The Story

This is one of three questions set in the exact same place, so bear with me here. Twenty years after the fall of the State and the Overseers, about thirty since the phase-gate to Ilus was opened, we follow a technician who is hired by the Sequoia R&D Alliance to work at this lab, Kitsuki. (Mostly as a chance to get away from the political mess that is earth after the fall.)

The inciting event is a malfunction of the fusion reactor, which eventually turns out to be sabotage, and frees the Typhon and Kit infected with a retrovirus that is part of the Keidran genetics program. And gets turned fluffy.

The Question

As a generalization, how good would fusion reactor be as a bomb? If you were to watch the TV incarnation of The Expanse, you'd be surprised by the bang the Donnager made when Yao scuttled it. Is that even close to realistic? I would assume that no, you need a proper bomb to make a proper explosion.

But as a generalization, looking at what a possible future ICF reactor does and how, could you even try to weaponize or sabotage it? Or should the saboteurs of the labs enormous power source (for the massive amount of quantum computers) be better off just getting a nuclear warhead down there and blowing the place the old fashion way.

Basically, if I wanted a boom, what do I do.

  • 1
    $\begingroup$ I assumed that the Donnager used a fission type nuclear explosion, not some hack of their reactor / warp core breach dealy. $\endgroup$
    – Willk
    Commented Feb 4, 2023 at 18:06
  • $\begingroup$ @Willk maybe they had explosives built into the reactor... somehow, as in the books they explicitly say it was the fusion reactor. IIRC. $\endgroup$ Commented Feb 4, 2023 at 18:12
  • $\begingroup$ @Willk Maybe they had conventional explosives mounted on the reactor, as real maritime warships have them too, but instead they are fixed all across the hull to sink the ship. BTW how is my big-web-o-questions looking? I practically wrote a story with them all. $\endgroup$ Commented Feb 4, 2023 at 18:18
  • 10
    $\begingroup$ The thing with nuclear fusion reactors is that nuclear fusion reactions absolutely cannot proceed in conditions where humans could even dream to survive. If anything bad happens to a nuclear fusion reactor, the inside of the reactor is immediately exposed to normal human conditions and the nuclear fusion reaction instantaneously shuts down. The only amount of energy which can escape is the amount of energy which was being produced at that very moment, which is about the same as the power rating of the reactor. If we could ever make them work they will be extremely safe power sources. $\endgroup$
    – AlexP
    Commented Feb 4, 2023 at 19:40
  • 3
    $\begingroup$ @AustinHemmelgarn - but all of those are situations where an enormous amount of stored energy were released all at once. Fusion reactors have to create very precise conditions to convince their fuel to give up its energy, so they make very poor bombs. $\endgroup$
    – jdunlop
    Commented Feb 6, 2023 at 18:27

10 Answers 10


Not significantly more than the reactor was designed to release in normal operation.

You mention ICF reactor, so I'm assuming you are referring to inertial confinement fusion reactors - an upgrade of the Lawrence Livermore National Ignition Facility that's been in the news for the first over unity fusion reactor.

Something like this is a very complex engineered solution. The facility is supposed to be 3.5 billion USD, and the output of the event was about 3 MJ (enough energy to boil 14 kettles of water). 1 Kg TNT releases about 4.1 MJ so, not much of an explosion as clearly the machine would necessarily be designed to handle such events frequently for a long time to be a practical reactor.

So you want a much bigger bang, there are only 2 possibilities 1) you increase the yield dramatically, 2) you implode a much larger pellet.

  1. Increasing the yield dramatically won't be possible to rig a much larger bang - if you knew how to do that, you would already have incorporated that in your design. Also your future ICF reactor would not be practical if the yield could be increased by orders of magnitude even with a magical 100% yield.

  2. A much bigger pellet. This won't even ignite (maybe a fizzle if your lucky and achieve some fusion). Your complex reactor simply won't be able to heat a much larger pellet fast enough to reach ignition.

Suppose you decide to use a pellet 5 times the diameter of the production pellet in the hope of achieve a bang 125 times normal. The surface area of the pellet will be 25 times the standard. There is absolutely no possibility that a practical design could be over-driven sufficiently to trigger fusion event that would require 25 times the normal energy input.

Controlled fusion is a very non-linear process, you need a sufficient combination of density, temperature and confinement time to achieve. Missing target conditions means yield drops dramatically.

