Hacker News new | past | comments | ask | show | jobs | submit login

I'd love to see some references on those three claims. None of them make sense to me.



There is a very well researched youtube video that explains a couple of these things:

https://www.youtube.com/watch?v=FrUWoywZRt8

I do not recall where I heard about the helium-3 situation, but I recall hearing some figures and the prospect of having enough to run fusion reactors was not good. Doing a search suggests that I had been only slightly mislead about the scarcity of helium-3. It is still extremely rare, but the US reportedly has used up to 70,000 liters of it per year:

https://en.wikipedia.org/wiki/Helium-3#Human_production

The density of Helium-3 at STP is presumably 0.1338 g/L, based on Helium-4's 0.1784 g/L. That suggests that the total annual US industrial demand is ~9 kg. This is definitely not as bad as I had thought, but it is still fairly dire.

As per wikipedia, a single 1GW nuclear fusion reactor that is 100% efficient at converting helium-3 into electricity would need 52.5 kilograms per year. 6.7 metric tons per year would be needed to power the entire US.

https://en.wikipedia.org/wiki/Helium-3

That assumes 100% efficiency, and if the conversion efficiency is anything like a nuclear fission plant, we would be lucky to see 3% efficiency, but we could be optimistic and assume something higher. Either way, it does not change the conclusion.

With an annual supply of ~9 kg, practical helium-3 fueled nuclear fusion reactors are not happening. Maybe the supply would be higher if you include other countries, and maybe we could get it somewhat higher if we try, but the reality is that helium-3 is an extremely rare isotope and the idea of using it as a practical fuel for a nuclear fusion reactor is a pipe dream unless the supply problem is solved and people figure out how to actually build a reactor that generates power using it, without encountering any of the other known problems that make this unlikely.

The difficulty in scaling supply is why are people discussing wild ideas like mining the moon, or even mining Jupiter. The supply situation is so constrained that the US government is reportedly buying 3 liters of it from a company that promises that it will strip-mine the surface of the moon for it with delivery by April 2029:

https://www.morningstar.com/news/pr-newswire/20250507sf81778...

I fully expect them to fail to deliver.

That said, the fairy dust remark was probably inaccurate, but the idea that our supply is short by orders of magnitude is correct according to mathematics.




Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: