The carbon dioxide phase diagram shows that at high pressure and normal temperature CO2 becomes liquid. Since it is heavier than water, I wonder if there are lakes of liquid CO2 at the bottom of the Earth's oceans, and if no, why?

enter image description here

  • 1
    $\begingroup$ Quick answer is no, however there are methane clathrates somewhere down there and possibly also carbon dioxide clathrates as well. Waiting for someone with better knowledge of the subject to answer it... $\endgroup$
    – Gimelist
    Commented May 19, 2015 at 22:46
  • $\begingroup$ it looks like, and it's hard to tell precisely just looking at the chart, but at 4 degrees C - about the temperature of the deep ocean, it looks like that would require over 1,000 ATMs (33,000 feet). It looks to me, looking at the chart, that even the deepest part of the ocean wouldn't have sufficient temperature/pressure combo to liquify CO2. If the oceans were twice or three times as deep - then it might be possible. Europa just might have pools of liquid CO2 . . . maybe. Maybe not with it's lower gravity. Mixing would also be a factor. $\endgroup$
    – userLTK
    Commented May 21, 2015 at 1:47
  • $\begingroup$ Isn't that diagram very misleading? It's for pure CO2. There must be a diagram for CO2-H2O-NaCl somewhere. $\endgroup$ Commented May 21, 2015 at 3:25
  • 1
    $\begingroup$ It's a standard phase diagram, so I don't think it's meant to be misleading. I think Anixx' question is, in essence, since CO2 can be a liquid under temperature and pressure similar to what exists deep in the ocean is liquid CO2 possible in the ocean. My hunch is no as it would just mix and stay dissolved in the H20, not separate. CO2 in H20 also sometimes becomes H2CO3 which is also prone to lend an H+ to an H20 and create light acidity - one of the problems we're having with greenhouse gas. I've also never seen a phase diagram for a mix of elements. It would be more complicated $\endgroup$
    – userLTK
    Commented May 21, 2015 at 5:29
  • $\begingroup$ @userLTK dissolution is possible only up to certain concentration, and there is a lot of CO2 available. $\endgroup$
    – Anixx
    Commented May 21, 2015 at 5:34

2 Answers 2


Yes, according to this paper in the Proceedings of the National Academy of Sciences. But they also say that this "lake" of liquid CO2 is covered by CO2 clathrate, so it's not quite your average brine pool.

Note about clathrates:

The following image is from the Wikipedia article on CO2 clathrate. It is a CO2 hydrate P-T phase diagram with experimental data (black squares) from a 1998 paper by Sloan listed in the references of the Wikipedia article.Sloan 1998 CO<sub>2</sub> hydrate phase diagram

Note the H2O phase lines drawn in. There is a relatively small habitable area for clathrate somewhere inside about 273-280 K (0-7 ºC) and 10-12 bar (or about 90-110m in ocean depth as calculated here). Above the phase boundary, the clathrate will dissolve to liquid. Below, to gas. What this diagram does not say (according to a knowledgeable colleague) is that clathrate will not readily form or remain in its semi-organized state given very slight amounts of mechanical agitation. Hence why it does not exist en masse under typical seafloor conditions.

  • 1
    $\begingroup$ Why these lakes are so rare? Given the amount of CO2 in atmosphere, all ocen floors should be covered with liquid CO2. $\endgroup$
    – Anixx
    Commented May 20, 2015 at 5:10
  • 4
    $\begingroup$ @Anixx: Why? Assuming liquid CO2 is somewhat soluble in water, then any lakes would slowly dissolve into the ocean when they come into contact with water. Similarly, you might ask why there are no solid salt deposits in the ocean. $\endgroup$
    – jamesqf
    Commented May 20, 2015 at 18:00
  • 2
    $\begingroup$ According to a colleague of mine, semi-organized clathrate-state CO2 pockets are extremely vulnerable to transformation back to their gaseous state if they are physically disturbed, in which case they would quickly escape to the surface as gas bubbles. This could be why you don't see more of them around. $\endgroup$
    – Ian
    Commented May 20, 2015 at 20:59
  • $\begingroup$ Or, more likely, travel upwards as gas until partially or completely dissolved into solution. $\endgroup$
    – Ian
    Commented May 20, 2015 at 21:07
  • 1
    $\begingroup$ @Anixx Those are idealized models. In reality, a bit of stuff doesn't have a tag that it's either "liquid" or "gas" (or another phase) - it's just a useful way to model stuff, not a physical reality. Introduce disruptions into such a lake, and you'll get a mix of water vapour, carbon dioxide, carbonix acid, some of them "liquid", some "gaseous", and who knows what else. Fringes are especially tricky, and your scenario is very much on the fringe here. $\endgroup$
    – Luaan
    Commented May 5, 2016 at 11:27

There are several reasons why the ocean floor is not a pool of liquid $\ce{CO_2}$:

  1. Solubility of gaseous $\ce{CO_2}$ in brine is low.

  2. Mass transfer of gaseous $\ce{CO_2}$ from the atmosphere is slower than the time to react with water to make carbonate/bicarbonate

  3. Carbonate/bicarbonate act as a buffer for atmospheric $\ce{CO_2}$ precipitating and dissolving as atmospheric $\ce{CO_2}$ fluctuates.

Liquid $\ce{CO_2}$ at the ocean floor must have a source of pure enough $\ce{CO_2}$ to liquefy and it must be in enough abundance, at a high enough pressure, so that it is not diluted or mixed with water or brine.

The data above shows that this has happened somewhere in the world... that is cool.


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.