One correction to Ed's statement - I'm not sure if EGMEA is available anymore -
the AZ family of resists moved away from this a while ago, due to some health
concerns with the solvent.  Instead, they switched to PGMEA for nearly all of
their formulations (thus, the currently available 4000 series resist is actually
called AZ P4620, etc.)

AZ EBR 70/30 is a good idea - this is 70% PGME (Propylene Glycol Monomethyl
Ether), 30% PGMEA (Propylene glycol monomethyl ether acetate)- the mixture dires
more readily than pure PGMEA alone.

PGMEA is actually the developer for SU-8, so it should not attack the patterned
SU-8 (that's kind of the point).

I have not seen much evidence of acetone swelling fully cured SU-8 (meaning,
having completed its appropriate PEB at 95ºC) - SU-8 seems pretty impervious to
all solvents.

You could also potentially use an oxygen plasma to strip the resist - the etch
rate of SU-8 is virtually zero at room temperature, and at best (with special
processing which I do not know the details of) 1 µm/min - the resist will etch
much, much faster.

Best Regards,
Chad Brubaker

EV Group       invent * innovate * implement
Senior Process Engineer - Technology - Tel:  480.727.9635, Fax:  480.727.9700
e-mail: [email protected], www.EVGroup.com


-----Original Message-----
From: [email protected] [mailto:[email protected]] On
Behalf Of Edward Sebesta
Sent: Wednesday, January 30, 2008 9:14 AM
To: General MEMS discussion
Subject: RE: [mems-talk] How to remove AZ photoresist in SU-8 microchannel?

Perhaps a less aggressive solvent. It could be AZ 70/30 or some EGMEA
solution. These are the solvents the novolak resin in dissolved in when
they manufacture resists. There are a variety compounds similar to
EthylGlycoMethylEtherAcetate. (I am assuming that AZ5214 is a novolak
positive resist.)

However in terms of solvents dissolving something, the rule of like
dissolves like applies. That is, the solid will be dissolved by a
solvent with a similar polar/non-polar nature. So if SU-8 is is similar
in nature to the solvent it will dissolve.

You might want to use a resist that is completely aliphatic in nature,
like a negative resist, if SU-8 has a somewhat polar nature. If SU-8 is
aliphatic in nature, then use a resist system like a novolak resist and
a gentle solvent as mentioned above.

I would also make the resist strip quick as possible. Since you are
removing an organic off an organic, the differentials of solubility will
be relative, not absolute, and some minor amount of dissolution of the
SU-8 might occur.

An additional complication is the nature of the OTS and therefore what
dissolves the OTS layer.

I would consider lowering the temperature of the softbake of the resist
to see if you could make it more soluable and thus making the resist
strip shorter. You can't go too far in reduction of the temperature
since you will lose photo contrast. You could make sure you avoid
hardbakes of the resist after patterning.  This again allows a milder
solvent.

Finally, you could perhaps process the SU-8 to make it much less
soluable. I am not familiar with it.

I would call AZ, they are fairly good in their advice.

Ed