Helium-3 is a substance that is very useful for nuclear fusion, which can generate a lot of power. Unfortunately we don't have a lot of it on Earth, but on the Moon it can be processed from basic moon rocks. Say that we figured out a way how to mine large amounts of the stuff in relative safety, but we still need to process the rock into Helium-3 Where would be the best place to do that, on the Moon itself or on Earth? The ratio of moon rocks to Helium-3 is not that spectacular (according to Popular Mechanics a patch of three quarters of a square mile to a depth of about 9 feet nets about 220 pounds of Helium-3), and moving enough moon rocks to Earth for processing is rather costly. Moving the processed Helium-3 back to Earth would require a lot less moving things around in space ships. However, it would require the facilities and knowledge to process this to be present on the Moon.

So which is the more viable option, to move all the moon rocks back to Earth and process them here, or have the processing done on the Moon and move the Helium-3 back here?

  • 9
    $\begingroup$ A lot of questionable assumptions around uses and demand and viable sources for He3 are built into this question imo. Speculation on the back of speculation to have He3 fusion reactors at all, and even assuming them, Earth based sources, including making it from tritium, will offer the option of not needing lunar sources. I am not convinced lunar He3 offers a commercial pathway to exploitation of non-terrestrial resources. $\endgroup$
    – Ken Fabian
    Mar 4, 2019 at 2:14

4 Answers 4


The Moon would be a much better place most likely. As you said, 220 pounds of Helium-3 in a mass of many many tons of rock, makes it so that even a few tons of equipment to be dropped on the Moon would vastly reduce the price to return it home. Lifting 220 pounds from the Moon to return to Earth is relatively easy, all of the Apollo missions did it, and then some. How much mass is in that 3/4ths of a square mile, 9 feet deep? Well, the density of lunar regolith is about $1.5 g/cm^3$ The volume of 3/4 square miles 9 feet deep is $4.370625*10^{12} cm^2$, or about 6,500,000 metric tons! I can't even imagine how much rocket power would be required to lift that much!

I suspect the machinery to process the rock, and the power, would amount to maybe 100 tons, if not a bit more. Even if I'm off by a factor of 100, it is still vastly cheaper to process the Helium-3 on the moon rather than return the rocks to Earth. According to this article, the main thing involved in the separation is heating the rock to 600 °C, which wouldn't require much in the way of expensive equipment.

Next, is there something inherit in the other Moon rocks that would be valuable? Not really. Most of it is similar to composition on Earth, including somewhat more aluminum and volcanic rocks. Also, the Moon is very dry.

The other argument is that the filtered Moon rock could then be used to make structures on the Moon, which could likely make things even easier to continue the processing in the future.

  • 2
    $\begingroup$ If you don't want to imagine the power needed to fly 6.5 million tons of rock around, imagine what kind of ship can regularly fly up and down to the Moon with a cargo bay full of moon rocks, then divide 6.5 million tons by that cargo bay size. $\endgroup$ Feb 2, 2016 at 13:24
  • 1
    $\begingroup$ @ThomasJacobs If there're He-3 fusion rockets around, space transports might not be much of a problem. But in that kind of scenario one could also cheaply bring a million ton processing industry to the Moon. A future cost/income trade off is inherently unpredictable (because all investors compete to second guess it) even more so than guessing what will be discovered next in engineering. $\endgroup$
    – LocalFluff
    Feb 2, 2016 at 14:37
  • $\begingroup$ There's one last question, how big and complex would the He-3 extraction machinery be? If it can be fit on a pair of trucks, there's no question, the Moon. But if it's the size of an oil refinery? $\endgroup$
    – SF.
    Feb 2, 2016 at 17:27

Two big ifs here. IF we achieved viable commercial fusion power (other than the sun) and IF Helium-3 was an indispensable part of this process. But for the sake of argument, let's say Helium-3 is the fusion fuel of the future.

I'll quote John Schilling's comment from Rand Simberg's Transterrestrial Musings blog.

Helium-3 mining on the moon simply does not pass the arithmetic test. The highest 3He concentration ever recorded in lunar regolith is fifteen parts per billion, and the process by which it is deposited is inherently resistant to geologic concentration. Assuming someone manages to invent a 3He fusion reactor that operates at 50% efficiency (giggle), that translates to net energy output of 4.5E6 joules per kilogram of high-grade regolith. The energy output of a kilogram of the lowest grade of coal burned in a good 19th-century reciprocating steam engine, is about 4.5E6 joules per kilogram. And that doesn’t change if you substitute dried peat for the coal. So, the proposal is to set up an enormous mining infrastructure on the Moon, and invent a fundamentally new kind of engine backed by fifty years of failed promises, for the sake of an energy source roughly as good as burning high-grade dirt in a type of engine obsolete for over a century. And no, that analysis doesn’t change significantly if we include accessible reserves or environmental impact. I understand that you want desperately to believe that there are immense riches to be had in space, as soon as the suits see the light and come up with the money. The good news is, this is probably true. But the list of great riches to be had in space, does not include lunar helium-3 (or helium-4, for that matter). The numbers do not add up, no matter what the glossy magazine articles say, and math trumps faith.

So a huge volume and mass of regolith that contains only minute amounts of Helium-3.

To answer your question, It'd be less expensive to extract the Helium-3 on the moon. That would save the transportation expense of delivering mountains of mass to earth's surface.

But I don't think Helium-3 will ever be an economic driver for lunar development.


The appeal of 3He-D fusion is the lack of neutrons, which waste energy (being uncharged and all, you can't collect their energy) and when the neutrons collide with the reactor, they make the reactor itself radioactive (so 'easy' D-T fusion creates nuclear waste).

Lunar 3He can be obtained essentially by scooping up the regolith and baking it to release solar-wind-implanted gases adhering to ilmenite grains in the lunar regolith. This is best done on Luna rather than elsewhere, as the rarity of 3He in the lunar fines means you'd be transporting tremendous masses of materials for processing elsewhere.

John S. Lewis' "Mining the Sky" discusses this in greater detail.

  • 1
    $\begingroup$ Neutrons might be uncharged and have no electronic stopping power, but they are certainly stopped anyway via nuclear stopping. "Use the (strong) force, Luke!" You can collect their energy and you would want to do so; it's called shielding, necessary to protect people, electronics, and some structural materials as well. $\endgroup$
    – uhoh
    Mar 4, 2019 at 2:33

Helium-3 is NOT a substance that is very useful for nuclear fusion. At least, not man-made fusion.

The easiest nuclear reaction, at the lowest energy, is Deuterium, tritium, and this is still not close to break-even despite decades and billions of dollars spent.

3He fusion is much more difficult.


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.