks is the release filter, known as FILTRA.

enter image description here

So — again — the issue will not be with nuclear power plants because they are, compared to the rest of our civilization, ridiculously well prepared for disaster compared to some other problem areas.

A final note: as someone both invested in the debate on nuclear power and nuclear technology in fiction, I personally feel that it is long past due that we got over tripe such as The China Syndrome, or the TV series "24"...

"Eeep! Terrorists stole the McGuffin that controls all our nuclear power-plants and caused all 104 of them to start melting down, and we can't stop it unless we get the thingie back! Oh noes!! It happened in a few sites, and everyone around them are now dead!!!"

...because this trope is as silly and unrealistic as to assume a light drizzle over New York causes the city to disappear under six feet of water within a few hours. Nuclear tech in fiction has been assigned magical properties of the blackest sort for the past 60 years, and it is time we got over it and stopped using nuclear as a lazy plot-generating device in apocalyptic/dystopian fiction.

MichaelK
  • 43,723
  • 6
  • 106
  • 189
3

Yes, they will

The critical event is loss of electrical power. After this event, diesel generators will start up (hopefully! there were examples of failure ...) and an automatic shutdown will be performed.

However: The ongoing nuclear decay in the fuel will demand further cooling. The diesel for the generators is limited (typically, for one day of operation). When the generators stop because of lack of fuel the nuclear power station will start to destroy itself.

Fukushima is an example of exactly this scenario: Power outage, diesel generators destroyed by the tsunami, self-destruction of the nuclear power plants. The human operator teams present at the site weren't able to stop it, because they could not restore the cooling in time.

Sir Cornflakes
  • 3,507
  • 15
  • 28
  • Your scenario hinges on the assumption that when the outbreak occurs — which OP says happens over 2 months from the start to having 10-20% survivors — electrical power suddenly and unexpectedly goes away. How did you come up with this Non Sequiteur? Also you assume that the generators cannot keep running until Cold Shutdown has been achieved. Where does that assumption come from? Essentially you are saying that there will be a sudden power loss and that no nuclear power station can deal with it. Those are some extremely pessimistic assumptions. – MichaelK Mar 15 '16 at 15:03
  • @Michael Karnerfors: When some virus is killing 80% or more of the total population, the survivors will probably have different thoughts and plights (think of burying the dead a caring for the ill ones) than keeping the electrical network up and running and fuelling the diesel generators of the Nuclear power plants. It is only realistic that electrical power suddenly (maybe not unexpectedly) goes away. A nuclear power station can deal with power loss, but only for a certain time (a few days in the best case). – Sir Cornflakes Mar 15 '16 at 15:10
  • Before Fukushima your scenario could have had some credibility. After Fukushima however, I am sorry but your scenario is outright unrealistic. Not only has emergency cooling been heavily scrutinized across the world, but mitigation measures for when a meltdown do(!) occur — such as hydrogen re-combiners and release filters — have also started to be utilized to much greater extents. Fukushima was a instant event that knocked out both cooling and emergency cooling in one stroke, which has set the baseline for post-Fukushima nuclear power. OP's scenario is not instant, but gradual. – MichaelK Mar 15 '16 at 15:24
  • @Michael Karnerfors: Sorry, but your arguments may hold water for Nuclear Power stations desinged and built after the Fukushima accident. But most of the Nuclear Power stations all over the world are decades old and later mitigations don't bring them to a state that they can control themselves for approximately one month (which would be probably sufficient for a safe shutdown, but we don't have an actual demonstration that nothing goes wrong. Some passive systems at Fukushima did not work as designed.). – Sir Cornflakes Mar 15 '16 at 15:34
  • Well I guess you better get on the horn then and inform all the nuclear regulatory authorities around the world — and their counterparts in the nuclear industry — that what they have been doing for the past five years is completely inadequate... for the sole reason that such an assumption makes for a convenient plot-generating device in a fictional world. I am sorry if this comes off as harsh but the trope of "My Nuclear Is Going Critical! The World IS DOOMED!!!" is ridiculous and totally over-used both in fiction and the public discourse since the 1970's. Get over it already. – MichaelK Mar 15 '16 at 15:44
  • @Michael Karnerfors: The basic assumption behind all nuclear safety is that nuclear power stations are attended. Leaving them unattended is not part of any safety plan implemented anywhere in the world. – Sir Cornflakes Mar 15 '16 at 16:09
  • Let us continue this discussion in chat. – MichaelK Mar 15 '16 at 16:11
2

Yes, the landscape is scattered with nuclear debris.

Most nuclear power plant designs will fail spectacularly with long lasting effect on the surrounding area.

The notion that nuclear reactors can run for extended periods of time without human intervention is simply not founded in reality.

