[labnetwork] Sputtering issue

Wed Apr 19 11:37:50 EDT 2023

Thanks so much for everyone’s helpful feedback and contribution to this discussion.

Here is the summary of the responses I received:

  1.  Check the magnetic field strength of the magnets
  2.  Use a thinner target (the Ta target I’ve been using 0.31” thick)
  3.  Use a conductive elastomer between the cathode and the target (I used indium foil that was suggested by the manufacturer)
  4.  There might be a chance of tantalum oxide on the surface of the target that one can get rid of by using pulsed DC
  5.  Make sure the housing is clean (This is a very valid concern but since the sputtering gun I am using have never been used it is not the case for me, the picture I attached that shows the particles were generated during the sputtering of the tantalum)
  6.  Test the sputtering gun with another material

Best,
Saba

________________________________
From: Minji <minji at seas.ucla.edu>
Sent: April 17, 2023 4:40 PM
To: Jing Guo <jeanne.guo at rice.edu>
Cc: Michael Yakimov <yakimom at sunypoly.edu>; Saba Sadeghi <saba.sadeghi at uwaterloo.ca>; labnetwork at mtl.mit.edu <labnetwork at mtl.mit.edu>
Subject: Re: [labnetwork] Sputtering issue

Hi Jing,

From my previous experience, this is most likely caused by the TaxOy
isolation layer built on the surface of the target, which then brings
a lot of arc to break down the dielectric layer.  Trying to lower the
base vacuum(Less O2 left in the chamber)before you send the Ar gas in
will help, another way is switching the DC sputtering power supply to
some kind of mid-Frequency power source, that will help to zero the
accumulated voltage potential on the surface of the isolation layer.
Many years ago, I designed a special pulsed sputtering power
supply(adding a very short positive microsecond pulse on top of
Negative sputtering voltage) which could solve the problem totally.

Best,
Minji

____________________________________________________________

Minji Zhu   |  Manager

Center for High Frequency Electronics

UCLA SAMUELI SCHOOL OF ENGINEERING
64-124 Eng IV, 420 Westwood PLZ
Los Angeles, CA 90095-1600
O: 310.825.4940 / M: 310.807.6168
www.chfe.ee.ucla.edu<http://www.chfe.ee.ucla.edu>

