Ion Source Proton / Deutron fraction enhancement, with H2O / D2O Vapor.

For the design and construction details of ion guns, necessary for more advanced designs and lower vacuums.
Post Reply
Posts: 188
Joined: Tue Dec 08, 2009 1:59 am
Real name:

Ion Source Proton / Deutron fraction enhancement, with H2O / D2O Vapor.

Post by lutzhoffman » Thu May 06, 2010 4:58 am

I found this paper quite interesting, by simply adding a small amount (Less than 1%) of H2O vapor, to the H2 gas in this microwave ion source, the proton fraction was enhanced from 75% to over 95%. While this source operates at a higher pressure than I would like, the effect is still very interesting, and profound.

What remains to be seen is if this same effect applies to other ion sources, or even in a fusor? If the answer is yes, and it should be if the article is to be believed, then even a fusor neutron count should go up with the addition of a small amount of D2O vapor? Check out the PDF below for more details, it could make for some interesting experiments.
(177.12 KiB) Downloaded 190 times

User avatar
Doug Coulter
Posts: 1312
Joined: Sun May 27, 2007 7:18 pm
Real name: Doug Coulter
Location: Floyd, VA, USA

Re: Ion Source Proton / Deutron fraction enhancement, with H2O / D2O Vapor.

Post by Doug Coulter » Fri May 07, 2010 3:12 pm

Well, to add my two cents or so, we've been running tests with various contaminants to our 99.999% (claimed, I measure one less 9 on the mass spec) D.

Though we've not seen huge increases in neutron output so far, what we have seen is that with a little of something else in there things that were kind of flakey at the margins get a lot more stable.

With contamination in the .5-2% range, we get lower and easier starting voltage at a given pressure, less uncontrolled current rise above a threshold we've noticed in other runs, and generally more smooth operation. In one sense this increases the neutron count in a run as we don't spend time either in un-lit mode or in serious current limit at low voltages. On the other hand, the pulsing modes we've found that have the highest Q simply disappear at fairly low concentrations (<1% or so) of "contamination". This could explain why no one else has duped our huge increases of Q in that mode -- no one else ever sees it with CP grade D (97% pure)....not surprising in some ways and this has happened before in science. Remember at one point silicon was rejected as a semiconductor material because the (leakage) conductivity of all known samples was too high -- and that turned out to be due to ppm and ppb impurities that "just couldn't be the problem" in conventional wisdom until really pure samples were made and tested. Then and only then did the theory come along to match reality and explain how such small contaminations could have such a large effect -- and here we all are using it right now!

We tried Ar (pure) Ar+CO2+He (welding mix) , Ne (also very pure), and damp air (which was the most effect of them all). We haven't tried plain water vapor, but I recall in the early days when we were electrolyzing heavy water that things worked "better than they should" and were more stable and produced neutrons at lower volts than with the pure stuff, and by golly that stuff coming off the electrolyzer was anything but pure -- air and water vapor (both heavy and light) were dominant over the desired product. The D output of that was only around 30% of the total. I have also tried another diatomic gas, N2 at one point, but got some very bizzare results that looked like stored energy in the nitrogen cloud (perhaps some oddball polymer of N atoms) that resulted in very bright flashes at strange points in the tank and loud noises -- when my tank makes sounds like being hit with a hammer and I'm only putting in 3kv at 5 ma, I stop -- something dangerous might be happening, to my equipment at least. We've seen this about twice with D2, but not like with the nitrogen.

These tests were fairly hard to do, but we may do them again. The reason they're hard is that of course the fusor operates at *far* higher pressure than our mass spec is happy with, so we have to do the test on fusor, then shut it all off and pump back down and look at what's left as we enter (and quickly pass through) the pressure range the mass spec is happy with. Obviously fraught with possible errors, which is why I've not posted about it with "real numbers you can take to the bank" yet.
For the moment, I think the qualitative results should be enough. I may be off by a few percent of what I am reporting, but not a factor of 2 for sure.

I happen to think that in most fusors, especially after light-off, that there are usually far more plain old neutrals and charged blobs of atoms than there are D+ ions -- at some point I think even an empirical observation would show that, as if it were all charged ions, pure D+, there'd be enough couloumbs stored up in there and enough E field from that that we would surely not see what we do. We also see this "divergent" phenomenon when current gets above some threshold (in the 10ma range on my gear) where it seem that indeed we do start getting the bulk of the atoms in there ionized over a period of about half a second, and current then rises without any obvious limit (without producing any more neutrons though as at 50ma or so we hit our current limit). The only recovery from this is to switch off the supply for a second or so or turn the current limit down well below the about 10ma threshold for this effect to occur -- it has hysterisis. We see this in tanks with a lot of excess volume, and in another setup with very little excess volume, and when it happens any ungrounded thing in the tank picks up one heck of a charge -- ungrounded feedthroughs arc over outside the tank, things like that.

After all, if your ion source is pushing in a couple of ma at most, and the fusor is going at 10ma or more -- most of what's going on isn't just from the ion source itself in the pressures normally run here.
I strongly suspect that even if the ion source is making a nice pure set of monatomic ions that they don't last long at all -- first time they hit a neutral we get diatomic (and bigger according to my mass spec) ion clusters instead....which simply waste power.

To really see this, we are going to try and run down to the limits of our microwave/ecr ion source and keep the currents down to where it's just those ions involved at pressures where the mean free path make that a possibility. This will take us to about 1e-6 millibar, where mean free path is long compared to where most fusors run. We are planning another grid to use the two in conjunction as kind of a mass spec type ion trap to do that, and the work is underway now, I think we are on to something re possible Q when we don't waste so much energy bashing neutrals around and knocking electrons off the grid and tank walls. Not to mention, the monatomic ions do go faster in the same net field and get to a better part of the DD cross section than is the normal case in fusors.

This old Farnsworth dog still has some tricks, so I'm not giving up on it yet, even though a simple borehole tube blows it away in both neutrons and Q -- for the moment. We've seen burst mode outputs far above the borehole "refererence" levels in both already in the Farnsworth. And this seems to happen at lower pressures and with a higher fraction of monatomic ions, for what it's worth.

Now that Carl has a nice new RF ion source, I'm waiting with baited breath for any new results from him -- he may be the only other guy who can put all this together as we have here -- Carl, try with some really pure D too! And get that magnetron magnet off there and use something closer to the field that satisfies ECR! Which since you're running 200 mhz instead of 2.5ghz is a lot less than that magnet makes. I rough calculate you want closer to 78 gauss than 980 for that. In my ion source, getting this right made a *huge* difference in both how hard to light off it was (less RF needed) and how low a pressure it would go to -- huge in this case being 3+ orders magnitude less pressure for it to run. I don't really understand the E field geometry you get with wound antenna over the tube, so ECR may not be a possibility with that setup, but it's worth a shot I think. The magnetron source I'm using has a nice clean straight E field directly across the tubing axis that makes this easy -- I can't speak to your design on that one.
Why guess when you can know? Measure!

Posts: 157
Joined: Fri Jul 02, 2010 5:41 pm
Real name:

Re: Ion Source Proton / Deutron fraction enhancement, with H2O / D2O Vapor.

Post by dbrown » Tue Jul 06, 2010 1:44 am

Not too surprising since microwaves will couple very effectively with water vapor (far better than neutral hydrogen gas) allowing a lot more energy to be pumped into the water molecules, breaking them and raising these atoms to a high excited state and then, in turn, these new ions transfering their energy to the hydrogen molecules convert them into more ions that further ionize other molecules.

Post Reply