Page 2 of 4

Re: Carbon nanotube assisted fusion?

Posted: Sat Mar 31, 2012 9:30 am
by Chris Bradley
> I totally concur that all ideas should be pursued. . .especially the fun ones.

Ray, the issue here is not at all to do with the sentiment that all ideas should be tried out. (Not least of the issues being who tries them out, when, where and with whose money?)

The issue *here* is that this is a forum for amateur experimentalists.

I've been criticised too in the past for not appearing to do enough to support the ideas I have put forward, but I don't believe I've ever proposed anything without it being a bite-sized chunk that it could be accessed in an amateur experiment.

The problem with your proposals, and @ Don B too, are that they are not only speculative but inaccessible by the amateur. This site does amateur practice, whereas physicsforums does current, extant science only, and does not countenance speculations and unpublished materials. Therefore, both this and that forum are not the right places to discuss this. It is therefore unreasonable to criticise either forum for not wanting to discuss this. Better to find another forum. Maybe the one Eric Lerner has set up, http://focusfusion.org/index.php/forums/ , is the one for you.

What you *might* say here, however, is "I am going to try to buy some nanotubes from XYZ, make a UV laser, and then zap it", or whatever it is your plan may entail.

Incidentally, homebrew free-air 'nitrogen' UV lasers are unlikely to be anything new to the forum congregation here. I've not built one myself as I don't feel my lack of experience with UV safety aspects is sound, but I dare say a few here have.

*Re-read* [I trust you already have?] viewtopic.php?f=15&t=7284#p49156 (and also re-read the rules before providing your introduction!).

Re: Carbon nanotube assisted fusion?

Posted: Sat Mar 31, 2012 11:00 am
by ray@aarden.us
" People who read the forums regularly will find nothing new or surprising. "

Re: Carbon nanotube assisted fusion?

Posted: Sat Mar 31, 2012 1:23 pm
by Chris Bradley
Ray Joseph wrote:
> " People who read the forums regularly will find nothing new or surprising. "


You are at liberty to quote and misunderstand all that you read. But maybe not forever.

Re: Carbon nanotube assisted fusion?

Posted: Mon Apr 02, 2012 3:24 am
by Doug Browning
Well, nanotubes can be purchased off the shelf for reasonable prices now. And they have specific conductivity better than silver, with max current carrying density 1000X that of copper equivalent. But I don't think they are normally available in nearly big enough diameter for our purposes here. The liquid filled reactants idea might help significantly, but finding liquid deuterium and tritium fuel is probably a show stopper for amateurs.

What would really be needed for gaseous reactants, to get multiple strings of close packed nuclei after firing/pinching, would be more like micrometer diameter carbon tubes. Graphene apparently wants to curl up thermodynamically on nanometer scales when unsupported, unless multiple layers are provided like in graphite. There are CVD techniques to grow multiple layers using saturated gases, but it would need a template "seed" ring of the right size to start them.

With that in mind, I can propose an amateur DIY technique to grow the micrometer carbon tubes. Start with the a CRT shadow mask. These have 40 to 80 um diameter holes for high resolution computer monitors (the usual precautions apply here for safely imploding CRTs!!! tube neck cover rags, cardboard box, window screen shield over all, outside only, safety glasses, safety gloves, safety dust mask, ear plugs, warning - toxic leaded glass in the faceglass..., glass shrapnel and splinters resulting from neck fracture and possible phosphor dust. I have heard that the safest initial fracture technique is to break the vacuum seal-off beneath the socket pins plastic insulator with a diagonal cutter, so as to make the smallest air leak hole and cause the least structural damage to the glass neck) .

One wants to obtain the hexagonal circular hole pattern type shadow masks from monitors with triangular arranged electron guns (not the Sony, Nec inline gun ones, they have slotted masks; the Philips monitor I still use here has the circular hex pattern). One can also purchase new laser punched sheet metal, with down to 3 um holes, I have found. Some used CRT shadow masks may have coatings of phosphors or insulator ceramic on one side from the CRT manufacture. Nickel or copper base may be suitable metals for the hole punched mask and CVD carbon growth base. (You don't want to end up with Schottky diode interface junctions or insulating junctions, as often occurs when grown on Silicon wafers for example. A high eV contact potential barrier metal is required to avoid the problem. And adhesion of the tubes could be helpful here, whereas the usual substrates are chosen for easy removal of the carbon nanotubes.) CRT shadow masks are likely to be Invar metal on a tensile stress frame.

