Edge Annealing

[TheRadicalModerate]

Could somebody explain edge annealing to me? I can't seem to make sense of this and it seems to be crucial to overcoming the various Rider-like thermalization arguments.

First, when we talk about edge annealing, are we talking about the electrons, the ions, or both?

Next, I assume that when we have various misbehaving particles flying out of the well, they are misbehaving in one of two ways:

1) They've developed some longitudinal velocity component. That's bad, because they won't return to the center of the well. I'm also assuming that it's rare, because to acquire such a vector you have to experience scattering somewhere other than the center of the well. How does annealing act to zero-out a longitudinal velocity?

2) They've acquired some radial thermalization, i.e., they don't come out of the well at the same speed they went in, due to some scattering event. This case in turn breaks down into two subcases:

2a) They come out with higher velocity (more energy) than they went in. Presumably, this is bad because, if they have a high enough energy, they'll fly out of the machine and their energy will be lost. How does annealing reduce their energy closer to the modal (i.e. designed) energy of the polywell?

2b) They come out with lower velocity than they went in. In this case, I'm not even understanding how edge annealing can act on them. Seems like lower-energy ions will ultimately poison the well by reducing the negative potential difference between the edge and the center, eventually blowing out the well entirely. Is there any mechanism that can boost these ions back up to the proper energy?

What I'm looking for here is mostly mechanisms. And, just to be particularly anal, here's a little summary tree of Things That Can Go Wrong With the Energy of Particles In a Polywell:

1) Longitudinal velocities
1.a) Ions with longitudinal velocities
1.b) Electrons with longitudinal velocities
2) Radial Velocities
2.a) Higher than mode
2.a.i) Ions with higher radial velocities
2.a.ii) Electrons with higher radial velocities
2.b) Lower than mode
2.b.i) Ions with lower radial velocities
2.b.ii) Electrons with lower radial velocities

If somebody can help explain either how these things get annealed, or at least why they're not a problem, I'd be grateful.

[jmc]

Discussed here by me among other things

viewtopic.php?t=289&postdays=0&postorder=asc&start=105

[MSimon]

TRM,

There are a lot of theories. Very little data.

I'm partial to beam bunching as the major (real?) factor in energy annealing.

However, I have nothing to go on at this time except some simulations and some hunches.

[Roger]

First, when we talk about edge annealing, are we talking about the electrons, the ions, or both?

Brem losses occur when you have hi energy Electrons and hi density electrons in the same place. I'm guessing electrons.

[93143]

The electrons are hot at the edge and cold in the core, which could result in an annealing effect, but the edge isn't where it would happen. I'm pretty sure edge annealing is just for ions.

jmc does a good job of explaining it in his link.

I posted on this topic a while back, describing what I understood Bussard to be saying about annealing in the context of gas kinetic theory:
viewtopic.php?t=258

Longitudinal (angular) scattering is one of the things that's supposed to be annealed out. There's no net angular velocity on the ion distribution, so maxwellianization in the edge should kill the detrimental effects of high-energy sideswipes. At least in this model.

I'm not sure you need precise ion focus for this to work. Someone once described the edge region as "paper-thin", which I'm pretty sure is unnecessary. It doesn't have to be perfect; all you have to do is keep the ions from being lost to upscatter until they can fuse. Either way, if POPS can be made to work, klystron beam bunching can't be a bad thing...

...not that I know anything about klystrons...

I think electron behaviour has been experimentally confirmed with WB-6. The claim was 1e5 transits before loss, with loss occurring prior to thermalization, and there were a lot of characterization tests before the fusion shots.

[drmike]

I think the electrons are going to thermalize. Langmuir showed in the 1920's that thermal electrons appear at rates 10^16 times faster than collision frequency would allow. It has been called Langmuir's Paradox ever since.

If the electrostatics can be maintained, the ions don't need to be thermal even if the electrons are. Near the MaGrid, they have close to zero velocity, near the core they have max velocity. Thermal electrons just maintain a uniform electrostatic potential. The system will work better if the two species are pretty much independent, but collisions will cause coupling. Getting a good handle on the collision frequency will really help with the modeling.

Thermal electrons don't hurt, lost electrons hurt. We don't really care about the electron energy distribution so long as we can hold them where we need them. The ions will then "do their thing", and we'll have function nuclear reactor.

The data will tell us for sure.

[MSimon]

I just looked up Langmuir's Paradox. Doesn't that argue in favor of the efficiency of beam bunching?

[drmike]

I don't know. I would say it makes assumptions about the electrons safe, so calculations on the ions can be more easily compared to measurements. If the ions can self organize into beams, and that stays stable, it would be great.

I'm assuming that POPS or something like it will be necessary to maintain that beam bunching over long times. But I really don't know.