I mentioned a topic at HEAS last year I'd like to discuss here, because I think it may be an important idea. And it's one of the few I can lay claim to. All my other fusion work is actually just plain old well worked over tech that most ignore -- stuff that was worked out for electrons that seems to be getting missed by the fusion crowd. After all, charged particles are charged particles, within some scaling factors, and look at the things we've done with electrons that no one has moved over to the domain of ions
Beam on target fusion devices currently kick Farnsworth types around the block in both efficiency and output, but they are what they are, and that's it.
So they receive less development work because of the widely held belief that you can't win that game. In other words, if enough people believe something
impossible -- it's self fulfilling because no one will even try.
One of the several issues of them is that you are basically shooting at golf balls 1/3 mile apart, from the air, with a shotgun.
Obviously most of what you shoot doesn't hit one doing it that way.
But --- how about if each golf ball effectively had a large funnel in front of it, so no matter where you hit the funnel, you hit the ball? While that isn't practical at atomic scales, something almost as good might be possible. What if your target was made of some crystalline molecule such that any beam ion that misses a target atom is scattered toward the next one? You'd obviously lose some energy in such encounters -- but a complete miss is already the worst case of that, and the most likely event in existing designs as they stand now.
Now, designing such a compound is going to be a trick, to be sure. You'd want some fairly high Z atoms to do the scattering, but obviously you need some fusion fuel in the funnel pipes to hit.
The problem is, the chemistry of most of the light elements is pretty simple -- you don't get to multiple valences till you get to boron, which makes putting them into the right alignment in a crystal a little tougher. For example, hydrogen and lithium don't have any real stable compounds with two bonds to them, they tend to "hang off the side" of some higher valence element.
And I'd prefer to use either a H isotope, or Li for my targets -- easier pickings and lower input energy required that way.
And while I do some chemistry, I'm hardly an expert and know way-not enough about crystallography, I just know what I want, not how to get it.
So, any crystallographers out there -- help!
I got clued into and/or this idea from a couple of different sources. In one of the more recent I was looking into ion implantation and noticed a warning that unless you have the substrate crystal aligned correctly to the beam, you get these variations in implantation depths even when all else is perfect. Even pure silicon has an "easy way" compared to other orientations, and it becomes obvious why when you rotate a 3-D model of a silicon crystal and observe.
Of course, the cold fusion boys are all over the idea of "lattice assisted fusion" themselves but I don't necessarily consider that an endorsement as I don't see them producing the good results so far. We are a lot closer to "boiling that cup of tea" than any of them are, at least insofar as repeatable. witnessed results by respectable people go where the only reports of cold fusion demand such sensitive gear to measure that it's doubtful they are even out of their noise yet with output. We're more at the "doggone radiation hazard" output levels and trying to keep it small enough to work around and live. Still not fantastic, but as I just said -- scaling it down all the time to keep it do-able for us.
This idea is another way to accomplish what I first thought of in about 1968 -- where in that earlier thought I figured I'd make more of a benchrest rifle array than a shotgun, and in a regular array (crystal) of fuel, once you know where a few of them are -- you know where they all are, so just don't shoot between the bullseyes. This saves all the energy you'd otherwise waste on particles destined to miss the targets. However, the technical issues with doing that, though they pass theoretical muster, are very daunting when you try to scale things up. Even with a cold target, there is motion around the lattice centers. So if you could build a perfect "shadow mask" to only let particles into your lensing system that were going to be imaged straight onto the target nuclei, you could only scale so far due to that. I suspect if you instead tried to raster-scan the target and turn the beam off between targets, you can't do it fast enough to keep ahead of the resulting heat and sonic disruption in the target. And that whole thing only works if you do all your selecting and gating on slow ions you haven't invested much energy into yet. The fact that they are going slow means the dimensions of any "gating grid" arrangment gets to be so small that you can't make it. So, I thought, why not do the selection at the other end, where there are other advantages -- like the molecules that do the guiding might be moving some, but still have a built in funnel that still points to the target?
If you lose a few hundred eV of energy per scattering event, you can build in a little extra going in, compared to the increase in successful interaction rate, I'd have to believe.
Note, FWIW I do have several rifles that can do this level of accuracy, and there the job has more variables -- wind, inconsistent things about the ammo, barrel vibrations etc. This should actually be easier without those problems.