Also note that making a practical ICF reactor would require major improvements to the event reported on 2022 Dec 13. Such as improving the cycle time for once per day to something like once per second (or faster), and the fact that the net energy loss was roughly 100:1 as the over unity figure was based on the amount of energy delivered to the pellet - not the total energy required to deliver energy to the pellet.

But, for a story - maybe there is a way.

Use the pellet as it was designed to initiate a fusion reaction.

Encase the pellet in a much larger shell, they has very precisely location holes in the shell. Holes that match the location of the incoming lasers so the lasers pass through the shell without affecting it, and causing the normal fusion reaction.

This fusion reaction then ignites the shell and you are ready for a big bang.

In real-life this is sort-of similar to a thermonuclear bomb that is ignited by a fission bomb. Note that this process requires a fission bomb, not something more like a stick of dynamite that the pellet represents. I don't expect any possible future where you ICF reactor could ever generate fusion events remotely close to being large enough to ignite an outer shell for a big bang.

OP asked in comment - Would all the radiation from a reactor do if you cut off the shielding system, which would have been the superconducting magnets that contain the core plasma?

In a true ICF reactor, there are no superconducting magnets that confine the core plasma - the plasma is "confined by inertia" - the pellet (or futuristic equivalent) simply can't explode fast enough to prevent fusion. The conditions that enable fusion (temperature and density) for ICF are so extreme that the time of containment, though measured in nanoseconds, is still sufficient for high-gain fusion, i.e., far in excess of the conditions at the core of our sun.

You could have magnetic confinement, not to confine the plasma for the purpose of fusion, rather to keep the plasma from damaging the walls of the reaction chamber one the plasma has passed the ICF stage.

I don't expect to see a hybrid reactor that combines ICF and magnetic confinement, I believe this would in practice be a magnetic confinement reactor that uses a novel form of plasma heating - except for magnetic confinement you need the plasma flowing in channels - not exploding from a central point. You would need something more like Star Trek style force fields (beyond known physics) that can contain a plasma that is simply exploding outward.

Though a 2nd question - I thought this was also interesting.

  • 1
    $\begingroup$ As to the answer, crud... As to tech advances, of course, we are nowhere near then, but if you looked at the linked questions, this is set about 200 years off from now, with massive leaps and bounds in tech. As to another question, what is the worst you can do to destroy a reactor if you really wanted to? What if you collapsed the magnetic bottle or try to fry the lasers or something like that? $\endgroup$ Commented Feb 4, 2023 at 18:43
  • $\begingroup$ As for the specs of the machine, it uses lasers, helium 3 and deuteron pellets, which are standardized and solid, fed into the machine from cases, just like in The Expanse. It uses magnets and the Lorenzo force to reclaim power, and assumes some enormous breakthroughs in laser optic efficiency (maybe so far as 50% efficiency, but its never outright stated) and so the reactor has a prodigious power output. $\endgroup$ Commented Feb 4, 2023 at 18:44
  • $\begingroup$ As to your comments, I generalized to ICF overall since that is what the question was actually asking. And making the answer standalone without additional context to those that never watched The Expanse. It is often fairly easy to rig high power equipment to self-destruct in some form or another - after all there is a lot of power available to se for destruction. $\endgroup$ Commented Feb 4, 2023 at 18:52
  • $\begingroup$ Well... if you did happen to know, what would all the radiation from a reactor do if you cut off the shielding system, which would have been the superconducting magnets that contain the core plasma? Would it cook the entire floor and melt people into the bulkheads? Or something a bit less dramatic. $\endgroup$ Commented Feb 4, 2023 at 18:58
  • 1
    $\begingroup$ You can't "shut off the shielding system", as the whole means of capturing energy from a fusion reaction is the shielding system, so it's integral and a hardware system. $\endgroup$
    – jdunlop
    Commented Feb 6, 2023 at 18:21

Breaching the reactor will let air in, leading to an immediate end to fusion. Some hot things may catch on fire. The reactor may occasionally make louder noises in normal operation. The reactor operators may wish they were dead in a massive explosion when they hear the whistling of oxygen-rich air entering the reaction chamber, though, since that's going to take a lot of work to clean up and Important People are going to be asking questions and looking for someone to blame.

