We have already discussed how difficult it would be to blow up a planet like Earth, basically deciding that you can't do that with any reasonable amounts of energy. But what if I don't need to blow up the planet (as in, cause a mostly-solids-to-near-solids planet to fragment), but rather want to just (at least mostly) get rid of a gas giant?

Let's say that I wanted to remove Saturn from our solar system, or at least reduce it to unrecognizability (allowing for a core to remain after the process is complete, but making it look very different). An obvious possibility would be to boil off a significant fraction of or ignite the hydrogen making up the bulk of its mass and somehow burn the gas giant's atmosphere up. (Jupiter has a lower fraction of hydrogen, but still considerable amounts of hydrogen and could work as well. This is not specifically about these planets; they serve more as examples than constraints on an answer.)

How could I do that? Would it even work (A.K.A. how much energy is required)? Are there other ways to achieve the same long-term effect of making the planet appear vastly different to a visual observer, which do not involve actually blowing up the planet?

I'm not necessarily looking for a big boom, but spectacular effects will not detract from scientifically sound proposals. Advanced but scientifically plausible technology is fine, but no magic, please. Technological solutions which are available are those that we currently have on and around Earth, plus anything that is plausible based on our current understanding of the sciences involved. Having it happen over a time scale of 10-100 Earth years is adequate; shorter is better, but not required.

  • 1
    $\begingroup$ Is this primarily about Saturn, or do you want a solution for any generic gas giant? $\endgroup$ Feb 12, 2015 at 15:36
  • 7
    $\begingroup$ "ignite hydrogen", do you mean thermonuclear "ignition", or what? $\endgroup$
    – hyde
    Feb 12, 2015 at 17:05
  • $\begingroup$ In games and simulations, the idea of a pure gas gigant may sound possible. But, if we want to stick to reality, a gas gigant or a star may not be a pure gas object, but, an object with a solid or plasma core, with a lot of surrounding gas ... $\endgroup$
    – umlcat
    Feb 12, 2015 at 17:07
  • $\begingroup$ What sort of technology is available? A Slaver Disintegrator (Known Space) or a Molecular Disruption Device (Ender's Game) could do the trick pretty easily. $\endgroup$
    – KSmarts
    Feb 12, 2015 at 17:27
  • 1
    $\begingroup$ @KSmarts: The Ender's Game MDD would destroy it, but it would pretty immediately reform as a big ball of hydrogen. $\endgroup$
    – Zan Lynx
    Feb 13, 2015 at 0:42

10 Answers 10


Blowing Saturn up would be hard.

Saturn has a mass of $568.3\times 10^{24}$ kg, or about 95 times the mass of Earth. To blow up Saturn, i.e. to place some sort of unimaginably huge bomb in its center and vaporize it, you'd have to energize all or most of that mass to escape velocity, otherwise it would just reform out of the cloud of gasses. The escape velocity of Saturn is 35.5 km/s, so this would take $1/2 \times 35,500^2 \times 568.3 \times 10^{24} = 3.6 \times 10^{35}$ Joules of energy.

For reference, this is the amount of energy the sun produces in thirty years.

Burning up all of the hydrogen also wouldn't work. You'd need to bring enough oxidizer to react with all of that hydrogen, which would weight 8 times as much as the hydrogen did in the first place. Then, once you lit it, the gas wouldn't have enough energy to escape into space, so all you'd really accomplish would be producing lots and lots of water.

So how do we destroy Saturn?

My first though was to toss a small star or black hole at it, but flying one over would probably take more energy than just blowing Saturn up.

Instead, let's push Saturn into Jupiter. The energy required is equal to the difference between the starting orbit for Saturn and its finishing orbit. (We'll move Saturn instead of Jupiter because it's a bit smaller) This is given by the equation $E = \frac{-GMm}{2a}$, where $GM$ is the standard gravitational parameter (of the sun, in this case), $m$ is Saturn's mass, and $a$ is the semi-major axis. The orbital energy of Saturn in Saturn's orbit is $-5.3 \times 10^{28}$. (The convention with orbital energies is that the planet has zero energy at an infinite distance from the sun.) The energy in Jupiter's orbit will be $-9.7 \times 10^{28}$, giving a total energy needed to throw Saturn at Jupiter of $4.4 \times 10^{28}$ joules, which, while enormous, is several orders of magnitude less than what we'd need to blow it up outright.

