[ncmncm]

People talking about fusion expect to "breed" tritium in their reactor. This takes the form of blasting GW of hot neutrons into a thousand (or ten-) tons of lithium hydroxide, and somehow extracting grams of tritium from it at parts-per-billion concentration.

There is no choice about that: it is the only way to get enough tritium to keep operating.

[pfdietz]

I don't know of any blanket design that uses lithium hydroxide.

[ncmncm]

Oops, lithium hydride. (Autocorrect strikes again!)

Lithium hydroxide is what you get, as thousands of tons of caustic vapor, when it catches fire.

What that does when you breathe it does not bear contemplation.

[pfdietz]

I don't know any that uses lithium hydride, either.

The usual designs are things like Li, PbLi, and lithium containing ceramics.

[ncmncm]

(As you pointed out before, elemental liquid Li or Pb would interfere with magnetic containment. LiH is an example of a diamagnetic Li-rich material resistant to radioactivation (other than the desired 3H). We need a great deal of Li in the neutron-absorbing blanket to breed tritium fuel.)

1000 tons of lithium deuteride (half 6Li, half 7Li, all 2H) would cost ~$2B for the deuterium, plus a smallish fraction of that for the 6Li-enriched lithium. Any deuterium that picks up a neutron would become tritium, adding to what is got by fooling with the lithium. Maybe you economize with half-H, half-2H, for only ~$1B.

You have many reasons not to let your LiH catch fire, beyond that it cost you $1-2B and would totally destroy your $50B reactor and be deucedly hard to put out. It burns in air to LiOH, Li3N and H2, and reacts with any water, CO2, or nitrogen you might have hoped would douse it. Li3N further reacts with the hydrogen making lithium amide LiNH2, thence various unpleasant peroxides.

Regular LiH is solid at a more-familiar operating temperature under 400C, and liquid at what might thought an extreme 700C. The deuterides would raise the melting point some. You really want something in there to scavenge any metallic lithium, if molten, because that corrodes steel and silica.

[ncmncm]

Lithium hydride is like a ceramic, i.e. a high-melting-point solid.

But whatever you use, you have to get extremely low concentration tritium out, somehow, to run the reactor on tomorrow.

[pfdietz]

The second sentence is a good reason NOT to use hydrogen in your breeding material, since if you do you have to separate the tritium from it, and do it very rapidly.

[ncmncm]

Not getting this. I understand you have to get it out fast because you need it for fuel tomorrow. Is it that you don't want your bred tritium floating in a sea of regular hydrogen, needing separation by physical rather than chemical means?

That seems like the least of our problems.