4 ms·

Who knows, fusion could be key to accessing the next order-of-magnitude level for society. Perhaps it'll be key to large scale carbon sequestration to stabilize

Who knows, fusion could be key to accessing the next order-of-magnitude level for society. Perhaps it'll be key to large scale carbon sequestration to stabilize future climate change effects. If we had cheap Fusion tomorrow we could replace much of fossil fuels with synth fuels with neutral CO2 cycle. We get to reuse existing infrastructure but without massive co2.

pfdietz 5y ago
It could, but IMO it's a very low odds bet. There are fundamental physical reasons why fusion reactors will struggle to compete. Fusion is a technology that IMO persists on cultural momentum, a meme technology that is part of the future because that's a story that we've been told since the 1950s. I think the most likely end will be that controlled fusion will go into a similar niche as dirigibles and vacuum tubes as a retrofuturistic anachronism. It will be recognized as something that used to be part of the future, but no longer is.
- Retric 5y ago
  It might be be cheaper than fission. More complex reactor vs less need for thick containment walls and likely fewer NIMBY issues. Decommissioning costs are likely significantly lower. Operating costs and downtime are more questionable as you don’t need nearly as much security or to deal with enrichment and spent fuel. No need for radioactive plumbing etc. Even just constructing fuel rods is shockingly expensive. D-T requires extracting tritium from a breeder blanket which will likely be very expensive. But the ratio of T:D can be lowered the more efficient the design, with pure D-D designs avoiding that issue entirely.
  - pfdietz 5y ago
    I don't think the containment (and a containment will be needed) will be significantly cheaper. The fusion reactor itself is much larger, and accessing it (like, lifting the top off an ARC reactor) will require considerable volume. ITER has a large volume around it where robotic servicing equipment can be arranged to get inside the reactor. All this volume will have to be hermetically sealed from the outside world, since it will become permeated with tritium. Keeping that tritium from escaping the building will be a major headache (polymer seals cannot be used on penetrations, as tritium permeates through polymers). This will be true in any fuel cycle using deuterium, not just DT, since D+D -> T+p reactions will be occurring.
    - Retric 5y ago
      > Keeping that tritium from escaping the building will be a major headache JET already had robust procedures for handling Tritium. It wasn’t that difficult because we are talking about such small amounts and the T is lighter than air so minute releases aren’t a major public hazard. Don’t forget Fission reactors actually produce Tritium.
      - pfdietz 5y ago
        A commercial reactor will involve far larger amounts of tritium, and the structure of the reactor will be so hot that tritium will permeate through it. The amount of tritium made and consumed in a 1 GW(e) DT fusion reactor in a year would be enough, if it were all released, to raise two months of the entire flow of the Mississippi River above the legal limit for drinking water. Containment is going to have to be extremely good.
        Retric 5y ago
        Their never going to have a full years tritium on hand so that’s kind of an odd benchmark. Getting the cost of Tritium down to $10,000 per gram is somewhat optimistic, significant effort is going to be used to recover it. As to the Mississippi River flow estimate that’s something like 6 orders of magnitude larger than what I am referring to. Still, the other way of looking at that statistic is if half of the Tritium used per day was dumped into the Missisippi every day it would be considered safe to drink when well mixed.
        pfdietz 5y ago
        That figure was to point out the large volume flowing through the system. Containment will have to be very good. Only a tiny percentage of that tritium can be allowed to escape. Realize also that what has to be constrained is the cumulative leakage from all fusion power plants, not just a single one. The world would need on the order of 10,000 1GW power plants, to displace fossil fuels.
  - simiones 5y ago
    Here is an article about the probably unavoidable downsides of fusion [0], from the Bulletin of Atomic Scientists. To summarize their points: - neutron bombardment, the unavoidable consequence of the only realistic fusion reaction, deuterium + tritium, turns every known material brittle and radioactive; and can be trivially used for uranium enrichment - tritium is almost impossible to contain, and requires fission plants to create - the fusion plants could theoretically create a surplus, but they would have to recapture essentially 100% of a very hard to capture gas (today they normally leak about 10% of the injected tritium) - fusion plants need to be massive to be even close to break-even - the energy drain of the facility consumes a sizeable portion of the generated energy; most of this power drain is still required while the facility is not operating the reactor, increasing projected costs of running a facility - ICF tech is very close to fusion weapon tech, so there is a massive risk for proliferation from there as well; MCF tech is not, thankfully [0] https://thebulletin.org/2017/04/fusion-reactors-not-what-theyre-cracked-up-to-be/ https://thebulletin.org/2017/04/fusion-reactors-not-what-the...

