How do tokamaks extract helium from the fusion plasma?

[chandrahas]

If tokamaks were to be run continuously, somehow, then fusion would produce a lot of helium. But since the charge to mass ratio of helium is exactly the same as deuterium, both of them behave almost exactly he same in electric or magnetic fields. And tritium on the other hand has a lower charge to mass ratio. So, if both deuterium and helium behave exactly the same, then how do tokamaks extract helium from the plasma for operation and make sure deuterium and tritium are both still in the tokamak?

Thanks

 

[mfb]

But since the charge to mass ratio of helium is exactly the same as deuterium, both of them behave almost exactly he same in electric or magnetic fields.

Only if they have the same speed. In a thermal plasma, helium nuclei are slower.

The solution is simple: Remove both, separate them at low temperatures, re-inject deuterium and tritium only.

 

[chandrahas]

But even at low energies, how would you separate helium from deuterium if they behave the same way? Does the method take advantage of stuff other than just electric and magnetic fields? How does it work? Your help is appreciated.

 

[Khashishi]

Currently, tokamaks operated in pulsed scenarios where the vessel is constantly pumped out, and fresh gas is pumped in during each shot. Future advanced tokamaks will try to operate in steady state. This means that there will be some fraction of the plasma ions will be helium ash, which dilutes the fuel. The helium is gradually removed as some plasma escapes the edge of the magnetic confinement. There are a few methods devised to accelerate the removal.

RF waves can induce transport of the helium ions. He1+ has a different resonant frequency than D+ and T+, so it can be driven out. But the difficulty is that most of the helium is in the He2+ state, which has the same resonant frequency as D+ and T+.

We can use magnetic coils to control something called edge localized modes which are instabilities in the plasma which pump out particles at the plasma edge. But to be honest, they pump out both helium and fuel ions.

There's something called I-mode in which energy is confined better than the particles. It's kind of weird to think about since particles carry the energy of the plasma, but there is some collective behavior which lower energy particles escape. The upside is that it can allow relatively fast removal of impurities and ash while maintaining a high temperature required for fusion. But some fuel will also leak out.

The helium and fuel that escapes the plasma is diverted toward a divertor, which can be used to separate helium and fuel to some degree. There are cyropumps in the divertor which can be optimized to pump helium. See https://arxiv.org/abs/1311.4689 for one concept.
This paper [http://eprints.lib.hokudai.ac.jp/dspace/bitstream/2115/42426/1/169_55-60.pdf] talks about using materials such as nickel in the divertor to trap helium ions selectively.

Unfortunately, existing methods are far from perfect, so there will be some steady state helium in the fusion reactor.

 

[mfb]

But even at low energies, how would you separate helium from deuterium if they behave the same way? Does the method take advantage of stuff other than just electric and magnetic fields? How does it work? Your help is appreciated.

If the better methods in the previous post don't work, just remove everything from the reactor chamber, then use distillation. Hydrogen has a much higher boiling point than helium.

 

[chandrahas]

Got that, but what exactly is the I mode?

 

[the_wolfman]

Unfortunately, existing methods are far from perfect, so there will be some steady state helium in the fusion reactor.

Impurity transport isn't my area of expertise so please correct me if I'm wrong. My understanding is we normally define a figure of merit based on ratio the helium particle confinement time to the global energy confinement time $\tau_{He}^*/ \tau_E$ and it's generally assumed that $\tau_{He}^*/ \tau_E \lesssim 10$ is required for ITER and a steady-state burning reactor. A number of experiments such JET, DIII-D, JT-60, TFTR, etc have demonstrated $\tau_{He}^*/ \tau_E < 10$ for a variety of operating regimes including standard H-mode discharges. It is my understanding is that there is reasonable confidence that we can obtain sufficient He ash removal in ITER and beyond.

However, He ash removal is a concern for certain operating regimes, such as elm free H-modes. But like you said, there are tricks, like applying RMP's, that we can use to try to decrease the He particle confinement time.

 

[the_wolfman]

Got that, but what exactly is the I mode?

The are a zoo of different confinement regimes in Tokamaks. The "standard" confinement regime is call the L-mode which is a low confinement regime. However, it was observed that with sufficient input power the tokamak plasma will spontaneously transition to a high confinement regime which we call the H-mode. The H-mode has both improved energy and particle confinement. Over the years we discovers a number of other confinement regimes. One such regime is the I-mode, which is an intermediate between the H-mode and the L-mode. The I-mode has improved energy confinement, but standard particle confinement.