The CVD process is then started by pushing the high carbon saturation CVD gases through the metal mask holes with a slight pressure differential between the two vacuum sides, and generating the electric discharge plasma on the gas exit side of the mask. Once the base "seed" tube structures have formed around the mask hole exits, the CVD process can change over to conventional mode with CVD gas supplied to the plasma side until the tubes have extended to the requisite length (waiting too long to switch over the CVD mode may thin the tubes down to a single layer, by gas starvation of the outer layers that fall behind in length). ( a couple of millimeters final tube length is typical, but some heroic efforts have reached almost 2 cm length) An electric field normal to the mask also helps direct the direction of carbon tube growth along the field lines so as to produce a "forest" of vertical tubes.

With the carbon tubes conveniently aligned with the mask holes this way, it becomes relatively simple to fill the tubes with the fusion reactant gases in the subsequent loading step by pressure feeding fuel mixture from the back side of the mask.

By starting the CVD process on a metal shadow mask, one has already made a common electrical connection for one side of all the tubes. A heavy metal plate can be brought in contact with the back side of the holed metal mask to provide a heavy current path from the ignition capacitor, and also to seal the tubes from gas loss on that side. Micrometer tubes are going to need something on the order of nanosecond 10,000 amp peak capacitor discharge currents, per tube, to fire them (this is arrived at by scaling down the Sandia scheme by radius; and larger diameter tubes will require proportionately higher peak current with prop. longer pulse length).

So some pointed ignition probe in high pressure gas insulation might work to fire individual tubes. Successful micrometer tube firing would produce serious radiation hazards with this amount of fuel, so appropriate shielding will be required. I would suggest a tubular metal reaction chamber submersed in a swimming pool. Serious attention will have to be taken to avoid exposure to high voltage electric shock with this setup!!! I would suggest a small lifting rig/winch that can slowly drop the prepared test setup into the pool just prior to capacitor charging and firing. (Count-down!) Results can be judged by Cerenkov radiation light evolution in the water. An additional lead brick wall would of course be highly useful for stand behind observers. Or another inflatable water pool barrier to kneel behind.

The firing of one microtube will likely disrupt the integrity of surrounding ones, so it may be advisable to slice a "wafer" up into small sections or ribbon, or just start with small pieces. Possibly the thin mask material could just be bent around a cylindrical metal electrode/rod to separate the tubes enough.

Since Sandia National Labs has remarked that they will try a reduced prototype (macro size tube, 26 mega-ampere firing current) by year end, I am waiting to see if their software simulation derived 100 to 1000 X power gain factor actually materializes before proceeding with any real work on this scaled down microtube scheme. New, unforeseen, problems may occur, that were not modeled in the simulation. Historically, that has always been the case. Adjustments may be needed.

Sandia, by the way is using a constant axial magnetic field thru the fuel, with an added solid frozen fuel center core, in their design with a metal outer sleeve. The outer gas layer is pre-ionized by a low power UV laser so that the ions are frozen in place in the magnetic field during the main pinch of the sleeve. The sleeve confined axial DC field increases in strength during the pinch by compression of the conducting sleeve. This provides a level of thermal insulation for the compressed/heated ions to prevent heat loss to the metal sleeve. This DC magnetic field can easily be added for the carbon tube variant if needed (rare earth permanent magnets below the mask). The faster rise time of the microtube pinch might ionize the gas, but if not, then a laser or microtube resonant microwave field could be used. The proportionately faster (nanosecond versus 10 microsecond) pinch pulse for the carbon microtube case might reduce/eliminate the need for this magnetic insulation technique since heat transfer time will be reduced by 10,000 X.

It might be noted that 1/10,000 the diameter tube only produces 1/100,000,000 the power output of the Sandia experiment (well, technically for the same tube length and fuel density, a shorter tube requires proportionately less firing voltage, so power in/out scaling is still preserved for shorter tubes), but current input is reduced by 1/10,000 and pulse length is reduced by 1/10,000. Net result is that power out versus power input ratio is maintained. So if Sandia gets 1000X power gain, so do we.

