Evaporation

Discuss deposting things and synthesizing things in vacuum. Also ion implantation.

Evaporation

Postby Doug Coulter » Fri Aug 13, 2010 9:36 am

I'm starting this thread to prompt others to chime in -- I've done OK with evap, but I'm betting some others here know a heck of a lot more than I do, so this should get things going.

I have done evap in two different systems -- one fairly crummy vacuum (my mini bell jar with big rubber seals and just a two stage mech pump) and my really good big system. To say the least, with Al, the results were far better in the big tank in terms of the deposit quality. Al seems to act like a pretty good getter while evaporating, but this means that a lot of what hits the target is oxidized on the way there....and of course I've found that pure Al canes from Lesker work a lot better than say, Al alloy MIG wire as a material source.

So far I've been using the coil type heat source from Lesker for Al as it wets the tungsten and becomes a nice uniform source when you premelt it as I do. And a little goes a long way when making mirrors or beam splitters this way -- you get a lot of runs before having to add more Al to the source. The Lesker part number for this is EVF53025W, and is one of the lower power ones they sell, but still needs plenty to run. One thing to realize is that in their catalog, they are listing the max power you can use and not melt it, but for things like aluminum, you need no where near that much.

As I'm just staring out with this, I made a fairly kludgy power supply for it, which is nevertheless complete overkill for this job. What I did was to re-wind the secondary of a spot welder from Harbor freight with about 7 turns of # 6 wire, and have enough taps so that I can have from .5 turn to 7 turns in 1/2 turn increments. I am using about 3 turns for the Al job, which is a bit over 3 volts on the heater, and have just a very simple control - solid state relay and pushbutton (and a lot of MOV type kickback protection on that). In use, it takes a few seconds to melt the Al, and another couple to reach fast evaporation temperature. I do pre-melt the Al before doing the deposition run. letting it flow into the spaces between the 3 wires that make up the heater. This seems to give the best results, and since I make more than one run per loading, it's what all the subsequent runs are anyway, so this is a little more uniform way to go about things. I would bet that a smaller transformer (like a rewound MOT) and some better control than bang-bang on/off would do better for this and be smaller too.

One thing for a beginner to be warned about -- this isn't a particularly well controlled thing with the way I am doing it -- and of course (in hindsight) the Al comes flying off there in every direction -- about like hitting a can of spray paint with a rifle bullet in effect. This means my first couple runs coated my tank nicely, ruined a few insulators in there, and so on.
Not good.

My source is kind of floating up in the middle of the tank, on a couple of 1/4" Cu feedthrough rods I can bend around some to position it where I want. I took an old chemistry set ring stand and added a low shelf to it, ground off the edges of the circle so a 6" thing would fit through my 5.9" door, and assemble it all inside the tank. The lower shelf holds a cut off coffee can to keep the Al from dropping into the tank on melting, and to catch any Al vapor going that way, with slots cut in it for the FT rods to come in. This fits up against the bottom of the ringstand fairly well as I made the shelf that sits on adjustable. I made a couple of target holders that sit on the top of the stand, basically pieces of sheet metal with holes in them for the Al vapor to come through that I can set substrates over for coating. This works pretty well and keeps the Al vapor off things in the tank. I can see how hot it is getting through the sloppy holes for the FT inputs. Which also means. some Al vapor comes out that way. One trick I've used is to cut some mica washers to slip over the rods and block that path to keep my main window clean, and with a little experience, you can tell by how quick they get opaque when your real deposition target has the right amount of Al on it. Of course, you have to get off the button a little early, as it doesn't stop right when you drop input power.

I am thinking of trying gold off this same setup, but haven't risked it yet -- that stuff isn't cheap, even if one loading of pure wire will probably do all I'll ever need -- if it stays on the wire and doesn't just drop off when I go to melt it, that is.

Let 'er rip fellas -- I've just barely scratched the surface here and am a real beginner at it, though I was surprised it works as well as it does with the cheezy setup I'm using.

I built a thing that holds three boats and has a shutter in a Cu flashing box to keep most of the heat and excess vapor, but it hasn't made it into the tank yet, as I'm doing mostly fusion in there at the moment....Picture later when I have a clue if it even works -- "looks good on paper" but who knows. The idea behind that one was to use the shutter to control preheat and so on, and get more accurate deposition, as the shutter can be closed fast with a small solenoid (I have some very small ones that seem not to mind intermittent use in vacuum and don't outgass much, as long as you don't get them hot).
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.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: Evaporation

Postby Jerry » Sun Aug 15, 2010 1:35 am

I have a couple books on deposition that I picked up at Powell's Technical Book store. They cam out of the Tektronix library and they are pretty well worn! Vacuum Deposition of Thin Films by Holland is the best one. 1956.

Evaporation likes to be in the -6 range. Main reason is to keep gas from getting trapped under the deposited layer.