For an actual explosion:

  • Starting from a cold reactor, run an oxygen line from the life support system, and feed in hydrogen or some other flammable gas from...well, find something. Build up as much pressure as the system will take (which won't be much, it's intended to house a near vacuum) and try to start the reactor. This should get you a reasonably loud noise and a heavily damaged reactor.
  • Sabotage the fuel storage. A D-T ICF reactor will use pellets filled with frozen deuterium and tritium, which are both essentially as reactive as normal hydrogen, and a boiling point only 20-some Kelvin above absolute zero. Look at Fukushima...a little loose hydrogen can easily blow the roof off. Bonus if the reactor uses tritium: contamination of the vicinity with radioactive material that will greatly interfere with short-term cleanup.
  • Engineer a massive failure of the superconducting magnets used to convert the bursts of expanding plasma into useful power. A superconducting magnet can "quench": part of it ceases to be superconducting, and the resistance converts all the energy stored in the magnet into heat. Controlled quenching is used to condition superconductors for use, but uncontrolled (or maliciously controlled) quenching can be quite destructive. The rapid collapse of the magnetic field itself can rip things apart (especially if it's supposed to stay in balance with another electromagnet), and the heat can rapidly boil off coolant in a BLEVE. And the boiling coolant displaces air, leading to an asphyxiation hazard that will require the area to be quickly evacuated.
  • 2
    $\begingroup$ I used to work with superconducting magnets. A 12 Tesla field has the same amount of energy per cc as TNT, but virtually no mass. There is a lot of energy. The magnets are typically shorted with a large shunt resistor at the top. If the cooling fails, this will glow brightly for many seconds as the superconductivity fails, and the energy is dumped into the resistor. This is still destructive but better than the alternative. I don't know how they do this with really large superconducting magnets: disconnecting a shunt or two might cause a chain-reaction quench. $\endgroup$ Commented Feb 5, 2023 at 11:01
  • $\begingroup$ @RichardKirk Or cutting the coolant supply to the entire reactor by, say, cutting holes in all the feed lines, it will asphyxiate everyone in the bay (if its liquid argon or helium) and then make the reactor implode and melt as it quenches, probably spewing out plenty of radiation in the process. $\endgroup$ Commented Feb 5, 2023 at 13:23
  • $\begingroup$ @RichardKirk Yeah, an uncontrolled quench triggered by a faulty connection happened early on in the LHC's lifetime, which resulted in quite a bit of damage despite a similar quench protection mechanism being in place (home.cern/news/press-release/cern/…). $\endgroup$ Commented Feb 5, 2023 at 14:40

Short out the capacitors

The one thing fusion reactors have in common, whether tokomak or ICF, is a massive start up power requirement.

A plasma fusion reactor may just need one burst of power, but a ICF needs periodic burst power, each time a pellet enters the chamber. This, for even our test rigs at the moment, means an absolutely giant bank of capacitors, flywheels, and all kinds of ways of storing power. A commercially viable one needs an even bigger bank.

Sabotage this, and, well, have you ever done that experiment in school, where you short out a fully charged capacitor, and the little thing explodes? Ramp that up by millions of times. It may not be a thermonuclear explosion, but it'll be a pretty good sized conventional one, and leave an absolute mess of burning electronics.

A high tech future fusion reactor would be likely to have advances in capacitor and energy storage tech, making the whole thing more energy dense, and therefore more explosive

  • $\begingroup$ I would point out that even in modern hi-density capacitors there are all sorts of measures to prevent the cap from popping like an overcharged motherboard component. Future, more energy-dense capacitors would likely have similar engineered mechanisms to prevent catastrophic failure. Not to say that they might not be possible to subvert, but I don't think it'd be an easy equivalency to letting the magic smoke out of a home electronics project. $\endgroup$
    – jdunlop
    Commented Feb 6, 2023 at 21:42
  • $\begingroup$ I'd agree with this, but I feel like as stuff gets more software defined, the possibility for doing this with something like the Stuxnet worm goes up $\endgroup$
    – lupe
    Commented Feb 6, 2023 at 22:28

As much as the movies want to convince you otherwise, explosions are hard. Do you know what the difference is between a charcoal briquette and a quarter stick of dynamite? Time. They both release the same amount of energy. Scale this down to a firecracker, and the only reason it goes bang is because the wrapper holds the gasses in until there is enough pressure to break the wrapper.

There is a huge structural difference between an explosive and a reactor of any kind. With nuclear weapons, for instance, the fissiles are compact into a small space and held there with a conventional explosion until most of the mass has had had the opportunity to convert to energy. In an equivalent power plant, the fissiles are spread out among moderators, coolant, and control equipment. You couldn't get it compact enough to blow up if you tried.

