Tungsten Parts for Ion Source Construction: Cheap Source !!!

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

Tungsten Parts for Ion Source Construction: Cheap Source !!!

Post by lutzhoffman » Mon Mar 08, 2010 9:56 pm

Hello:

Most RF based ion sources need a charged extraction electrode to pull the ion beam from the source bottle. In most commercial designs they tend to use a cone shaped, or tapered electrode setup, in order to help shape the extracted beam. As far as the material goes you want something which is very resistent to sputtering, for a long source lifetime.

HVEC uses a Tantalum source canal, in part because Ta is more easy to machine than W, and it is slightly more corrosion resistant. Ta and W are essentially the same, as far as being sputter resistant goes. Both are way better than even Ti, by a factor of two. Most amature designs settle on copper, because where are you going to get the Tungsten or Tantalum? Then how the heck are you going to shape and work it?

I have found what I believe to be a great potential source of ready made Tungsten parts, many of which are close to being the correct shape for us. The source is of all places: The fishing supply world, due to Tungsten being so dense it has found an application here. We may find some very nice parts for our applications from their offerings. Take a look at this part for example:

http://www.basspro.com/webapp/wcs/store ... rchResults

-or-

www.tungstenbythepound.com

Just burn off any coating and bingo, pretty close to what you need for an extraction electrode, you can now Silver braze it to copper if you clean it up.

For a source canal liner there are some ready made 1/2 " long x 1.8mmID tungsten tubes on ebay, also for fishing, see item # 220568115989 from a UK seller.

I have only covered this from an ion source part perspective, I am pretty sure that many folks out there, could find many other uses for these parts, like for inside of their fusors to hold grids, terminate a feed though, etc. Just avoid any of the porus stuff which is designed to absorb "scent" for the fishies.

So the next time that you need a sputter resistant, or high temperature part, you may consider going fishing, you never know what you may catch : )

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

Re: Tungsten Parts for Ion Source Construction: Cheap Source !!!

Post by Doug Coulter » Mon Mar 08, 2010 10:17 pm

Believe it or not, nothing I know of actually sputters less than plain old aluminum. I am using that for the negative extraction electrode here, and at 10 ma or so -- it's run about 200 hours with zero visible sputtering.

Sputtering is weird -- it's not directly related to evaporation temperatures or anything like that -- plenty of high temp metals sputter easily, but Al doesn't.
And sputtering is not much trouble at about a kilovolt and in pretty pure hydrogen -- it's the heavier gases that make things sputter fast (mercury is a really heavy gas for example, and even that won't make aluminum sputter much....). After aluminum, only magnesium has a lower sputter rate.
(as far as metals go)

I guess you could grind that tungsten but that's about what you can do with that stuff. You can braze nickel on it fine. There are "silver solders" with nickel that will wet it (basically the same ones recommended to use on stainless steel).

Until I have actual trouble with my aluminum electrode, I'm fat, though. I'm using the end of a SS capillary tube for the other polarity (positive) and no troubles at that end either.

Here's some old data on what sputters easy and what does not. Tantalum looks good too, and you can get that from many old transmitting tubes...it was used as anodes for them.

I have noticed very little sputtering on the pure graphite I use for parts of some grids as well.
This would be the new stuff from McMaster that I get in rod form -- it's pure graphite, no binders, machines like a dream too -- comes off the lathe shiny. (but you are now black all over).
Attachments
Sput2.gif
Sput2.gif (29.16 KiB) Viewed 3589 times
Sput1.gif
Sput1.gif (32.31 KiB) Viewed 3589 times
Why guess when you can know? Measure!

lutzhoffman
Posts: 188
Joined: Tue Dec 08, 2009 1:59 am
Real name:

Re: Tungsten Parts for Ion Source Construction: Cheap Source !!!

Post by lutzhoffman » Wed Mar 10, 2010 9:02 am

Hello:

I am just wondering about the age of the sputtering data? The modern figures from 2 different companies which specialize in sputter coating and supplies list W and Ta as the ultimate best? There figures are based on inert gasses to be fair, so this could explain part of the difference. A chemical reaction could also influence the results I suppose.

I am sure there must be a reason that they use Ta, and SS, to line almost every ion source canal I could find a spec sheet on. HVEC uses Aluminum as the base material, but then they add a Ta liner for high atomic # species, and SS for H light elements, for "longer source life".

Since the W is already of the right shape, with the right size hole, plus it is under 10 bucks to boot. I will try it, especially since I do not have a lathe. Actually Ta is not that expensive either, if someone needed a small amount it can be salvaged from many power vacuum tubes. So do not throw away your burned out tube from that VT Tesla Coil.

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

Re: Tungsten Parts for Ion Source Construction: Cheap Source !!!

Post by Doug Coulter » Wed Mar 10, 2010 3:26 pm

That's old stuff, but physics of the elements doesn't change with time much.

I suspect a number of things here. I mean, the Al does work great....
Al is cheap -- no money in it, and of course nothing is perfect.
Al has a low melt point, and in ion sources like a duo-plasmatron, the electrodes get hot (very).
In mine, they don't.
Al can form film of oxide, which for low volt low current stuff, makes it a lousy electrode (for things like microscopes). For me -- it's not a problem, and for that matter, we're working in a very reducing atmosphere.

I mean, I have kind of proof it's fine for what I'm doing with it -- it's doing it for hundreds of hours at a whack no problems yet.

In the pic for the W weight, looked to me like the channel was far too long for it's diameter, and the only way to fix that would be to grind the stuff. Do-able, but a pita even in a machine shop compared to working with something easier. Too small a channel (too many calibers long) means ions will go through a focal point someplace inside, then expand again and smash the walls. Shape is really important here...

