Questions Once Again

[Art Carlson]

I was actually talking with my professor about this a while ago, he said electrons are kinda fast and small, and may be harder to confine than protons and whatnot.

Is that right? I thought since electrons are a lot lighter, it would actually be easier to confine them.

You will often hear in this community that electrons are easier to confine magnetically since their gyroradius is smaller (by a factor of 40 compared to protons, and more compared to other ions). But it's not that simple for a lot of reasons. In the collisional regime (admitedly not relevant for fusion), the diffusion constant for electrons is much higher than that for ions. Up to a few years ago it was not clear that transport in tokamaks was related to cross-field diffusion. The data also allowed the possibility of parallel loss along chaotic field fluctuations, which would be faster for electrons. Turbulent transport occurs through the EXB drift and is by its nature the same for electrons and ions. At the end of the day, ions and electrons will be lost at the same rate (assuming they are supplied at the same rate by ionization of neutrals in some form). The only question is how large the electric field is and how that effects the loss mechanism of each species.

[Tom Ligon]

Art,

I recall this was pointed out to Dr. Bussard around 1996, I think by a grad student he'd been corresponding with. He was downcast for a couple of hours, and vanished into his office to work on it.

His original analysis had suggested a single-collector system, while wasteful, would still probably be far better than any steam cycle system, but that it would probably be possible to add a "grid" to allow the lower-energy alphas to be picked off seperately. When he realized there would actually be an energy spread in these alphas, he looked at the problem again, and decided if it was possible to put in one grid, it would probably be possible to put in several, and decided the efficiency would only drop a few percent. I believe his demonstration of this to me was a simple sketch of a distribution curve with some steps cut in to represent the grid energies. He estimated the energy waste from the notched areas.

The larger problem is how to actually engineer the grid system, even for one grid. This was never in the scope of work (we were still looking for the first Polywell neutron), and he assumed power collection in a working p-B11 machine would be non-trivial. I believe he would have been happy to get the reaction going, then turn it over to the engineers and go for a nice cruise in his sailboat.

Wasting some of the KE of the alphas will almost certainly result in a lot of heat in any machine rated at hundreds of megawatts. I don't think we're talking about the thin stainless steel wire used for the grids of my fusor (which typically run red hot at fusion conditions). Any system of incremental grids to accomodate the spread of energies will require wasting some of that KE.

[Robthebob]

so are electrons easier or harder to confine? Generally.

[drmike]

Electrons are easier because they are lighter. But that means for the same energy, they go faster and farther than ions, and that helps drag ions away from confinement. It's complex because everything interacts globally, but you get shielding effects locally - a large current in one area can affect the whole plasma via a resulting magnetic field, but the sheath around the current can protect it from a nearby wall.

It's safe to say people have been beating their heads on this problem for 100 years and nobody has yet figured it all out. So nothing is really "easy".

[Art Carlson]

so are electrons easier or harder to confine? Generally.

I don't think the question makes sense. If you want to confine one species or the other, say in a Penning trap, you would make some different design choices, but I don't know how you want to measure "easier". Dollars per particles per second?
But why would you care about confining a single species, anyway? To get densities high enough to be interesting for fusion power, you need quasineutrality, that is, you have to confine an equal number of electrons and ions that are tightly coupled, and you need something close to thermal equilibrium, that is, the energy cost of losing particles is about the same whether they are electrons or ions.
The closest I can come to asking a sensible question is whether the electric field points inward or outward when you feed electrons and ions at the same rate. I think in most systems with closed field lines it points inward, but I'm not sure. If you want, you can translate that to mean that the electrons are easier to confine, but I wouldn't.
Why do you care? What is it you really want to know?

[rnebel]

Why would you want to feed in ions and electrons at the same rate?

[rnebel]

Art et.al.
We haven't been looking at direct conversion schemes. These were built by LLNL for continuum conversion in the 70s for mirrors. My recollection was that they got conversion efficiencies over 50%, but I might be wrong about that. I suspect that Ralph Moir knows the answer to your questions.
If Helium buildup is a problem, there are ways to selectively remove Helium from the plasma.

[Art Carlson]

If Helium buildup is a problem, there are ways to selectively remove Helium from the plasma.

Interesting. How?

[tombo]

This relates to the design>"alpha collector geometry" thread (8/15/08 )

I don't understand how a simple multiple gridded system would work.
What is to collect the alphas that are slowed to an almost stop at each grid?
If the grid is transparent, won't they just fall back down the potential well, taking back their energy and polluting the well?

