Hacker News Reader

Comment by sbierwagen | Hacker News Reader

  Safe: With no possibility of melt-down, or hazardous nuclear 
  waste, fusion dose not suffer the drawbacks that make 
  fission an unattractive alternative.

Eyeroll. D-He3 fusion produces less neutron flux than D-T, but it's hardly aneutronic-- otherwise they couldn't breed He3 from it. After a decade, the whole reactor will be nuclear waste.

I wonder what the service interval on an installed reactor will be? Or are they doing isotope separation on the produced helium on-site?

(Fun fact: you can get deuterium from water[1] but helium comes from natural gas fields![2] It's literally a fossil gas, which is why we're running out of it: http://en.wikipedia.org/wiki/Helium#Occurrence_and_productio... )

1: Commercially, heavy water is a byproduct of electrolysis plants: since deuterium doesn't electrolyze quite as easily as light water, it tends to accumulate in electrolysis stacks

2: The same rock formations that trap natural gas also trap helium.

View on HN

>D-He3 fusion produces less neutron flux than D-T, but it's hardly aneutronic-- otherwise they couldn't breed 3He from it.

You can't "breed" 3He; you have to breed tritium, which decays to He3 with a half-life of 12 years. D-3He reactors do produce some neutrons, roughly 1% of the flux of D-T, and almost entirely from D-D side-reactions.

Actually, a big problem with D-3He as of today is the lack of neutrons -- since you only get one neutron per 100 (e.g.) reactions, you can't possibly breed enough tritium to break-even. You have to come up with some other neutron source, or send rockets to Jupiter.

>After a decade, the whole reactor will be nuclear waste.

Fusion reactors are generally made from boron carbide -- a material unique for the fact that it does not become very radioactive when irradiated with neutrons. 14C which is produced can be extracted by oxidation and centrifuging, and in any case is not nearly as scary as, say, 90Sr. No other concerning radioactive isotopes exist with atomic masses between 10 and 21 (cf 22Na).

>commercial helium comes natural gas fields! It's literally a fossil gas, which is why we're running out of it:

Nope, that's 4He. There's basically no 3He in natural gas (because the helium there is radiogenic), so obtaining fuel that way wouldn't work even if we wanted to.

  You can't "breed" 3He; you have to breed tritium, which 
  decays to He3 with a half-life of 12 years.

Huh, I didn't know that. Helion directly claim they've "developed [a] complete self-supplied 3He fuel cycle", though. How did they manage that? Are they just massaging the truth, by saying that warehousing the tritium exhaust for a decade, then putting the 3He back in the reactor, still counts as a closed fuel cycle?

I imagine you could produce 3He by hitting deuterium with a proton beam, (or doing something exotic with 4He, or 6Li) at incredible cost per litre of produced 3He, but that's not really a "fuel" cycle, since the power reaction isn't involved in any way...

  Fusion reactors are generally made from boron carbide

??? Source? I've never seen a vacuum chamber cast from boron carbide, and you certainly can't make magnets out of it, which have to be centimeters away from the reaction, and therefore get irradiated. Are there any running fusion experiments that use internal boron carbide cladding to shield the vacuum chamber walls?

>Helion directly claim they've "developed [a] complete self-supplied 3He fuel cycle", though. How did they manage that?

I have no idea.

The D-3He reaction is: D + 3He >> p + 4He + 18 MeV. The D-D side-reaction is D + D >> n + 3He. Maybe they're recovering 3He from the side-reaction? But D-D is harder to ignite than D-3He.

This would hardly be aneutronic (you'd have one neutron per two nuclear collisions, instead of roughly per 100 with exogenous 3He), but it does provide a way of making 3He. I should stress that I am only guessing.

>Are there any running fusion experiments that use internal boron carbide cladding to shield the vacuum chamber walls?

Germany's new stellarator, Wendelstein 7-X: http://www.sciencedirect.com/science/article/pii/S0022311504...

[1] http://www.sciencedirect.com/science/article/pii/S0920379697...

Actually, ITER uses beryllium[1]. This has the same high cross-section and low propensity to create dangerous radionuclides as B4C, but is expensive and toxic. It is, however, much easier to make things out of, because it is just a metal. Older projects use tungsten, which does create dangerous radionuclides (181W and 185W). I was mostly aware of research re: B4C, but had been ignorant of beryllium.

So I should have said "modern fusion reactors generally use first-wall materials like boron carbide and beryllium, which do not become radioactive when irradiated". In practice, it's not worth holding up experiments on containment to make safe walls (ITER isn't built to last). In any case, the question of irradiated reactor walls is slowly becoming a solved problem.

http://en.wikipedia.org/wiki/Plasma-facing_material

http://en.wikipedia.org/wiki/International_Fusion_Materials_...

According to wikipedia D-D is the second easiest reaction behind D-T. Half the reactions produce 3He directly. The other half produce a proton plus tritium, which decays to 3He with a 12-year half-life.

You'll get some D-T from the tritium you produced, but the pulsed reaction probably helps a lot.

"Assuming complete removal of tritium and recycling of 3He, only 6% of the fusion energy is carried by neutrons."

http://en.wikipedia.org/wiki/Fusion_power#Deuterium.2C_deute...

Great reply. The wiki also has a long discussion about Helium 3. http://en.wikipedia.org/wiki/Helium-3 Fusion produces neutrons either in the first or secondary reactions, but there are ways to minimize the amount of them and their energy (and damage/radioactivity)to where you don't generate "nuclear waste". There is an interesting continuum of fusion reactions from pure D-D (which produces little energy, but lots of lower energy neutrons) to D-He3 (that produces some neutrons and lots of energy) to pure He3-He3 (that is called 'anuetronic').

D-D fusion makes Tritium (that decays into He3), Helium 3, or Helium 4 through the fusion process itself, with no breeding.

We believe that there is a correct ratio called Self-Supplied in which you have a small amount of 2.4 MeV neutrons, only deuterium as an input fuel, and the majority of the energy is from the Helium 3 fusion. The hard part is how to separate out the right isotope mixture from the exhaust between pulses.

He3 is never added? How long can a reactor run without supplying He3?

(This thread is having a hard time settling on a single isotope notation.)

The best way to make Helium 3 is with Deuterium fusion. Its a very interesting bootstrap question -- how do you build new reactors, if you have to have working reactors to generate fuel?

So really it's a mixed-mode D-D/3He-D reactor? How much more common are D-D reactions than 3He-D reactions while the reactor is running normally?

How much does the initial charge of 3He cost, given that the stuff costs $7,000 a gram?

> You have to come up with some other neutron source

Just throw chunks of Californium 252 at it ;-)

(That is one of the most expensive practical materials on earth, probably hundreds of dollars per microgram)

Actually the D + D reaction has two equally probable branches (and one very improbable one that makes an alpha particle and a gamma ray, which is mostly ignored), one makes Tritium and a Proton and the other makes D + He3. So the D + D reaction will breed He3 fuel, along with Tritium. The resulting proton from the Tritium branch can also combine with Deuterium and create Helium 3 as well in the right conditions. Under carefully control, the right conditions, and with some engineering work I would wager you could bias the reactions and side reactions towards creating more Helium 3.

Reply on news.ycombinator.com