[labnetwork] Sputter problem

Fri Jun 24 13:50:10 EDT 2011

HI,

How is your base pressure and leak back rate? If you have a leak it may
be adding extra O2 which the Ti just gobbles up. What machine are you
doing this in?

How is your water cooling flow, if there is reduced flow it will affect
the process, stuff will heat up, the heat may cause the target to
de-bond, thus creating a bad RF circuit. That recently happened to us on
a sputter tool, the target de-bonded slightly and the rate dropped.

Are you doing a reactive process or do you have a TiO2 target? Is your
pressure control working properly, is the gate valve actively controlled
? if so if there anything wrong with the pressure signal to the gate
valve?

Hope this helps.

Rick

Rick Morrison

Senior Member Technical  Staff 

Acting Group Leader Mems Fabrication

Draper Laboratory

555 Technology Square

Cambridge, MA  02139

617-258-3420

From: labnetwork-bounces at mtl.mit.edu
[mailto:labnetwork-bounces at mtl.mit.edu] On Behalf Of Steven Foland
Sent: Friday, June 24, 2011 11:05 AM
To: Gupta, Su
Cc: labnetwork at mtl.mit.edu; Keith Bradshaw
Subject: Re: [labnetwork] Sputter problem

Hi Prof. Gupta,

We are doing RF sputtering, not DC. Cleaning the target longer does
increase deposition rate temporarily, but it drops back down quickly. We
are most likely operating at the "oxidized" rate, but still are unsure
why we are experiencing this gradual decline in film thickness.

Here is my data from yesterday:
1st run: 5 minute clean cycle, 5 minute coat cycle: 250 Angstroms
2nd run: 5 min clean, 10 min coat: 198 Angstroms
3rd run: 10 min clean, 5 min coat: 118 Angstroms
4th run: 5 min clean, 5 min coat: 96 Angstroms
5th run: 5 min clean, 5 min coat: 80 Angstroms
6th run: 5 min clean, 5 min coat: 90 Angstroms
7th run: 20 min clean, 5 min coat: 120 Angstroms

So you see, we have a gradual decline in dep rate, but can increase the
dep rate slightly by running a longer clean cycle.

If enough time has passed (a day or two) between runs, the rate
increases back to its original value of ~250 Angstroms in 5 minutes.

Thoughts?

Thank you,
Steven Foland

On Thu, Jun 23, 2011 at 11:54 PM, Gupta, Su <sgupta at eng.ua.edu> wrote:

Hi Keith:

Can you give me some more details about your sputter system and the
processs? For instance, are you sputtering reactively from a Ti target
or is it TiO2? If you are doing DC reactive sputtering from an elemental
target, then the target voltage is the best indicator of what is going
on with the process. For instance, the target may be oxidizing further
with each run (even with the preclean) and you may be dropping down the
slope of the hysteresis loop from the 'metallic' rate to the "oxidized"
rate, which would be accurately reflected in a drop in the target
voltage. A small leak or outgassing could also cause this type of
problem, but it probably would not be as regular and systematic as what
you are observing.

Regards,
Su Gupta
Assoc. Prof, MTE
Faculty Director, uamcro
Univ. of Alabama
________________________________
From: labnetwork-bounces at mtl.mit.edu [labnetwork-bounces at mtl.mit.edu] On
Behalf Of Keith Bradshaw [bradshaw1234 at gmail.com]
Sent: Thursday, June 23, 2011 5:09 PM
To: labnetwork at mtl.mit.edu; Steven Foland
Subject: [labnetwork] Sputter problem

We are sputtering TiO2 .

We begin the day with a 250 angstrom rate....each subsequent run is
reduced by 10-15% in rate until we are at a 125 angstrom rate.  We are
not changing anything.

Next day we begin again at 250.

Tried argon clean between runs, tried 3 hour wait between runs, we are
using a load lock and vacuum looks stable , still rate drops on each
run.

Any ideas?

cordially,

Keith Bradshaw
Dallas

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20110624/467a1200/attachment.html>