I made a stab at a design that would push them sideways (electrostatic) to a radially oriented collector once they were slowed down.
I used the naive assumption that the energy spread for each alpha type was the energy of the well plasma.
Art Carlson's result here makes the spread even worse.

My conclusion there was that adding one grid to collect the low band alphas also collected 1/4 of the high band alphas leading to a heating/cooling (of the grid) problem.
To reduce that problem there may be a more complicated arrangement to catch the alphas on the second pass past the grids as they start to fall back down the well, but that would require a more sophisticated model of the fields and particle motions.

Is there a simple grid system to separate energies, that the old vacuum tube guys figured out eons ago?

[rnebel]

Art:

Take a look at the POPS papers, particularly the Phys Rev Letter. Deuterium was removed from the well by resonantly heating it. The scheme is mass and charge selective.

[Art Carlson]

Art:

Take a look at the POPS papers, particularly the Phys Rev Letter. Deuterium was removed from the well by resonantly heating it. The scheme is mass and charge selective.

I don't have a printout of the POPS paper handy. Is it available online? "Resonant heating" sounds like ion cyclotron resonance, but the magnetic field in a polywell varies so much (where there is one at all) that I can't make sense out of that idea. Since the POPS is pulsating, you could be talking about resonance with the pulse frequency, but that wouldn't apply to the polywell either.

[rnebel]

Art:

The idea is pretty simple. You oscillate the amplitude of the virtual cathode at the resonant transit frequency for the desired ion species you want to pump energy into. It selects species by q/m. For a harmonic oscillator potential, it's pretty trivial. It's described by Mathieu equations (driven harmonic oscillators). We demonstrated this experimentally (that's what's in the Phys Rev Lett.). This will be more involved for potentials other than harmonic oscillators, but it is probably doable.

[Art Carlson]

The idea is pretty simple. ...

Got it. The basic idea seems sound. It should work reasonably well for harmonic potentials because the frequency is independent of the amplitude. With a 3-D harmonic potential, there are additional complications (e.g. How do you pump particles on circular orbits?), but I could imagine that a useful effect would still remain. The potential of a polywell (so I understand) would be more nearly flat-bottomed. In that case, the oscillation frequency will be proportional to the square root of the energy. If I take a low-energy alpha and start pumping up its energy, it will move out of resonance before it gains enough energy to leave the well (in all likelihood). I suppose the next thing to try would be to chirp the frequency. I can't say off hand whether that would work or not. Anyway, it's an interesting idea, so we can put ash buildup on the long list of problems to worry about if decent confinement scaling is ever demonstrated.

By the way, what are you doing here, Rick? I thought you had a final report to write. Can you give us any hints as to the timetable to hear about what you have learned in the lab?

[MSimon]

Art,

I invited Rick and Tom to comment when I saw your post.

As to what Rick should be doing: only Rick can answer that one.

[Robthebob]

Oh my god, my thread is becoming a battlefield...

What I was asking before is base on our current technology, is electrons easier or harder to confine than ions, of course I understand with different particles, you design different devices, I mean with plasma, even my professor, I dont know if you guys know him, his name is Edward Thomas, this may be a bad thing to disclose his name online, please dont bother him if you know him, anyways, even with plasma, which there have been so many people that've worked on it, it still elude us, because as Dr. Thomas said, plasma is a "complex system" the behavior of each particles may be easy to predict, but as a whole, 10 to the too many powers particles, it's hard to work with them, we dont have computers that fast that can calculate that many particles that fast.

Now, I'm just saying if the particles were 10 to the 3 power lighter, perhaps moves faster, although from what I can gather from the website, the ions will be moving faster than the elections, they were charged, because plasma as a whole is neutral, I think even when the polywell is running on full power, the electrons to ion ratio inside the well is still a lot more than 1:1, so the system overall would be negatively charged, correct me if I'm wrong please, I can be easily wrong. Under the above setting, would a mechine used to confine such particles be easier or harder to build, that was really my question.

I have yet another question now tho, and this isnt really theory here, but I already have a thread going so... wasn't the projected cost of a polywell powerplant a lot lower than any of the traditional powerplants, may be it fossil fuels, wind, solar, nuclear fission, plasma fusion, etc. If so, by how much?