Dave,

     A contact 1 micron in diameter or larger should be easy to wet
etch. Contacts in oxide with resist openings of less than a micron can
be wet etched easily with superwet BOE. A micron is much larger than
either a cyanide or iodine complex. So it isn't the size of the etching
chemical species.

This is a long shot, but perhaps some copper is electroplated under the
small resist features, but it can't get under large resist features.
Only a monolayer of copper would be enough to block the gold etch.

As the copper is electroplated around the small resist feature it might
stress or pull on the resist. Perhaps there is slow chemical penetration
at the base, perhaps the electrical potential under the resist pillar
causes a lateral electric gradient of potential resulting in copper
deposition under the resist.

Basically the size is not the issue. A 1 micron contact should be easy
to wet etch. It isn't the wet chemistry either. I am fairly sure that if
you has a resist pattern on the gold, with 1 micron openings in the
resist, you could etch the gold. I think it is something to do with the
copper electroplating or it being a copper/gold layered structure. I
haven't read a lot about etching fine patterns in multi-layer films of
different metals. The copper might be functioning as a sacrificial
anode.

One thing you might try is to bake the resist pattern at a high
temperature as possible without melting the resist. Or do a Fusion
treatment and bake on the resist to make it more inpenetrable to
solutions.

Or you might try a chemical etch back on the copper a 100 angstroms
roughly.

I would also try imaging a resolution test mask on to one of your wafers
with the copper/gold layers. Of note would be how lines and spaces pitch
structures etch out, how do isolated lines behave.

Also, for the larger openings which you can etch, I would do a partial
etch of a resist patterned copper/gold sandwich and analyze the edge of
the larger opening. See what the profile is.

Best of luck,

Ed


-----Original Message-----
From: mems-talk-bounces@memsnet.org
[mailto:mems-talk-bounces@memsnet.org] On Behalf Of Dave Lewis
Sent: Monday, October 19, 2009 3:13 PM
To: Dave Lewis; General MEMS discussion
Subject: Re: [mems-talk] Etching gold through copper micro-holes


All,

I have a silicon wafer and I have deposited gold to the silicon by
sputtering (this gold substrate is less than 1 micron thick).  My gold
substrate is electroplated with copper after photoresist has been
patterned onto the gold.  Once the copper is on the gold, I remove the
photoresist which leaves behind holes in the copper that allow access to
the gold.  At this point I would like to remove the gold from the wafer
and have my copper film remain.  I have successfully done this, but only
with >5 micron sized holes in the copper.

As soon as I get our RIE up and running again, I will attempt O2 plasma
descum step prior to etching, but I want to know if the problem could be
something other than an organic contaminant layer.


Regards,

Dave Lewis
Lake Shore Cryotronics
Ohio
Dave.Lewis@Lakeshore.com