Keeping a nuclear reaction from becoming critical is a feat within itself. It's balancing act of maintaining a state that is less than critical.

It will fail within a few months to a few years working under the assumption that it is not getting it's cooling water replenished which is entirely plausible.

Most nuclear power plants in the United States require are light water reactors or pressurized water reactors and require the circulation of water in order to stay within a 'steady state' of operational limits. The two failure scenarios that come into play under an unattended state both involve overheating and a change of state to critical. Other incidents occur on the timeline to critical such as hydrogen explosions but these events are essentially ancillary and aren't central to root cause.

Background

There are two types of nuclear reactors that can suffer from water related failure.

Failure Scenario - Lack of Water

Both water cooled reactor types suffer cataclysmic failure due to a lack of water. Some reactors can be more robust than others depending on the design. Many reactors derive their cooling water directly from their environment using ocean, lake or river water. These reactors are prone to having their water intake ducts clogged with debris thereby restricting the flow of cooling water to their segregated cooling systems. A lack of human intervention in these reactor types can lead to failure.

Cooling towers are used with reactors that are not in close proximity to ocean, river and lake water. Palo Verde Nuclear Power Plant in Arizona is one such example as was Three Mile Island.

In the event of an overheating reactor in the United States the Federal Government only requires a 30 day supply of cooling water. This cooling water, called a UHS (Ultimate Heat Sink) is a finite source and dissipates over time due to a number of reasons including evaporation, steam release and lack of re-circulation of primary cooling circuits due to radiation (the water for cooling is used one time only, in some cases, due to the amount of radiation contamination of the water)

As the water supply for cooling dwindles and water pressure decreases enough for flow rates to diminish beyond preset thresholds, the reactor undergoes an automatic shutdown called a SCRAM. A SCRAM event does not require electricity. Neutron absorbing control rods are held in place by electromagnets above the fissile pile and upon loss of electricity the electromagnets lose their magnetism and the rods are dropped into place bringing fission to a near halt in the core. These systems are automated and do not require human intervention, however, the continued decay heat of the fissile material continues to create issues with cooling and with a finite water supply, eventually, the reaction pile becomes exposed to air which cannot cool the fissile rods due to simple lack of density.

At this point water changes it's state to a gas and becomes steam. The heat increases further and more steam and pressure buildup in the chamber. Water becomes superheated and takes on properties and attributes more akin to an organic solvent. The pressure is so high within the chamber that it eventually prevents the water from boiling. This superheated water's hydrogen bonds are eventually broken and the chamber becomes filled with superheated highly pressurized hydrogen which eventually explodes due to combustion or the failure of the reactor pressure vessel to contain the extremely high pressures. This failure has been made famous by the hydrogen explosion at the Fukushima Daiichi reactor containment vessels in Japan due to the tsunami which did not damage the reactors, it damaged generators that prevented the circulation of water for the cooling systems.

The loss of integrity to the reaction chamber is the penultimate step to the catastrophe, all systems are essentially destroyed and now the fissile material can become molten and eventually melt through any concrete casements that require cooling features in order to prevent the molten nuclear pile from actually melting through it. Once this molten pile of nuclear material hit's moisture an explosion can occur sending nuclear debris into the atmosphere and contaminating the surround landscape with fallout. Mind you, this isn't a nuclear explosion it's just an explosion....but you see what the problem is here, it's called a meltdown.

Conclusion

Water inlets that use water from the environment (rivers, lakes, oceans) for secondary cooling systems require regular maintenance to prevent debris from clogging their inlets.

US located closed loop cooling reactors require only 30 days of backup cooling water.

Manual events such as steam or pressure release to prevent containment vessel explosions won't be occurring. There are disaster prevention events that require human intervention. See this IAEA root cause analysis document and salivate over it's plethora of manual events and whether staff followed procedure and a logical decision tree before, during and after a disaster.

I will close this scenario with, there's nuclear waste all over the place.

Other Failure Scenarios

There are hundreds of nuclear reactors across the world. Educational and research reactors are sprinkled across the topography of nations. Some are mercury and graphite cooled. Others use molten salts and even molten sodium. Fast reactors requires considerable human intervention and are used primarily in naval ship propulsion and in some cases the production of electricity in Russia. There are a plethora of scenarios where these reactor types can go bad in their own unique and beautiful ways.

All in all in this scenario that I put forth is feared. There are many other failure points that I have failed to mention. The list is too long and nuanced. With the framework you have presented.

Here is the official US Government Nuclear Reactor Regulation Response Plan to Pandemic Notice the emphasis on staff and staff skills and manning the facility to prevent failure. With all the dead engineers it will be difficult to staff these facilities, chaos will ensue attempting to man the reactors as they slowly begin to fail one by one.

So to answer your question again.