(Note that the energy change is negative, but assuming that we use a Hohmann transfer orbit, we'll be burning the same amount of energy in our velocity changes to decrease our orbital radius as we would need to increase it. Maybe. I think. Orbital mechanics was 5 years ago, now.)

  • 3
    $\begingroup$ How about a giant space fan? Fans work in space right? Haha anyways great answer - I never thought it would be this tough but your logic makes sense. $\endgroup$
    – thanby
    Feb 12, 2015 at 17:07
  • 24
    $\begingroup$ Wait, doesn't this just make Saturn bigger and stronger? You're feeding Saturn, not destroying it! It's a trick, @ckersch is a Saturn agent. $\endgroup$
    – Samuel
    Feb 13, 2015 at 0:48
  • 16
    $\begingroup$ Vaporize a gas giant. I think that word does not mean what you think it means. It would take very little to vaporize a gas giant, SINCE ITS A GAS ALREADY. $\endgroup$
    – Aron
    Feb 13, 2015 at 4:17
  • 3
    $\begingroup$ @Aron The term "gas giant" isn't the best, since gas giant (at least Jupiter and Saturn) interiors are mostly highly pressurized liquid metallic hydrogen. Very non-gaseous. $\endgroup$
    – ajp15243
    Feb 13, 2015 at 5:12
  • 5
    $\begingroup$ "all you'd really accomplish would be producing lots and lots of water" Sounds like you just answered Would a planet made completely of water be possible?, too! $\endgroup$
    – user
    Feb 14, 2015 at 16:58


Saturn at its brightest has an apparent magnitude of about -0.24. If we can darken it to about 6.5, it will become just about invisible to the naked human eye -- it will apparently vanish. We can do this by interference with Saturn's low orbital region or upper atmosphere, at an energy cost many orders of magnitude below that required achieve the same effect by removing most of the planet.

To do this the simple way will require reducing its albedo by a factor of about 500, from 0.34, its current albedo, about the same as sand, to 6.8 x 10-4, roughly 100 times blacker than vantablack, the blackest material yet known. If we can invent something that black, we can paint the rings with it and either put a cloud of it in orbit, or load the upper atmosphere with balloons (containing hydrogen or vacuum) painted with it.

If we can't find a material that black, we can exploit geometry. Black material arranged in a fractal "pine tree" shape is considerably darker than a flat surface of the same material. The amount of material needed would increase by one or two orders of magnitude.

Angled mirrors can reflect incident sunlight off the ecliptic and keep Saturn in the dark. A spinning disk of aluminum-coated mylar or other thin polymer film, perhaps a kilometer across, makes a fine mirror. It will require very small ongoing attitude adjustment to correct for torque due to tidal forces. Forty billion of them will about do the job, neglecting the rings.

Don't want to lug that much material to Saturn? Try biotechnology. Engineer a hydrogen-blimp creature made mostly of hydrocarbons than can float in Saturn's upper atmosphere, photosynthesize with its dark fractal upper surface, and reproduce. Make a microscopic organism to infest the rings. Get some spores to Saturn for a mere pittance (safe atmospheric injection will be the only real hurdle), then sit back and watch the planet dim.

The above methods can be combined in various ways, and none of them come anywhere near the cost of removing a significant fraction of Saturn's mass.

EDIT: On reflection, this is probably much too literal an interpretation of "change it beyond recognition". If this answer is considered cheating-and-not-in-a-good-way, I'll delete it.

  • 3
    $\begingroup$ I wanna see it painted, painted black // Black as night, black as coal // I wanna see the sun blotted out from the sky // I wanna see it painted, painted, painted, painted black $\endgroup$
    – Nick T
    Feb 13, 2015 at 23:10
  • 4
    $\begingroup$ @NickT: I see a gas giant and I want it painted black... $\endgroup$ Feb 14, 2015 at 18:34
  • 17
    $\begingroup$ It might be cheaper to just paint every telescope lense black... $\endgroup$
    – NPSF3000
    Jun 23, 2015 at 4:14
  • 1
    $\begingroup$ +1 for creativity and for reading the question thoroughly. The only problem would be getting rid of the infrared signature. $\endgroup$
    – MichaelK
    Dec 26, 2016 at 13:34

Never let a good gas giant go to waste

I would use massive hydrogen scoops to fuel my colonization efforts. Given that accelerating a small ship to relativistic speeds takes several hundred times mankind's current yearly power output, getting all that reaction matter is eventually going to make a dent even in a gas giant like Saturn. Over the millennia, the gas giant will be slowly depleted. You can also use the moons as dynamos, generating vast amounts of electrical power by draining the powerful magnetospheres. These will power your laser sails.