We would be operating at (1/10,000) cubed the power level of the Sandia experiment when reduced tube diameter and length are both taken into account (even less with fuel density factored in too). But then we aren't trying to run a power plant. And the wafer of tubes (1000,000 tubes?) could be fired off in rapid fire mode, as fast as the discharge capacitor can be recharged and the point electrode scanned.
You get 1000X the power out, of whatever you can manage to put in. Maybe even a power plant using continuous microtube ribbon production.

Re: Carbon nanotube assisted fusion?

Posted: Mon Apr 02, 2012 12:42 pm
by ray@aarden.us
I really like the extrusion method. If there is interest in comparing experimental designs with varying diameters, there might be a possibility to electrolytically plate additional metal onto the shadow mask and thus fill in some of the diameter of the aperture. At the suggested size, there is probably little to no gain for the holes to be hexagonal. At 40 to 80 um, that is about 10,000 times the size of the minimum diameter tube; there are about 12 atoms around the circumference of the tube. If we use a nominal 1 Angstrom bond length, the circumference is 12 A. Since the circumference is 2*pi*r or pi*d, the diameter is about 4 A. The atoms in the sheet (graphene) that make up the tube are arranged in a hexagonal grid. At that, the tube is not hexagonal.
http://en.wikipedia.org/wiki/Carbon_nanotube

Yes, liquid hydrogen would be a task. The nice thing is that the liquid does not need to be pure hydrogen. We are only looking for protons. By some definitions, acids are proton donators. Water gives up protons readily. You can’t stop strong acids such sulfuric from donating their protons. There are many other acids, H2SO4 is fun because the anion (SO4) is so big, it will not fit very well inside the tube. In fact, once a proton is in the tube, the anion can help stabilize the energy by being just outside the tube.

Re: Carbon nanotube assisted fusion?

Posted: Mon Apr 02, 2012 4:17 pm
by Doug Browning
Interesting idea on the acidic sources for pre-ionized protons. This would seem to fit with the original tiny size nanotubes approach. With only a dozen atoms across though, these tiny tubes will only collapse down to a single file string of trapped protons after pinch, with few fusions occuring I fear (nuclei being around 1000X smaller than atomic dimensions, they might even just pop back out through the carbon lattice in the early compression phase, multiple layer nanotube could maybe prevent that). A curious possibility for nano sized tubes would be some possible electron tunneling effects (fast rise time induction voltages along the length) thru the confined nuclei, maybe lowering the electrostatic repulsion for some fusions. If that were the case, maybe just high pressure gas or high acidity liquid would push the reaction. Well, one can hope at least, not too likely for Kev/Mev barriers.

The nanotube size also could be useable with the longitudinal beam compression approach. A colliding beam induction accelerator essentially, within the nanotube. The earlier objection to the nanotubes being destroyed by the fusions could be overcome by making huge quantities en mass via CVD on sheet.

A difficulty with the colliding beams in nanotubes approach is the need for a center point of the tube being at a negative potential and the tube ends at a positive potential for the acceleration fields. This can be done easily for macroscopic sized parts, but nanotubes would require some innovation. I suggest a modification of the CVD processing of metal holed mask sheet.
The tubes could be grown in sequence on each side of the sheet from the laser punched hole edges. So co-aligned nanotubes formed on each side.

The sheet then forms the central negative terminal for acceleration. The tube ends (which would require controlled equal length growth or some trimming procedure) could be covered with metalization on each side of the sheet/tube assembly to provide the negative terminals. A high voltage pulse would then accelerate absorbed protons to the center points in the holed sheet from each tube end. This would require nanosized holes in the sheet though, so the latest UV laser or wafer lithography processing would be required to make the holed sheet. Also, the pulse width must be short enough to only excite the inductance of the tubes and not the massive conductivity currents. Perhaps some magnetic ferrite or nickel plating on the outside of the tube surfaces could be helpful.

