Ah, you guys on the other side of the pond get the jump on me -- wonderful
Of course, the logical limit of "going dense" is solids and beyond, and in my view may have merit -- see my
smart target ideas for just one example. I think there are a lot more possibilities than that, this is just the one that got my attention -- and in fact, was at least partly inspired by something I think John does -- ion implanting, where it turns out they go deeper into some lattice orientations than others -- showing the effect, which in that case is maybe not wanted -- but confirms the possibility of ion channeling in solids. The question there is do you lose too much in scattering vs the increased probability of interaction. In other words, by the time you get a hit, has the incoming particle lost too much energy to sill do good? I've already thought about how to deal with lattice damage that all that implies.
Here is a cheezy wikipedia entry that mentions channeling, and I've been talking to chemists and crystallographers about how we might make such a lattice that has the fuel atoms in the right places to try this. What you want is a molecular framwork that makes the channels, then the fuel atoms in the middle of each one. It turns out to be a bit of a challenge to do with monovalent fuel atoms from a chemists point of view, but maybe not impossible. Here, you get "effective" density in lieu of actual per-cc density, but I don't see that this matters too much.
The laser boys are working that at some level, but I think they are missing something important, or I am (who knows?). A long time back there was an interesting article in SciAm about doing interesting things acoustically. The basic idea was like this -- suppose you had an explosion in the middle of a regular solid (a sphere is the simple case). This would produce a certain pattern of shock on the surface sometime later, with reflections back and forth and so on. Why not instead put a time-reversed shock wave into the surface and have it propagate and concentrate in the middle instead? I think this falls apart at some energy density where you can only make the shock "so loud and no louder" but the concentration idea seems a good one. Of course, "between us girls", we know the laser boys don't really have an interest in fusion power as such, they are part of the nuclear weapon stewardship program, and trying to find ways to do nuclear tests without violating some treaties.
Their green spin is mostly that -- spin. They sure do get some fun toys to work with out of that.
I think to explore these further we'll have to find or work out some scaling laws -- to better than first order. My own work has been more with improving the percent of accelerated particles that do interact per pass, worrying about any recirculation later. It would seem interaction probability would go up faster than linear with density, as each incoming particle has more other ones to maybe interact with.
But I'd bet there's another side of that story -- above some density, they start hitting each other on the way in, obliquely and so on, at low center-mass energies (and making for more thermalization, which issue I know Cris relates to). And if you are talking about ions I believe the space charge repulsion builds up quicker then linear (heck I know it does, I'm just too lazy to go find the math right now) so you wind up having to overcome that with more input energy. Or do something like Chris's cool recirculating beam-collider-re-gatherer. One must never forget that mean free path means something utterly different for particles with charge on them than it does for a neutral gas.
Chad is now working with what would be a pretty tiny fusor, and with a little more work there we will get a data point. He's just getting going on a good vacuum system, and will need a lot of grid work etc to really become a contributer there, but at least lives close enough that we can meet FF and his fusor is tiny after all -- so it goes in the back of a car. He was here last weekend, and we got a good increment of progress on his device, and got some old b10 tubes working for us both. He is trying to do this in a 2.75" CF cross...we ran Paschen's law, but due to not screening off the side arms for field control, got it wrong that time -- Chad, you listening? Fix that with some SS screen over those long paths! If that is done, he should be running at much higher pressure (about 1 mbar) than a normal fusor and that would be instructive. We may have to get him setup in a grid that is more in proportion to the conditions -- right now it's far bigger than the normal size ratio most fusors would do best at, but...it's not a normal fusor either, and some other scaling laws may affect what's best there. He is just going to have to try some things and find out. To make a grid the same ratio as I use in the big guy, it's going to be really tiny, have to be thin conductors, and need a lot of precision to make. My guess is that he's going to need a bit of help with that one -- and that with the tiny grid, getting heat out of it is going to be interesting -- not much surface area to radiate it from easily. But see the key word -- guess.
For sure, we are quite fortunate the sun is such a lousy fusion device. Else, boom, and we never existed.
I think at some point worrying about energy density goes to a scaling problem (actually, a bunch of them depending on which detail you're concentrated on at the moment). To simplify my learning process, what I've been concentrating on in Q, gain, whatever you want to call it -- interaction probability for a given input energy, feeling that once we've got that -- the rest is one of those nasty "exercises for the student". Or not, but if you don't have Q, why bother with having a lot of whatever? The old saw "I'm losing money on each one, but making it up on volume?" Of course, in that search for Q, one always has in the mind what it would take to get big with it -- any engineer will work like that to make the trade-offs possible to hold in the head while thinking about other details. So maybe you don't go down paths you can't see a way to scale, but you leave that for a later checklist and another design rev.
I have run the back of envelope calcs for a silly idea already (with the help of a guy who really has the math at some point in the past). Could one not envision building a device that fired single fuel atoms at one another with such precision as to get a 100% fusion rate? The answer is, yes, it's theoretically possible if the atoms are reasonably cold at the time, nothing in uncertainty theory, or just plain classical physics prevents this if the optics are large enough to smooth out individual atom vibrations in them for example, and in this case with only two particles and fine aiming means the absolute minimum of space charge issues -- they repel one another but only straight on. With that interaction rate -- you don't need as many. Now extend that to still a single file, but one following the others -- bunching on the smallest level of just one per "bunch". Now your scaling limit is how fast you can have bunches follow one another.
As soon as you try to have multiple per bunch, you run into a lot of other issues, whether they are soluble or not, I dunno -- could go either way. Really, a lot of the problem of electric fusion boils down to this model someplace at the bottom of things. Do you try for more per bunch and deal with in-bunch repulsion, maybe with pre-distorting optics? Do you just try the "Shiva" kind of thing? I've never liked that latter too much as it seems you get a lot of particles interacting with low CM energies, but maybe at some point it becomes synergistic and confining? Doesn't "feel right" to me, though.
My own take on density issues is this, at least in re ions. Since it appears you want it, but can't easily get it in a simple rig due to space charge and other issues, why not gather a bunch of ions into a more diffuse space at first, bunch them up in groups,
then and only then fire them all at some focus. By the time they get to the inner focus, their trajectories are set, the energy is already in them, and there's nothing else in there yet to repel them back out. This divides the space charge issues up into parts that can be managed separately. While diffuse, you can collect them around say, an outer grid, and bounce them back and forth between the wires with RF, kind of like an ion trap or mass spectrometer does (I'm in the process of learning that math). In my case, I'm visualizing cylinder grids (because that's what works best otherwise here) end on, and seeing orbits in figure 8 shapes between each rod pair -- all this is done at fairly low energies so far, and the cloud spends most of it's time in one of the big loops in each figure 8, passing through a sort of lousy focus as it goes through the wires. Then at some point, you fire a pulse into the center grid, and you've got all these pre-aligned ions already going the right way and from almost point sources to focus down into the middle. Just a thought, but it's the one I'm backing with money and time just now -- gearing up to try that one, because whether it works or not, I am sure to learn a heck of a lot, not just "head knowledge" but in the all important (to me) feel for how things work empirically.
I've found that once I get that, my own brain is about the world's best simulator of what would happen if I tried this or that -- but not before getting that feel. I've learned a lot over the past few years, but that's just taught me I need a little more yet. My hit rate on predicting what will happen if I try this or that is improving, but I don't feel like I'm "there" yet.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.