I do this etch a quite often using AZ 2035, a negative resist, and strip
the resist with acetone. I have the same problem with a 'veil' of
crosslinked resist not being removed, especially over the top of more
reflective areas such as gold pads.

One solution is to use a thicker resist layer, another solution is to
use a positive resist under the negative resist (positive doesn't
crosslink), reducing the exposure time seems to work, as does increasing
the undercut of the negative resist by increasing the develop time (but
I don't know about S1813).

My conditions are 50W RF, 250W ICP, 80 sccm O2, 20 mTorr. Etches
polyimide at about 100 nm/min in an oxford plasmalab 80 plus. Resist
thickness of about 5 microns to etch 2 microns of polyimide.

Jason Milne
Microelectronics Research Group
The University of Western Australia

-----Original Message-----
From: mems-talk-bounces@memsnet.org
[mailto:mems-talk-bounces@memsnet.org] On Behalf Of Zotl Ernst
Sent: Wednesday, 9 January 2008 11:50 PM
To: mems-talk@memsnet.org
Subject: [mems-talk] Dry etching of polyimide with PR mask

Dear all,

I have a problem when dry etching polyimide with a resist (S1813) mask:
After stripping the resist with MEK it seems that some crosslinked
resist covers part of the PI. This problem is especially visible around
pads (gold) which are etched free (seems like the crosslinked resist is
clinging to the PI or resist sidewall).
We run this etch on an RIE tool with 300W, 100 sccm O2 and a pressure of
150 mTorr.

I had some success in avoiding this kind of residue by mixing the O2
with 20% CF4, reducing the power and increasing the pressure to about
250 mTorr, but this dramatically affects uniformity, with very slow etch
rate in the center and high etchrate on the edge of the tool.

I would be glad if anyone has some hints how to avoid this problem.