With the hydrogen bomb, things are even worse. They use a conventional bomb to set off a plutonium bomb, and use THAT to set off the hydrogen bomb. It takes an immense amount of temperature and pressure to make the hydrogen fuse, and the biggest challenge is maintaining that pressure for long enough for the hydrogen to fuse.

Over and over again, we go back to the firecracker that needs a wrapper to be an explosive. Power generation equipment always gets in the way.

Let's look at what we think might be a scaled down equivalent: a car exploding. Note that this isn't the engine exploding, it's the fuel tank, and even that only happened with frequency in the movies.

Spontaneous explosion is a quality that is peculiar to chemical energy. That's the basis of all conventional explosives. With all other power sources, it's a challenge to get enough potential into a small enough area.

  • 2
    $\begingroup$ Correct. Just as it's impossible to turn a nuclear powerplant into a nuclear bomb, so it is impossible to turn a fusion reactor into a fusion bomb. And because of the general lack of radioactive material in a fusion reactor you can't even turn it into a hellfire of burning fission products spewing out radioactive fallout like happened at Chernobyl (and which with any sane reactor design is impossible, as shown by TMI and Fukushima among others, where no breach of the reactor vessels happened despite meltdowns). $\endgroup$
    – jwenting
    Commented Feb 7, 2023 at 7:40
  • $\begingroup$ @jwenting well you can, just dig a tunnel underneath the facility and place an actual fission/fusion bomb there ;) Jk, but that'd be pretty much the only way to make an explosion bigger than a couple of magnets crunching each other $\endgroup$
    – Hobbamok
    Commented Feb 8, 2023 at 9:29
  • 1
    $\begingroup$ @Hobbamok that'd still not turn the fusion plant into a bomb, it'd just blow up the fusion plant. $\endgroup$
    – jwenting
    Commented Feb 9, 2023 at 8:33
  • $\begingroup$ @jwenting totally but I wanted OP to have some sort of decent-sized explosion in the end. And the stuff usually in a fusion reactor is not really gonna deliver $\endgroup$
    – Hobbamok
    Commented Feb 9, 2023 at 13:12
  • $\begingroup$ Yea, that's not "fusion reactor exploding," it's a scuttling charge. The Donnager literally carried nuclear bombs, and they would have provided a bigger boom, and a more even distribution of damage. However, OP didn't ask how to make one blow up, he asked if it was realistic. $\endgroup$ Commented Feb 9, 2023 at 16:52

Lard up your fusion with some fission. And antimatter!

Yup yup yup. Reactors give off energy little by little. When they break they stop giving off energy.

But: how are you powering this fusion reaction? The ones we have on Earth compress regular small atoms until they fuse and then the energy is used to generate electricity which then charges our earth toned electric scooter.

Perhaps though you are using fusion in a spicier way, and inviting its rowdy brother Fission and their weird uncle Antimatter to this party?


The power of fusion

The fuel for such a fusion-driven spaceship would likely consist of many small pellets containing deuterium and tritium—heavy isotopes of hydrogen that harbor one or two neutrons, respectively, in their nuclei. (The common hydrogen atom has no neutrons.)

Inside each pellet, this fuel would be surrounded by another material, perhaps uranium. A beam of antiprotons—the antimatter equivalent of protons, sporting a net electrical charge of minus-1 rather than plus-1—would be directed at the pellets.

When the antiprotons slammed into uranium nuclei, they would annihilate, generating high-energy fission products that ignite fusion reactions in the fuel.

This scenario has boom potential. Nuggets of fissile material all ready to go boom when the antimatter comes. "Don't store the antimatter next to the uranium nugs!" they said. But there was no room for the foosball table and so the nugs got relocated to the antimatter closet. Then when the antimatter got loose the uranium nugs were right there.

  • $\begingroup$ Assume anti-matter is the objective, rather than as a continuous mains supply, as anti-matter provides a means of storing huge amounts of energy for time when larger bursts are needed (Alcubierre drive-ish warp fields?). All that anti-matter on the Enterprise has to be coming from somewhere; fusion reactors seem an ideal place to start. Once you have a large store of anti-matter, the Fourth of July is going to be spectacular. $\endgroup$
    – JohnHunt
    Commented Feb 6, 2023 at 2:40

Not big at all. Fusion reactors exploding is a popular media phenomenon that does not match the actual science.