So I think other things come into play -- when you're making so much profit on something, why not use the most expensive parts if they help it sell? Even if you can have something 90% as good for 10% the price, what would their motive be to offer that -- it's in the noise to their profit margin on sales actually made -- that's all labor money. In this case, you might not even use something 120% better for half the price if the market has a perception that the precious metal is key. They need not be correct about that for it to matter to a company making this stuff.
Why guess when you can know? Measure!

John Futter
Posts: 1460
Joined: Thu Apr 22, 2004 2:29 am
Real name: John Futter
Contact:

Re: Tungsten Parts for Ion Source Construction: Cheap Source !!!

Post by John Futter » Wed Mar 10, 2010 6:25 pm

Doug is right
aluminium is one of the hardest materials to sputter
we used to use tantalum apertures on our sputter ion sources but the films had hi Ta content we now use Al.
Also the KN acc @ work which uses a CCP ion source the canal tube is Al and it lasts around a year before it has to be refurbished ie around 3000 hours of operation.

to get Al to sputter it has to be hit with ions with energy over 10keV I have a graph somewhere that shows the yield vs eV and you can also simulate in SCRIM

User avatar
Chris Bradley
Posts: 2931
Joined: Fri May 02, 2008 11:05 am
Real name:

Re: Tungsten Parts for Ion Source Construction: Cheap Source !!!

Post by Chris Bradley » Wed Mar 10, 2010 6:44 pm

Interesting info, Doug, well worth posting.

I have used copper for the electrodes in my exp on the basis of feeding them with RF. But given this info, and the apparent, and surprising ease (I'm only running 10's of uA...), with which copper atoms seem to be moving around my kit, I think I will be trying out Al electrodes in future - once I find any time to do any more (next Christmas hols again, I guess??).

In fact, maybe I'll try two Al electrodes and two Cu, and see whether one metal tends to coat the other!

John Futter
Posts: 1460
Joined: Thu Apr 22, 2004 2:29 am
Real name: John Futter
Contact:

Re: Tungsten Parts for Ion Source Construction: Cheap Source !!!

Post by John Futter » Wed Mar 10, 2010 10:23 pm

interesting reading for you all

link to where I found these
http://www.npl.co.uk/nanoscience/surfac ... eld-values

We use these tables @ work every day
Attachments
arsputterplots.pdf
(113.07 KiB) Downloaded 148 times
arsputtergroups.pdf
(445.66 KiB) Downloaded 147 times

User avatar
Chris Bradley
Posts: 2931
Joined: Fri May 02, 2008 11:05 am
Real name:

Re: Tungsten Parts for Ion Source Construction: Cheap Source !!!

Post by Chris Bradley » Wed Mar 10, 2010 11:28 pm

Indeed, very interesting...

It seems to fit a relationship of [ionisation energy,Z]. The ionisation energy relationship doesn't surprise me much (how far into the cumulative 'electron cloud' in the substance an ion can penetrate), but the lower Z having lower sputtering seems somewhat counter-intuitive as more energy would be transferred to atoms of similar masses, rather than lower masses.

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

Re: Tungsten Parts for Ion Source Construction: Cheap Source !!!

Post by Doug Coulter » Thu Mar 11, 2010 12:22 am

I've not seen theoretical backing for any of this, but that's why I said it's counter-intuitive, and practice backs it up pretty well. Cu sputters for me a lot more than Al does FWIW. Some of my earliest experience made me some nice Cu mirrors that way.

Of course, the silver lining on this cloud is that the difficulty sputtering Al onto telescope mirrors (and the troubles with Ag) are the reason the evaporation business got going, so we now have both as great (and easy to use) tools in the lab.

Nature even helped this along. Al dissolves tungsten. So, a smart guy figured you could just make the tungsten really thick, not put much Al on it, and it would live, and work, rather than the filament dissolving and breaking. This was after all attempts to evap Al from various material crucibles had failed -- Al above its melting point is kind of reactive...

This has side effects that make it work better, strangely. With the very thick tungsten, it no longer matters that the Al is really conductive electrically -- it's down in the noise compared to the fat W wires that are usually twisted so the Al can run between them by capillary action. Were you able to do this with thin tungsten, the moment the Al flowed over it, it would become a relative short circuit and be very difficult to control at that point as things changed quickly. But as things are -- no sweat at all, the small amount of Al (particularly hot) doesn't affect the total conductivity as much since there's a lot more grams of W in the circuit. So it worked for them right off, and with some refinements we now have the slick trick.

Nature wants to help, you just have to listen up when it speaks.
Why guess when you can know? Measure!

lutzhoffman
Posts: 188
Joined: Tue Dec 08, 2009 1:59 am
Real name:

Re: Tungsten Parts for Ion Source Construction: Cheap Source !!!

Post by lutzhoffman » Thu Mar 11, 2010 9:17 pm

Very Interesting!!!

Things are not always what they seem, even when it looks crystal clear on the surface. It is very interesting that Al does that well. Can this have something to do with the alumina surface coating which forms pretty fast, on even a fresh machined piece? I am pretty much converted on this issue now to Al, for low to medium temp use which an ion source is.

One benifit of Al also would be the very easy removal of any film which does occur, just a little HCL, and its gone! Just like that, a very easy way to clean the quartz.

Does anyone have any experience in using the low melting Mg based Alumaloy type of products to join Al to SS, or Cu. I have never tried it for disimilar metals, but it works great for Al to Al.

Thanks Folks: I leaned something : )

Post Reply