pfdietz 5y ago

It could, but IMO it's a very low odds bet. There are fundamental physical reasons why fusion reactors will struggle to compete. Fusion is a technology that IMO persists on cultural momentum, a meme technology that is part of the future because that's a story that we've been told since the 1950s. I think the most likely end will be that controlled fusion will go into a similar niche as dirigibles and vacuum tubes as a retrofuturistic anachronism. It will be recognized as something that used to be part of the future, but no longer is.

Retric 5y ago
It might be be cheaper than fission. More complex reactor vs less need for thick containment walls and likely fewer NIMBY issues. Decommissioning costs are likely significantly lower. Operating costs and downtime are more questionable as you don’t need nearly as much security or to deal with enrichment and spent fuel. No need for radioactive plumbing etc. Even just constructing fuel rods is shockingly expensive. D-T requires extracting tritium from a breeder blanket which will likely be very expensive. But the ratio of T:D can be lowered the more efficient the design, with pure D-D designs avoiding that issue entirely.
- pfdietz 5y ago
  I don't think the containment (and a containment will be needed) will be significantly cheaper. The fusion reactor itself is much larger, and accessing it (like, lifting the top off an ARC reactor) will require considerable volume. ITER has a large volume around it where robotic servicing equipment can be arranged to get inside the reactor. All this volume will have to be hermetically sealed from the outside world, since it will become permeated with tritium. Keeping that tritium from escaping the building will be a major headache (polymer seals cannot be used on penetrations, as tritium permeates through polymers). This will be true in any fuel cycle using deuterium, not just DT, since D+D -> T+p reactions will be occurring.
  - Retric 5y ago
    > Keeping that tritium from escaping the building will be a major headache JET already had robust procedures for handling Tritium. It wasn’t that difficult because we are talking about such small amounts and the T is lighter than air so minute releases aren’t a major public hazard. Don’t forget Fission reactors actually produce Tritium.
    - pfdietz 5y ago
      A commercial reactor will involve far larger amounts of tritium, and the structure of the reactor will be so hot that tritium will permeate through it. The amount of tritium made and consumed in a 1 GW(e) DT fusion reactor in a year would be enough, if it were all released, to raise two months of the entire flow of the Mississippi River above the legal limit for drinking water. Containment is going to have to be extremely good.
      - Retric 5y ago
        Their never going to have a full years tritium on hand so that’s kind of an odd benchmark. Getting the cost of Tritium down to $10,000 per gram is somewhat optimistic, significant effort is going to be used to recover it. As to the Mississippi River flow estimate that’s something like 6 orders of magnitude larger than what I am referring to. Still, the other way of looking at that statistic is if half of the Tritium used per day was dumped into the Missisippi every day it would be considered safe to drink when well mixed.
        pfdietz 5y ago
        That figure was to point out the large volume flowing through the system. Containment will have to be very good. Only a tiny percentage of that tritium can be allowed to escape. Realize also that what has to be constrained is the cumulative leakage from all fusion power plants, not just a single one. The world would need on the order of 10,000 1GW power plants, to displace fossil fuels.
- simiones 5y ago
  Here is an article about the probably unavoidable downsides of fusion [0], from the Bulletin of Atomic Scientists. To summarize their points: - neutron bombardment, the unavoidable consequence of the only realistic fusion reaction, deuterium + tritium, turns every known material brittle and radioactive; and can be trivially used for uranium enrichment - tritium is almost impossible to contain, and requires fission plants to create - the fusion plants could theoretically create a surplus, but they would have to recapture essentially 100% of a very hard to capture gas (today they normally leak about 10% of the injected tritium) - fusion plants need to be massive to be even close to break-even - the energy drain of the facility consumes a sizeable portion of the generated energy; most of this power drain is still required while the facility is not operating the reactor, increasing projected costs of running a facility - ICF tech is very close to fusion weapon tech, so there is a massive risk for proliferation from there as well; MCF tech is not, thankfully [0] https://thebulletin.org/2017/04/fusion-reactors-not-what-theyre-cracked-up-to-be/ https://thebulletin.org/2017/04/fusion-reactors-not-what-the...