Mass spectrometry, Tritium question measurments

[Doug Coulter]

I'm editing this, and will remove this line when this is complete. It's taking a couple of hours to chew up a ton of video and get it happy up on youtube so I can put it here.

A discussion about whether we made tritium on the fusor run data area had some suggestions for further tests, and I did most of them and am reporting them here. After all, this is the area for discussing measurements. I found out some things of interest that are puzzling me, so y'all chime in with observations once you see the vids and comments I'll make.

This first video is a series of clips connected that show some basic mass spectrometry, what I see when I do this, and what happens when some powerful quartz halogen lamps are turned on inside the tank to blast gas off the walls, as well as what happens when one turns of the spectrometer's ion source filament.
Here's the url in all its glory if you want to go straight there.

Sorry about the fuzzy focus. That camera is even worse than my old eyes are at close range. Perhaps I should try taping a reading glasses lens over it and see if it helps that as much as it does me. At any rate, the fuzzy blobs on the left axis are powers of 10 in arbitrary units (more or less related to sensed ion current). Unless it's just scaled, the lower left is e-15 and the fuzzy blob about 1/3 of the way up is e-14 -- somethings. Amps? Not quite, but close.

Here's the url for the next video, when I let in some D and see what I see. At this point no power has been applied to anything but the spectrometer itself in the system. Now, Chris, before you get too excited, also see the last one, which is what puzzles me..
I didn't get the best and final D purity in the video, but I did get it in a screenshot, which by golly, you can actually read the numbers in.

D purity at e-6 mbar or so

Note that a line a 6 appears when I let in the D (along with some air and water, yuck, but I get most of that flushed out). No power other than the ~1kv levels in the mass sweep, 100v or so ion source, has yet been applied. I've never seen this before and I've checked this gas for purity basically at least one time for every time I fill one of the little bottles I use. But there it is.

Heres the stuff after running a discharge for 10 minutes or so, roughly 5kv at roughly 10ma. The neutron tube did click a time or two, but not enough above what it does on cosmic rays to say this made any neutrons.


And a screen shot.

After the fake run

I'll probably have a couple more comments to add once I've watched these myself (I just finished taking them and haven't played them back yet!), and gotten the screenshots off the other computer up to here, but the main one that strikes me is that yes, while I saw suspicious lines here -- pretty obvious -- once I began the real pumpdown again, the 5,6 lines went away more or less instantly, and they didn't do that after the fusion run, they stayed in proportion all the way back to tank base pressure. So, keeping one's nose up and trying to good science here, seems to mean I have to dupe a fusion run and look at that one, which I wasn't very careful (no more than normal) to do last time, else this isn't settled, in my mind anyway.

Edit: Comparing these to the screenshots I took on the fusor run, I can't easily tell the difference in the ratios of the 5,6 lines and the 4 lines from one to the other. Next time I do a more accurate measurement. The software lets you put the cursor over something and get an accurate numerical output on screen, which I didn't do these times. Though if I can't see it, it's probably not there, considering noise and other variables. Noise isn't bad at these levels (at e-13 and up) but...if we really made a bunch of T, you should see it without that I think.
Of course, I could do a really long run... So the only difference I can notice, and doggone it despite all that work with video didn't really catch it, is that on this fake run, the 5,6 lines more or less blinked out of existence as soon as I pumped down, much more quickly than in the real fusor run. No journal paper yet! (not that I want one)

The other comment worth noting -- even in this very clean system, and at pressures much lower than a normal fusor run, the mean free path is short enough to foul up the operation of a quadrupole mass spectrometer over a 4" long path, and is causing beam deflection errors in the 1/2 inch range at similar energies to those proposed for Cris's idea. This doesn't bode well for things like focus over long paths, Coulomb forces completely notwithstanding -- that would be another error source. And note how much other "junk" there is even in a super-clean system! Well, for hobbyist it's clean, for pro, not so much. I'd have to get rid of all viton and get an ion/sublimation pump to go truly "pro grade" here. And probably take a few things out of the tank that are in there now, such as a couple of solder joints on my multi-wire feedthrough connectors inside the tank. Yes, we've seen the lines for tin when the thing is warm...not hot, mind you, just warm. Zinc you can see at room temperature. And Cadmium. Anyone building a system, watch the silver solder alloy if you use any, many have Cd in them...

For completeness, here's what it looks like right now (a couple hours of pumping I guess, no baking)

After pumping back down again to almost base pressure, this is 3e-8 mbar (indicated)

I caught this right at the end of a sweep when it was doing that screen color flip you see in the videos. It nicely shows pretty closely with lines are likely "real" -- coincidence, but after this week, I'll accept luck.

[chrismb]

Hi Doug. I'd not go so far as you and say I was 'right', because I was just highlighting a check to be done. Besides, this doesn't disprove your post-fusion mass-6 line was T, just that some of that line is likely to be a 3D+.

For me, the tell-tale is therefore the mass 5. If you have a clean system and eliminate as much H as poss, then fusion would give you DT+ and DDH+ ions. If you were H-clean before the straight ionisation run, then you'd not get 5.

I think the key to working out the percentiles of that mass 6 line, if that is something you really want to do, is to do a sweep of D:H, as I mentioned, and to then see if you can tease out of that map of numbers the way these things stick together in tri-atomic ions.

Bottom line for me is; a) you've just repeated Tomson's 1911 experiment and shown it holds true for deuterated molecular deuterium and, b) you don't get mass 5, so that is your tell-tale which showed up before due to fusion.

..Nice work... a little more data across a parameter sweep and maybe you've got journal-publishable material for a paper there!

