Let's say that my colony ship crash-landed on an atmosphere-less planet. Sure, there might be sub-surface pockets that have evolved primitive life, but sulfur and phosphorous are at a premium in the self-sufficient underground habitats us colonists have excavated, and we're not giving any of it over to the purpose of farming the native algae for fuel.

Our 3D printer failed, since it wasn't advanced enough to print its own replacement parts or replacement printer (a way of avoiding rampant AI, long story). Assume the planet is identical in composition to Earth, with several big changes: no atmosphere, no biosphere, and essentially none of its hydrosphere (minus sub-surface pockets of water), and that it has never seen any form of widespread life - it has no fossil fuels or hydrocarbons to burn. Also, assume we have knowledge of a 21st-century level of technology on hand but have largely been set back to the beginning of the Industrial Revolution in terms of what we can actually produce.

How do we re-invent the steam engine under these circumstances?

  • 3
    $\begingroup$ How much power do you require ? For what ? A steam engine sounds like an odd chocie under these circumstances. There are probably batteries and solar panels on the ship or enough gear to make e.g. a heat pump. What do they eat because presumably that's a potential fuel. You told us what the planet has not got, but not what is easily available. $\endgroup$ Nov 14, 2021 at 16:55
  • 16
    $\begingroup$ If you are stranded on a lifeless, airless world and your production capabilities are no better than late 17th century (= "beginning of the Industrial Revolution"), then you are going to die. Sorry. $\endgroup$
    – AlexP
    Nov 14, 2021 at 16:57
  • 8
    $\begingroup$ You mean that right now they have self-suficient habitats. But they cannot make spare parts, because their production capabilities are limited to late 17th century. Nothing made by mortal human hands has an infinite life. The habitats will fail, and they cannot repair them. They are dead. $\endgroup$
    – AlexP
    Nov 14, 2021 at 17:07
  • 1
    $\begingroup$ Modern parts and components require modern a industrial base to drive it. Many key components will be impossible with any 3D print technology as such a device can't do everything (and you have in fact ensured this because it can't build a clone of itself). Knowledge of how to do something does mean you can do it in a practical sense. In the absence of lots of (pure) water many processes would become close to or actually impossible, meaning they'd need to be reinvented. $\endgroup$ Nov 14, 2021 at 17:32
  • 1
    $\begingroup$ "The problem is recapturing the vital elements once you burn something.": no, that really isn't the problem, or even a major problem. The problem is that steam engines are just a way to convert whatever power source you use to produce the fuel (and the oxidizer, since the planet has none) into mechanical power, and you appear to want to use these steam engines as your power source. That's not going to work. $\endgroup$ Nov 15, 2021 at 21:52

7 Answers 7


You have three ways to do this, all fairly well-established current tech.

First, if your planet is tectonically active*, you can have geothermal power plants. These are basically steam engines, though in most modern tech the working fluid is a hydrocarbon with a lower boiling point.

Second, you have a solar plant, with a large array of mirrors concentrating sunlight on a boiler: https://www.eia.gov/energyexplained/solar/solar-thermal-power-plants.php

Third, if you can locate some uranium or thorium ore, you can leverage power from the first two to build nuclear power plants. Yes, nuclear plants are just steam engines :-)

However, you have another technical hurdle to overcome. Like any heat engine, steam engines work on temperature differences. Since your planet lacks air or water to serve as a heat sink, you'll need a large radiator to maintain the needed temperature difference. You might consider the radiators on the ISS as a model: https://en.wikipedia.org/wiki/External_Active_Thermal_Control_System

*I've read that plate tectonics might depend on plentiful water, but I don't know enough to expand on that.

  • $\begingroup$ No surface or free water doesn't necessarily mean no water - just all water trapped as hydrating water in the substance crystals/molecule - remember how CuSO4 is white if you heat it to eliminate the water and blue when just crystalized from a solution? Amount of water in Earth's mantle $\endgroup$ Nov 14, 2021 at 23:40
  • $\begingroup$ Nuclear energy with 1850 technology - just forget it when stranded on a deserted island. $\endgroup$ Nov 15, 2021 at 0:05
  • 1
    $\begingroup$ @Adrian Colomitchi: Sure, but the problem is how to use that hydrated water as an effective heat sink. As for creating a nuclear reactor (not a bomb) with 1850s tech, you actually don't need tech at all if you have an appropriate ore body: en.wikipedia.org/wiki/Natural_nuclear_fission_reactor $\endgroup$
    – jamesqf
    Nov 15, 2021 at 5:00
  • $\begingroup$ "Sure, but the problem is how to use that hydrated water as an effective heat sink" Yeah, nah, mate. I didn't say you should, that was just in regards with "I've read that plate tectonics might depend on plentiful water" which I interpreted as you saying that no surface water would inhibit the place tectonics (shouldav been clearer, sorry for that). As for the nuclear reactor, without at least concentrating the ore, the energy production requires a massive volume of moderator (you want it penetrating in every nook and cranny) and then the heat will be too diffuse to create enough steam. $\endgroup$ Nov 15, 2021 at 5:36
  • 1
    $\begingroup$ Nuclear power is not as easy as putting some ore together and building a society on the power. The natural reactor you talk about was an abnormal amount of uranium getting washed up together and burnt for 100's of thousands of years and you probably couldn't boil a tea kettle off of it as the energy was dispersed over a very large area. Plus, if you managed to create a sustainable reactor pile, to get energy out of it, you will need to shield it to prevent issues with radiation. Last, Thorium is not fissile. a breeder reactor is not simple to make or process the fuel $\endgroup$
    – Sonvar
    Nov 15, 2021 at 20:13

Since you have been able to dig underground and you have found some food source, just keep digging.

Assuming a geothermal gradient similar to that of Earth, 25–30 °C/km, "just" reaching 3-4 km underground will give you enough to boil water and start getting more energy so that you can dig even deeper and reach higher temperatures, with even higher yields.

  • $\begingroup$ How do they harness deep geothermal power, with 17th century industry? $\endgroup$
    – PcMan
    Nov 14, 2021 at 17:34
  • $\begingroup$ You don't need to drill deep to generate power. You can use horizontal geothermal systems. $\endgroup$ Nov 14, 2021 at 17:35
  • 4
    $\begingroup$ @StephenG did you get the fact that "horizontal geothermal systems" is not a power generation unit, but a thermal mass to be used with geothermal heat pumps? Like in "Cheap and efficient application of reliable Ground Source Heat Exchangers and Pumps" and "Cheap and efficient application of reliable Ground Source Heat Exchangers and Pumps" by the same group? And that operating a heat pump requires one to consume energy and doesn't enable one to generate it? $\endgroup$ Nov 15, 2021 at 12:50

While you still have your 3D printers print mirrors. Solar power. Note that you need them anyway to smelt ore to make the metal to make your steam engine.

And note that the scenario is impossible--if you can grow food you have biomass.

  • $\begingroup$ Sure, you can produce biomass, but you can only produce so much of it at once, because the elements required to make biomass are in short supply. $\endgroup$
    Nov 15, 2021 at 2:34
  • 2
    $\begingroup$ @KEY_ABRADE If the supply of elements is that limited their supply will soon be lost to the environment and they'll die out anyway. $\endgroup$ Nov 15, 2021 at 2:44
  • $\begingroup$ Not if they mine more out of the crust, though. But to do that, they need power. $\endgroup$
    Nov 15, 2021 at 3:20

Step 1: Gathering Water

Option 1: Ice

Even on places we think of as desolate environments, like Mars, we still have massive polar ice caps filled with water. Quarry the ice, bring it back to base camp.

Option 2: Underground deposits

Even with no hydrosphere, there might still be some underground deposits of water, left from eons past. Use those as your source of water.

Option 3: Extreme water recycling.

We still need to use some water, but if we use extreme water recycling, maybe we can make it work. Space is cold, and without an atmosphere, your planet will be too. If we take the steam from our generators and pass it through a low-insulation tube in the cold outside, it will condense into water (or, without careful management, ice.)

Step 2: Boiling The Water

Option 1: Use sunlight

There are a few power plants today that use mirrors to boil water, in order to generate electricity. Even with no atmosphere, we still have a star that can be used. An array of mirrors that reflect light onto a source of water will quickly boil it, allowing for turbines to be used. It's the same as using a magnifying glass to light things on fire.

Option 2: Geothermal Energy

Underneath the earth, it gets HOT. Hot enough to melt metal, or boil water. Simply pipe your water (or drop your ice) into a deep shaft, and let the planet take care of the rest.


In The Martian (Andy Weir book, is great, read it), a ball of incredibly radioactive stuff is used to generate heat. Place a highly radioactive ball of stuff into a metal case, and drop it into a water tank. Voila, you have steam (and also cancer, if you're not careful)

Edit: Option 4: Burn Metal

Metal, when it's fine enough, burns. If you somehow powder up some of the metal you have, it can be burned for a carbon-free fire.

Notes and Considerations

  • You'll have to pressurize the water, because, like in space, it will boil off with no atmosphere.
  • It'll be hard to work with this stuff in an spacesuit, you might have to make a better thing

There's probably more I haven't thought of, but it will do.

  • 1
    $\begingroup$ "Voila, you have steam (and also cancer, if you're not careful)" That gave me a little chuckle. $\endgroup$
    – Sonvar
    Nov 15, 2021 at 20:16
  • $\begingroup$ solar heating actually works a lot better without an atmosphere but cooling gets much harder and a turbine needs both $\endgroup$
    – John
    Nov 16, 2021 at 1:09

Not a scientist, but if you don't have your heart set on steam, you probably have plenty of temperature differential to create Stirling engines. If you've got no atmosphere, I imagine outside temperatures drop pretty low out of sunlight and get pretty high in the light. I've heard they aren't very efficient, but they don't use fuel and should run for most of the time day or night with the differential between outside temp and underground temps. Connect them to electrical generators and go from there.

The only downtime would be as the temperature outside swings from hot to cold and vice versa and crosses the temp of your habitats. If the engines have any tendency to not self-start, it may give you issues. But I also believe I've read in other stories that even on the day-side of the moon and such, it's still freezing or lower in the shadows, so maybe just shade the "south" side with a wall or dunes and the tops stay frosty.


No steam needed

I assume that the star your planet is orbiting is close enough to provide as much heat as the sun on earth. However, without an atmosphere the energy reaching the ground will be a lot more, and the thermal gradient between light and dark areas will be huge. In these conditions thermocouples will provide enough energy even if they are not efficient.

First you should smelt some metals concentrating solar power with mirrors; metals with low melting point like lead and zinc would be enough. They would provide the material also for the cables even if they are not the best conductors, in any case without an atmosphere they would not oxidize.

Then you should find a canyon with steep cliffs that shield the bottom from the sun light. Place the receiving couple on top of the cliff and the heatsink on the bottom and you'll have enough energy.

For the night time initially you'll have to find a way to store the energy in the ship systems, but then you can dig some wells and use the the thermocouples to exploit the geothermal energy.

On earth the efficiency would be 2/3%, with the gradient in space it would be more, but not much, on the other hand it will be very simple to build, even people no longer used to craftsmanship because they 3D print everything could do it and there would not be mechanical parts that in an environment with high thermal stress would break easily.


Heatless steam engine.

Note that in a vacuum water will boil immediately. You do not need to heat it. You have no atmosphere on the surface and so water will boil.

But you are not on the surface. You are / in the self-sufficient underground habitats us colonists have excavated /. I am going to assume you have found places that have atmospheric pressure below ground.

You will run your steam engine to harness the differential pressure between the surface vacuum and your subsurface pressurized habitats.

A connecting pipe down in your habitat is opened to the surface then closed. Water in the pipe will boil and produce pressure within the pipe. This pressure will drive a piston and produce energy. When the water has finished boiling, the conduit to the surface is closed and below ground pressure is allowed to compress the water back into a liquid.

The trick would be using the pressure below ground in such a way that it could replace your piston but did not entail venting your atmosphere topside. I am the architect and that is a question for the contractors to figure out.

  • 5
    $\begingroup$ Water will indeed boil under a vacuum... and it will produce steam at the vapor pressure of water. That depends on temperature, of course; for example, 0.02 atmospheres at 20°C, which is the pressure of a column of water 20 cm high. (And once the superficial layers of water boil off, the water under them will freeze, as being deprived of any fast moving molecules.) $\endgroup$
    – AlexP
    Nov 14, 2021 at 19:03
  • 4
    $\begingroup$ There's no such a thing as "heatless steam engine" water will need heat to boil anyway. Thou shall not pass the latent heat of vaporisation, not even in vacuum. $\endgroup$ Nov 14, 2021 at 23:48
  • $\begingroup$ This steam engine is going to run on the hate! $\endgroup$
    – Willk
    Nov 16, 2021 at 2:35

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .