[labnetwork] [EXTERNAL] Gold e-beam deposition - carbon contamination

[Ryan Rivers]

Here's what I've personally seen at our lab:

1) do not use fabmate for gold. Spring for a tungsten crucible. The reason
thermal evaporation is so clean is because you're in tungsten boats.
Tungsten and gold are immiscible.

2) you need to fill your gold to 60% or more of the crucible volume, and
then get your gold hotter when you precondition your melt on melt
formation. Should be a flat surface in the tungsten. Gold, when properly
and fully liquefied, will not dissolve carbon more or less at all. The
carbon will float to the top of the melt as it is lighter.

3) Pure carbon black is not jumping off your walls into your melt. Vapor
pressure isnt there. It's not possible. What's happening is you have VOC
content from your photoresist and/or surface contamination of your refill
pellets (they come in plastic bags, usually, and no amount of scrubbing
will fix this). That's getting into your melt, usually under yellow hot
conditions. Above 600C in a vacuum pretty much all forms of hydrocarbon or
polymer dissociate and the carbon becomes carbon black, which floats on top
of your melt.

4) when there is carbon black on your melt and you hit it with an ebeam, it
has way more resistance than gold. (I^2)R proceeds to tell us it's going to
get really hot. If you see a white block floating on your gold, that's
slag. Cool down the melt. Remove it, wipe it off with a nitrile gloved
fingertip. The underlying gold will shear and the slag will come right off.
(Again assuming a tungsten crucible.)

Long story short - get a tungsten crucible and everything is easy.

Best of luck,

Ryan

On Sat, Feb 1, 2025, 4:47 PM Demis D. John <demis at ucsb.edu> wrote:

> Attached is a paper that mentions and cites the reference regarding Ta
> pellet addition. We have good success with this method in a crucible.
>
> (Thank you Tom Reynolds for sharing that, and for reading this mailing
> list even after your retirement!)
>
> -- Demis (contact info <https://wiki.nanotech.ucsb.edu/wiki/Demis_D._John>
> )
> *Reminder*: The NanoFab has a publications policy
> <https://wiki.nanotech.ucsb.edu/wiki/Frequently_Asked_Questions#Publications_acknowledging_the_Nanofab>
>
>
> On Sat, Feb 1, 2025 at 07:43 Demis D. John <demis at ucsb.edu> wrote:
>
>> Adding a tantalum pellet to the Au source will getter the carbon and
>> reduce spitting.
>>
>> I can’t quite find the original paper, but here are some web pages that
>> reference it:
>> 1)
>>
>> https://compoundsemiconductor.net/article/116629/Tackling_the_foot#:~:text=concern.-,The%20combination%20of%20the%20crucible%20liner%20and%20tantalum%20%E2%80%98getters%E2%80%99%20looks%20like%20a%20win%2Dwin:%20the%20deposition%20rate%20increases%20while%20reducing%20gold%20spitting.,-But
>>
>> 2)
>>
>> https://www.memsnet.org/memstalk/16896/#:~:text=look%20for%20a%20paperwritten%20in%20the%201970s%20that%20suggests%20adding%20a%20small%20piece%20of%20tantalum%20to%20yourgold.%20%20The%20Ta%20will%20not%20melt%20but%20it%20will%20control%20the%20spitting.
>>
>>
>> I don’t have measurements of carbon contamination with this - although we
>> do this in our e-beam evaporators.
>>
>> -- Demis (contact info
>> <https://wiki.nanotech.ucsb.edu/wiki/Demis_D._John>)
>> *Reminder*: The NanoFab has a publications policy
>> <https://wiki.nanotech.ucsb.edu/wiki/Frequently_Asked_Questions#Publications_acknowledging_the_Nanofab>
>>
>>
>> On Sat, Feb 1, 2025 at 05:17 Czwakiel, James <czwakj at rpi.edu> wrote:
>>
>>> Youry
>>>
>>> Once contamination is detected…. Clean clean clean
>>> All metal covers removed and blasted and ultra sonically cleaned
>>> Plus …. Make sure crucibles are not touched even with gloved hands…..
>>> always use tweezers
>>>
>>> James Czwakiel
>>>
>>> Semiconductor Equipment Engineer
>>>
>>> [image: Image]
>>>
>>>
>>>
>>> Get Outlook for iOS <https://aka.ms/o0ukef>
>>> ------------------------------
>>> *From:* labnetwork <labnetwork-bounces at mtl.mit.edu> on behalf of Youry
>>> Borisenkov <yb2471 at columbia.edu>
>>> *Sent:* Friday, January 31, 2025 11:59:09 AM
>>> *To:* Fab Network <labnetwork at mtl.mit.edu>
>>> *Subject:* [EXTERNAL][labnetwork] Gold e-beam deposition - carbon
>>> contamination
>>>
>>>
>>> *CAUTION:* This email originated from outside of the organization. Do
>>> not click links or open attachments unless you recognize the sender and
>>> know the content is safe.
>>> Hi All and happy Friday,
>>> I know it is a topic that has been discussed multiple times, however,
>>> I'm still lacking a solution that works and will appreciate you sharing
>>> your experience.
>>>
>>> Currently we are using a Fabmate crucible, and getting some carbon
>>> contamination .
>>> For clean Gold layers we use thermal evaporation. The films are good but
>>> it requires a lot of gold.
>>>
>>> So far we tried using a Molybdenum crucible with and without a spacer.
>>> The issue with this approach was that eventually after some time, we get
>>> carbon contamination back. I believe it's present in the chamber and
>>> eventually a critical mass is built on top of the gold in the crucible,
>>> coming from the chamber.
>>>
>>> Were anyone successful in overcoming carbon contamination in
>>> their e-beam deposition overtime?
>>> What are the procedures you are following? Maybe cleaning more
>>> frequently?
>>>
>>> --
>>> Thank you,
>>> Youry Borisenkov
>>> CNI <https://cni.columbia.edu/columbia-university-clean-room>
>>> Columbia University
>>> CEPSR 1017, New York, NY, 10027, United States.
>>> <https://sustainable.columbia.edu/crown-commuter>
>>> _______________________________________________
>>> labnetwork mailing list
>>> labnetwork at mtl.mit.edu
>>> https://mtl.mit.edu/mailman/listinfo.cgi/labnetwork
>>>
>> _______________________________________________
> labnetwork mailing list
> labnetwork at mtl.mit.edu
> https://mtl.mit.edu/mailman/listinfo.cgi/labnetwork
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20250201/6cb17d5d/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Image.png
Type: image/png
Size: 11688 bytes
Desc: not available
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20250201/6cb17d5d/attachment.png>