On long enough cosmological time scales, hydrogen and helium nucleii will become scarce in the Universe. It seems to me that any advanced civilisations that might exist in that epoch would have the motivation to try and prevent the stars from using them up, in order to burn the fuel more slowly and extract a greater proportion of the energy as usable work.

One might say that, on time scales measured in trillions of years, the stars are an unsustainable use of the universe's fuel. This question is about whether such civilisations would have the means to do something about it. My questions are:

  1. What would be the most energy-efficient way (using known physics) to blow apart a star or otherwise prevent or greatly slow the rate at which it performs fusion? We're assuming this civilisation has access to vast amounts of energy but doesn't want to waste it unnecessarily, since the aim is to access energy from the hydrogen the star would have burned. In order for this to be worthwhile, the energy gained from doing this would have to be substantially more than the energy the process takes.

  2. What would be the astronomical signature of such an activity? If it was happening in a distant galaxy, would we be able to detect it from Earth?


I'm still interested in this question. There are some great answers below, but currently I don't feel that any of them approach the most efficient way to solve the problem, i.e. a way likely to be attempted by a civilisation with the resources and motivation to do so. Luboš' answer makes it clear that the energy requirements are not too stringent in themselves (you just need to skim off a small fraction of a nearby gas giant and fuse the hydrogen into heavier elements; this question already presupposes the ability to extract power from fusion on a literally astronomical scale) but points out that no "bomb" made of atoms can enter a star and blow it up. However, this does not rule out other methods, such as increasing the star's angular momentum through some means (as in AlanSE's answer) or somehow (perhaps electromagnetically) removing plasma from its surface rather than blowing it up from the inside.

If removing plasma from the surface is possible then ideas based on this have a certain appeal, because the resulting hydrogen could be fused, releasing energy that could be used to skim off more, and so on exponentially. (Kind of like the opposite of that bit in the film 2010 when a self-replicating monolith turns Jupiter into a star.) Some of the resulting energy could be put into increasing the star's angular momentum as well. However, I don't know enough about plasma physics to know whether there's a way to suck the plasma up from the star's surface. (It can be done gravitationally, of course, but then the hydrogen just ends up being depleted by runaway fusion on the surface of another star, so this isn't very useful.)

Given this, my current specific questions are:

  • Is there a plausible way to remove material from the surface of a star at an appreciable rate, either electromagnetically or through some other means? If so, is there anything other than the (virtually unlimited) availability of energy that limits the rate at which this could be done?

  • Given a sufficiently huge supply of energy, is there an effective way to rapidly increase a star's angular momentum? It seems (per AlanSE's answer) that redirecting the star's radiation would take too long, but perhaps firing massive objects or particle beams at it would accomplish the same effect in a shorter time.

  • Aside from these ideas, would adding heavier elements than hydrogen to the star suppress fusion by absorbing neutrons? What quantities would be required in order to affect the star's evolution by this method? (A civilisation that eats stars will generate a huge supply of stable nucleii as a waste product, so maybe it makes sense to simply dump some of them into the next star.)

  • $\begingroup$ I realise this question is speculative, but hopefully it has enough of a basis in "real physics" to be on topic. $\endgroup$
    – N. Virgo
    Commented Sep 21, 2012 at 9:02
  • 11
    $\begingroup$ Throw a wet towel over it. $\endgroup$
    – MoonKnight
    Commented Sep 21, 2012 at 9:22
  • 3
    $\begingroup$ I don't understand what you mean by "I'm concerned that the stars are using up hydrogen nuclei at an unsustainable rate.". $\endgroup$
    – user12345
    Commented Sep 21, 2012 at 10:28
  • $\begingroup$ @user16307 it was kind of a joke (but only kind of). Over very very very long time scales, the stars use up H nuclei by turning them into thermal radiation. If you were a very advanced galactic-scale civilisation, you might get concerned about that after a few tens of billions of years, because you'd realise that your civilisation could live for a lot longer if that H wasn't being used up so inefficiently. $\endgroup$
    – N. Virgo
    Commented Sep 21, 2012 at 11:53
  • $\begingroup$ Regarding part (2) of your question, I believe that en.wikipedia.org/wiki/… may offer some of the info you are looking for. (You may also want to take a look at the references and external links given there.) $\endgroup$ Commented Sep 21, 2012 at 15:29

7 Answers 7


Burning (and fusion) is "unsustainable" by definition because it means to convert an increasing amount of fuel to "energy" plus "waste products" and at some moment, there is no fuel left.

