I've been reading about "cryogenic" vs. "storable" propellants, and specifically the issue of boil-off: If the storage temperature gets above the propellant's boiling point, the propellant boils off and is lost. Apparently this is why "storable" propellants like hydrazine are used for satellites, where there's no support for keeping cryogenic propellants cool.

Does this mean the thermal control systems of a typical satellite can't radiate away enough heat to keep the propellant storage cool enough to prevent boil-off? What's the typical temperature of a satellite orbiting the Earth, and why can't it be kept low enough to use cryogenic propellants?

  • 1
    $\begingroup$ To cool a spacecraft there are passive and active devices depending on the amount of heat they have to 'remove'. The amount of heat depends on the operational modes of the spacecraft (therefore on which devices are working and for how long) and then you can have part of the orbit in shadow since you are orbiting the earth (but you could design a orbit to be always in light thus avoiding big thermal changes). So it is hard to say which is the usual temperature... $\endgroup$
    – Rhei
    Jan 15, 2015 at 17:07
  • 1
    $\begingroup$ And another thing: usually cryogenic devices need power to keep the temperature low therefore you need to provide it either increasing the solar panels area or increasing the amount of propellant on board. In both cases you are increasing the mass at launch and therefore the costs $\endgroup$
    – Rhei
    Jan 15, 2015 at 17:11
  • 1
    $\begingroup$ See also: How is heat dissipated from a satellite or any metal in space?. $\endgroup$
    – mins
    Jan 22, 2015 at 0:15

1 Answer 1


Short answer:

Overall temperature of a satellite around Earth is more or less the same than on the ground due to heat from the Sun and heat from systems.

Cryogenic propellants must be stored at very low temperature, or they evaporate. This is not effective for a satellite which must stay in orbit for years (unless the satellite orbits Saturn where the equilibrium temperature starts to be very low, and below boiling points).

They must be kept in heavy tanks to avoid deformation caused by gas pressure.

Storable propellants are more easy to use, and are robust in case the cooling system is temporary ineffective.

More details:

The closer the spacecraft to the Sun, the hotter it may become.

enter image description here
Black body temperature according to its distance from the Sun (ignoring planet albedo.)
(source: Spacecraft Thermal Control Handbook, David G. Gilmore et.al.)

Off the protection of the atmosphere, this will really be a problem, the spacecraft must radiate the heat received, in addition of radiating any heat produced by the equipments on-board.

Temperature inside a spacecraft is the result of two opposite effects:

  • heat radiation capture (from Sun and planets albedo) and own heat production
  • internal heat extraction and rejection by radiation (by thermal control).

enter image description here
(Source: Spacecraft Thermal Control). Note that energy received from orbited planet will vary with altitude (e.g. GEO vs LEO), this can be ignored here for simplification. Effects like Sun eclipse and satellite spin will increase complexity for the regulation, but will be ignored too here.

The result depends on how much heat is captured, or produced, and how efficient is the heat extraction (by conduction) and heat rejection. This resulting temperature must be in the range required by the systems aboard: e.g. sensors, solar arrays and communication equipment. Actually different areas of the spacecraft may have different temperatures.

enter image description here
(source: Fundamentals of Space Systems, Vincent L. Pisacane)

Whatever this temperature is, it must be regulated.

  • If there is any problem with the cooling system, any usual liquid propellant will start to boil. A portion of the gas must be released to vacuum to avoid the tank destruction. This is similar to what happens on a pad while the vehicle is waiting its launch, but there is no possibility to refill the tank.

enter image description here
STS external tank with gaseous oxygen evaporation under beanie cap.
(source: Nasa)

LH2 has a boiling temperature of -183°C, this is indeed a cryogenic fuel. Hydrazine (UDMH) boils only at 63°C. It is storable at Earth temperature.

  • More LH2 will be released in case of temporary cooling failure. The tank must be more robust to contain the greater pressure which results from evaporation. Robustness means usually additional mass. For these reasons, a storable propellant is used for attitude control and reboost.

In addition the propellant may need to be oxidized to produce energy. There are two types of oxidization: either spontaneous -- hypergolic -- like for UDMH with N2O4 or requiring an igniter like for LH2 with LO2.

  • Non-hypergolic combustion adds complexity, an engine is required to mix the fuel and the oxidizer and ignite the mixture. This is a problem when the engine must be fired a lot of times during the mission to maintain attitude and/or altitude. Failures would occur, for this reason only hypergolic are used for satellites.

  • Some propellants don't require additional oxidizer to release energy. They are named monopropellants, and work by exothermic chemical decomposition.

The propellant in the end is ejected thru the thrusters.

enter image description here
Monopropellant thruster - Source: Moog)

There are two types of thruster feeding: tank pressure and pumps.

  • Tank pressure is the only system used for orbit / attitude control of satellites. Pumps are too complex and to heavy for this use.

enter image description here Hydrazine bladder tank, returned to the sender (source: Daily Mail Online)

Designers will usually choose a mean of propulsion which is storable and hypergolic and can work without pumps. Taking into account that the propellant must be efficient (specific impulse) some combinations are often encountered on satellites:


  • Bipropellants are most commonly used on GEO orbit spacecrafts, they are used for transfer from launch orbit to GEO.

  • Monopropellant systems are more simple, but they are less powerful.

  • Cold gas are ever more simple, but very less powerful.

  • $\begingroup$ the conclusion that every satellite inevitably has the same temperature seems incorrect, after learning from space.stackexchange.com/questions/5246/… especially "Because the sun is very bright in space, we usually therefore want to create a surface that emits a lot in the infrared and absorbs little at the sun's most powerful wavelengths." and "That way they have the lowest possible ratio of αs/ϵIR, which means that the equilibrium temperature of the satellite stays the lowest possible." $\endgroup$
    – szulat
    Jan 22, 2015 at 2:10
  • 1
    $\begingroup$ @szulat: The first table (temperature as a function of Sun distance) gives the equilibrium temperature of a black body sphere. For 1 AU, this temperature is 255 K. See section 2.4 of Spacecraft Thermal Control, Meseguer, I Pérez-Grande, A Sanz-Andrés. When the spacecraft produces heat, this equilibrium is changed. The idea is to control the new value by using heat dissipation and reflection (as you describe) to obtain the temperature required by systems on board. $\endgroup$
    – mins
    Jan 22, 2015 at 7:03
  • $\begingroup$ So, based on this answer, is it right for me to conclude that silicon devices have enough free electronics to operate in the ambient temperature of space? Or do they need additional heating to create free-electrons? $\endgroup$
    – Lord Loh.
    Sep 22, 2017 at 19:53
  • $\begingroup$ @LordLoh. Semiconductors (usually) have a negative temperature coefficient, resistance increases when temperature decreases. As explained, temperature in space depends on thermal control and difference between how many photons are generated or captured and how many can be radiated. If your device is off (no heat generated) and in the shadow of the Sun (no heat captured), then it will be close to 0°K and can't be restarted without first heating it. This topic deserves its own separate question. $\endgroup$
    – mins
    Sep 22, 2017 at 20:34

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.