Non-Tokamak Perfect Magnetic Bottle

Discuss how polywell fusion works; share theoretical questions and answers.

Moderators: tonybarry, MSimon

Post Reply
kamyers1
Posts: 1
Joined: Thu Oct 30, 2014 4:20 pm

Non-Tokamak Perfect Magnetic Bottle

Post by kamyers1 »

Let me preface this by saying that I am a non-nuclear engineer, not a physicist. It has also been over 30 years since I studied any of this in college. That being said...

I was reading about plasma confinement difficulties in the tokamak, and confinement time concerns with the polywell and similar designs that involve cusp or curvilinear plasma leakage paths. As a result, I came up with a relatively simple design for a theoretically "perfect" magnetic bottle. Unlike the polywell, it would have no cusp or curvilinear leakage paths. Unlike the tokamak, it would not accelerate the plasma around the circumference of a torroid, leading to confinement difficulties. Note that although this device is not a tokamak, it would have pseudo-radial symmetry vaguely similar to (but different from) a tokamak, rather than spherical symmetry like a polywell. I should also note that this design should not have the assymetric magnetic field issues on the inside and outside of the toroid that are encountered with a tokamak.

As I thought this idea through further, I quickly became concerned with ion thermalization. I could get the ions into the bottle with an appropriate ion gun, but without a mechanism to maintain their kinetic energy, they would rapidly thermalize. Obviously just confining the ions in the magnetic field doesn't do much good if you can't maintain the necessary energy levels for fusion to occur.

I have listened to and read some of Bussard's presentations and papers on the polywell, but I don't completely understand the necessary requirements for forming the potential well, and I have very little idea what Bremsstrahlung radiation is exactly, or how to control/avoid it. But I am thinking that something similar to the potential well described by Bussard might work with my magnetic bottle configuration, albeit adapted for radial rather than spherical symmetry.

I think this group has many folks who are a lot more well versed in most of this than I am. So my questions for you are basically as follows:
1. Would a non-tokamak, magnetic bottle configuration that is theoretically capable of "perfect" magnetic plasma confinement as described above really even be useful?
2. Does a lack of ion/electron leakage and recirculation paths necessarily doom it to failure?
3. In such a device, would it be adequate to hold the outer wall of the device at a high positive potential and supply an inner grid at a substantial negative potential in order to help form the "potential well", and would that keep the confined ions at an adequate kinetic energy level to produce fusion reactions?

Thanks in advance for your responses.

hanelyp
Posts: 2261
Joined: Fri Oct 26, 2007 8:50 pm

Re: Non-Tokamak Perfect Magnetic Bottle

Post by hanelyp »

A "perfect" magnetic bottle would be useful, though it would have to run in pulsed mode to extract fusion products. In the real world plasma in a magnetic bottle experiences cross field diffusion if it has enough density to be interesting as a fusion reactor, so no perfect bottles. But if you think you have something that is otherwise perfect confinement, describe it so we can see if it really works.
The daylight is uncomfortably bright for eyes so long in the dark.

prestonbarrows
Posts: 78
Joined: Sat Aug 03, 2013 4:41 pm

Re: Non-Tokamak Perfect Magnetic Bottle

Post by prestonbarrows »

It is very easy to make a magnetic bottle that 'perfectly' confines single particle orbits. The problem is in a real plasma you have much more than one particle and they are all interacting with each other.

The classic shortcoming of any bottle is the concept of a 'loss cone' and the fact that the particles undergo collisions with each other causing them to diffuse through phase space such that you are always loosing a significant number out of the loss cone.

You would have to describe your scheme in more detail if you want any replies which are not just speculation.

What you describe in #3 is essentially an Inertial Electrostatic Confinement (IEC) device. These are well studied and their classic shortcoming is losses due to collisions with the negative physical grid.

ohiovr
Posts: 431
Joined: Mon Mar 02, 2009 6:36 pm
Contact:

Re: Non-Tokamak Perfect Magnetic Bottle

Post by ohiovr »

prestonbarrows wrote: These are well studied and their classic shortcoming is losses due to collisions with the negative physical grid.
Actually if that one little problem could be solved, it would change the face of civilization.

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Re: Non-Tokamak Perfect Magnetic Bottle

Post by D Tibbets »

ohiovr wrote:
prestonbarrows wrote: These are well studied and their classic shortcoming is losses due to collisions with the negative physical grid.
Actually if that one little problem could be solved, it would change the face of civilization.
Indeed, that one little problem, along with 3-4 others.

Bremsstruhlung heat loss, cross field diffusion if any magnetic component, up scattering overcoming any reasonable electrostatic containment, MHD instabilities if any magnetic component, exacerbation of many instabilities with increasing useful densities, etc., etc., etc......

Then you confront the engineering challenges necessary for the physics to be applicable. On the surface, a torus or especially a Polywell or mirror machine may have simple physical elements. But making all of the physics play nicely with each other is challenging.

Perfect containment of charged particles in useful densities is impossible. The goal is to make it good enough, The Lawson criteria allows for a lot of lee way in how you mix the components to get the best compromise- which may be good enough!.

Dan Tibbets
To error is human... and I'm very human.

ohiovr
Posts: 431
Joined: Mon Mar 02, 2009 6:36 pm
Contact:

Re: Non-Tokamak Perfect Magnetic Bottle

Post by ohiovr »

D Tibbets wrote:
ohiovr wrote:
prestonbarrows wrote: These are well studied and their classic shortcoming is losses due to collisions with the negative physical grid.
Actually if that one little problem could be solved, it would change the face of civilization.
Indeed, that one little problem, along with 3-4 others.

Bremsstruhlung heat loss, cross field diffusion if any magnetic component, up scattering overcoming any reasonable electrostatic containment, MHD instabilities if any magnetic component, exacerbation of many instabilities with increasing useful densities, etc., etc., etc......

Then you confront the engineering challenges necessary for the physics to be applicable. On the surface, a torus or especially a Polywell or mirror machine may have simple physical elements. But making all of the physics play nicely with each other is challenging.

Perfect containment of charged particles in useful densities is impossible. The goal is to make it good enough, The Lawson criteria allows for a lot of lee way in how you mix the components to get the best compromise- which may be good enough!.

Dan Tibbets
Lets see. If I have a fusor, and magically take way the grid's opacity, while allowing its negatively charged goodness, there would still be problems? I think at that point we can all give up and enjoy fossil fuels while they still last. :D

hanelyp
Posts: 2261
Joined: Fri Oct 26, 2007 8:50 pm

Re: Non-Tokamak Perfect Magnetic Bottle

Post by hanelyp »

ohiovr wrote:Lets see. If I have a fusor, and magically take way the grid's opacity, while allowing its negatively charged goodness, there would still be problems? I think at that point we can all give up and enjoy fossil fuels while they still last. :D
You still have upscattering of some ions when they collide with each other, requiring another mechanism to even out ion energy to avoid higher energy ions boiling out.
The daylight is uncomfortably bright for eyes so long in the dark.

ohiovr
Posts: 431
Joined: Mon Mar 02, 2009 6:36 pm
Contact:

Re: Non-Tokamak Perfect Magnetic Bottle

Post by ohiovr »

[quote]You still have upscattering of some ions when they collide with each other, requiring another mechanism to even out ion energy to avoid higher energy ions boiling out.[quote]

Do fusor ions hit the walls of the container? If so, how often?

If fusor ions are leaving the core, just increase the anode voltage.

prestonbarrows
Posts: 78
Joined: Sat Aug 03, 2013 4:41 pm

Re: Non-Tokamak Perfect Magnetic Bottle

Post by prestonbarrows »

ohiovr wrote:If fusor ions are leaving the core, just increase the anode voltage.
Electrostatically confined particles will still upscatter to an energy higher than whatever voltage you apply to the grid. That is basically the definition of upscattering; some subset of the population will gain an energy greater than the acceleration voltage due to thermalizing collisions and escape the potential well.

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Re: Non-Tokamak Perfect Magnetic Bottle

Post by D Tibbets »

Look upMaxwell Boltzman thermalization. Graphs show the distribution of any thermlized plasma or gas given the average temperature. The average temperature would be the temperature/ KE introduced by the voltage of the grid. A substantial portion of the particles is above this energy, and thus will travel further away befor being reversed, and actually escaping the electrostatic containment entirely and flying to the walls with what ever KE they have remaining. Also, keep in mind that there will be many electrons present also, whether they come from the cathode, ionization of neutral gas, or whatever. There has to be very many electrons if there are very many ions- ie: ion densities useful for meaningful fusion. Physics precludes a strong concentration of any one electrical charge polarity without stupendous applied potential (voltage). You have to have to have confinement for both electrons and ions. some mix of magnetic and electrostatic, or inertial confinement needs to give an end result of confinement of both.

Thermalization is unavoidable given enough time, and this time may be very short. You can prevent a plasma from thermalizing through several means. If the ions fuse before the initial monoenergetic ion population can thermalize it will work. But given that fusion collisions are at best ~ 1/10th as frequent as coulomb thermalizing collisions for D-T at the fusion cross section peak, of closer to 1/1000th as frequent for other fuels, it is essentially impossible to prevent thermalization through this means. You can extract the up scattering (hotter) ions. This loses energy, but may have some limited benefits in other areas. You can reset the energy of the ions faster than they thermalize by some means. The Polywell may do this under certain conditions. It is called annealing. It is basically taking advantage of unavoidable thermalization of a subset of the ions. Because of the potential well, the ions on the edge, at the top of their potential well, are moving very slowly. They are cold. As such the coulomb cross section increases logarithmically, and thermalization can occur very quickly within a small area on the edge. The significance is that though thermalized, this edge ion temperature range is very small relative to the temperature of the accelerating voltage (at the bottom of the potential well). So long as the ions do not fully thermalize in one pass across the machine, they will be reset to ~ zero radial energy each time they reach the edge, at least this is the theory. Another approach may be to actively cool the up scattered ions in some region of the machine through some possible (?) means. Down scattered ions are not really a concern as they are effectively self restoring and not as painful for other reasons.

Dan Tibbets
To error is human... and I'm very human.

prestonbarrows
Posts: 78
Joined: Sat Aug 03, 2013 4:41 pm

Re: Non-Tokamak Perfect Magnetic Bottle

Post by prestonbarrows »

Yup, or just distribution functions in general. If you could measure the energy of each particle individually in an instant of time then make a histogram of the results, you will always end up with some Gaussian-like bellcurvey graph.

There will always be some particles with much higher-than-average energy that can escape the potential well. Even after those fast guys escape, new particles get bumped into the higher energy levels and the well constantly leaks at some rate. This is the same idea as a glass of water evaporating at room temperature even though it is well below the boiling point. Those few fast moving molecules in the distribution have enough energy to escape into the air.

Random Google image search below
Image

You won't get quite the same curve from a fusor, but the same idea holds. You always end up with some finite spread of energies. You can see that even for a temperature of '1eV', there is a significant fraction at 2-6eV.

Post Reply