I am not sure whether the word "unsustainable" was used as a joke, a parody of the same nonsensical adjective that is so popular with the low-brow media these days, but I have surely laughed (because it almost sounds like you are proposing to extinguish the Sun to be truly environment-friendly). The thermonuclear reaction in the Sun has been "sustained" for 4.7 billion years and about 7.5 billion years are left before the Sun goes red giant. That's over 10 billion years – many other processes are much less sustainable than that. More importantly, there is nothing wrong about processes' and activities' being "unsustainable". All the processes in the real world are unsustainable and the most pleasant ones are the least sustainable, too.

But back to your specific project.

When it comes to energy, it is possible to blow a star apart without spending energy that exceeds the actual thermonuclear energy stored in the star. Just make a simple calculation for the Sun. Try to divide it to 2 semisuns whose mass is $10^{30}$ kilograms, each. The current distance between the two semisuns is about $700,000$ kilometers, the radius of the Sun. You want to separate them to a distance where the potential energy is small, comparable to that at infinity.

It means that you must "liberate" the semisuns from a potential well. The gravitational potential energy you need to spend is $$ E = \frac{G\cdot M\cdot M}{R} = \frac{6.67\times 10^{-11}\times 10^{60}}{700,000,000} = 10^{41}\,{\rm Joules} $$ That's equivalent to the energy of $10^{24}$ kilograms (the mass of the Moon or so) completely converted to energy via $E=mc^2$, or thermonuclear energy from burning the whole Earth of hydrogen (approximately).

You may force the Sun to do something like the "red giant" transition prematurely and save some hydrogen that is unburned. To do so, you will have to spend the amount of energy corresponding to the Earth completely burned via fusion.

But of course, the counting of the energy which was "favorable" isn't the only problem. To actually tear the Sun apart, you would have to send an object inside the Sun that would survive the rather extreme conditions over there, including 15 million Celsius degrees and 3 billion atmospheres of pressure. Needless to say, no solid can survive these conditions: any object based on atoms we know will inevitably become a plasma. A closely related fact is that ordinary matter based on nuclei and electron doesn't allow for any "higher-pressure" explosion than the thermonuclear one so there's nothing "stronger" that could be sent to the Sun as an explosive to counteract the huge pressure inside the star.

One must get used to the fact that plasma is what becomes out of anything that tries to "intervene" into the Sun – and any intruder would be quickly devoured and the Sun would restore its balance. The only possible loophole is that the amount of this stuff is large. So you may think about colliding two stars which could perhaps tear them apart and stop the fusion. This isn't easy. The energy needed to substantially change the trajectory of another star is very, very large, unless one is lucky that the stars are already going to "nearly collide" which is extremely unlikely.

Physics will not allow you to do such things. You would need a form of matter that is more extreme than the plasma in the Sun, e.g. the neutron matter, but this probably can't be much lighter (and easier to prepare, e.g. when it comes to energy) than the star itself. A black hole could only drill a hole (when fast enough) or consume the Sun (which you don't want).

However, if you allow the Sun to be eaten by a black hole, you will actually get a more efficient and more sustainable source of energy. Well, too sustainable. ;-) A black hole of the mass comparable to the solar mass would have a radius about 3 miles. It would only send roughly one photon of the 3-mile-long wavelength every nanosecond or so in the Hawking radiation and it would only evaporate after $10^{60}$ years or so. It would be so sustainable that no one could possibly observe the energy it is emitting. However, the black hole would ultimately emit all the energy $E=mc^2$ stored in the mass.

If there are powerful civilizations ready to do some "helioengineering", they surely don't suffer from naive and primitive misconceptions about the world such as the word "sustainable" and many other words that are so popular in the mentally retarded movement known as "environmentalism". These civilizations may do many things artificially but they surely realize that the thermonuclear reaction in the stars is a highly efficient and useful way to get the energy from the hydrogen fuel. Even some of us realize that almost all the useful energy that allowed the Earth to evolve and create life and other things came from the Sun.

The Sun may become unsustainable in 7.5 billion years but according to everything we know about Nature, it's the optimum device to provide large enough civilizations – whole planets – with energy.

  • 1
    $\begingroup$ Sorry Lubos, But SE allows only a single upvote..! BTW, I like that novel man..! $\endgroup$ Commented Sep 21, 2012 at 11:49
  • 1
    $\begingroup$ Thanks for the answer :) I did indeed use "sustainable" as a kind of joke, although I do think the word has its place. I realise that the universe's H supply will inevitably run out, but maybe it would last a lot longer if it was fused in a controlled way rather than burnt in stars. We're talking "sustainable" on a 10 billion year time scale but not a 10,000 billion year one. A star doesn't seem to me like a very efficient way to convert H into energy, unless you build a Dyson sphere around it, but even then there would surely be conversion losses in extracting usable work. $\endgroup$
    – N. Virgo
    Commented Sep 21, 2012 at 12:06
  • 2
    $\begingroup$ Thanks, Crazy Buddy. ;-) Nathaniel, I see. When you worry about the solar radiation lost to wrong directions, it will still be cheaper to surround the Sun with solar panels. ;-) Note that the Sun's size is of order a million of kilometers "only". And one thing I didn't mention yet: if you care about the Hydrogen, why don't you just fly somewhere and grab it from nebulae or interplanetary gas etc.? The Universe has no shortage of fuel. It has a much more serious shortage of good ideas and "things at the right place". $\endgroup$ Commented Sep 21, 2012 at 12:17
  • 5
    $\begingroup$ BTW a reader of my blog points out that the required technology was invented a long time ago. It's called the Sun Crasher starwars.wikia.com/wiki/Sun_Crusher $\endgroup$ Commented Sep 21, 2012 at 12:39
  • 3
    $\begingroup$ Right. It strikes me as much more energetically efficient to actually gather hydrogen from nearby astrophysics sources and bring them back to the sun for fuel. Coupled with a suitable Dyson sphere, you would have amongst the more efficient (in the Carnot sense) stellar factories that you could think off. Fusion on the human scale will never be as efficient as what goes on in the sun. $\endgroup$
    – Columbia
    Commented Sep 21, 2012 at 13:26

The most efficient way to save hydrogen for future use by a very advanced civilization is not to try to stop current stars from burning up their hydrogen, but rather to make sure that the star generation rate in the galaxy drops to zero. Basically, give up on current stars as a lost cause and just prevent new stars from forming. This works because the amount of hydrogen in gas clouds in the galaxy is orders of magnitude higher than the amount inside stars.

To do this, the civilization would need to monitor all of the gas clouds in space so they can notice clouds that are getting close to the stage of creating a proto-star. If they can explode a sufficiently energetic bomb near where the expected star would be born, they would be able to increase the pressure and prevent the collapse into a star for a significant period of time. I don't have any calculations of the energy required but it must be very much significantly less than the energy needed to take a significant amount of hydrogen out of the gravitational well of a star that has already started burning hydrogen.

They would have to be very careful that the explosions they create are not too energetic since a bomb that is too big could trigger more star formations if the expanding gas cloud runs into other stationary clouds. This whole scheme would require a lot of monitoring and simulations of all the gas clouds in the galaxy, but I don't think it is impossible from a physics point of view.

Another difficulty would be to monitor stars and to predict supernova explosions since they can also trigger nearby gas cloud to collapse and generate new stars. They would have to either gently move the gas clouds out of the way or come up with someway to prevent the supernova explosion.