[Doug Coulter]

Well, I'm still editing the original post at this point! I have some screen shots where you can actually read the numbers. Sadly, I have no source of pure H without spending real money. It's considered so hazmat that my welding gas guy can't even get it! Electrolyzing gives really crummy stuff, and though we have a purifier someplace, it's the size of a small car (well, a steamer trunk) and in storage, and needs a couple hundred PSI input to work. So I won't be doing that one soon. I'll get some screen shots up in that original post today or tommorow (they're on the wrong computer and I'm not sure what directory they went to, so I have to do some looking) and we can look at those. They are a lot more readable than these movies.

What was interesting to me, is yes, the 5 line, but also the fact that as I pumped on down, the 6 from this run just disappeared instantly and it didn't on the fusor run. So perhaps it's the mass spec itself making d3+, but only under some conditions! That is what I expect, but I can kind of show that by doing a similar movie after a real fusor run. Which I'll do at some point, real soon now. If for no other reason than to make sure I can trust my memory of what I saw last week! I try, but you know how it is.

I note that we should also notice that the O in the original gas largely went away, and without looking (because I'm on the wrong computer now) I'd bet the water lines did something -- the O probably got eaten by D to make some heavy water. That'd be another place to look for T, as well. There's always some water!

In order to keep the data up here, and most of the comments below, here's some pictures I took while doing this.

System overview

discharge

Sigh, close focus failure again. I was trying to get close enough to get fewer reflections off all the layers of glass, but failed at that too. I used my second, pear shaped grid for the discharge, out in the main part of the tank, so I could get a discharge at lower pressures, but it's too close for the camera even on macro mode. Going to have to try some things with low diopter optics, I can see that coming.

One last pic of a full mass scan, all the way to mass 200. Heavy stuff doesn't fly well, as you can see. Looks like 1/F noise to me and kind of defines the current noise floor. The gage was reading 3 e-8 mbar at the time. It'll probably be tomorrow morning to get to 1e-8.

Full scan a few hours later on, not baked yet

[Doug Coulter]

I do note something very unusual here. In the third video above, at about 0:51, right after I say I'm going to open the foreline valve, (which I did), there is a sudden large change in the amount of 5,6 vs 2,3,4 lines. I have never seen that before. I don't believe I've seen that after a fusor run, but now I know what to look for next time for sure! Seems the mass spec, in the 1e-5 range, itself makes a 5,6 line! Now that would have to be some agglomeration of various H isotopes, since I know I didn't make significant fusion; in fact, my pear shaped second grid doesn't make very much even at full power and voltage -- wrong shape all the way for conditions, and in normal use, it's my "loud" ion source and doesn't really contribute much of anything else.

Stranger and stranger -- nothing else went down quick when I started pumping....Hmmmm.

[Joe Jarski]

Nice series of videos! It's enlightening to see the mass spec in operation and how the make up of the gasses change under different conditions.

[Doug Coulter]

I should probably do more of that and make a nice tutorial, since most people won't manage to own one of these, and it is pretty eye opening about what is going on in a system.
This started to turn into that, which is why I put it in the forum I did -- we have to think of newbies who are trying to learn from all this, and it's good to make it easy for them to find things.
The thing IS fun the play around with, and does a lot more than I've yet shown. One nifty feature is you can "park" it at mass 4 and run around with He coming out of a tube and find all leaks in moments. You can also spread out the mass range around a center, and even see sub-integer atomic weights. That is, once I recalibrate the mass scale, it's off about .4 protons at the moment.

I'm not going to suggest that people try to get one (heck, this one was close to 20 grand -- there was a time I had "too much money", but I've since fixed that!). The reason is -- once you've used one for awhile, you learn enough to not need one anymore unless you go around and diagnose vacuum problems for a living or need to find isotopic ratios or whatnot. I'll say this -- in a lot of cases an optical spectrometer is almost as useful if you learn how to use that, at least anytime you've got a glow happening. Pretty small impurities change the optical output a lot. You get to the point where your eyeballs can do it moderately well, or a hand-held (decent) spectroscope is very useful. You at least need the "teacher" version, not the student one.

Back in the day, the old masters would almost always make a little discharge tube, and put it in the foreline (where the pressure is high enough for it to work) and observe that to know if something odd was going on. I've done it, and it's really useful. Price/performance ratio for an observant person is really good for that one. I finally got a middling good optical spectroscope, but darn, it's hard to use and doesn't have a marked scale, which would make it a lot more useful. One of those things sold for high school physics teachers. A good one of that ilk would probably be ideal -- I don't see needing a "real monochromator" for that kind of thing.

Just a little example of why a real mass spec is nice, but only needed once. I have a tiny DC gear-head motor in the tank right now, one I cleaned as well as possible of oil, to drive a conductance valve I made (no longer need it, and it's coming out of the tank soon, though). I initially had none of the problems the books tell you about -- it didn't pollute vacuum, it didn't take a week to pump down the little trapped gas pockets in it, or not so I could tell, and it's generally a nice-working thing. But, leave those kilowatt bake-out lamps on too long, and zowie, the mass spectra gets real complex with hydrocarbons and gawd knows what else, and that mess takes a really long time to clear out of there. Now that I know that, I'll either not put stuff like that in there anymore, or take even more care to keep it cool no matter what. Which isn't real completable with baking everything out, but...you don't always need that.

[Joe Jarski]

That's good to know about the spectroscope. I haven't played with one since high school, but I've been thinking about getting one - I just didn't know at what quality level they actually become useful.