Eventually, you'll get rid out the outer layers and can send in reinforced robodiggers to get at the nice juicy core material, which will make for a nice wreath of solar collectors when accelerated to the factories in Mercurian orbits. Heck, by the time you're done with it, it might be such a tiny remnant, you might consider terraforming it, although it might be a tad cold (nothing a few giant mirrors can't solve, of course).

  • 3
    $\begingroup$ the rocky core will go liquid when you remove the outer layers, the sole reason they are solid is because the pressure of the gasses. $\endgroup$ Feb 12, 2015 at 14:33
  • 1
    $\begingroup$ @ratchetfreak Nah, there's bound to be a bit of a rock/metal core down there too, crushed under the metallic hydrogen. $\endgroup$ Feb 12, 2015 at 14:34
  • 1
    $\begingroup$ If you remove enough hydrogen, then the planet will deorbit from the change of mass. $\endgroup$
    – Nick2253
    Feb 12, 2015 at 15:39
  • 11
    $\begingroup$ @Nick2253 The orbital mechanics don't change based on the mass of the orbiting body (except when the orbiting body is nearly as massive as the thing it's orbiting). A less massive object is easier to derail, but it won't automatically derail itself. $\endgroup$
    – Brilliand
    Feb 12, 2015 at 17:43

Destroying a gas giant is hard, moving it may be easier. As soon as you can move it, you can throw it into the sun, collide it with the next gas giant, or remove it from the solar system.

A good way to move it would be one or more fusion rockets floating in the atmosphere. The fuel can be taken directly from the atmosphere.

There is a good description of of such a device in the webcomic Schlock Mercenary

  1. Build a fusion candle. It's called a "candle" because you're going to burn it at both ends. The center section houses a set of intakes that slurp up gas giant atmosphere and funnel it to the fusion reactors at each end.

  2. Shove one end deep down inside the gas giant, and light it up. It keeps the candle aloft, hovering on a pillar of flame.

  3. Light up the other end, which now spits thrusting fire to the sky. Steer with small lateral thrusters that move the candle from one place to another on the gas giant. Steer very carefully, and signal your turns well in advance. This is a big vehicle.

  4. Balance your thrusting ends with exactness. You don't want to crash your candle into the core of the giant, or send it careening off into a burningly elliptical orbit. When the giant leaves your system, it will take its moons with it. This is gravity working for you. Put your colonists on the moons.

  • $\begingroup$ If you affect the orbital velocity (which you can) destroying comes for free. $\endgroup$
    – Chris H
    Feb 13, 2015 at 15:26

While ckersch's answer is correct on the energy needed there's a deeper issue involved here that I think you are missing:

The main part of the energy needed to destroy a planet is the energy needed to push the components to escape velocity. For the back of the envelope calculations people are doing on here the nature of the planet doesn't matter.

There are two basic sources of inaccuracy in the calculations: They ignore the chemical binding energy and they assume each bit of mass takes the same energy to boost away.

The first simply can't be calculated as it's based on an unknown value. The smaller the bits you turn the planet into the more energy that is going to be needed (and remember even gas giants have a rocky core, they still have some chemical binding energy.)

The second is a much bigger source of error. The inside of a spherical shell is in zero gravity. To correctly calculate the gravitational binding energy you have to blow off a series of infinitely thin shells, recalculating the escape velocity after each one. In practice this will be impossible for anything other than Earth as you need a density profile of the planet.

The escape velocity of a sphere is $$\sqrt{\frac{2GM}{R}}$$ thus the simple answer is $$\sqrt{\frac{2GM}{R}}M$$ but the actual gravitational binding energy of a uniform sphere is $$\frac{3GM^2}{5R}$$ Note that even here how much you break it up is a big deal--that's 80% more than the energy to split it into two parts receding at escape velocity. I won't even dream of tacking the binding energy of a non-uniform sphere.