The astronomical signature would be a galaxy with no star formation and no supernova explosions over an extended period of time.

  • 1
    $\begingroup$ Interesting idea - and the SFR is a good place to start. I doubt, though, that adding shocks to giant molecular clouds will do anything but trigger star formation, but maybe you can direct it toward smaller stars. As for the amount of energy, keep in mind the standard calculation that 30 million years' worth of the Sun's luminosity about equals its gravitational potential energy with respect to all particles being at infinity. That's not quite supernova level, but it's still rather large. $\endgroup$
    – user10851
    Commented Oct 5, 2012 at 7:57
  • $\begingroup$ It would obviously have to be a very advanced civilization that could simulate the effects of their preemptive explosions very accurately to make sure they are big enough to prevent star formation and yet not too big to trigger more star formation... Yeah, probably impossible... $\endgroup$
    – FrankH
    Commented Oct 5, 2012 at 10:16

Here is another solution, somewhat resembling the answer from AlanSE: surround the star with mirror shell. The megastructure for this purpose, solar wrap was considered in this paper:

Beech, M. (2010). A Dark Sun Rising - Its a Solar Wrap. Journal of the British Interplanetary Society, 63, 104-107. (online, but paywalled)


The structure of a solar mass star artificially induced into an isothermal state is examined. The process is imagined to form part of a possible large-scale, solar engineering project that our distant descendents might attempt to perform in order to stave- off the Sun's future red giant phase. It is found that stable, solar-mass, isothermal configurations having radii between 2 and 10 solar radii with no internal energy generation through nuclear fusion reactions can be constructed. It is suggested that future humanity might choose to inhabit the outer surface of a solar wrap and tap the thermal energy of the isothermal Sun as a long-lived power supply.

It may seem counterintuitive that reflecting radiation back into the sun will cool it off enough to stop the fusion reaction. Remember, however, that a star as a gravitating object has negative heat capacity, its core is hot because there are temperature gradients in outer regions, so by returning the energy lost through radiation and inducing isothermal conditions its temperature (constant within its volume) will be much lower than the current core temperature, low enough to stop fusion.

The solar wrap is somewhat similar to Dyson sphere, however, its purpose is to reflect all radiation back into the star, so it could have much smaller radius, about $10\, R _\odot$.

As the result of achieving isothermal condition the Sun will expand: its radius will be $2 - 10 \, R _\odot$, its temperature will be about $10^6\,\text{K}$ (so reflecting the radiation from it would be nontrivial engineering feat).

This and other ideas, such as mechanisms for reducing the mass of the sun, are also the topic of the book by the same author:

Beech, Martin. Rejuvenating the sun and avoiding other global catastrophes. Springer, 2008, Google books.

  • 1
    $\begingroup$ Great, I think we might actually have a winner! I'll check out that paper once I'm at work, where I should be able to access it. $\endgroup$
    – N. Virgo
    Commented Nov 24, 2013 at 9:00
  • $\begingroup$ I don't understand this, can you elaborate? "It may seem counterintuitive that reflecting radiation back into the sun will cool it off enough to stop the fusion reaction. Remember, however, that a star as a gravitating object has negative heat capacity, its core is hot because there are temperature gradients in outer regions, so by returning the energy lost through radiation and inducing isothermal conditions its temperature (constant within its volume) will be much lower than the current core temperature, low enough to stop fusion." $\endgroup$
    – d-b
    Commented Dec 9, 2020 at 15:37

What would be the most energy-efficient way (using known physics) to blow apart a star or otherwise prevent or greatly slow the rate at which it performs fusion?

Spin it until fusion stops. Do this using the sun's own energy.

To accomplish this, I will have to ask you to envision something like a Dyson Sphere, but the primary function of the matter encircling the star will be mirrors. I will be making statements about the force balances and basic physics, how this could be actually done in practice is out of the scope of this answer.

