Tokamak - actual 'thermonuclear' reaction rate?

[chrismb]

This is ref the postings in the thread viewtopic.php?f=38&t=212&start=10

The plot I usually post up is figure 6 of a paper you can download here; http://www.iop.org/Jet/fulltext/JETP98036.pdf

This is one of the most powerful tokamak runs ever. Only JET and TFTR are equipped to perform DT operation. JET sits behind metres-thick concrete, and work inside the vessel can now only be done by robot.

The figure shows actual results versus simulation. It's a bit tricky to work out whether an experimental neutron came out of a thermal-thermal fusion reaction or a beam-background reaction, so I will take it as read that the simulation below describes it accurately, as it is accurate in the other parameters (power output, &c.).

So if you eyeball the thermonuclear fraction, you'll note that there is very little at the start of the pulse. The fusion power, as much as it is, is dominantly from the fast ions circulating in the toroidal current. This remains the case for almost the entire duration of the pulse!! Overall, it looks to me like most fusions for the whole pulse are beam-target.

Clearly, the tokamak guys have high hopes of pushing this into dominantly thermonuclear with a longer run and they may well achieve that, but for now the beam-target fusion that goes on at the start of the pulse is actually a necessary heat input for the plasma to get it to the thermal temperature required for thermonuclear reactions.

[Doug Coulter]

Looks pretty nasty for those guys, they seem to need that miracle. I recently saw a movie of a pulse, in visible light. All the light was from around the edges of the plasma, near or touching the walls. They said something like "there's no light in the middle because the energies there are all too high". Do you suppose one should interpret that as "no light" or just "no visible light"?
I do know that when they got rid of the tungsten diverters, and went to carbon, they claimed huge reduction in losses from Hi-Z stuff emitting extra photons. But you'd see some lines from carbon, no? And we of course see the sun...

My objection to thermal/neutral stuff has always been spreading the energy out to too many degrees of freedom (thermal) and losing too much to photons (electrons). But I admit surprise that the hottest part of the plasma (supposedly, but they do measure that) doesn't emit light at all, or enough to see. Those are a lot higher densities than solar corona, and we see that.
Could I have dropped a scale factor here?

Wonder if we get any credit if we solve their problems for them? Doubt it, but I also think the approach doesn't have much promise anyway. On the other hand, any chance to learn seems to pay off later when I run into some sort of related problem elsewhere.

[chrismb]

They said something like "there's no light in the middle because the energies there are all too high". Do you suppose one should interpret that as "no light" or just "no visible light"?

That is [generally] no visible light. There is a mass of X-rays pouring out of the thing and probably a dominance of 13.6eV photons, I'll wager.

The gas is so thermally hot that electrons won't simply dive back into an atom and settle there. However, there will be some visible stuff baised towards the smaller wavelength emissions lines because at any one moment you might find a hot nucleus and a hot electron both going in the same general direction with about the same speed. In such a case, a 'cool' recombination event may take place [in their inertial frame] and so there is a low level emissions profile that is used as a diagnostic. However the dominant visible emissions are from the edge of the plasma, and because you are looking through this edge to look into the plasma, such recombination light from within the really hot plasma is swamped by the recombination light from the cooler plasma edge.

The plasma edge is where the interaction between the physics outside of the plasma confinement region and the confined plasma meet. One can picture what free neutrals do as they fly towards the plasma volume, they reach the region of higher plasma density and ionise, recombine, re-ionise, &c., but in any case the influx of neutrals will keep the edge cooler and so you see that edge. As those ions then fall deeper into the plasma mass, the recombinations drop off due to the temp. IOns coming back out of that plasma meet the cooler plasma at the edge and therefore have a bigger chance of recombination.

You see this as a visible edge to the plasma. You know when you've got a real hot plasma because you can see an edge. If you only see a continuum of plasma with no 'structure' then it is a low temperature, and thus likely a low ionisation fraction, plasma. If the plasma is hot, you should see edges and surface structures in the plasma.

[Doug Coulter]

I am supposing that they are hoping for collision events with the non neutron reaction products for enough heating to get to "thermonuclear" mode? Obviously, the neutrons aren't going to contribute much heating. I am also guessing that (most of) the X rays come from near-misses of electrons and ions, braking radiation? From your picture, that looks like the spherical version that seems to do so much better than the ITER design that they are proposing using it for materials testing? I forgot the name of that project, drat. Link?

[chrismb]

I don't think they are explicitly looking for non-thermonuclear-fusion heating, per se. They'd view it merely as a side effect from driving a pinch current through the plasma.

I forgot the name of that project, drat. Link?

Well, you could point your mouse over the image and you'll see I gave it the title 'MAST', which is a bit of a clue.... But actually I think that is a red-herring because looking at the central column it looks like it is START.

START was the first [spherical] tight aspect ratio [tokamak] and gave results better than expected so they went ahead and built a dedicated experiment [MAST]. This wasn't a 'dedicated' kit in that START was built from equipment scavenge along the lines of many a fusor setup, being made up of what was lying around at Culham. One of the reasons this is such a 'good' image is that the chamber used is actually a large cylindrical process chamber that was lying gash and was 're-used'. Though they would've liked to have a dedicated toroidal chamber, the use of the cylinder meant that they got a superb angle shot of the whole plasma, which was in no small measure a direct factor in them securing funding for MAST. (Getting good imaging of your work is not a particularly useful scientific function, but I think it's one not to overlook to generate interest and support which is why I have tried hard to do that.)

The START experiment was around a £100,000 cost, and gave ~10^13 neutron output pulses, so it worked out cheaper $/neutron that any fusor project also.

The spherical tokamaks give a higher beta than the lower aspect ratio toroids, but they don't have the same confinement times [I think that is the downside - don't quote me on that] so are not yet appearing to be better net energy candidates than regular tokamaks though they can pump out the neutrons readily enough in short pulses. There are some opinions that spherical tokamaks will ultimately show tokamaks the way to go and there are a number of programmes trying to get these running as the high beta should mean they won't need to be very big by comparison.