I would avoid mig wire. I think you want as pure aluminum as you can get. With mig wire you get a lot of other things that you may not want, The most common 4043 alloy has Cu, Si,Fe, Mn, Mg, Cr, Zn and Ti, though Cu and Si are the main alloy elements are Cu and Si, with the Si in the 4-6% range.

There are tungsten filaments with the aluminum already wound in with them like these:
http://cgi.ebay.com/ws/eBayISAPI.dll?Vi ... 500wt_1154
http://cgi.ebay.com/ws/eBayISAPI.dll?Vi ... 500wt_1154

Are you using a shutter over the source while it is warming up? Thats one of the reasons I am installing ferrofluid feedthroughs.

One nice thing about depositing aluminum you can get rid if it easy on insulators and the like using just about anything caustic. Oven cleaner, Zep Purple, etc. They all contain lye which will dissolve the aluminum and leave the rest alone. If you are using crystal dep sensors you can reuse them too by dissolving the aluminum.

I used to know a guy who had a gold deposition system. The source was a tungsten rod, maybe .040 to 1/16 coated with gold. Pretty simple.
Jerry
 
Posts: 573
Joined: Sun Jul 18, 2010 12:07 am
Location: Beaverton, OR

Re: Evaporation

Postby Doug Coulter » Sun Aug 15, 2010 9:32 am

Yes, I found out about the MIG wire early on -- it's not good for this. The reason is that the components don't evaporate at the same rate as they exist in the source material.
Also, they freeze so fast on hitting the deposition target that you get some pretty weird "crystalline" structure that is just random atom arrangements -- perhaps with a layer or gradient as the different components come off the heater. Might work out differently if you heated the target to near the melting point of the deposition stuff, that is, if it would live through that.

As Lesker points out in their words on this topic (everybody go get that catalog! Gold!) vapor pressure and melting points aren't that closely related. So I switched to some
little pure Al canes from Lesker and that made it all good. You just pinch them over a few of the turns, melt (in high vacuum, raising temp slow at first) and they wick into
the coils of the heater fine. After that, you are good to go a number of times before a recharge is needed (but be careful, the W coil becomes very brittle).

I've gotten by for proof of concept without the shutter so far -- a pre deposition on a junk target first, then change out for the real thing gets it done, with that hassle.
The shutter thing I built hasn't been put in and tried yet, but the shutter works off a tiny solenoid (1/4" dia, 1/2" long) to save an expensive feedthrough, using a little toggle
mechanism we came up with. Placing the solenoid out of the heat area makes it workable. I did put it in a vac system and check for outgassing, and it seemed fine there, just haven't
put it into production at this point, as I got stalled on the mask making I needed to do to make things like many-couple thin-film thermopiles out of things like Bi and Sb from boats.

evapsource.jpg
Evap source under construction


Here is the new boat source in the process of being built. Yes, a kitty cat stepped on the boat on the left and broke it. I later fitted the top with a shutter. Provision is made
to spot weld a type C thermocouple to each boat to have some closed loop temp control and ramping. Work in progress, not tested much yet.

evapxfrmr.jpg
Rewound spot welder transformer for evap


Here's the spot welding transformer I rewound for this work. Overkill by a lot, but it works. MOT rewound in a similar fashion is about right. Low resistance is key, as if you just
try to use a too weak source overdriven, the volts are held low till the heater starts getting hot, then rise too fast as the resistance goes up in the heater faster than you can turn
the input down.

WireEvap.jpg
Wire evap sources


This is what I'm using now, when I do this. The lower heater has the clamps that screw to the feedthrough rods on it, and is loaded with pure Al (or it was pure when I removed it from the tank, now it looks a bit dirty). This one has made maybe 10 or 12 runs and been loaded 2 or 3 times. Above that is one wound with gold ready to go when I have the guts. I will pre melt the gold with a dummy glass target and do one evap on that target before "going for the gold" on something more real.

I got reasonable resolution with mechanical masks, and went off to work on a good photolith process for making really good ones, and since that's not done yet, I've not revisited this. From my point of view at this time, the masks are the limiting path -- this other stuff seems to work fairly easily for me.

I am making progress with a PVA/dichromate resist I am tweaking here, but it's not quite ready for prime-time in making the masks I want (and will never be semiconductor kinds of resolutions except for big sloppy devices perhaps).
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.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: Evaporation

Postby Jerry » Thu May 19, 2011 2:42 pm

Got this thing from my friend at Surplus Gizmos last night. Its a Varian evaporator control. One side is a SCR power control and the other is a 4 position, 200A, 600v switch. Bad thing is it is for 208/240. One thing I didnt know though, the output of the transformer is a lot higher than I thought, up to 40v. So I am thinking of picking up an old chinese mig machine off craigslist. The unit also has a bunch of 4/0 cable coming off the switch.

Image
Varian evaporator control by macona, on Flickr
Jerry
 
Posts: 573
Joined: Sun Jul 18, 2010 12:07 am
Location: Beaverton, OR


Return to Vacuum deposition

Who is online

Users browsing this forum: No registered users and 1 guest

cron