3 ms·

You could absorb neutron flux in a thousand tons of molten 6Li/7Li, and then try to separate grams of tritium from that thousand tons of molten, extremely infla

by ncmncm 4y ago

You could absorb neutron flux in a thousand tons of molten 6Li/7Li, and then try to separate grams of tritium from that thousand tons of molten, extremely inflammable metal, on a continuous basis. Good luck with that. Renewables do not incur such operating expenses. Put renewables and a fusion reactor on the grid, and fusion would never, ever win a bid. "Fusion .. will generate ... less ... than fission"? The whole damn ten-thousand-ton reactor becomes radioactive waste in short order. Fortunately, none will be built, so it is only a theoretical problem.

ASalazarMX 4y ago
I dug more about this because it's an interesting topic. Basically we don't know how long a fusion generator will last before radioactivity forces its decommissioning because none exist yet, but fortunately it's short-life isotopes, about 50-120 years before they become safe. Orders of magnitude better than fission waste. I don't know about fusion vs renewables, because renewables are better, but they need a lot of non-renewables to be manufactured. If a fusion reactor lasted 100 years, it would be a much better option than renewables.
- ncmncm 4y ago
  The reactor would need to be replaced frequently as the chamber walls and molten-lithium pipes weaken from neutron bombardment, every two years, maybe as long as five. By the time they could build one, there might be some way to make it last ten? It needs even bigger containment than a regular nuke because reactive molten lithium bursts into flame on contact with air. The oxidized product is an exploding cloud of vaporized radioactive drain cleaner. A thousand tons of lithium make a lot of drain cleaner. And again, by that time all our power will be cheaply supplied by renewables with no disastrous failure modes, and maintained just by taking bits down and putting up new bits, in shirtsleeves when weather favors it.
- marcosdumay 4y ago
  > Basically we don't know how long a fusion generator will last before radioactivity forces its decommissioning That was one of the main questions the ITER was supposed to answer, wasn't it? AFAIK earlier designs were quite short-lived, but some small changes fixed it. We will know how long they last as soon as we don't have an ITER anymore.
  - ncmncm 4y ago
    It is scheduled to be done in 2035, which means in practice for projects like this 2040 or 2045, and at $25B meaning $30B+. It will surely be abandoned before then, and we still won't know. What were these small changes that fixed it? Did they have anything to do with the cost going up by 3x? Anyway, inducing radioactivity is not what would use it up. That just makes it super-expensive to do repairs. It would get used up through neutron bombardment weakening the crystal structure of the metals it is constructed of.
  - pfdietz 4y ago
    No, it wasn't. ITER uses materials (like CuCrZr) that are unsuitable for a production reactor, ITER will not perform tritium breeding (except in a few experimental modules that won't be sufficient to prove that technology), and ITER will only operate at full power on DT for at most a few weeks total over its lifespan.
marcosdumay 4y ago
Hydrogen does not seem to dissolve very well in molten lithium. It is quite easy to separate on the lab, but yeah, the practical consequences of convection currents and putting it through a heat exchanger may change this. Anyway, I imagine the people studying fusion have an answer to this, since it only requires normal centuries-old chemical knowledge. I also do not believe fusion will ever be a competitive power source for stationary applications on Earth.
- ncmncm 4y ago
  Despite their fabled immiscibility, a few grams of oil would dissolve very thoroughly in a thousand tons of superheated water. The numbers matter. The people studying fusion are carefully restricting their attention to the immediate problems of getting it to work at all. E.g., ITER will have no lithium blanket. They figure on building another whole reactor to be completed (initially guessed) 15 years later at (initially guessed) 4x cost to begin tackling practical difficulties of extracting useful thermal energy.

by ncmncm 4y ago

ASalazarMX 4y ago

I dug more about this because it's an interesting topic. Basically we don't know how long a fusion generator will last before radioactivity forces its decommissioning because none exist yet, but fortunately it's short-life isotopes, about 50-120 years before they become safe. Orders of magnitude better than fission waste. I don't know about fusion vs renewables, because renewables are better, but they need a lot of non-renewables to be manufactured. If a fusion reactor lasted 100 years, it would be a much better option than renewables.

ncmncm 4y ago
The reactor would need to be replaced frequently as the chamber walls and molten-lithium pipes weaken from neutron bombardment, every two years, maybe as long as five. By the time they could build one, there might be some way to make it last ten? It needs even bigger containment than a regular nuke because reactive molten lithium bursts into flame on contact with air. The oxidized product is an exploding cloud of vaporized radioactive drain cleaner. A thousand tons of lithium make a lot of drain cleaner. And again, by that time all our power will be cheaply supplied by renewables with no disastrous failure modes, and maintained just by taking bits down and putting up new bits, in shirtsleeves when weather favors it.
marcosdumay 4y ago
> Basically we don't know how long a fusion generator will last before radioactivity forces its decommissioning That was one of the main questions the ITER was supposed to answer, wasn't it? AFAIK earlier designs were quite short-lived, but some small changes fixed it. We will know how long they last as soon as we don't have an ITER anymore.
- ncmncm 4y ago
  It is scheduled to be done in 2035, which means in practice for projects like this 2040 or 2045, and at $25B meaning $30B+. It will surely be abandoned before then, and we still won't know. What were these small changes that fixed it? Did they have anything to do with the cost going up by 3x? Anyway, inducing radioactivity is not what would use it up. That just makes it super-expensive to do repairs. It would get used up through neutron bombardment weakening the crystal structure of the metals it is constructed of.
- pfdietz 4y ago
  No, it wasn't. ITER uses materials (like CuCrZr) that are unsuitable for a production reactor, ITER will not perform tritium breeding (except in a few experimental modules that won't be sufficient to prove that technology), and ITER will only operate at full power on DT for at most a few weeks total over its lifespan.

marcosdumay 4y ago

Hydrogen does not seem to dissolve very well in molten lithium. It is quite easy to separate on the lab, but yeah, the practical consequences of convection currents and putting it through a heat exchanger may change this. Anyway, I imagine the people studying fusion have an answer to this, since it only requires normal centuries-old chemical knowledge. I also do not believe fusion will ever be a competitive power source for stationary applications on Earth.

ncmncm 4y ago
Despite their fabled immiscibility, a few grams of oil would dissolve very thoroughly in a thousand tons of superheated water. The numbers matter. The people studying fusion are carefully restricting their attention to the immediate problems of getting it to work at all. E.g., ITER will have no lithium blanket. They figure on building another whole reactor to be completed (initially guessed) 15 years later at (initially guessed) 4x cost to begin tackling practical difficulties of extracting useful thermal energy.