[db48x]

Don't get too starry-eyed; this is deuterium-tritium or deuterium-deuterium fusion they're talking about. It produces neutron radiation so you still need shielding, and it really complicates the materials engineering. All the properties of the material you're using to do some job depend on what kind of atoms are in it, and all those neutrons are busy transmuting those same atoms. The first working fusion reactor will probably have crazy maintenance requirements. (It's already a bit of a problem with fission plants, but with the simpler fusion reactions as much as 80% of the energy is produced in the form of neutrons.)

The holy grail that you're thinking of is aneutronic fusion, usually deuterium-helium3 or deuterium-lithium6. He3 is super rare though, so a D-He3 reactor using it might need to get the He3 from a bigger fusion reactor using D-D fusion, which produces it (and that pesky neutron).

Even better still might be proton-boron fusion; it needs temperatures an order of magnitude higher than D-D fusion (and magnetic confinement two or three orders of magnitude stronger), but produces far fewer neutrons (there are fewer undesirable side reactions). Alas, this route will produce four orders of magnitude less energy than the much simpler D-D reaction.

Given the scale of the engineering problems, we could even end up harnessing fusion power by building dyson spheres; a star might be the only feasible, stable way to build a fusion reactor. Hopefully our universe was set on an easier difficulty setting than that when it was instantiated.

[mcv]

> The holy grail that you're thinking of is aneutronic fusion, usually deuterium-helium3 or deuterium-lithium6. He3 is super rare though

Finally a good reason to go to the moon!

[patrickk]

Thanks for the information. I was thinking of the implications of the dream scenario, not that I think it's actually plausible. From what you've said it really sounds like a compact, low maintenance fusion reactor is still way, way off, perhaps a century or more (if it's feasible at all.) Maybe we need a working quantum computer to run simulations to build a fusion reactor, another dream project that is also constantly a decade away, or a super intelligent AI to tackle the problem ;)