The one issue I would worry about here is that the nano size tube diameter may not be sufficiently small to guarantee a good beam collision cross section. The conductivity of the graphene walls along with induced eddy currents from the charges passing down the tube will tend to center the protons. But normally, (for close proximity type walls) the walls of accelerator tubes have a specific finite conductance (matched to the beam impedance) to provide critical dampening of the sideways scatter of the beams. And that requires some beam channel length for adequate dampening. We would be limited to millimeter tube lengths (2 cm max presently, which would be slow to make), which is very short by conventional accelerator standards. But it may scale down here with a nano channel, which would be much more rapidly interacting at lateral dampening, due to the close charge to wall proximity.

There are readily available twisted lattice nanotubes, which might provide some resistance tuning advantage. Not sure if they are easily made en mass on a holed sheet though. The near perfect conductivity of straight graphene wall would likely cause the protons to just bounce back and forth off the walls. There are also the chemical surface modification techniques of graphene to graphane forms, but these generally seem too extreme, producing outright insulators. Maybe metal film coated insulating graphane tubes.

For the final collision point, at the center of the nanotube length, it would be advisable to return to the pure high conductivity (almost superconducting level) of pure graphene wall, since that would prevent coulomb scattering of the colliding protons. The superconducting wall induced eddy currents would freeze the protons to the central axis (assuming they are already centered there by previous eddy current dampening effects), effectively enlarging the cross section for a head on collision. It would be of some interest to know how long a (wall dampening) nanotube is required to assure near certain head on collision. After all, we only get one shot at it, it better be good.

Then there is an issue of defects in practical nanotube synthesis and how much effect that might have on our final effective cross section. Pictures of nanotube forests on wafers look like crumpled carpet fibers, we'll need a lot better synthesis quality than that! (maybe a magnetic axial field added to the E field during synthesis, or a spinning orthoplanar one? Ferroxplana microwave ferrite is made that way sometimes.) Also, how good will the alignment be between the nanotubes synthesized in separate procedures on opposite sides of the nano holed sheet. Background magnetic fields could affect beam accuracy too. If it were just one collider nanotube, a B field might be used for collision alignment, that could even be used for the individually fired case of a sheet full of tubes. But if the fusion reaction destroys the tubes every time, we don't get much opportunity to usefully align them.

And another odd question arises here, just what will the final fusion product look like if a string of protons accurately piles up in one ball at the center?
------------------------------------------------------------------------------------------------------

On the larger microtube version, the CRT shadow masks with hexagonal pattern have circular holes, they are just in a hex pattern layout on the sheet, per the phosphor dots on the screen. Purchased perf metal sheet could get that down to 3 micrometer holes.

(I don't see where that couldn't be reduced all the way down to the 20 nm realm currently used for IC lithography if it were really useful. A matter of cost no doubt. CVD tube deposition could be slowed by gas diffusion rates through such small holes and tubes.)

Micrometer diameter tubes will necessarily require multiple layers to thermodynamically stabilize the bending tendency of the graphene film, and to provide sufficient current carrying capacity for the pinch collapse.

Re: Carbon nanotube assisted fusion?

Posted: Mon Apr 02, 2012 6:14 pm
by Carl Willis
The crucial ingredient missing from this thread, if it is to be potentially useful to the community, are references. References substantiating the fusion-relevant properties of nanotubes alluded to, references about the growth techniques alluded to, etc., etc.

I presume none of this stuff follows from personal experience (certainly not anything that has been shared here before). Therefore, the ONLY tool a reader has to differentiate between informed commentary and bongwater-inspired commentary are references to the original sources of information.

-Carl

Re: Carbon nanotube assisted fusion?

Posted: Mon Apr 02, 2012 6:49 pm
by Doug Browning
Well, Wikipedia is a good start for carbon nanotubes info and CVD. A lot of other sources or articles are on line too. I just Googled away until I found the specific info I was looking for.

Clearly significant challenges exist in the quality of nanotube synthesis techniques at this time, particularly for an advanced particle beam in tube scheme which needs to be defect free. But it is early in the game. Trying to put integrated circuits on the first silicon boules would have been folly too. It is still useful to see what developments become possible with certain synthesis qualities improved. When a shortfall to a productive technology area is identified, efforts can be made to bridge the gap. Right now we have battery and capacitor interests funding a good slice of the research, and they like kinky tube arrays with lots of area. If one could show that X and Y properties would solve the energy crisis, I think we would see X,Y tubes available in short order.