Yes, the landscape is scattered with nuclear debris.

Community
  • 1
Citizen
  • 472
  • 6
  • 16
  • 1
    The poster set the conditions that the apocalypse would happen over 2 months, with 10-20% surviving, not that everyone died within 2 minutes of the outbreak. Getting a reactor into so called "Cold Shutdown" takes at most 1 week, because the decay heat fades quickly. The 30 days of which you speak is ample margin to get into Cold Shutdown. After that there will not be meltdowns and hydrogen explosions. Your scenario is only ever viable if 1) people died instantly or 2) people figuratively stick their head in the sand and keep running the reactors at full speed, pretending as if all is well. – MichaelK Mar 15 '16 at 08:48
  • This also suppose that there are no failsafe for unsupervised shutdown. – MakorDal Mar 15 '16 at 14:49
  • 1
    This is incorrect in almost every way. The NRR Response Plan is for maintaining regular NRC operations (responding to license amendment requests from the plants, performing evaluations, etc) and has nothing to do with staffing of the plants. The cooling tower part is irrelevant because when the reactor is shutdown (which will happen automatically at some point) the cooling tower is not even used--it's just there to help extract more energy from steam and minimize impact on surface water temperature during operation. – nvuono Mar 15 '16 at 20:01
2

Damage, yes. Major damage, no.

We already have a perfect example of what will happen: Fukushima. That happened because the reactor was tripped and not getting any power to deal with the residual heat of the nuclear fuel.

The computer should be able to keep things running until something goes outside acceptable operating parameters. (This will inevitably happen as the fuel gets too contaminated with decay products even if nothing actually breaks.) The first reactors to fail will be ok--they will continue to draw power from the grid for cooling.

Eventually, however, too many will fail. One of the things that will cause a trip is the lack of two separate sources of power. As too many plants on the grid fail this will eventually cause a cascade that takes down every remaining nuclear plant. They'll fall back to local generators to keep the fuel cool--but eventually those run out of fuel. The rods boil their storage dry and you get a mess.

In practice the failure will happen pretty quickly as many of the external sources of power they rely on are fossil fueled--and those will shut down pretty quickly due to a lack of fuel.

All the safety improvements are based on buying enough time for the emergency crews to put things back together properly--if the emergency crews aren't coming they'll eventually fail.

Loren Pechtel
  • 30,827
  • 1
  • 37
  • 92
  • Fukushima is a bad example, because containment efforts started immediately and are ongoing five years later. Without human intervention, the failures of the cooling systems may have had much worse consequences in terms of radiation released. – user2727 Apr 26 '16 at 08:32
0

Some misconceptions here. Most reactors would scram (automatic shutdown). But scram is not a fail safe. It's merely one safety measure. It's assumed humans step in and do the rest. The reason for this is time. It takes months if not years to bring a reactor that has been running back to a safe temperature. It's a full time management process, not a light bulb. That won't happen in a catastrophic situation so the core will start to heat up. Meltdown and breach of containment will occur within days followed by massive release of radiation. Some people talk about radiation like you can escape it being say 100kms away. No you can't. One reactor can spread deadly radiation over an enormous area. That's just one reactor! Seriously if people really knew how dangerous these things were there would be riots in the streets.

Strangelove
  • 17
  • 1
  • 1
    I am sorry but now you are speaking complete nonsense. Every year, LWRs are undergoing what is called "revision". It's like a yearly checkup where inspections, maintenance and — most importantly — defueling, refueling and rearranging the core is done. A quick revision is done in about a month. So by then the reactor has been turned off, gone into cold shutdown, been opened up, had its fuel elements manipulated (some of them moved to the storage pool), closed, then started up again. So you are making stuff up here, just because you want to portray nuclear power as evil black magic. Not cool. – MichaelK Apr 28 '16 at 13:00
  • Having lived in europe during Chernobly, the guy has a point. And it wasn't the active rods in Fukishima, it was the cooling pools for the spent rods. The reactor did SCRAM. It wasn't the reactor that caused the explosion, it was the spent rods that were cooling down in the adjacent pool that caused the hydrogen buildup. – James Aanderson Dec 31 '18 at 05:00
-5

The containment will be breached within hours, not hundreds of years. The decay heat will melt through the containment like it's butter. The containment is designed to contain the core under circumstances where it's being cooled. Nothing can contain that heat without active cooling. It's laughable what is being suggested here.

Jim Bob
  • 1
  • Hello Jim, can you please elaborate on your answer? Currently I can not see, what makes it different from these answers: http://worldbuilding.stackexchange.com/a/38114/13987 , http://worldbuilding.stackexchange.com/a/38087/13987 and http://worldbuilding.stackexchange.com/a/38026/13987 – T3 H40 Apr 28 '16 at 05:45