Both BFR (Big Falcon Rocket) booster and BFS (Big Falcon Spaceship) are to run on Raptor engines, fueled with liquid methane and liquid oxygen - cryofuels.

The trip to Mars will take at least a couple months, and BFS requires fuel for powered landing. And cryofuels normally boil off over time - tanks so durable as to hold their vapor pressure would be too heavy for space. Not a critical problem for a mission that lasts a week or two, like a flight to the Moon, but arriving at Mars in several months and finding the tanks dry would be a problem...

So how does SpaceX plan to deal with it?


3 Answers 3


This is a really good question and the answer is probably not 100% known, even by SpaceX at this moment.

No doubt they will have some active cooling to minimize boil off.

Structurally there are tricks they can play. For example the landing fuel is stored in a smaller tank, which is submerged in the main tank. Thus the surface area to warm up over is seriously reduced.

You can see that in this image from the 2017 ITS design:

2017 ITS Design

In the section labelled Propellant Tanks, there is a smaller set of tanks inside.

Instead of a small amount of fuel at the bottom of the tank (Or maybe floating all over it, while in transit) they will have a small amount of fuel in a small tank, inside the another tank that was cooled by the fuel/oxidizer there before (plus any left over).

I expect we will see many changes to the design before the first flight, for this and many other reasons.


I think the most important factor in avoiding boiloff is not using hydrogen.

The atmospheric boiling points of the chosen propellants are as follows:

Substance Boiling point
Oxygen 90 K (-183 °C, -287 °F)
Methane 111 K (-161 °C, -258 °F)
Hydrogen 20 K (-253 °C, -423 °F)

Space is a place of temperature extremes: roasting in the sun, but pretty cold in the shade. Wikipedia gives a minimum (presumably night time) temperature at the lunar equator of 100 K, and 35 K in some polar craters. These temperature figures suggest it should be theoretically possible to keep an unmanned spacecraft at around 100 K with a simple sunshade, which would enable methane to be kept liquid indefinitely at slightly below earth atmospheric pressure and oxygen at slightly above earth atmospheric pressure.

Hydrogen is traditionally considered the best upper stage fuel, despite requiring large tanks due to its low density, because of the exceptional specific impulse it achieves. It would have been simpler to generate hydrogen propellant on Mars than methane as it would only require a water electrolyser without the additional step of the Sabatier process to convert hydrogen and carbon dioxide into methane. But after considering hydrogen briefly in the earliest stages of Raptor engine development, SpaceX quickly abandoned it and I suspect that the main concern was boiloff on long duration missions.

Other than choosing not to use hydrogen, I doubt that SpaceX have given a lot of thought to the issue of avoiding boiloff on Mars missions yet. While the figures above show that it is theoretically possible to keep the chosen propellants liquid without an active system, there are many factors to be considered and I believe some active systems will be required. An entirely passive system would probably require the ship to be pointed directly towards/away from the Sun in order to keep the sunshade down to 9m diameter, and that may be unacceptable. Also, the crew area will need to be considerably warmer than the propellant tanks and some heat may leak through. As Geoffc has noted in his answer, small inner tanks for containing the landing propellant (which among other things would help reduce the insulation requirements) have appeared in some of SpaceX's designs.

Note that the James Webb Space Telescope will be positioned in a halo orbit about the Earth-Sun Lagrange point and will maintain a temperature of 50 K apparently by using only a passive sunshade.

  • $\begingroup$ What should be pressure inside 2 header tanks during entire Mars trip, 6 bar like in the main tanks or different for every tank. If header tanks will need some active cooling, what are their options. Using LOX to cool CH4 header tank and liquid nitrogen to actively cool LOX header tank. $\endgroup$
    – David Cage
    Commented Oct 7, 2021 at 10:53
  • $\begingroup$ Didn't SpaceX plan always to use subcooled LOX, CH4 for Starship/Super heavy launches. I doubt they will try to sub cool propellant on Mars, but they can for launches from Earth. So will CH4, LOX inside 2 header tanks be subcooled and can it help to avoid boiling off. Or it wouldn't have big effect because 1 day after launch temperature inside header tanks quickly drops back closer to CH4 and LOX boiling points. $\endgroup$
    – David Cage
    Commented Oct 7, 2021 at 11:11
  • $\begingroup$ @DavidCage The first published starship design (the ITS) was in 2016 and considered CH4. But raptor development goes back much further and in the early days back in 2009 they were considering H2. There's no doubt they would subcool on earth because they were already doing that for Falcon. But Falcon has O2 in the top tank while Starship has O2 in the bottom. This concerns me a little because CH4 freezes at -183C and you'd think subcooled O2 might freeze CH4 in the pipes. It's possible the CH4 contains impurities (e.g. C2H4) that lower the freezing point but if so they'e keeping it secret $\endgroup$ Commented Oct 7, 2021 at 21:34
  • $\begingroup$ @DavidCage On Mars they's presumably use the starship itself to store the manufactured propellant and that will likely require both insulation and refrigeration. In space you can use the fact that temperature is close to absolute zero in the shade to keep propellant cool. On Mars you can have daytime temperatures above water freezing point. At night it's -100C but that's still too high to prevent propellant boiloff. Refrigeration may be done with a heat transfer coil in the tank, probably N2 loop (boilng point -196C at 1atm.) $\endgroup$ Commented Oct 7, 2021 at 22:03
  • $\begingroup$ @DavidCage Interesting point about header tank pressure - it could be higher than 6 bar but needs to be a lot higher to make much difference to the boiling point. Spacex is going to have its hands full with NASA's moon plans for a while and that may help with experience in mining ice - which is essential to the propellant factory for Mars. I think it would be good to make propellant on the moon by water electrolysis. Possibly just store the O2 and throw the H2 away as O2 is 80% of the propellant mass for CH4 fueled rockets, 89% for H2 fueled. No known way of making CH4 on the moon I guess. $\endgroup$ Commented Oct 7, 2021 at 22:13

Liquids boil when their vapor pressure exceeds ambient pressure. The structure of Starship tanks limits ambient pressure to 6 bar. Therefore, to prevent boiling you must refrigerate to below -170 °C (for O2) and -145 °C (for methane).

When Starship is refueled in Earth orbit, the fuel can be sub-cooled, but not below -182 °C since that’s the freezing point of Methane. To prevent boil-off, Methane must be maintained between -182 °C and -145 °C for the duration of the voyage. The equivalent numbers for O2 are -219 °C and -170 °C. If the two tanks are in thermal equilibrium, the range is -182 °C to -170 °C to keep both liquid.

enter image description here

Left unregulated, are tanks going to heat up or cool down on a Mars trip? Brylle Reyes in https://www.academia.edu/934756/Thermal_Control_Handbook has an interesting concept: a standard 1 m Black Body sphere with absorptivity = 1.0, at thermal equilibrium with space. If the sphere is 1 AU from the sun, equilibrium temperature is +6 °C. At Mars distance, the temperature is -47 °C. Not surprisingly, these temperatures are close to the corresponding average planetary temperatures.

So, if we spray Starship with BBQ paint and spin it with axis perpendicular to the ecliptic, its temperature would approach these equilibrium temperatures. During the entire trip, skin temperature would decrease, but tank temperature would increase . The BBQ black Starship will need cooling to maintain tank temperatures in the “Goldilocks Zone”,.

Of course, Starship is not BBQ black. The heat shield is high absorptivity and highly insulated. The shiny side is lower absorptivity but uninsulated. It’s like a sleeping bag with one side made from a space blanket and the other side black, fluffy lamb fleece. Lay too close to a hot campfire on a very cold night, and you are in the same dilemma as Starship. If you roll the insulated side toward the fire, the insulation will protect you from overheating and you will freeze your buns on the other side. Roll over, and you will warm up again.

This is likely what Starship will do: use attitude control for passive thermoregulation to maintain tank temperatures in the Goldilocks Zone.

There are other tricks: For Black Bodies, Absorptivity = Emissivity = 1.0 at all temperatures. Real materials have differing emissivities at different temperatures. This graph shows emissivity profiles for polished SA508 steel, and with 3 different oxidation treatments. https://www.sciencedirect.com/science/article/abs/pii/S0017931017325802

Emissivity of oxidised SA508

Solar panels can do double duty as sunshades. Turning the nose towards the sun can warm the payload while shading the tanks.

If active-ish cooling is needed, liquid CH4 or O2 can be circulated through a radiator on the shaded side where it can radiate to the void, trying to equilibrate with the microwave background temperature. If Starship were turned nose-to-the-sun, Methane could be pumped through the bells of the Raptors as radiators. The tank contents could be cooled while the payload is warmed in the sun at the other end of Starship.


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.