The holed shadow mask like substrates is my idea, but there is information on line about various metal substrates and various CVD strategies like gas composition or electric fields for influencing nanotube growth properties. CVD generated nanotubes/graphene/graphane is an actively evolving area, what is conventional today, will likely be obsolete tomorrow. They were peeling the stuff off graphite pencil marks with adhesive tape just a few years ago. Some boat hulls (and electrolytic super capacitors, and batteries too) are even being made from CVD nano stuff now.

Accelerator technology is abundent online also, but a good reference book is "Theory and Design of Charged Particle Beams" by Martin Reiser. Or "Principles of Charged Particle Acceleration" by Stanley Humphries (available as a free download online, along with "Charged Particle Beams" by Humphries). The beam density regime here, for nanotube colliders, is at the extreme limit for the usual particle ensemble statistical techniques used for accelerators. We are dealing with almost single file proton streams, with limited charge defocusing.

Our case is maybe closer to what are sometimes called crystalline beams, but those imply extreme (transverse) cooling of the particles which we don't have, at least not initially. Being a linear (ie non re-cycling) topology with little charge defocusing, we are freed from many of the beam instability issues from focusing and guidance imperfections. My approach here is to just view the particles individually in the transverse momenta plane, and as repulsive spring coupled in the beam axis dimension. An correctly impedance matched beam wall should do well here at axial guiding of the individual single file particles, by means of EM induced eddy current imaging, once the resistive wall dampening factor removes the transverse momenta.

The real question for this nano-collider scheme is how long a nanotube does it take to enlarge the effective collision cross section to near certainty by dampening/centering. (and there is the practical question of inevitable defects in the nanotube synthesis which could spoil this.)

Bremsstrahlung radiation from the protons interacting with the periodic carbon lattice electron fields should be checked, but I think that is unlikely to be a serious issue in a one shot linear case, with the protons centered in the nano channel of several atomic diameters at only sub-Mev energy. For a cyclic collider, it could be an issue, but I suspect the hexagonal lattice symmetry and overlapped orbital bonds of graphene provide a near constant total interaction.

The Sandia news was mentioned in the news here recently:
http://www.sciencedaily.com/releases/20 ... 161505.htm

I haven't gotten a copy yet, but the original source was:
Phys Rev Lett 108.025003 Jan 13, 2012 by Steve Slutz and Roger Vesey

Re: Carbon nanotube assisted fusion?

Posted: Tue Apr 03, 2012 2:47 pm
by Edward Miller
You have posted a lot of complicated/complex, and hard to test ideas. I'm not trying to crush your dreams but you have to think them down into actual experiments. Once you start running into experimental problems I'd be happy to help. I've done a lot of experiments carbon encapsulated deuterium and I can tell you it's not as cheap/easy/simple as it sounds.

Re: Carbon nanotube assisted fusion?

Posted: Tue Apr 03, 2012 4:45 pm
by Doug Browning
I am under little illusion as to the difficulties to be overcome in nanotube/microtube synthesis yet. This nano area is at the development stage comparable to point contact transistors right now, and fusion applications will require integrated circuit like level technology. But at this stage, I am trying to identify the most likely productive approaches, requiring the least amount of, or very directed, development to get there. A broad sweep of all conceivable possibilities is needed first, then winnowing them down to the best shots.

The colliding beam in a nanotube scheme is obviously extremely demanding of defect free tubes and long tubes, and so any hope for that is well into the future. But it may have the advantage for advanced fuels eventually. The pinch microtube scheme might be do-able now (I haven't seen any info on micrometer diameter tube fabrication yet, this is a big if), and the Sandia results will give some guidance on magnetic thermal insulation by year end. It could be that nanotube CVD growth will compete with the desired microtube CVD path for fabrication around a mask hole, but there may be enhancement and suppression techniques, I have some in mind.

One other thing that would need to be looked at is whether a micrometer diameter pinch has problems with radiation loss due to high surface to volume ratio and whether the shorter pinch time to compensate (nsec) is feasible and adequate. Sandia's magnetic insulation scheme will not help for radiation loss here.