Fission reactors are known for their potential to melt down because all it takes to start a fission reaction is assembling a big enough piece of enriched uranium in one place. Once you've done that, the reaction is self starting, self sustaining, and even exponentially amplifies itself for as long as sufficient fissile material continues to be present. Because of that, one of the most major parts of any fission reactor's design is ways to limit, control, and stop the fission reaction.

Fusion bombs are known for producing powerful explosions because they are specifically designed to explode powerfully. They produce the conditions needed for fusion in a destructive manner by detonating lesser explosives in extremely precise calculated arrangements, and the fusion only lasts a brief moment before the necessary conditions dissipate. The result is so powerful because the total lack of any need for the device to remain intact makes it feasible to design it so that the momentary fusion goes far, far beyond the bare minimum conditions for fusion.

Authors of science fiction stories took the dangers of fission reactors, and the relative power of fusion bombs vs fission bombs, and assumed that using fusion instead of fission would amplify the danger of a reactor in a ratio similar to how it amplifies the power of a bomb.

In reality, however, the sustained non-momentary fusion reaction needed for a fusion reactor to produce power requires great effort to actively maintain the conditions under which fusion can occur. The difficulty of making fusion happen at all, not the danger of it, is what makes useful fusion reactors so difficult to create.

A fusion reactor running at peak capacity is already creating and maintaining the strongest fusion reaction that the machinery is capable of. Malfunctions or sabotage would only make the fusion reaction weaker, or stop it completely.

An extremely abrupt rupture of the reactor might result in a small explosion, but the power involved would be no greater than what the reactor ordinarily keeps contained, and it would stop almost instantly. You would get a lot more destruction done by hooking up the power output of the reactor to an actual weapon designed to use that power destructively.

  • $\begingroup$ mind that fission reactors ALSO can't be turned into nuclear bombs. And a sanely designed one can't even turn into a conventional bomb (even TMI didn't do that, though because of severe operator error there was a serious buildup on hydrogen gas in the reactor vessel, but never enough pressure to rupture it). $\endgroup$
    – jwenting
    Commented Feb 7, 2023 at 7:41
  • $\begingroup$ @jwenting Chernobyl did produce an actual explosion with vaporization of some core material, but it took a terrible design being terribly mistreated to achieve that. And the Kiwi TNT test involved rigging a Kiwi nuclear thermal rocket core to explode, which required modifying the control drums to rotate at 4000°/s instead of the normal 45°/s. The resulting explosion...was equivalent to about 90-140 kg of black powder. $\endgroup$ Commented Nov 12, 2023 at 2:53
  • $\begingroup$ @ChristopherJamesHuff that wasn't a nuclear explosion. It was a steam explosion, completely different thing. The idea that a nuclear power station is a nuclear bomb is utterly and completely wrong at every level. $\endgroup$
    – jwenting
    Commented Nov 12, 2023 at 13:45
  • $\begingroup$ Chernobyl was predominantly a steam explosion, with some contribution from hydrogen formed in the process, but it's not accurate to say the nuclear reaction didn't contribute. Kiwi TNT was purely a nuclear explosion. I'm not saying power stations are nuclear bombs, I'm saying that achieving such an explosion requires mishandling of an inherently unsafe reactor design (being simply impossible with most reactor designs), or deliberate modification, and even then the explosive yield is utterly pathetic, in Kiwi TNT's case equivalent to a fraction of its mass in black powder. $\endgroup$ Commented Nov 19, 2023 at 15:45

Not the reactor itself.

Many people have mentioned that fusion reactions won't just happen outside of a reactor and it takes a huge amount of engineering effort to make the reaction happen at all, unlike a lump of plutonium which will explode if you squish it fast enough, or hydrogen which occurs as a byproduct form overheating fission reactors.

So instead of damaging the reactor and having it blow up, leave it running. Boost the power output as high as it can go without breaking.

Then (having put nails in all the fuse boxes earlier) short-circuit the output. Blow up the transformers. Melt the wiring. Make some giant sparks.

Or jam the voltage regulator on the high setting so all the lightbulbs and appliances in the whole building go bang.

None of that is going to blow the place apart, though. It's a crazy scientist's lab, right? What else is in here? Got some spare phase gate parts? I'm sure one of those holophasic gluon field thingies has plenty of destructive potential. Okay, that's a bit far-fetched.