I propose that mirrors would at some distance from the sun, and the focal length of these mirrors would be equal to this distance. The idea will be to reflect the sun's light back onto the sun in a way that makes it spin faster. We would like to redirect all the light nearly tangentially to the edge of the sun, but we can't do this because of entropic limits. Remember, nothing can be focus the sun's rays to heat something hotter than the surface of the sun. This is why I select the parameter of $R=f$.

R=f Focal length equal to distance, per Wikipedia

With this type of mirror we could focus the sun directly back onto the sun. We will assume sufficient distance from the sun to treat it as a simple circle. To make the sun spin faster, we will redirect the image right of the center, so the image will have a center a distance of $d$ to the right of the object. In order to get the optimal location to direct the reflection, we will

  • Integrate the distance from the axis of rotation from the lower bound of the image circle to the upper bound of the object circle
  • Integrate this value between the two intersections of the circles
  • Find the greatest value of this moment integral over all valid values of $d$

I actually did that calculation. I obtained $d=0.836 R$. To summarize, the proposal is to refocus the light back onto the sun so that it looks like a Venn diagram.

Venn diagram

By doing this, we trash a lot of the radiation. I calculate the average radius of interaction to be $0.418 R$, but if we dilute that number by the number of photons lost, we will get a multiplier of $0.202 R$ to convert the photon's momentum to the average torque exerted. I believe this is a fundamental, entropic limit, and if I'm wrong, it's at least conservative.

At this point we're still not finished. That's because this mirror isn't balanced. If it always reflected light in this way, it would not have a stable orbit since the gravitational force can only act radially and there is a tangential component to the photon's force on the mirror. One could compensate for this quite simply by holding a flat mirror at a $45^{\circ}$ angle to the sun's radiation, directing them in the other tangential direction. The torque from that balancing mirror, however, would depend on the distance from the sun. Because of that, we can't numerically correct for it here. If the distance of this satellite from the sun was large compared to the sun's radius then the loss from this balancing mirror would be negligible.

For simplicity I'll assume all the radiation from the sun is photons (it's only about 98%). The power output of the sun is:

$$P = 3.846 \times 10^{26} W$$

The photonic momentum (which is totally isotropic normally) can be found from $E=pc$ applied to the above power output. This gives the total momentum of photons emitted per unit time, or in other words, the isotropic photonic force.

$$F = P/c = 1.282 \times 10^{18} N$$

The moment of inertia of the sun can be found from the forumula for the moment of inertia for a solid ball.

$$m = 1.989 \times 10^{30} kg$$

$$I = \frac{2}{5} m R^2 = 3.848 \times 10^{47} m^2 kg$$

Using the methods I've laid out here, I can calculate the torque. This is assuming that all photons from the sun are used as efficiently as possible.

$$ \tau = F \bar{R} = (1.282 \times 10^{18} N) ( 0.202 R) = 1.801 \times 10^{26} N m $$

How fast would it need to spin in order to stop fusion, and then how much to break it apart? This is a difficult question to answer. However, one thing we can say is that if you have enough energy to completely disociate everything in the sun gravitationally you have enough energy to do both of the tasks of stopping fusion and breaking it apart less spectacularly. The energy to fully spread out the sun's mass over all space can be calculated. This is similar to a recent question, in short, the full dissociation energy is the half of the potential integrated over the entire volume. If I've done this right, the full disociation energy is:

$$ E_{diss} = \frac{3 G M^2}{8 R} = 1.423 \times 10^{41} J$$

It is also difficult to estimate the time needed for the available torque to dissociate the sun. But let's do a limit case where the sun doesn't deform due to the increased rotation. In that case we can seek an angular velocity that is equivelant to the above energy of dissociation, then ask how long it would take the available torque to accelerate it to that point.

$$ E_{diss} = \frac{1}{2} I \omega^2$$

$$ \omega = 0.00086 \frac{rad}{s}$$

This seems small, but consider that this would be the state of rotating once every 2 hours. Now, how long would the given torque take to get it to this state?