  • 1
    $\begingroup$ I'm not seeing the integral you seemed to be building up to... $\endgroup$
    – Samuel
    Feb 13, 2015 at 0:53
  • $\begingroup$ @Samuel Note my last sentence--even if I had the density data I'm out of my depth at this point. We do not have the density data for anything but Earth, anyway. $\endgroup$ Feb 13, 2015 at 0:55
  • 1
    $\begingroup$ Ah, well I'm an engineer, not a scientist, so I'd assume a big ball of hydrogen. Both the energy to remove a shell and the density are a function of the remaining mass. $\endgroup$
    – Samuel
    Feb 13, 2015 at 1:04
  • $\begingroup$ @Samuel Even a big ball of hydrogen will have a major variation in density due to the pressure. $\endgroup$ Apr 3, 2017 at 17:39
  • $\begingroup$ Variation from what? The density will increase as the pressure increases. This follows from the ideal gas law: PV=nRT. If you look at a stack of cubic meter volumes (constant volume) from outside to the center and assume constant temperature, as the pressure increases due to more and more mass pushing down from above the 'n' (molecules in a given volume) must also increase. It's chemistry/physics 101 stuff, a good approximation until there is a phase change. $\endgroup$
    – Samuel
    Apr 3, 2017 at 17:48

General Fusion has a neat approach to achieving nuclear fusion. By using large pistons, they will send a sonic shock wave from the edge of a sphere toward the center, where some fusion fuel lies. The way that these shock waves interact create an amplified version of the wave at the absolute center. Apparently, this amplification is so great that even conventional machinery can create pressures and temperatures so great that nuclear fusion can happen. We're looking for a much bigger boom here. So what if we start out with thermonuclear fusion? Oh my.

Could we create fusion at the center of Saturn? No. The center is made of rocky materials. Iron can fuse under certain conditions, but we would rather not go there. Additionally, the phase changes might present a barrier to the wave propagation, as waves tend to partially reflect at density boundaries. Also, the waves will somewhat diminish in intensity traveling over such long distances.

So where could we create fusion? As deep as we can go! Human ingenuity can create submarines that can go deep to pressure levels of 3000 psi or greater. The electronics and other systems probably won't need to be in a pressure vessel anyway. The higher pressure will affect the dynamics of the shaped charges, but they could affect it for the better (I mean, higher yield). An advanced society might be able to go much deeper. Actually, due to the lower molecular mass, the scale height is much larger than on Earth, maybe 60 km. You could easily go 400 km or more below the 1 bar level. Liquid Hydrogen should exist around 1000 km, and this will be even better for our fusion.

You need to get greater depth because in order to actually blow up Saturn, you need reaction mass to push off against. A wall of 100 km of relatively high density gas is okay to push off of, but not fantastic. You'll need a big boom for this blast to actually decimate the rest of the planet.

Using the General Fusion approach, we will arrange thermonuclear weapons in a large sphere, perhaps 100 km in diameter, deeper than 500 km depth. By the way, we will need a lot of bombs. They should also be high yield. Also, they need to be timed with absolute perfection so that the blast waves all culminate in one giant fusion ball in the center. Doing so will require correcting for the pressure gradient too, but I consider this challenge similar to what General Fusion is already dealing with.

We need to produce, at minimum, the dissociation energy of Saturn. This won't be enough to turn the planet into a cloud of gas (due to the specific physics of the explosion), but it might be enough to satisfy the OP. Fusion tends to release less than 1% of the $mc^2$ energy of matter, but the specific reactions are complicated. You might seed the fusion area with some Boron, Tritium, and other stuff to keep it interesting. But our goal is to have this area at such a high pressure and temperature that it becomes a "Mr. Fusion", combining whatever atoms that release energy, and even some that don't.

Optimistically, we would need a fusion volume of a box about 2,800 km on the side. This is about equal in size to the practical limitation of our giant thermonuclear fusion trigger spheres. But only a small fraction of that volume (at the center) will actually fuse.

So we'll need several of these spheres. Also need to make sure they all trigger at the same time. Best case scenario, you'll need a few 100. Each one of them will probably release more yield that all nuclear weapons that humanity has ever built. But the fusion that they trigger at the center of their spheres will buy you many orders of magnitude beyond that.

