I've asked a similar question here but the answer given shows the behaviour of water under general conditions.

I'd like to know what the behaviour of water is like as pressures increase towards infinity without being able to escape it's confinement.. i.e. a ball of water at the core of a galactic mass.. maybe this question is more for theoretical physics since we can't really measure or experiment?

  • 4
    $\begingroup$ Nuclear fusion? $\endgroup$ Commented Mar 13, 2015 at 12:34

4 Answers 4


Ron's answer gives us a good idea of what might happen in terms of "normal" chemistry, but if you really mean "increases to infinity", some very strange stuff happens. Suffice it to say, it doesn't stay water after a certain point. The intense temperatures created by the compression will cause the water to break apart, eventually no longer even having oxygen atoms due to nuclear reactions. Because we're talking about an externally applied pressure, the Chandrasekhar limit doesn't apply, so there is a point at which electrons and protons combine (when the electron degeneracy pressure is overcome) and a mass of neutrons remains. Neutrons themselves also have a degeneracy pressure (though we don't have good models to predict the exact pressure that has to be overcome). From here, we don't know what happens with as much certainty, but the formation of quark matter has been predicted.

Eventually, we reach a singularity. We can think of this as all the matter we had before being compressed into an infinitesimal volume with infinite density and our applied pressure ceases to mean anything. If we started with enough water, this would behave much like any other black hole, though micro black holes are hypothesized to have some special properties.

  • $\begingroup$ Great answer, Michael. I see how the lattice of water rearranges itself up to a point and wonder if that process continues through fusion to ever-denser lattice or would you think water will never separate under these conditions and remain H2O until some of the H2O in the middle begins to shmush into a clump of dense neutrons? In your last sentence 'If we started with enough water..', how much water might be enough to kick-off a singularity? Thx $\endgroup$
    – irth
    Commented Nov 27, 2014 at 12:04
  • $\begingroup$ When you actually reach singularity, there is no pressure anymore (such physical quantities are undefined there). But approaching singularity we have +∞ limit for pressure, and when gravitation became strong enough an external (applied) pressure becomes irrelevant, right. $\endgroup$ Commented Nov 27, 2014 at 16:44
  • $\begingroup$ @irth It wouldn't take much to actually get some kind of singularity; basically anything over the Planck mass (~22 µg). Very small black holes are thought to evaporate very quickly from Hawking radiation, though, which is why I distinguished them. As for when it stops being water, certainly once the nuclear electrostatic repulsion is overcome, fusion will occur, which should happen well before we're converting things to neutrons. (in white dwarfs, it takes the gravitational force of ~1.4 stellar masses to overcome this) $\endgroup$ Commented Nov 27, 2014 at 17:42
  • $\begingroup$ @MichaelD.M.Dryden, thanks for extrapolating. Sounds like singularities could be so common that they could be at the centre of any mass, be that atom or galaxy. wild to ponder. just to clear up this question tho, will water fuse with other water before hydrogen and oxygen split in this quesiton's conditions? $\endgroup$
    – irth
    Commented Nov 28, 2014 at 0:02
  • $\begingroup$ The only reason we can think of micro black holes in this thought experiment is because we're talking about a magical external pressure. There's no reason to expect a singularity would occur at the centre of just any mass in particular. As for whether fusion would occur before water breaks into individual atoms, I don't know. Pressure presumably has some effect on water's thermal decomposition, but I don't have enough information to speculate. Whether the heat produced by the compression is removed or not probably makes a big difference. $\endgroup$ Commented Nov 28, 2014 at 2:42

The phase diagram for water (reported in your link and reproduced below) is a good starting point.

phase diagram

The diagram shows us that at pressures around 1 terapascal (about 10 million atmospheres) ice is a solid, at least up to 400 C. It has been predicted (reference_1, reference_2) that at higher pressures, somewhere between 1.5 and 6 tera pascals, solid ice will undergo an insulator to metal transition and display properties typically associated with metals (band structure, electrical conduction, etc.). That's around 15-60 million atmospheres.

  • 7
    $\begingroup$ Would you happen to know whether all materials eventually undergo metallization at sufficiently large pressures? It seems to be a universal feature, and it makes sense (core electron orbitals are forced to interact and overlap due to proximity, creating lots of overlapping electron bands), but I've never seen that mentioned explicitly. $\endgroup$ Commented Nov 26, 2014 at 10:24
  • 5
    $\begingroup$ @NicolauSakerNeto Interesting question. I don't know the answer but I poked around and found this, [Under sufficiently high pressures, just over half of the nonmetals, starting with phosphorus at 1.7 GPa,[92] have been observed to form metallic allotropes](en.wikipedia.org/wiki/Nonmetal#Allotropes). Maybe at even higher pressures the other non-metals would also show metallic behavior. I see that they predict helium would be metallic on massive planets. If helium can be metallic, then wouldn't you expect everything to be metallic at high enough pressure, like you suggest? $\endgroup$
    – ron
    Commented Nov 26, 2014 at 14:31
  • $\begingroup$ Any idea what kind of temperatures would exist between 1.5 and 6 tera pascals of pure water. Trying to understand if H2O would still possess movement. $\endgroup$
    – irth
    Commented Nov 27, 2014 at 12:19
  • $\begingroup$ @irth The above chart suggests that all temperatures would be accessible at 1 TPa, so my guess would be the same for somewhat higher pressures if adequate temperature control is available. $\endgroup$
    – ron
    Commented Nov 27, 2014 at 14:27
  • $\begingroup$ Reading up on the research, at those high pressures, actual temperature stops playing a role in increased pressure is found through increasing the speeds of molecules, this makes the molecules more dense and increases pressure. Another interesting topic on the subject is triple points. I understand most elements and molecules have various triple points and wonder if the chart above is accurate in showing only one for water at 0 celcius. $\endgroup$
    – irth
    Commented Nov 28, 2014 at 0:06

For example here ICE VII - phase of water on Super-Earthes - how behaves itself water under big pressure.

Ice VII is a cubic crystalline form of ice. It can be formed from liquid water above 3 GPa by lowering its temperature to room temperature, or by decompressing (D2O) ice VI below 95 K. ©Wikipedia


This could be equivalent to the compression generated by a black hole, hopelessly water as any other fluid is compressed to a high density solid state


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