[labnetwork] Regarding : Aluminium E-Beam Evaporation

[Owain Clark]

Doesn’t always work we find, eventually some enterprising user will find a way to melt an Al ingot onto a cooled hearth which should in theory not happen. But I do agree if this is not an issue and you can dedicate a pocket then this is the best method. Some electron beam gun cooling channel design and water flows will get you a lot farther then others in this respect.

In the last year I spent some time personally trying to improve our standard long throw Al evaporation process. We use 40cc pockets and there is no choice but to use a crucible because multiple materials use the pockets and ultimately it makes multi user/project operation easier. Al is cheap to refill, and the odd crucible is just noise in the tool budget too.

What I found ultimately was that alumina was the best. Power to deposit was significantly lower than any other crucible for a given rate because due to the thermally insulating properties of Al2O3 the entire Al crucible load melts and becomes totally molten, this gives a very easy to control deposition with a large vapour emitting area and even with a 1m+ throw distance rates of 2-5A/s at the wafer were no trouble for ~1kW of input power. The downside is that Al2O3 always cracks, no matter now gentle you are with preconditioning. But considering the overall budget of tool operation a new crucible every few months is not significant. The crucible outer wall stays cold and any Al that moves through the cracks cools and does not transmit further – this gives the cracked crucible physical stability. Eventually after several months and many um of Al you see the power for a given deposition rate slowly start to increase. This is the cue that Al moving through cracks has begun thermal contact with the pocket and is a sign to prepare to swap the crucible for a new one. Al in this crucible stays perfectly shiny following deposition for its entire lifetime.

For graphite, coated graphite or fabmate the crucible will eventually crack, and the Al will react with the liner. The liners I tested like this gave a yellow-ish hue to the Al after only a few depositions which I presume is C contamination (although I have not yet tested it via elemental analysis). The colour change is obvious by eye. Power vs Al2O3 is approximately double for a given rate due to increased thermal conduction through the crucible, and you cannot melt the entire Al contents.

Using metallic crucibles such as W the Al will creep eventually ruining the deposition by demanding significantly more power required for a given rate as it goes over the top of the crucible, only a small spot can be melted, and the power required is 2-3x more than Al2O3. Crucibles are physically unreactive and Al stays pure but it does not take long before creep renders them useless.

So all in, if you must use a crucible for Al I would recommend Al2O3 or a similar thermally insulating unreactive material. And I would always take any material vendor suggestions with a pinch of salt and just a starting point. Tool vendors I listen more closely too but seeing is always believing. Try the options yourself and draw your own conclusions based on expected theory. Al2O3 has been our standard crucible for over a year now and both research and enterprise users of the tool are happy. Keep a pre melted 2nd crucible ready to go as needed and all is fine.

Owain.

From: labnetwork <labnetwork-bounces at mtl.mit.edu> On Behalf Of Steffen, Paul
Sent: 29 August 2022 14:01
To: 'Chandan H B' <chandanachar95 at gmail.com>; labnetwork at mtl.mit.edu
Subject: Re: [labnetwork] Regarding : Aluminium E-Beam Evaporation

CAUTION: This e-mail originated outside the University of Southampton.
Chandan,

Don’t bother with the crucible liner. Just melt a big chunk of aluminum directly in the pocket. It will not adhere to the walls so you will be able to remove it if you need to swap materials.

-Paul

From: labnetwork <labnetwork-bounces at mtl.mit.edu<mailto:labnetwork-bounces at mtl.mit.edu>> On Behalf Of Chandan H B
Sent: Monday, August 29, 2022 2:04 AM
To: labnetwork at mtl.mit.edu<mailto:labnetwork at mtl.mit.edu>
Subject: [labnetwork] Regarding : Aluminium E-Beam Evaporation

Dear Lab network community, We are currently facing an issue in the deposition of Aluminium in one of our Electron beam evaporation tools, we tried intermetallic, fab mate, copper, glassy coated graphite, and graphite crucibles for the evaporation. 

Dear Lab network community,

We are currently facing an issue in the deposition of Aluminium in one of our Electron beam evaporation tools, we tried intermetallic, fab mate, copper, glassy coated graphite, and graphite crucibles for the evaporation. It seems that none of them are working out.
We see crucible gets broken at the initial deposition quite often.

Are we missing out on any parameters unchecked?
Kindly recommend us a few parameters or solutions for the same.

Any suggestions/Inputs are highly appreciated.
Thanks in advance!

Regards,
CHANDAN H B
THIN FILM ENGINEER

P.S: Here are a few parameters that are provided to the tool.
Power: 10KW
Voltage: 10KV constant
Current: Variable
Rise1 & Soak1: 5min & 1min /8min & 2min
Rise1 Power: 4% (for intermetallic crucible)
Rise2 & Soak2: 5min & 1min /8min & 2min
Rise2 Power: 8% (for intermetallic crucible)
Ramp down: 5min
Rate of deposition: 0.1nm/sec
Beam Pattern: Spot Beam at the center of the crucible
Crucible Volume: 20cc
Material fill %: As recommended 67-75%


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20220829/ec329666/attachment.html>