For a rocket engine that runs on (deeply) cryogenic fuel like LH2, e.g. SSME, is the fuel kept liquid throughout the entire length of the regenerative cooling circuit in the nozzle? Or is it allowed to boil or go supercritical at some point in the circuit? I imagine it would be a really bad idea to do so because generally, gases are worse heat conductors than true liquids.

  • $\begingroup$ I've seen this discussed here several years ago, but I can't remember how or when. The conclusion was no, but I can't find any reference to it yet, it may have been only in comments. $\endgroup$
    – uhoh
    Commented May 8, 2019 at 4:51
  • $\begingroup$ Possibly the question @uhoh is thinking of: space.stackexchange.com/q/22065/26446 $\endgroup$
    – DrSheldon
    Commented May 8, 2019 at 5:30
  • $\begingroup$ @DrSheldon thanks, what I think I'm remembering is a discussion of liquid to gas phase change within tubing that is embedded in or attached to the nozzle of a rocket, as it is here as well. It might have been related to the Saturn V F-1 engine, but I'm not sure. Whether said comment/discussion actually exists or not is another matter. $\endgroup$
    – uhoh
    Commented May 8, 2019 at 5:37
  • $\begingroup$ Critical point of hydrogen is 33K and 13.5 bar so except for maybe pressure fed engines it won't boil and simply go supercritical. It's used in expander cycle engines like Vinci. Critical point for e.g. Octane is 569 K and 25 bar so most of the times hydrocarbons will be super critical, too. For large hydrocarbon engines where we have a big mass flow compared to chamber surface we might not go supercritical. $\endgroup$
    – Christoph
    Commented May 8, 2019 at 6:58
  • $\begingroup$ @Christoph that sounds like a good answer $\endgroup$
    – GittingGud
    Commented May 8, 2019 at 12:12

1 Answer 1


At least for the SSME, the hydrogen exiting the nozzle cooling circuit was a supercritical fluid.

Data I used when working on a simulation of the SSME shows at 104% throttle setting the hydrogen was at 5911 psi (40.7 MPa) and 445 deg R (247 K).

This slide doesn't show the nozzle cooling circuit exit properties but it does show the mixer outlet properties (purple arrows), 5336 psi (36.7 MPa) and -183 F (153 K). This slide is also from a newer generation of the SSME than I worked with and tends to run cooler and lower pressure.

enter image description here

Here is a phase diagram of hydrogen from here.

enter image description here

Reference for the properties slide (it's slightly different from the copy I have and shows 5310/-193 for the mixer outlet properties, but I can't be bothered to make a new screenshot and do the unit conversions again).

  • $\begingroup$ Then how well does this supercritical fluid absorbs heat from the nozzle sidewall? I read somewhere that if the fuel truly boils inside the cooling circuit it is really bad, because of the heat insulation effect of the thin layer of gaseous fuel immediately inside the cooling circuit inner wall. If the pressure is added so that all that's inside is supercritical, does this mean the fluid will be uniform inside the circuit? $\endgroup$ Commented May 8, 2019 at 22:28
  • $\begingroup$ I agree that boiling is bad in a heat exchanger. "Boiling" is a two-phase phenomenon (gas and liquid phases co-exist) and can't happen in a supercritical fluid. I would say that it works quite well given the long and successful flight history of the SSME. $\endgroup$ Commented May 8, 2019 at 22:31
  • $\begingroup$ is the hot gas channel leading from the two pre-burner to the main combustion chamber also cooled by a small stream of LH2? (as with the two pre-burner's walls) $\endgroup$ Commented May 9, 2019 at 0:30
  • $\begingroup$ That component is called the 'hot gas manifold' and yes, it is actively cooled by H2. You can read about it on page 24 and 25 of this document: large.stanford.edu/courses/2011/ph240/nguyen1/docs/… $\endgroup$ Commented May 9, 2019 at 0:37
  • 1
    $\begingroup$ thanks for the info! $\endgroup$ Commented May 9, 2019 at 1:33

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.