Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Can a solid planet become a star?

  1. Oct 15, 2013 #1

    I had this doubt when I was 14-15 years old, and I waited for many years (I'm 49 now) to ask about it, as I always thought this to be a silly question.

    In case someone can help with this, that's a thought experiment. It starts with this: if you have a sphere of gas, and then apply pressure on its surface, then the thing will heat up. Whatever gas, it will heat up, and that's because of pV = nRT, right? It doesn't matter how much time it takes - the only way that the sphere of gas under a huge pressure to have a certain volume V, is by having a high temperature.

    So, if you have a sphere of any gas, if you keep putting weight on it, it doesn't matter how slow that is done, the gas will heat up and ignite a star, there's no way around that.

    Now, assume I have a lot of blocks of say solid carbon, and then I slowly put them together. There will be some heating due to friction, but if I do that slow enough, the heat will dissipate, as as far as I know solid carbon under huge pressure will not be required to have a big temperature in order to have a certain volume -- it can well be at say -200F.

    So, if I keep doing that slow enough, I'll accumulate a huge weight of carbon, but the center will still be cold, and therefore it will not ignite -- although I think eventually, with enough weight, the center will become some sort of degenerate matter, or even a black hole.

    Is this true? That is, a huge, huge solid planet will not necessarily become a star, even if its mass becomes closer to (or perhaps even exceeds) the Sun's?

    Yeah, I know, it's a silly question. I wondered about it at that age, when I was dreaming of becoming a sci-fi writer, and thought of a sunless shell planet, heated up by a small star in its very center. Yeah, that would have been a silly story, but just think of the visuals :)
  2. jcsd
  3. Oct 15, 2013 #2


    User Avatar

    Staff: Mentor

    I believe you are correct. The fact that nuclear fusion occurs inside stars is because when the gas cloud that they form from collapses under gravity, the energy generated from the collapse can't escape fast enough and builds up. Eventually the star reaches a temperature high enough for nuclear fusion to occur.

    Done slowly enough I believe you would form a degenerate carbon object similar to a black dwarf. (A white dwarf that has cooled off) Continue the process and you would eventually cause a core collapse supernova, forming a neutron star.
  4. Oct 16, 2013 #3
    It will heat up while it is contracting. Not when it is at constant volume.
    Heat is transferred by conduction and radiation from warmer gas to cooler, given enough time.
    Neglecting self-gravity, you can give arbitrarily big volume to gas at a low temperature simply by having enough of the cold gas. If you do have self-gravity then there is a maximum volume cold gas can have. Only hot gas can have bigger volume - but if it has no heat source then it will over time cool and contract.
    Two ways around it. If the weight is added slow enough then heat conduction can keep up with compressional heating, and the temperature stays too low to ignite. Also there are gases which are very hard to ignite. There is absolutely no way to ignite nickel - if you heat it to nuclear reactions, these will only spend energy, not produce any, so put too much weight on nickel and it will collapse into a neutron star, not ignite. Nickel is not a gas, though. You could get tiny amounts of energy by igniting argon or krypton and converting them to nickel, but it is a small amount and you will need extreme temperatures to force even these reactions to happen.
    Same applies to any gas. They also solidify at sufficiently low temperatures and high pressures.
    White dwarfs consist mostly of carbon, though they are not so cold. Piling up carbon is just a roundabout way to create a similar object.
  5. Oct 16, 2013 #4


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    What is your planet composed of? Stars are mostly H, they have lots of potential energy available for fusion. The rocky planets in our solar system are mostly Fe, which means that the material has been "burned" in several stellar furnaces. Fe requires energy to fuse or fission so there is little if any nuclear energy available to power a stellar furnace. A Fe planet could be seen as stellar ashes. So no a Fe planet cannot turn into a star.
  6. Oct 16, 2013 #5

    Not sure that I get that. If you have a sphere of gas with a certain volume, and that sphere is submitted to a certain pressure, and this is steady state, then it will have a certain temperature, correct?

    It doesn't depend on which process was carried to have that sphere at a certain volume and pressure? That is, it may have expanded, or may have had a phase transition, or radioactive decay or chemical processes may have generated that sphere?
  7. Oct 16, 2013 #6
    Thanks all for the input.

    So let's see if I got this right - if I stack up iron or heavier, whether solid, liquid or gas, up to the mass of the sun, it will not ignite, right? If I keep on stacking iron upon iron, it will eventually be so heavy that will become something else, but not a star?

    Now, if I do the same with a lighter element, it all depends on how that's done. If I take say silicon oxide, and pile it up, but slowly enough to keep it cool, it will not ignite even when I reach the mass of the sun. But if I take say silane (SiH4, melting point 88K), and start to pile it up in some way that keeps its temperature above 88K, it will definitely ignite when it reaches somewhere around the mass of the sun.

    Is that right? Thanks all once again for the input.
  8. Oct 16, 2013 #7
    By the way, an associated thought experiment: if you have a sun-mass rocky planet, cold and made of silicon oxide, then there must be a way to ignite it, right?

    Say we explode a small atomic bomb at its very center: it will vaporize some silicon and generate a bubble of gas, which will be pressurized up by the rest of the planet and ignite, vaporizing more silicon around it, which will also ignite and so on - hopefully under the right conditions the whole planet could ignite and become a star?
  9. Oct 16, 2013 #8


    User Avatar

    Staff: Mentor

    Keep stacking it up and it will eventually go supernova.

    No. You have to add it fast enough to make the temperature increase to the point that significant fusion occurs.
  10. Oct 16, 2013 #9


    User Avatar

    Staff: Mentor

    No, not at all. Even if the bomb increased the pressure and temperature in the core high enough to start fusion, it will immediately stop when the detonation is over with and the pressure returns to normal. Fusion is NOT the reason stars give off energy. They'd do that without fusion. What happens is that gravitational potential energy is released as heat as gravity pulls all the gas together. This is the same reason that Jupiter releases more heat than it receives from the Sun. It is still collapsing under its own mass and releasing that potential energy. What fusion does is it replaces the energy radiated away by the star. Only once it is out of fuel for fusion does it start to collapse again, raising the temperature until it reaches the point that another round of fusion can occur with the next fuel. (In stars massive enough at least)
  11. Oct 16, 2013 #10


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Gravity will do the ignition all by itself. When sufficient mass is present your molecules of SiO will breakdown, there will be a release of energy that accompanies this breakdown. This will happen LONG before any nuclear fires can start. Now add signifiganly more mass then the electron shells break down thus allowing nuclear reactions to begin, more mass, more reactions, more heat. How much heat?? I am not sure.

    Add even more mass gets you a neutron star, even more mass a black hole.

    I suspect that the only time you will see a "star" as we know them in the night sky is when the Hydrogen fires are burning.
  12. Oct 17, 2013 #11
    Correct only for a given amount of gas.

    But compare a piston in a cylinder.

    Compress air in it and it will heat up. IF your compression is so fast that heat conduction is negligible then the final temperature is independent of the exact compression speed. But it still depends on the initial temperature, for the same initial and final volumes.

    But heat conduction is still not negligible. Your metal cylinder will get very hot, and you have to cool it.

    Also just because air gets to a certain temperature does not mean that it will ignite. Maybe you are heating your air to a certain temperature but are for some reason not injecting any fuel. The air still gets hot on compression but of course will not ignite. Or maybe you are putting some fuel in through carburetor but it is a too lean or too rich mixture and also cannot burn. Or maybe you have the fuel all right, but your ignition is not making sparks (and it is not dieseling). All these cases you still heat gas by compression, but it will not ignite. Or maybe your input gas is too cold. PV=nRT still holds - you will actually have bigger n, the gas still heats up, but its final temperature is lower and again not enough to ignite.

    Now remembering that your cylinder is a well cooled, conductive metal one. Suppose you compress air as usual - but then stop the piston at top dead centre and leave it there. Assume that your piston seals are perfect and air does not leak out of the cylinder.

    Then your air gets hot as usual - but because of cooling metal, it will cool down. This goes on until, at steady state, it comes into equilibrium with outside temperature.
    PV=nRT still holds. As the gas in cylinder cools down, V is constant, but when T falls, so does P. It is still higher pressure than outside - exactly because PV=nRT - but it is a lower pressure than it was when it was newly heated by compression.

    Now imagine that you are not moving the piston fast and then stopping it dead, but moving it slowly. The heat conduction takes place while the gas is heating. It still heats, but reaches a lower final temperature. So if you are trying to run a diesel engine too slow, then because your air cools down due to conduction, it does not heat enough to ignite the diesel fuel.

    If your diesel engine were NOT cooled and instead were perfectly insulated against heat conduction, then you could compress air and leave hot air at top dead centre forever. It could never cool down, and would ignite any time you add fuel.

    This also means you can run a perfectly insulated engine as slow as you want and it still would ignite.

    If your engine is very well insulated, but not perfectly, then it would cool down - slowly. You can run it slowly - but not arbitrarily slowly, because if slow enough it still would cool down and stop igniting.

    Stars are big and well insulated. If you piled up mass slow enough, they still might be cooled by conduction so they do not ignite.
  13. Oct 17, 2013 #12
    There are a few misconceptions in your original post which I don't think have been picked up, so I'm going to go back to the beginning...

    I am also 49, but there is no such thing as a silly question [:smile:]

    1. p and t are not the only variables in this equation: if you increase p what else could change to keep it balanced if T stays the same?
    2. pV = nRT is known as the 'ideal gas law' because it is only true for a theoretical, 'ideal' gas, but it is in general a good approximation to the behaviour of real gases under a wide range of conditions. But as pressure increases the fact that real gases are not 'ideal' (in particular their molecules have a physical size and may attract each other) makes the approximation less accurate.

    But when you apply a given pressure you don't simultaneously constrain the gas to a given volume, the gas 'works this out for itself'. And if you apply pressure slowly, or wait a long time after applying the pressure so that the temperature gained is radiated away then the volume will decrease and you will end up with a relatively cold and dense blob of gas like Uranus. In fact Jupiter is doing exactly that now - heat radiated away from the surface decreases surface pressure and so it shrinks by about 2cm a year.

    The word "ignite" is misleading in relation to stars, when we say a star ignites we really mean "commences nuclear fusion reactions". Whilst fusion reactions do occur at high temperatures, an equally important condition is high pressure. But at the high temperatures and pressures required for nuclear fusion the substance stopped behaving like an ideal gas long ago - in fact it is no longer even a gas, it is a plasma.

    When a celestial body forms, mass collects together around its centre of gravity. As it does so it loses potential energy which is converted to heat. So the temperature of a young celestial body is largely determined by the mass of that body, not its composition.

    Why are you happy to say that "the heat will dissipate" from a solid body but not a gaseous one?

    No, the carbon at the centre would be really hot - but the temperature and pressure required to fuse carbon are much higher than for hydrogen or helium. What happens next would depend on a number of parameters, but it is unlikely that anything like this happens in reality as there is just not enough carbon around (or anything else) to form a body of solar mass that isn't mainly hydrogen and helium.

    At the current stage of evolution of the Universe the question doesn't arise - there is still much more hydrogen around than anything else so anything that is really big is nearly all hydrogen and helium. This is likely to be the case until the universe is 10,000 times as old as it is now.

    It is true that heavier elements wouldn't make good stars, but this is not because they are solids at temperatures we normally encounter - as mentioned above, anything that is going to undergo fusion is going to be so hot and dense that it is going to be a plasma anyway - it is because their fusion reactions don't liberate much energy (and once you get to iron this becomes zero).

    I hope that clears up a few "dead ends" in your thought experiment - but more importantly opens up even more avenues for exploration!
  14. Oct 17, 2013 #13
    I believe what will happen with carbon is:
    1) At a certain, diamond will form at the center and the reduced volume will create a planet quake as the layers collapse towards the center and heat up.
    2) A carbon plasma will form at the center causing further collapse and heat.
    3) Finally, even if all of the heat generated by the progressive collapsing is dissipated, the plasma will begin to fuse and you will have ignition. At sufficient pressure, room temperature fusion ignition is possible.
  15. Oct 20, 2013 #14

    Wow, that's amazing. So you can create a star even at low temperature, provided the pressure is high enough! That's incredible.

    If I got it right, from yours and the previous posts, as mass is slowly added to the planet the sequence would be:

    (a) cold big rock of carbon -> (b) cold layered carbon planetoid -> (c) cold carbon planet with plasma core -> (d) cold black dwarf -> (e) supernova

    Appreciate everybody's input. This has been very instructive.
  16. Oct 20, 2013 #15


    User Avatar
    2017 Award

    Staff: Mentor

    What makes you so sure about that?

    Pressure on earth is sufficient to create diamonds even in the mantle. A pure carbon pile would produce diamonds long before pressure and temperature are even close to the ranges needed for fusion.
  17. Oct 20, 2013 #16

    Hmm... I've been thinking about these points, and indeed they are troubling questions for my thought experiment.

    If I understand your explanations right, the problem with the assumption that "a sphere of gas submitted by a large enough pressure will necessarily have a large enough temperature for fusion" is the density of that gas, right? That is, you can have a cloud of gas with mass equal to say 10 Suns, but its center may still not ignite if the density is small enough. But then, wouldn't the pressure at the center be very low too? I mean, how can you have high pressure but low density? If the pressure is high, won't the density be high too?

    It's because I have trouble in imagining a sphere of gas, say hydrogen, that is at high pressure but low temperature. The mental image that I have is say butane - even if the thing is at room temperature, you compress it just a little bit and it becomes liquid. Even more I imagine it would become solid.

    So I reckon the same thing happens with hydrogen. Therefore, I couldn't imagine a way to get a cold, gaseous planet to remain gaseous and cold, and yet be big enough to become a star. The core of cold gaseous planet should solidify at low temperature and high pressure, and then behave the same as if that core was made of say silicon to begin with.

    Therefore, the only way I could see that a sun-mass sphere of gas to remain a gas is by having high temperatures - that's why my assumption that a sphere of gas submitted at high enough pressure will necessarily be hot and will become a star (that is, if it was cold, it wouldn't be a gas a high pressure).
    Last edited: Oct 20, 2013
  18. Oct 20, 2013 #17
    Not equally. Fusion reactors occur easily at high temperature and low pressure. The matter is that low pressure conditions easily allow the heat to dissipate.
    Plasma behaves as ideal gas at high temperatures.
    At low temperatures and high pressures, substances behave not as gases, but as liquids or solids.
    There are exactly such objects! And they are pretty common!

    They were already around when the Universe was 10 times younger. Carbon, to the amount of solar mass and slightly more, is readily formed by fusion of helium. Just look at the Pup!

    But although the Pup has no energy source, it is not called a "planet". It is called a "star".

    If a white dwarf of carbon were to have gas piled up on top, and fused to carbon - so slowly that the heat is dissipated and the dwarf is cold all the way to the core. Would the carbon be stable all the way till the star crosses Chandrasekhar limit and collapses? Or would carbon fusion be possible at any appreciable rate purely due to tunnelling of carbon 12 nuclei into each other at high densities?
  19. Oct 20, 2013 #18
    But at the centre of a star the pressure is very high which allows fusion to take place at much lower temperatures than would otherwise be the case, which is what I mean by "equally important".

    But then fusion won't start.

    But carbon-oxygen white dwarfs were not formed from clouds of carbon and oxygen, they were formally main sequence stars. And they are not cold, they are really hot: the universe is only a fraction of the age required for them to cool enough to become black dwarfs.
  20. Oct 20, 2013 #19
    I think we are both getting a bit confused about what we are actually talking about here. The point that I am trying to make is that you are ascribing properties to gases and solids, pressure, heat dissipation etc. based on your everyday experience of materials on Earth and this is leading to some invalid conclusions when dealing with masses and therefore pressures and temperatures many orders of magniture greater. And you need to drop the idea that the temperature of a body of gas is sustained by pv=nRT - it isn't.

    Oh something else I didn't notice in your orignal post - about planets enclosing a star, have you heard of the (science fiction) concept of a Dyson sphere?
  21. Oct 22, 2013 #20
    If high pressure allows fusion at low temperatures then it allows fusion in liquids and in solids. Crystal order would not stop nuclei from tunnelling into other nuclei, like it does not stop electrons from tunnelling around metal conduction band.
    Temperature isn´t. Pressure is. That is what defines gas.

    By having high temperature.

    At low temperatures and pressures, you have
    V=n/ρ, independent on P and T
    At intermediate temperatures and densities, you have
    At high temperatures, you have
    P=cT^4, independent on n

    Correct. All substances save helium (both isotopes) solidify at low temperature and pressure. Even helium freezes around 30 atmospheres. And the freezing point will increase with increase of pressure.

    But solids are actually better heat insulators than gases. It is just that they are not heated much by compression.

    Deep sea is icy cold. Below +2 degrees.

    Ice will not be heated by compression at all - to the contrary, compression will cause ice to melt and cool. Cold fresh water will contract on heating - and for that very reason, will cool on compression.

    Warm fresh water, and salty water, do warm on compression - slightly. As stated, deep sea is still icy cold. On compression to 1000 atmospheres, it warms by merely 1,2 degrees!

    On the same compression, 1 atmosphere to 1000 atmospheres, air should heat by over 1000 degrees!

    This means that liquids and solids are easier to ignite. Thermal runaway is not slowed down by heat losses.

    A gas will expand on heating - and therefore cool by expansion. This limits the warming by any added heat.

    A liquid as mentioned expands only slightly and therefore the heat loss on expansion is negligible. The added heat is spent on heat capacity alone - so the liquid heats up more and is more liable to thermal runaway.

    A solid, like liquid, expands only slightly and is cooled by heat capacity alone. But a solid, unlike a liquid, is solid. Therefore even if the solid does expands, it is rigidly in place and cannot convect upwards like gas or liquid can. Heat can only be transferred by conduction, which is extremely inefficient. Water at 10 km depth is icy cold, just 1,2 degrees warmer than surface. Rock at 10 km depth is everywhere hot - over 150 degrees everywhere.
  22. Oct 23, 2013 #21

    Wow, that's amazing. Learning a new thing every day!

    Actually learned two: that materials can fuse at high pressure and low temperature, and pressure of gas at high temperature does not depend on density.

    Quite surprising information - it's incredible to have fusion at low temperatures. Whammy wow!

    That completely changes my thought experiment, uh? If I accumulate enough cold carbon, and do it so slowly that it remains cold, eventually it will start fusion at the center at cold temperatures and huge pressures, will explode the outer cold junk and leave a small neutron star behind, uh? What an extraordinary process!
    Last edited: Oct 23, 2013
  23. Oct 23, 2013 #22
    Yes, because the pressure of light is independent on the amount of matter. At high temperature and low density, the pressure is mainly that of light.
    Probably not. Type Ia supernovae are thought to come completely unbound and leave nothing behind.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook