Joachim Oberhammer wrote:
> I have a 6 um thick layer of AZ4562 photoresist, which was used to
> pattern a 1 um thick Si3N4 layer, etched in CF4 plasma for about 35
> min. The resist is cured for 45 min at 110 degC. There must be
> something bad going on with the top of the resist layer, since it is
> not removable in Aceton (which works fine, usually, with that kind of
> resist).

   Is your RIE power very high? I've had to limit the power during
nitride etching to about 100W on a Plasmaquest RIE to avoid hardening a
5um layer of AZ4620.  At 200W I saw a lot of charring and the resist
could not be removed with solvents - sounds like your problem.  My RIE
parameters were 20mtorr, 15sccm CF4, 2sccm O2, 15sccm He, and the etch
rate was about 400A/min.
   If you can't find a way to strip the resist, a lower etch power or
bias might do the trick.  Good luck,

John Maloney
MEMS Engineer

MicroCHIPS, Inc.
http://www.mchips.com
http://john.maloney.org