To recap: We will start with thermonuclear weapons. That means that a fission bomb will detonate, and its radiation pressure will detonate fusion fuel next to it. This happens simultaneously for many bombs arranged in a large (several 100s or 100s of km) sphere within the top of Saturn's liquid Hydrogen layer. These trigger a sonic wave which amplifies at the center of the sphere, creating a massive fusion ball which releases many times more energy than the original bombs. Now, this giant fusionsphere happens in 100s of places in Saturn's liquid Hydrogen layer in one hemisphere. The energy released is enough to blast the planet to bits, but most of the energy goes to releasing gas into space at high speeds. The outer layer above the bombs is entirely thrown out into space. This could provide impulse to push the remaining core out of the solar system or somewhere else. Generally, the solid core will not be blown away, but most of the light elements probably will be.

While grand, our solar system is almost certain to have the necessary fission and fusion resources to accomplish this. Substantial technology hurtles still remain, but president gives good confidence that they can be solved by an interplanetary society. The scale is a few orders of magnitude beyond human cold war activities, but this doesn't sound like a problem either.

Historical thermonuclear bombs have, in fact, used liquid Hydrogen. Saturn's liquid Hydrogen layer isn't the right isotope, but it might not matter, and you might be able to refine Deuterium out anyway.


Honestly I have no clue if this is at all feasible, but the end result might be Saturn dissolving like a bath bomb.

Saturn (more-so Jupiter) as much as we like to classify as a planet is potentially more accurately defined as a failed star...a stellar body that failed to gain enough mass to ignite itself. All the components are there for it to become a sun, just not enough volume to do so. It's not to say it couldn't ignite if the conditions are right and who says we can't artificially do this for long enough for fusion to start up at the core of Saturn?

Couple options:

1) Hydrogen burns well...pumping a large amount of oxygen to the 'surface' of the liquid metallic hydrogen covering the surface all the way around and simultaneously ignited.

2) Nuclear bombs with the blast towards the core from several hundred (thousands?) or the devices. Once again, hitting all sides of the metallic core simultaneously to rapidly increase heat and pressure

Might be able to get this effect with a big enough asteroid impact? but I'm not sure there...honestly I'm not sure about any of this.

If we are successful in starting the fusion process within the core of Saturn, the energy released should be enough to shoot the majority of its atmosphere into space leaving behind a burning (fusion) ball of liquid hydrogen. This of course would go out with the pressure and heat components for fusion no longer being met and without the huge atmosphere, the ball of metallic hydrogen would quickly dissipate into a ball of hydrogen gas that solar winds slowly blows over space.

It's my 'bath bomb Saturn' plan...a quick burst of its atmosphere off, followed by a pretty center that expands and slowly dissipates.

  • $\begingroup$ I think you are underestimating the scale of a gas giant. You'd be talking billions or trillions of nuclear devices. $\endgroup$
    – Tim B
    Feb 13, 2015 at 9:24
  • $\begingroup$ I doubt any chemical reaction on the surface will be enough to create fusion conditions. $\endgroup$ Feb 13, 2015 at 11:19
  • $\begingroup$ And even if you did create fusion conditions, the gas giant would just rapidly expand, cooling itself over time. Fusion is anything but self-sustaining, it will fizzle out very quickly. Sure, the minute "ignition" might launch some of Jupiter's mass away from it, but I don't think you'll get much more than changing the atmospheric patterns for a while - something you can do a lot cheaper with a few asteroid impacts :) $\endgroup$
    – Luaan
    Feb 13, 2015 at 15:37
  • $\begingroup$ @Luaan - I figured as much, Perhaps there is some slightly futuristic technology to induce cold fusion in the liquid metallic Hydrogen? Asteroid impacts are fun and all, but there's something a bit more spectacular in momentarily lighting Saturn up like a star to eject it's atmosphere $\endgroup$
    – Twelfth
    Feb 13, 2015 at 18:16

To ignite the hydrogen you still need an oxidizer or just a general alkaline material to absorb the hydrogen.

To just burn with oxygen to create an ice rock you would need ~8 times the mass of the hydrogen in oxygen. That is a lot of gas.