On Mon, Apr 17, 2023 at 8:45 AM Jing Guo <jeanne.guo at rice.edu> wrote:
>
> Hi Saba,
>
> The strike pressure (2e-3 mbar which is about 1.5 mTorr) sounds too low for me. Usually from my sputter system, the strike pressure would be 20~30 mTorr, then we would adjust it below 5 mTorr for deposition.
> If the pressure is low, sometime there will be some difficulties to start the plasma. It might be the reason it cannot support the stable plasma status especially for brand new targets.
>
> If your system can show you the current, voltage for the sputter gun including ‘50W’, you may be able to get more information once the plasma is struggling.
> The flakes on the target could be generated by too much arching during the sputtering process. You could test a different easy target like Al on the same gun to see what is the performance.
>
> I also agree with Michael about the magnets. What does the magnet look like, a ring magnet? If you have a Gauss meter, you can check the magnets’ condition. If not, check it as Micheal suggested. If the sputtered mark on the target is not a regular ring, you’d better check the magnets underneath.
>
>
>
> Best,
> Jing
> ---------------------------------------
> Jing Guo Ph.D.
> Research Scientist
> SEA Cleanroom (SST 017)
> Rice University
> Houston, TX
> jeanne.guo at rice.edu
> 713-348-8227
>
>
>
>
>
>
>
> On Apr 15, 2023, at 1:55 PM, Michael Yakimov <yakimom at sunypoly.edu> wrote:
>
> I would check magnets' condition. Once upon a time, I had a ring magnet on a similar one to partially lose magnetization. The result was the strange shape of plasma and a few other funny things. Once removed, it was fairly obvious that the screwdriver sticks to some areas better than others.
>
> _______________________________________________
> Michael Yakimov
> Research scientist
> College of Nanoscale Science and Engineering
> SUNY Polytechnic Institute
>
>
>
> 253 Fuller rd.
> Albany NY 12203
>
>
>
> Phone: 518-437-8609 lab
> e-mail: yakimom at sunypoly.edu
>
>
>
> ________________________________
> From: labnetwork <labnetwork-bounces at mtl.mit.edu> on behalf of Saba Sadeghi <saba.sadeghi at uwaterloo.ca>
> Sent: Friday, April 14, 2023 4:15 PM
> To: Jing Guo <jeanne.guo at rice.edu>
> Cc: labnetwork at mtl.mit.edu <labnetwork at mtl.mit.edu>
> Subject: Re: [labnetwork] Sputtering issue
>
> Hi Jing,
>
>
>
> Thanks a lot for getting back to me and the helpful discussion.
>
>
>
> This was a pure Ta sputtering. I could ignite the plasma at 50 W, 30 sccm Ar, 2E-3 mbar pressure but then the glow spots appear after a few minutes which led to the plasma extinguishing! I could reignite the plasma again by ramping down the power and then up to 50W! by the two hours of sputtering, I meant the sputtering collectively lasted two hours until it shorted.
>
>
>
> The sputtering gun housing were clean before the sputtering, but after it looks like particles accumulated on the surface of the target! and eventually ended up in the housing and caused the short (picture attached)
>
>
>
>
>
> Thank you.
>
>
>
> Best,
> Saba
>
>
> ________________________________
> From: Jing Guo <jeanne.guo at rice.edu>
> Sent: April 14, 2023 10:25 AM
> To: Saba Sadeghi <saba.sadeghi at uwaterloo.ca>
> Cc: labnetwork at mtl.mit.edu <labnetwork at mtl.mit.edu>
> Subject: Re: [labnetwork] Sputtering issue
>
> Hi Saba,
>
> Was the process a pure Ta sputtering or reactive sputtering? For a pure Ta DC power sputtering process, usually it shouldn’t generate that much particles. Ta coating on the sputter gun house/chimney should have a good adhesion. Any accumulated particles on the Ta target surface after ~ 2 hours sputtering?
>
> If the chimney or clamp ring for the target has been coated too much materials, you need to clean the whole kit and measure the gun to the ground layer by layer once you install a target. Another issue could be that some magnets got demagnetized which cannot provide strong magnetic field on the target surface.
>
> What was the strike condition, like pressure, power, gun voltage/current? Did you see whether the voltage/current can stabilize once the plasma was generated?
>
> There are several ways to figure out it is the sputter gun issue or a Ta target issue. Normally targets should be fine.
>
> Thanks.
>
>
>
> Best,
> Jing
> ---------------------------------------
> Jing Guo Ph.D.
> Research Scientist
> SEA Cleanroom (SST 017)
> Rice University
> Houston, TX
> jeanne.guo at rice.edu
> 713-348-8227
>
>
>
>
>
>
>
> On Apr 13, 2023, at 6:55 PM, Saba Sadeghi <saba.sadeghi at uwaterloo.ca> wrote:
>
> Dear Labnetwork community,
>
> I am having an issue with magnetron-sputtering of tantalum with DC power. After a few minutes of sputtering Ta glow spots appear on the surface of the target that eventually cause the plasma to distinguish.  I have loaded two fresh 0.31”-thick Ta targets, but both times the same problem occurred, and both times it generated enough particles in a short amount of time ( about 2 hours of sputtering, accumulatively) that led to shorting of the source.
>
> I suspected that this might be a cooling issue, but there is enough flow rate at the sputtering source,and I use indium thin foil for thermal contact between the cathode and the target (recommended by the company- MeiVac).
>
> Have you experienced such a problem? I am wondering if this could be an issue with the sputtering source or could somehow the two Ta targets I loaded have been compromised?
>
> Thank you very much.
>
> Best regards,
> Saba Sadeghi
> IQC, University of Waterloo
>
>
> _______________________________________________
> labnetwork mailing list
> labnetwork at mtl.mit.edu
> https://urldefense.com/v3/__https://mtl.mit.edu/mailman/listinfo.cgi/labnetwork__;!!BuQPrrmRaQ!kxlcKJiirMsFBnN6_pix8dvETfN-5vXnpaJXAA_4g3Jh5zmL_s-vuowA7xmr2s77ZKDztvk-khToRLa3a5M1qDmnShX9c4XwH0zL-Q$
>
>
>
> Get Outlook for Android
> _______________________________________________
> labnetwork mailing list
> labnetwork at mtl.mit.edu
> https://urldefense.com/v3/__https://mtl.mit.edu/mailman/listinfo.cgi/labnetwork__;!!BuQPrrmRaQ!k2g1scR3qzaeb1tQiLClZN7JqMS_fLnM-wyk1ysARAGyBpfKIZWZuoqQZhz12L8yMBYH7O--WSID-TelN2s3r6647R5jRGI$
>
>
> _______________________________________________
> 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/20230419/1ca847ef/attachment.html>