$$ I \omega = \Delta t \tau $$

$$ \Delta t = 58.23 \text{ billion years} $$

Now, if someone started spinning the sun with the sun's own power, eventually the fusion would stop, but the radiation wouldn't stop right away. In order to see if the stored energy is sufficient to break the sun apart, we'll consider the thermal energy stored in the core alone. The core's temperature is about 15,000,000 kelvin. This region goes out to around 0.25 solar radii. The average kinetic energy of a helium nuclei in the sun's core would then come from $E_k = 3/2 k T$, coming out to

$$ E_k = 3.106 \times 10^{-16} J $$

The average molecular mass in the sun is about 1.67 http://web.njit.edu/~gary/321/Lecture7.html I can use this to find the number of nuclei in the core of the sun. I can then combine that with the previous value for energy per nuclei to find the total stored kinetic energy in the sun.

$$ E = E_k N = E_k m (0.25)^3 / (1.67 amu) = E_k (1.12 \times 10^{55} \text{particles} ) = 3.481 \times 10^{39} J $$

This is the total thermal energy stored in the sun's core. Roughly. Divide this by the normal power output to get a $s$ valued number for the sun's power.

$$ E / P = 9.05 \times 10^{12} s = 6.9 \text{million years} $$

We conclude that the stored energy of the sun is insufficient to fully break it apart by about 4 orders of magnitude. The described method would still be viable to stop the fusion reaction.

  • $\begingroup$ 58 billion years? The sun only has about 5 billion years left before it goes off the Main Sequence and becomes a red giant. So it appears your scheme will only work after it has already exhausted all of it's hydrogen fuel. Still +1 for all the hard work! $\endgroup$
    – FrankH
    Commented Sep 23, 2012 at 11:21
  • $\begingroup$ @FrankH Exactly, I only just now realized how stupid that sounded. I feel good about the numbers, so maybe we should conclude instead that it's not a great method and an advanced civilization would probably seek improvements. For instance, deflect the light at a large radius and use ion drives to push against the spin of the sun. I wanted to cover the obvious "passive" approach here, for better or worse. $\endgroup$ Commented Sep 23, 2012 at 13:08

The only way is to capture all or most of the radiated energy of a star is using a Dyson sphere.

I wouldn't call it a realistic option from an engineering standpoint, but it is posible in principle.


Hardly the easiest way, and my background in physics doesn't extend beyond High School ...

Anyway I'd start with the assumption that any star is basically a ball of hydrogen having a surface of plasma. It may, therefore, be possible to calculate the cavitation frequency of that ball. The practical aspect may be an engineering challenge though ... generating the necessary amount of power to penetrate the plasma, constructing rugged delivery device/s, moving the devices into position ...


A civilisation advanced enough to be contemplating direct manipulation of stars, probably wouldn't place the same emphasis on the passage of time as we do. We are hard wired to think of a year in terms of what a human can experience/do in a year. But a sufficiently advanced civilisation will be able to modify themselves to allow arbitrary levels of information processing and/or work. If a human brain operated twice as fast, then it would subjectively experience two years for every one year that passed. Even today, we are already conceiving of a future where mind uploading could one day allow exactly this scenario.

The remaining time that we have to use the available energy from all possible future stars results in a constraint on the total amount of information processing and work we can do, due to there being a time limit.

Right now we are wasting all but an extremely tiny fraction of the available usable energy from stars. Your proposal is that we extend the time limit by shutting down the stars we're not using now, so that we can turn them on again when we need them at some future date.

I would argue that it is more effective to instead try to use up as much as possible of the available energy right now e.g. through the use of Dyson spheres. The end result is the same - an increase in the total amount of information processing and work that we can carry out before the heat death of the universe. It makes no difference whether that work is carried out over billions or quadrillions of years. The benefit of doing it quicker is that we don't have the added inefficiency of wasting energy on stopping/starting stars.

Note: While my answer doesn't address how to stop or slow down a star, in considering whether or not it could still be on-topic I took into account this was part of the OP: "... the stars are an unsustainable use of the universe's fuel. This question is about whether such civilisations would have the means to do something about it."


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.