The other option is to harvest all the gas by skimming the outer layers with scoops.

  • $\begingroup$ Burning the hydrogen also wouldn't drive it off into space. If anything, it would reduce the rate of gas loss into space since all the hydrogen would be bound up into much heavier molecules of water. $\endgroup$
    – ckersch
    Feb 12, 2015 at 14:47
  • $\begingroup$ @ckersch unless you put enough reagents in there to make it explode in a exothermic reaction $\endgroup$ Feb 12, 2015 at 14:47
  • 2
    $\begingroup$ It won't be exothermic enough. For hydrogen + oxygen, the reaction generates 286 kJ per 18 gram mol of water produced, resulting in a velocity of around 4km/s if all of the energy goes into velocity away from Saturn. The escape velocity of Saturn is 35 km/s $\endgroup$
    – ckersch
    Feb 12, 2015 at 15:31
  • $\begingroup$ But maybe you could let the hydrogen react to form something with a much higher melting point, so the liquid / solid result falls to the surface? $\endgroup$
    – Burki
    Feb 13, 2015 at 14:58

Speaking of the change it beyond recognition part of the question.

I don't know exactly what would happen, but this is way more feasible than blowing anything up.

Putting giant mirror arrays in space focusing a large amount of sun on Saturn. Compared to the 10^35 units of energy required by other methods mine is tiny. Maybe just maybe you could focus enough solar rays to raise the temperatures equivalent to near that of earth orbit.

Surely given Saturn is -270F a significant increase would set in motion a series of changes that would make the planet look totally different.

Ideally going to 40F, a crazy change of 310F. The liquid gas would now all just be gasses. Everything frozen would melt. Things in the atmosphere would combine, and you might even get an ocean.

I will leave it to someone else to tell me how impossibly large the mirror array would have to be.


Borrow another dimension, relocate the bits of the planet out that you don't want to other locations. Beyond the energy used for the dimensional rift, no additional energy is required and no laws of physics are violated - well, as far as the relocation part goes - we do have to assume that opening a dimensional rift isn't actually going to violate any laws of physics.

The simplest explanation is to use the concept of swapping two variables in a computer program. To swap two variables A and B in the easiest way possible, a programmer generally creates a third, temporary variable C and does the swap this way:

C = A
A = B
B = C

Most modern compilers will optimize at least C into a register on the CPU. For simplicity, A and B reside in RAM, C resides in a CPU register. If you were a bit in RAM observing the data in A and B (and both were equidistant to you), the data would appear to swap places instantly. From RAM's perspective, the CPU is somewhere else affecting A and B but doesn't violate any of the laws in which A and B operate. Within RAM, the same amount of data exists both at the beginning and at the end of the operation.

The same principle can theoretically be applied to our dimension. We "borrow" another dimension for C, copy A to C, copy B to A, and copy C to B. Ideally, all of this happens instantly and then C is able to be reused for something else. For it to work without violating the laws of physics in this dimension, the other dimension should not have "time" applied to it so that the operation against this dimension can happen instantly from this dimension's perspective.

At this point, you have created a way to swap two areas of any size with a constant amount of energy required to "move" those areas. You could swap equal sized spaces containing an atom, a molecule, an apple, a person, a planet, half of a person, half of a planet, half of a sun, a whole solar system, or entire galaxies. Swapping the space containing any of those uses the exact same amount of energy. The only requirement is that the two spaces containing what is being swapped is the exact same size and shape. Amazon and businesses would love you for improving package delivery. Terrorists (and governments) would love you for making it easy to wipe out all of civilization.

How to open a dimensional rift? That is mildly off-topic for the question. However, quantum physics has some really odd behaviors that may be better explained by symptoms of multidimensional artifacts. Quantum mechanics may be the knowledge gateway by which we learn how to open rifts to other dimensions to accomplish the above. Exploring our universe without the need for spaceships like the Iconians in Star Trek OR just stupidly destroying ourselves - both options may become possible, but, given the choice, the latter is more likely to happen.

  • 2
    $\begingroup$ Welcome to Worldbuilding (and SE as a whole, I think)! The extra-dimension idea - while one of the most creative - isn't science-based, like the question asks for. $\endgroup$
    – HDE 226868
    Feb 14, 2015 at 15:40

You must log in to answer this question.

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