Will fusion provide >2% of US electrical power before 2050?
➕
Plus
65
Ṁ61k
2049
19%
chance
Close date updated to 2049-12-31 11:11 am
Get
Ṁ1,000
and
S1.00
Sort by:

https://news.mit.edu/2024/tests-show-high-temperature-superconducting-magnets-fusion-ready-0304

Press release, but they've consistently been putting out press releases stating they are meeting their timeline, on time.

predicts NO

IMHO, if fusion ever gets commercialized it'll be first and primarily as a neutron source for high-burnup fission reactors and breeders. Each 24MeV D-T fusion releases a very high-energy neutron that can trigger a 200MeV fission. If you're doing fusion alone, those valuable neutrons mostly go to waste.

predicts NO

@JonathanRay To elaborate on this, what’s a better use of a neutron:

  1. Try to get deuterium to absorb it (extremely low cross section) so you get tritium which is difficult to fuse for 24mev to get your 1 neutron back. Maybe you can just barely break even on neutron economy here using expensive beryllium’s n-2n reactions if everything is perfectly designed to conserve neutrons.

  2. get any of a zillion fertile heavy isotopes to absorb it (large cross sections) to get a fissile isotope which is relatively easy to fission for ~200mev and get 2-4 neutrons back

I doubt fusion alone will ever be an economically competitive utility scale electricity source

@JonathanRay you're absolutely spot-on with this -- even if we solve the tremendous technical challenge of Q>1 engineering breakeven, the enormous capex + opex of a fusion plant and fusion fuel makes economic breakeven seem very unlikely. I would be shocked if by 2050, commercial fusion power hit the ~20 c/kWh needed to compete with even gas peaker plants, let alone ~10 c/kWh baseload nuclear fission or <2 c/kWh peak solar/wind. Technological advancements can make a lot of previously unimaginable things possible, but at the end of the day fusion won't make a dent on the grid if it's not profitable to do so.

The saying has been "fusion is 30 years away" for 50 years, but I think we've reached "fusion is only 20 years away". As an optimist, I take that to mean a 20% chance of actually getting there within 30 years.
Interesting question, thanks for asking! The main thing I knew about Helion is that Sam Altman invested, but I'm wondering if there's a way to more specifically profit off your belief that they are the next Theranos, especially through Manifold. Baseline, you could ask questions like "Will Helion hit X target by Y date." More interesting might be some index that tracks Helion's eg number of positive press mentions or valuation/size of next round? And then ideally find a Helion believer to take the other side of the trade.
Starting very low because D-D requires much higher temperature than D-T and we can't even break even with D-T yet and T/He3 fuels are prohibitively expensive until you solve D-D fusion. The number of neutrons produced from fusion will be nowhere near enough to breed as much T from Lithium as it consumes. Maybe excess neutrons from fission plants can will be used to produce enough T for a few reactors, but fusion is probably never going to provide a large fraction of grid power. I want Helion to succeed but based on my understanding of the physics I'm 98% confident that they're the next Theranos. They claim they'll be able to break even with D-D but nobody else has even solved D-T yet, and that's orders of magnitude easier.