If there existed a planet that was made completely of liquid water, no solid matter at all, what would the core of such a planet be like? Let's assume that this planet has conditions similar to earth, so that it could support liquid water at its surface, with ice caps at the poles. But there's no matter of any kind except for liquid water, pure H2O. The surface is basically one big ocean. But what would happen if you dove down into that ocean? There would be no solid ocean floor in the traditional sense, but weird things would probably start to happen due to the pressure. Would the water be pressurized into ice? What abnormal properties might this core-ice have? How gradually would this happen? Would the core be hot or cold? If you could survive the extreme pressure, would it be possible to make it all the way to the core?

  • $\begingroup$ The deeper you go in the oceans, the higher the pressure. This means that even colder water temperatures are needed for the water to turn into ice! This is why the deep oceans are NOT a frozen mass of solid ice... $\endgroup$ Commented Sep 4, 2016 at 22:02
  • 1
    $\begingroup$ @EveryBitHelps -- the deep oceans (on Earth) can't be a "frozen mass of solid ice" because normal ice floats. So if some of the water turns into ice, the ice will rise to the surface. (Also, the oceans are salty, which lowers the freezing point.) $\endgroup$ Commented Sep 5, 2016 at 0:49
  • $\begingroup$ According to this xkcd comic [xkcd.com/1561/], the core will be "David Bowie & Queen". $\endgroup$ Commented Sep 5, 2016 at 0:52
  • 1
    $\begingroup$ I don't think so. Physics SE or Astronomy SE might tolerate it as a model-building question but I feel like more likely they'd provide you with a list of reasons a pure H2O planet could never form in our universe and consider it addressed. Worldbuilding accepts a broader notion of possibility. $\endgroup$ Commented Sep 6, 2016 at 8:08
  • 1
    $\begingroup$ Very similar to this question. worldbuilding.stackexchange.com/questions/4969/… $\endgroup$
    – Twelfth
    Commented Sep 6, 2016 at 15:59

2 Answers 2


For planets the approximate size of Earth You would find Ice VII, X, or XI. Possibly even liquid water existing as a supercritical fluid.

This is a surprisingly hard question, and depends on the radius and mass of the water planet in question. A water planet of Earthlike mass would definitely not have an Earthlike radius, and it's difficult to say exactly what its radius would be precisely because water is so damnably weird with so many different crystalline structures at different temperatures/pressures.

Have a look (and mind the logarithmic nature of the Y axis, and the moderate range of temperatures covered):

Phase diagram of water
By Cmglee - Own work, CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=14939155

Every labelled region on that diagram has different structural properties. As a point of reference, pressure at our planet's core is estimated to be in the neighborhood of 330–360 gigapascals. Since Earth is made of much denser stuff than water, we can probably take this a comfortable upper bound on an earth-radius water world's internal pressure. As you can see from the phase diagram, water at this pressure would be Ice X (assuming a moderate temperature, which is a poor assumption at a planet's core, but these are the data I could find), distinguished from Ice I–IX by its novel crystalline structure. In fact, Ice VII, X, and XI are the only ones one this diagram suggests could exist in a planet's core (which would, again, be very hot as well as under great pressure). The boundary between fluid and solid phases of water also goes to higher pressures as temperatures increase. In a particularly hot and small world, you could find plain old liquid water at the core, or more likely water existing as a supercritical fluid. 2

It is hypothesized that at even greater pressures, well above 1000 gigapascals, water would take on metallic properties. You'd need a truly massive water world for that to occur naturally at its core, however, perhaps approaching the limit of stellar fusion. Jupiter is thought to have internal pressures in excess of 3000 gigapascals, but Jupiter contains some materials significantly denser than experimentally verified phases of water, and is as you know, quite large. In addition, at least one theoretical prediction regarding the metallic phase of ice expects it to occur only at pressures in excess of 5000 gigapascals. (However, those endorsing more moderate theories placing it nearer 1000 gigapascal do indeed expect metallic ice to exist within Jupiter.) This site (http://www1.lsbu.ac.uk/water/water_phase_diagram.html) has extensive information on phases of water, including a more expansive phase-diagram that includes water as a supercritical fluid and the metallic phase of ice, though the latter is only theoretical.

As for the possibility of traversing the core, you can pretty much forget it. It's hard to imagine pulling together enough handwavium to help a human body survive pressures on the order of 100 megapascals or above. Even considering an autonomous drone core-explorer, it doesn't look good. For a solid core of exotic Ice VII, X, or XI, as you dig into it the supercritical fluid water above would immediately flow into any hole you dug and freeze into the depth-appropriate phase of ice, arresting your progress. (However, some exotic ices are metastable at temperatures and pressures other than those at which they are created, so if you have need of exotic ices, you might be able to mine them with a handwavium digger.) Even traversing the supercritical fluid would be an interesting engineering challenge, as supercritical fluids are, well, kinda weird. We actually use supercritical water as a means of rapidly oxidizing hazardous wastes 3 that would otherwise persist effectively forever on the scale of our civilization, so good lucking carrying a sacrificial anode strong and large enough to offset that.

  • 1
    $\begingroup$ There's actually some exploration of this in a paper from 2003 [arxiv.org/abs/astro-ph/0308324], though the idea was that there would still be a silicate mantle and an iron-rich core under all the water. Their 6-earth-mass model planet had liquid water giving way to Ice VI several dozen kilometers down. $\endgroup$ Commented Sep 5, 2016 at 0:51
  • 1
    $\begingroup$ I wonder if a particularly large waterball (one large enough to have a metallic water core) far from its host star might possibly exhibit every phase of water. The necessary variation of pressure at constant temperature to access some phases might be within reach for a very oblate (rapidly spinning) waterball. $\endgroup$ Commented Sep 5, 2016 at 1:04
  • 1
    $\begingroup$ Can you add a link to metallic phase? $\endgroup$
    – JDługosz
    Commented Sep 5, 2016 at 12:13
  • 1
    $\begingroup$ I'm going to agree with the idea of metallic water as a necessity. In a world where the surface is liquid water and it's water all the way down I would think that basic heating would induce a loss of water to space over time until the planet was no more. A "metallic water core" could at least (with some hand waving) potentially spin to create a magnetic field around the planet to help reduce strikes by charged particles. $\endgroup$
    – GrinningX
    Commented Sep 5, 2016 at 20:08
  • 3
    $\begingroup$ Supercritical water is neat stuff. Last I looked into it, it was capable of dissolving all known plastics. There were people in late 1990s trying to create supercritical cleanup systems for ocean waste. Didn't go anywhere. But it suggests this water world might have no macroscale life because there's no life cycle: anything that dies sinks and drains nutrients into the unrecoverable depths. $\endgroup$
    – SRM
    Commented Dec 9, 2016 at 23:53

It is speculated that it would be just ice and compressed water (different from ice):

"Their abyssal depths would be so deep and dense that even at high temperatures the pressure would turn the water into ice. The immense pressures in the lower regions of these oceans could lead to the formation of a mantle of exotic forms of ice."


You must log in to answer this question.

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