But wait, fusion reactors run on hydrogen - an explosive invisible odourless gas. Mix it with air in the right proportions and ignite it, there's your explosion. Storing hydrogen is tricky, so this facility doesn't store it. It stores water, and extracts the hydrogen as needed, using electrolysis. You sabotaged this and left it running for months, so you have all the explosive gas mixture you need. You can't let the whole lab fill with hydrogen over time, though, because any little spark would set it off, though. You'll have to store it somewhere (doesn't matter if it's a little leaky) and then release it not long before you want the explosion.


Reality questions: (I have no idea whether these questions even make sense)

With a tokamak-style reactor, how much damage would be done to the "supporting structure" if the plasma somehow got loose? I'm not thinking of a "boom" here (probably not), but just the damage and rebuilding cost.. I wonder if that's figured into the "fusion-as-ultimate-green-energy" scenarios?

Yeah, it's great if you can sustainably contain the plasma, but what happens if it gets loose? Would it basically just melt whatever the outer shell of the ring is (assuming there's something outside that ring that can manage to deal with that much energy), or would it basically melt down the entire building, meaning you'd have to start over building a.. 2+ billion dollar reactor, and maybe be able to salvage some of the superconducting material from the magnets?

I suppose with an ICF design could you have a large amount of space between the lasers and the target? I'm assuming in a tokamak that would be hard or impossible, unless you could get an incredibly high field strength.

  • 1
    $\begingroup$ I'm oddly qualified to answer this :p - a close relative is a former engineer on one of the test tokomak style reactors. I asked him, and a loss of the magnetic containment would and has damaged the wall of the vessel (and, umm, got to see a delightful video with a stray paint can in a tokomak). The damage is fairly minor though - the plasma is uncontained, so stops fusing immediately, cooling as it expands. It's possible a bigger reactor could do more damage, as the plasma would be hotter $\endgroup$
    – lupe
    Commented Feb 5, 2023 at 23:18
  • $\begingroup$ @lupe - would the plasma actually be hotter? Temperature of fusion is temperature of fusion; running hotter just seems like it's potentially wasting extractable energy. $\endgroup$
    – jdunlop
    Commented Feb 7, 2023 at 1:10
  • $\begingroup$ @jdunlop it's true you need a certain collision energy for things to fuse in the plasma, but a very small percentage of collisions currently reach that - temperature goes up, more collisions result in things fusing. We are also using temperature as a substitute for pressure in getting self sustaining fusion - we can't mimic the pressure of the sun, so we raise the temperature past the sun's $\endgroup$
    – lupe
    Commented Feb 7, 2023 at 7:43
  • $\begingroup$ @jdunlop the small test tokamaks do not reach the temperatures needed for fusion, they're designed to investigate the behaviour of the plasma in order to better understand the requirements for building hotter and larger tokamaks. $\endgroup$
    – jwenting
    Commented Feb 7, 2023 at 7:43
  • $\begingroup$ @jwenting - fusion occurs in something like the JET tokomak, and even in quite small ones - what is missing is the temperatures for a sustainable fusion reaction (one that keeps going without massive injections of power) $\endgroup$
    – lupe
    Commented Feb 7, 2023 at 13:28

Either as big as you want or no more than Chernobyl scaled for the energy output

Fusion reactors don't exist.<citation needed>

Several of the answers are using the present-tense to explain what they think a fusion reactor will do. That's amazing, since you don't explain anything about the reactor. You don't explain its nature, its size... nothing. Therefore, you really only have two options:

1: How many angels can dance on the head of a pin? As many as wanting.

When it comes to undefined reactors rigged to explode, your best bet is to simply define the damage that you want and move forward.

2: Use Chernobyl as a point of reference.

Yes, Chernobyl was a fission nuclear reactor. It's as close as we can get without an actual fusion reactor. But, let's start with a truth:

Myth # 2: A nuclear reactor can explode like a nuclear bomb.

Truth: It is impossible for a reactor to explode like a nuclear weapon; these weapons contain very special materials in very particular configurations, neither of which are present in a nuclear reactor. (Argonne National Laboratory)

But this doesn't mean a reactor can't explode...

Nuclear reactors can't explode like nuclear bombs. But they can explode, because an out-of-control fission reaction generates a boat-load of heat. Build up enough heat, and everything from water to metal vaporizes, creating pressure. Get enough pressure, and you get a boom.

Yeah... but how much of a boom?

And we're back to having no idea the type or size of your fusion reactor. Don't know how it's designed, what safety features are in play, we don't even know what fuel your reactor will be using, but we can take a guess:

The current best bet for fusion reactors is deuterium-tritium fuel (U.S. Dept. of Energy)

Deuterium is chemically identical to hydrogen (Source). So it's no more dangerous than hydrogen.

Tritium is also chemically identical to hydrogen (Source), although it is a radioactive isotope. The radioactivity could be a believable problem, but the referenced source suggests it would take a lot of it.

Having said that, hydrogen can explode. The Fukushima nuclear power plant experienced a hydrogen explosion. How many people died? None. Five workers were injured. Yes, hydrogen "explodes," but it doesn't explode like TNT or nuclear bombs. Think about the Hindenburg. When it exploded there was a massive fireball — but there wasn't a glass-breaking concussion.

Now, the Hindenburg wasn't contained like a concrete-and-metal encased reactor — but it did have access to abundant oxidant. Whether or not a fusion reactor had access to significant oxidant would be very circumstantial.


I could be wrong, but I don't see a fusion reactor doing anything more than what Chernobyl did. (And, realistically, it might even be safer than Chernobyl due to the lack of all that radioactive fuel — uranium.)

1We're not talking about a star here. A large-mass star will burn hydrogen and helium. When it's gone, they'll burn carbon. If they're large enough, they'll burn neon after the carbon. But once a star gets to iron the process of fusing iron absorbs energy. But we don't have that mass. So when a fusion reactor and its fuel mass loses containment the fusing part of the process would die very quickly, leaving the usual nasty mess.


Sabotage the charging injector. Inject the fuel for 1000+ cycles all at once

At more than a million of degrees there is no material that can sustain that temperature, the magnetic field can contain the plasma, but not the heat. The fusion reactor will work in pulses. No single pulse will be long enough to let a big amount of heat to escape the magnetic field, this is the current design and will be the future design there is no way not even in the future to solve the 1 million degrees problem.

To turn pulses in a continuous supply of electricity there will be more reactors working in parallel, but there will also be a system that will recharge automatically the reactor between one pulse and the other in the fastest possible way. That system will have available the fuel for a big number of pulses. Sabotage that system, inject all the fuel available in one go. In this case what was the reactor power?

A 1 GW reactor designed to be active 20% of the time for 1000 days has the fuel to generate 200GWh. According to this link that power all at once is a little bit more than 180 kilotons.

Edit: by answering to @Christopher James Huff I realised that I forgot to take into account the efficiency. The power in the explosion might be more than double than the electrical output because it would contain all the thermal power. But still it is not very much. The story will have to beef up the numbers.

  • 2
    $\begingroup$ "In this case what was the reactor power?": probably zero. The fusion in an ICF is triggered by hitting a pellet containing frozen fuel with lasers from all sides, or some other mechanism for applying a great deal of energy focused to a very small volume for a very short period of time. I don't see that working with a thousand pellets rattling around the reactor vessel. $\endgroup$ Commented Feb 6, 2023 at 16:42
  • $\begingroup$ @ChristopherJamesHuff For sure enough pellets for 1000 day would not fit inside the vessel. Those numbers were an example that will have to be reworked to make the story plausible. But a future (probably far far away) real reactor would be rated anyway in the old way for a simple reason: the power generated in a short pulse will have to be gathered in some way and sent to the heat to electricity conversion step. Whatever will be used in that step will determine the actual output power and rating. $\endgroup$
    – FluidCode
    Commented Feb 6, 2023 at 16:48
  • 2
    $\begingroup$ An ICF reactor is likely to have a system for dispensing a single pellet into the chamber with the precise positioning and timing required for it to be ignited. Additional pellets won't have that precise positioning and timing, and won't ignite. The most explosive result of throwing more pellets into the reactor will be if they don't interfere with operation, and that single properly-loaded pellet ignites as it would in normal operation. If the added pellets do affect operation, it will be with a severely negative effect on the power output. $\endgroup$ Commented Feb 6, 2023 at 18:40
  • $\begingroup$ @ChristopherJamesHuff Basically you are saying that the rector cannot be sabotaged. In this case there is no story for the QA. $\endgroup$
    – FluidCode
    Commented Feb 7, 2023 at 8:23
  • $\begingroup$ No, I'm saying this isn't a viable way to sabotage it with the intent to cause an explosion. $\endgroup$ Commented Feb 7, 2023 at 14:01

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .