Is it possible to melt regolith from lunar orbit with an energy beam?

Suppose we wanted to create landing pads, building foundations, and roads on the moon that were relatively free of lunar regolith. Could we do so by putting satellites in orbit around the moon that emit energy beams that melt lunar regolith. The molten regolith would then harden, and we would have hard and dust-free surfaces to land on, build on, and drive on.

The energy beam could be lasers, large mirrors that focus sunlight, microwaves, etc. The beam needs to be steerable. And it's OK if we need several satellites (but less than 100) working together either to get a strong enough beam or to cover enough area.

The process can be slow. If the system can only produce like 10 cubic meters per day of foundation that's OK. We also don't need to be able to hit every part of the moon. Even being able to put stuff anywhere on like 25% of the surface is OK.

Also assume we have a spaceship with a 9-meter payload diameter and 100 tons of capacity to get the satellites to the moon.

Some issues to consider are...

  • How much energy per cubic meter does it take to melt regolith and can reasonably sized satellites gather and emit enough energy to melt enough of it.

  • How deep do we need to make the foundations (and can we melt that deep). Assume we might want to land something like NASA HLS weighing 100 tons onto a 15m x 15m pad made by these satellites.

  • $\begingroup$ In a hard vacuum no material melts, it only vaporizes. Liquids do not exist at zero pressure. It might be possible to sinter the surface dust, but I doubt that would create a better landing pad. $\endgroup$ Commented Feb 2 at 14:38

2 Answers 2


Lunar soil is the name for the portion of lunar regolith I imagine you're thinking of - The dust that gets everywhere and causes problems. A minor thing, but having the proper search terms should help you with further research.

According to this article, (https://www.theguardian.com/science/2023/oct/12/streets-on-the-moon-lunar-dust-could-be-melted-to-make-solid-roads), scientists from the ESA made a substitute for lunar soil and used a laser to turn it into solidified triangular tiles. A 10m X 10m landing pad would take 100 days using this method (1 square meter per day for one laser). To replicate this on the moon, a ~2.4m^2 lens would work. Unclear whether these lenses would be on land rovers or satellites, but given that the article cites concerns of the lenses being covered by lunar soil, I think they're thinking about land rovers.

Given that the European Space Agency is doing this research, I think their tiles should be sufficient for landings.

Hope that helps! The paper is in Nature magazine so you know it's legit: (https://www.nature.com/articles/s41598-023-42008-1#:~:text=Lunar%20regolith%20can%20be%20sintered,quality%20of%20the%20manufactured%20parts.)


According to this, Lunar Soil melts between 1370 - 1650 Kelvin. Working with hot stuff in space is always hard because of waste heat. Satellites equip with Lasers is probably your worst choice because Lasers have terrible efficiency.

As you said, using parabolic mirrors to focus sunlight on the surface is a significantly better option. Since Mirrors can be 99+% efficient. And are much simpler to build.

At the moon, the Solar energy per m² will be about the same as on Earth. So 1350 W/m². The specific heat capacity of Lunar Soil is, according to the same doc, 1512 kJ/Kg K. Using this calculator you can guestimate it will take about 700000 Watt Hours to melt 1 kg of soil. I.e raise the temperature to the melting point.

So a single Satellite with an area of 520 m² would need 1 hour to melt 1 kg. Call it 500 m².

500 m² may sound like a lot, but thats only a 12,5 meter radius circle. We have solar sails rn bigger than this. And with a bit of creativity you should be able to make something drastically bigger. If, for the sake of argument, you were able to make 1 kilometer radii mirrors, each would have an area of 3141592 m². Or in other words, be able to obliterate 6283 kg of Lunar Soil per hour. Or, roughly, 1.74 kg/s.

Thats still not very fast. 1m³ of soil weighs 1500 kg. So even this mirror would work on this for 15 minutes. But if you make the road only say 10cm deep, then you only need to melt 150 kg. Which the laser can do in 1.4 minutes. And thats for a 1x1x0.1 meter patch.

My guess is you would do this differently. Use a bunch of Mirrors that kind of "Scanline" over the surface 100s of times per second. They gradually heat up a huge area until it all melts at once and bam there is your street.

Now of course, you still need to melt the same volume but this way you might avoid heat expansion issues.

We have also been assuming the lunar soil is at 0 Kelvin itself. Which it isnt. In reality, it gets like idk 500 kelvin probably. So you only need 2/3 the work. So it might take as little as a minute per 1x1x0.1 meter block.

If we, for the sake of argument, wanted to make a 100 kilometer long road, 4 meters wide and 0.1 meters deep that works out to a lot of soil. Like... 40000 m³, 60000000 Kg. With one mirror that might take 27 years. But with 100 of these mirrors, you can get the job done in 100 days. Again assuming the soil is at 0 Kelvin which it isnt. Might only take 70.

Now, how good this actually works is up for debate. There is some credence to melting rock, but it just has never been done on the moon. So idk, but it should be good enough for any story.

  • 1
    $\begingroup$ Not just specific heat, you alse need to figure in heat of fusion. $\endgroup$
    – Jasen
    Commented Jan 14 at 10:29
  • $\begingroup$ @Jasen Also, since heating takes time, probably heat loss into the surrounding soil. But then the calculations start getting a lot harder. $\endgroup$
    – user4574
    Commented Jan 15 at 3:30
  • 1
    $\begingroup$ just to get more detailed: The moons surface reaches up to 400K at the equator during (moon)daytime. The poles are as cold as 20K. And everything in between... well in between. But yeah you'll want to focus the melting operation on the area that is currently fully exposed to the sun to increase speed. $\endgroup$
    – datacube
    Commented Jan 15 at 12:45
  • $\begingroup$ @user4574 the units are in Kilojoule / kilogram / hour. Time is already taken into account. We are assuming the net energy gain /s is greater than the loss due to radiation. Otherwise the rock wouldnt heat up at all. $\endgroup$
    – ErikHall
    Commented Jan 15 at 12:49

You must log in to answer this question.

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