If Starship carries 100 passengers to Mars, they will consume about 6 metric tons of oxygen and produce about 10 tons of CO2 on each leg of the trip. Much more if they exercise.

CO2 could be converted back into O2 using algae bioreactors. There is a demonstration unit on the ISS. According to NASA data, it takes 1000 liters of bioreactor media per person to convert the CO2 back into O2. That's one metric ton per person. A one-person Bioreactor consumes about 2.8 kW. So for the whole Starship crew, 280 kW would be required. The panels on the ISS top out at 120 kW

On US nuclear submarines, they use oxygen "candles" to provide crew with oxygen. But because the "candles" rely on iron as the reducing agent, they are heavy. It would require 46 metric tons of oxygen "candles" to provide the Starship crew with O2 for one leg of a Mars trip.

The LSR demonstration system on the ISS reacts CO2 with Hydrogen to produce water and methane (Sabatier reaction). Some of the water is consumed in the device and some is available for other uses (such as electrolytic production of O2 and H2). This is an open cycle with reported 50% loss of O2 each cycle. So it could reduce O2 demand by 50%

Any ideas how SpaceX is going to solve the O2 supply problem for the crew? Is the MOXIE demonstrator on the Perseverance rover a promising technology?

  • 5
    $\begingroup$ Oxygen candles on nuclear submarines are only an emergency backup option if the main oxygen source -- electrolysis of water -- fails, and if the first backup system -- high pressure oxygen storage tanks -- is depleted. Under routine conditions, oxygen candles are never used. $\endgroup$
    – MTA
    Commented Oct 22, 2021 at 17:05
  • $\begingroup$ You are right. I was groping for a method for oxygen generation. "Candles" obviously would not be suitable. $\endgroup$
    – Woody
    Commented Oct 22, 2021 at 19:51
  • $\begingroup$ If you want an explanation of how oxygen candles work, How Do Nuclear Submarines Make Oxygen?- Smarter Every Day 251 has a good in-depth explanation and demonstration starting at 9:45. Oxygen candles have been used on spacecraft, but, of course, may not be an appropriate choice, depending on various trade-offs. They are, obviously, not a renewable source of oxygen, or at least not easily renewable. $\endgroup$
    – Makyen
    Commented Oct 22, 2021 at 22:55
  • 2
    $\begingroup$ Since the question already has a thorough and very well-received (+18/-0) answer that steers clear of random opinions and presents a nice mix of facts and extrapolations, I don't think it's necessary to block answers by closing for " likely to be answered with opinions rather than facts', so voting to leave open! This will allow others the opportunity to post further answers. $\endgroup$
    – uhoh
    Commented Oct 23, 2021 at 12:15
  • $\begingroup$ The Bioreactor seems to be very inefficient. An astronaut at rest produces about 100 W of heat by using the oxygen and the food to produce the energy need for live. If we assume a mean of 300 W for the day of the astronaut and 25 % efficiency of the bioreactor, we get about 1.2 kW. $\endgroup$
    – Uwe
    Commented Oct 25, 2021 at 10:58

1 Answer 1


How does SpaceX plan to supply or recycle respiratory gasses?

As with most questions about SpaceX, the answer to this question is essentially that SpaceX is a private company, and unless either SpaceX or Elon Musk have made a public statement about this, the answer (IFF there is an answer at all, i.e. if they even have a plan), we simply do not know and cannot know.

In fact, even if there was a plan, and even if they publicly announced it, we have seen dozens and dozens of times that their plans change – and this is not an exaggeration – quite literally on a daily basis.

We can, however, make educated guesses. Elon Musk recently gave a two-hour long interview and tour to Tim Dodd (aka Everyday Astronaut), and over the course of this interview, he repeatedly said that at the moment, all their efforts are focused on how to get to orbit as fast as possible, since they have tested everything they can test so far, and really need to go to orbit to achieve further test objectives such as refining the heat shield, testing the hypersonic, supersonic, and transonic sections of launch and reentry, testing the stage separation "non-system", and so on.

At one point in the interview, Elon Musk even seemed annoyed about a question, when Tim Dodd asked him two consecutive questions about the oil rigs, and he answered both with "We're not thinking about it."

So, my guess is that the answer to your question

How does SpaceX plan to supply or recycle respiratory gasses?

is "They're not thinking about it".

And by that I don't mean that they don't plan to keep their astronauts alive (obviously), I mean that they simply ignore the problem until and unless it needs to be solved.

We have seen this a couple of times already:

  • Landing legs: it was clear that the landing legs used on Starhopper, Starship Mk1, Mk2, SN1, 3–6, 8–11, and 15–16 were always just placeholders, designed to do the minimal thing required for testing a single landing. And people constantly asked "When are they going to start designing the real thing?" – and in the end it turns out, they simply designed them away. If they had invested significant effort into designing the landing legs, all of that effort would have been wasted. Obviously, they will still need landing legs for the first couple of flights to the Moon and Mars, until they can install catching towers there, but those those can actually be significantly simpler and lighter than the ones needed on Earth due to the lower gravity.
  • Stage separation system: there had been questions about this. But until now, they never needed to separate stages, so why waste time in designing one? And in the end, it turns out, they don't need a stage separation system.
  • Hot-gas thrusters: they have been part of many official renders, and were mentioned by Elon Musk many times, but it took very long for them to appear in the form of fit check test articles. They were installed for a short time, then removed again, because they are actually not needed for the currently planned flight profile. And now, it turns out that using ullage gas venting for reaction control may very possibly be enough, so they don't actually need the hot-gas thrusters at all.
  • Lunar landing engines: they are shown on the official NASA and SpaceX renders, but when Tim Dodd asked Elon Musk about them, he said they are not even sure they need them at all. Depending on how exactly the landing profile of Starship turns out, and how exactly Lunar regolith reacts, landing on a single throttled down Raptor engine may actually be fine. Again, if it turns out they don't need them, why would they design them?

So, while I am sure SpaceX does not want their astronauts to suffocate, I find it plausible that "they're not thinking about that" until they actually get ready to fly a hundred astronauts for a period of months. Right now, the longest flight they did was about 8 minutes, was within breathable atmosphere the whole time, and had no astronaut on board anyway.

  • 4
    $\begingroup$ This sounds solid. First priority: get the pathway to orbit as cheap, as reliable and as soon as possible. With the ability to lob multiple tons into orbit cheaply, a lot of the other problems can be brute-forced out of existence. For example, they will use 6 tons of O2 for breathing? Just carry 6 tons of surplus O2 along. Extract and dump the CO2. $\endgroup$ Commented Oct 22, 2021 at 7:20
  • $\begingroup$ @PcMan, and how many tons of water? How many additional tons of food? How many tons of spare just in case the journey has to be aborted near to Mars and the "free-return" option has to be activated? This is not to say that they can't do it, but to have a rule of thumb to estimate how many tons of "life supporting" stuff a human need to carry with him for a long journey in space (Christopher Columbus had to solve such a problem too). $\endgroup$
    – Ng Ph
    Commented Oct 22, 2021 at 9:20
  • 3
    $\begingroup$ @NgPh Water you recycle. It's not too hard, and water is heavy. Food, you carry along. Unless you plan to stay for 5years++ its easier to carry it along than to grow it. Oxygen usage for respiration is so tiny, you just carry it along and barely give it a second though, mass-wise. Per-Person Yearly usage on the ISS, with no recycling: Oxygen= 310 kg. Food=908kg. water= 1606 kg. Of these, recycling water is "easy"-ish. Growing food is hard, and likely not worth the effort shortterm. Recycling Oxygen is a distant third to look at. Eventually. Maybe in the budget after next year's budget. $\endgroup$ Commented Oct 22, 2021 at 12:02
  • 7
    $\begingroup$ And again, let's not forget that SpaceX' motto is "design only as far as you need to". At the moment, they are flying 0 humans for 8 minutes. Soon, they will fly 0 minutes for 90 minutes. On Artemis, they will fly 2 humans for a week. 100 humans for 9 months, is waaay off in the future. $\endgroup$ Commented Oct 22, 2021 at 15:55
  • 1
    $\begingroup$ This is a sound policy for a couple of reasons, the main one being that it's much easier to figure out such problems when you have a reliable simulated environment readily available (i.e., not competing for space on the ISS). It's amazing how many things you realize that you need/don't need once you deploy a system. No point in designing something until you can start testing/using it immediately. $\endgroup$ Commented Oct 22, 2021 at 22:01

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.