3 ms·

Although I happen to disagree, that is a different, more reasonable critique than your original comment.

pfdietz 8y ago
My original comment is entirely reasonable. Fusion faces generic obstacles that are likely unsolvable, and certainly unsolvable in the near term. Consider the problem of tritium breeding. Fusion reactors have to make their own tritium -- T from fission reactors would be far too expensive -- but tritium breeding blankets cannot be tested without a working fusion reactor. And the requirements for the breeding blankets are seriously constraining. They must achieve some neutron multiplication (from (n,2n) reactions of energetic fusion neutrons) to offset losses, not lose too many neutrons to penetrates/parasitic capture, and be able to give up the produced tritium fairly quickly. And there are very strict limits on how much tritium can be lost to permeation into reactor materials (or lost from the reactor exhaust, since ~99% of the injected tritium doesn't fuse). No one knows how to do this, or if it can be done economically, or even if it can be done at all. And I repeat: testing it is going to require a working fusion reactor! This circular dependency in development will be painful to run around, and won't happen quickly. Then there's the matter of materials that can withstand prolonged exposure to fusion neutrons. These materials have not been developed, in part because this again requires something close to a working fusion reactor (at least, something with Q at least ~0.1), and much painful rebuilding as that reactor, built with less adequate materials, destroys itself over time. This material development will not be fast, so we will not have working, practicl fusion reactors soon. Even when all that is solved, the fundamental problem of fusion -- low volumetric power density -- remains. That alone seems insurmountable. When you see a fusion reactor design, ask what this power density is. For ITER, it's 0.05 MW/m^3. For ARC or Lockheed's design, it's around 0.5 MW/m^3. These are very bad compared to fission reactors. A final point: the complexity of the irradiated part of a fusion reactor (the part that's too radioactive for hands-on maintenance) will be MUCH higher than that of the similar part of a fission reactor. Reliability is going to be an enormous problem. Making fusion reactors reliable enough for them to be practically interesting, even if all other problems were solved, will take much experience and time.

My original comment is entirely reasonable. Fusion faces generic obstacles that are likely unsolvable, and certainly unsolvable in the near term. Consider the problem of tritium breeding. Fusion reactors have to make their own tritium -- T from fission reactors would be far too expensive -- but tritium breeding blankets cannot be tested without a working fusion reactor. And the requirements for the breeding blankets are seriously constraining. They must achieve some neutron multiplication (from (n,2n) reactions of energetic fusion neutrons) to offset losses, not lose too many neutrons to penetrates/parasitic capture, and be able to give up the produced tritium fairly quickly. And there are very strict limits on how much tritium can be lost to permeation into reactor materials (or lost from the reactor exhaust, since ~99% of the injected tritium doesn't fuse). No one knows how to do this, or if it can be done economically, or even if it can be done at all. And I repeat: testing it is going to require a working fusion reactor! This circular dependency in development will be painful to run around, and won't happen quickly. Then there's the matter of materials that can withstand prolonged exposure to fusion neutrons. These materials have not been developed, in part because this again requires something close to a working fusion reactor (at least, something with Q at least ~0.1), and much painful rebuilding as that reactor, built with less adequate materials, destroys itself over time. This material development will not be fast, so we will not have working, practicl fusion reactors soon. Even when all that is solved, the fundamental problem of fusion -- low volumetric power density -- remains. That alone seems insurmountable. When you see a fusion reactor design, ask what this power density is. For ITER, it's 0.05 MW/m^3. For ARC or Lockheed's design, it's around 0.5 MW/m^3. These are very bad compared to fission reactors. A final point: the complexity of the irradiated part of a fusion reactor (the part that's too radioactive for hands-on maintenance) will be MUCH higher than that of the similar part of a fission reactor. Reliability is going to be an enormous problem. Making fusion reactors reliable enough for them to be practically interesting, even if all other problems were solved, will take much experience and time.