Outside the ISS in the orbital night, the temperature can reach -250 degrees Fahrenheit (-157 degrees Celsius). The ECLSS (Environmental Control System and Life Support System), more specifically the OGA sub-system (Oxygen Generation Assembly) vent overboard hydrogen in the electrolysis process to produce oxygen from water, and the sub-system CDRA (Carbon Dioxide Removal Assembly) vent CO2 overboard ISS, too. At orbital night temperature, or even in the penumbra / terminator area, could these gases (I'm not sure if the H2 is maintained gaseous or liquid before it is about to be vented) become flakes of ice when ventilated outboard? Could CO2 become flakes of dry ice?

(EDITED by the comment of a friend below): The solidification point of the hydrogen gas is close to -259° C. And the solidification point of carbon dioxide is -56 ° C. But I am not sure if other factors (pressure, vacuum...) would interfere with this process.

I found this question here also refers to H2 and CO2 vent outboard. Why vent CO2 and H2 waste products to space on ISS?

Ammonia is also ventilated in some cases, and this one I have no doubt that forms flakes of ice, as seen here (vented) and here (leak)

Just for reference, before part of hydrogen was used in a Sabatier reactor, but it was removed and returned to ground tests (2017) Why was the Sabatier system removed from the ISS USOS?

  • $\begingroup$ You should look up the temperature for hydrogen to be solid. $\endgroup$
    – Uwe
    May 26, 2020 at 21:30
  • 1
    $\begingroup$ I know, The solidification point of the hydrogen gas is close to -259° C. And the solidification point of carbon dioxide is -56 ° C. But I am not sure if other factors (pressure, vacuum...) would interfere with this process. $\endgroup$
    – Apaiss
    May 26, 2020 at 21:33
  • 1
    $\begingroup$ You are conflating the temperature that an exposed, unpressurized part of the ISS might attain at nighttime with the temperature of the thermosphere at nighttime. These are two very different things. The thermosphere does not get that cold. $\endgroup$ May 27, 2020 at 11:28
  • $\begingroup$ I do not think like that. From where you got "The thermosphere does not get that cold"? You could be clearer or bring some source? $\endgroup$
    – Apaiss
    May 27, 2020 at 18:20
  • $\begingroup$ Carbon dioxide freezing point is not -56 °C, it is -78 °C. The -56 °C is the boiling point. $\endgroup$ May 26, 2022 at 0:50

1 Answer 1


Summary: The solidification temperatures given for CO2 or H2 in the original post are for atmospheric pressure. At the very low pressure at the altitude of the ISS, both CO2 and H2 will be gases at equilibrium. However, depending on the details of how they are released, they may form solid particles (likely due to rapid depressurization and evaporative cooling rather than the "cold" of space) that eventually sublime.

You are correct that pressure matters. At the altitude of the ISS (408 km), the pressure is at most about $10^{-7}$ Pa NASA. The phase diagram for H2 below shows that at $10^3$ Pa ($10^{-2}$ bar) H2 is a gas at even 20 K. The solid-gas coexistence curve drops with a high slope below 10 K, but space is no colder than about 3 K and a gas with most of its field of view covered by nighttime Earth will likely be much warmer than this.

Phase diagram for dihydrogen

Graph from: DOI: 10.1007/s00114-004-0516-x

I haven't been able to find a phase diagram for carbon dioxide that shows the low pressure behavior in fine detail, but it should also sublime at the pressure at ISS altitude for any reasonable temperature. We can at least say that CO2 likes being a solid more than H2. In general, as the pressure decreases toward an idealized perfect vacuum, the gas state becomes more and more thermodynamically favored at any nonzero temperature. (Of course, for large covalent network solids or ionic solids (rock), it could take many lifetimes of the universe to evaporate. We are talking about molecular solids here, though.)

Carbon dioxide phase diagram

Frozen particles of CO2 or H2 gas may form initially due to rapid depressurization and evaporative cooling. @OrganicMarble pointed out this SE answer showing solid O2 forming outside of a thruster. This apparent solid O2 persists for at least a few minutes (note the times on the video screenshots). However, the shape of this formation changes. It's not clear how much it might be accumulating new material and how much it might be subliming. How long would it take to sublime fully? I'm not sure how to predict that and it would depend a lot on the size and shape of the solid particles and other conditions.

Screenshots from SpaceX video showing apparent frozen oxygen crystal.

So, what timescale are you interested in? The particles can probably last a few minutes, but probably do not last years. I don't yet have enough evidence to make claims beyond that.

  • 2
    $\begingroup$ So to the OP's actual question "...could become ice flakes in the extreme temperatures outside the ISS?" are you answering "yes" or "no" for each gas? Remember that this isn't a closed system in a box; the gases are being vented to space and as the gas quickly expands the number of collisions can quickly decrease to zero. There are no walls in this case. $\endgroup$
    – uhoh
    May 27, 2020 at 0:48
  • 4
    $\begingroup$ WaterMolecule The answer to this question space.stackexchange.com/q/15152/6944 states that solid O2 "snow" is forming on a vent pipe attached to a rocket engine. It sounds like you are saying vented vapor will always remain a gas, that appears to not be the case. $\endgroup$ May 27, 2020 at 1:07
  • 3
    $\begingroup$ @OrganicMarble But in Earth orbit the O2 snow will be a transient state only. The solid snow will sublimate fast. $\endgroup$
    – Uwe
    May 27, 2020 at 15:29
  • $\begingroup$ WaterMolecule (I liked it nick) So considering the vacuum and the low temperature, does that mean the gases will become ice flakes? It would be good if you were more objective in your answer, with a "yes" or "no" for each gas, following the information already provided by you. $\endgroup$
    – Apaiss
    May 27, 2020 at 17:09
  • 1
    $\begingroup$ @Apaiss I found this paper which might help a bit: doi.org/10.1016/0094-5765(92)90079-X Water can probably be assumed to evaporate slower than hydrogen or carbon dioxide. $\endgroup$ May 27, 2020 at 22:42

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.