Sadhana,
       It sounds like the old resist peeling problem. Some brands seemed
to have it. It is the very surface skin peeling off the resist features
near the edges, and they can roll up and look like threads.

However, let me give you the usual things to try.

1. Do you softbake heating through conduction. That is the wafer goes on
a chuck and heat from the chuck conducts through the wafer and drives
desolvation that way. I seem to remember the top down heating had a
tendency to make a crust for oven convection and IR softbaking.

2. Look at the exhaust on your bowl. You may be trapping solvent below
the resist surface through premature drying of the surface. Perhaps
there is too much exhausting during softbake. How much is too much. For
the spin bowl you have the problems of striations in the resist and
non-unformity in thickness when you have too much exhaust.

3. Try a post exposure softbake. This redistributes photoactive
compounds which is its purpose, but it can also tend to equalize your
solvents in the vertical direction in your resist. This is overcoming
the problem.

4. Make sure your softbake is long enough. This should be done by
checking the thickness versus time. Plot resist thickness versus time
and when you get to the point that the thickness isn't changing, add 50%
to 100% to the time. Don't skimp on softbake time.

Your softbake should be between 95 to 110 deg. C. Undersoftbaking resist
is one of the big no-no's.

5. Peeling resist used to be a curse of photolithography engineers.
Changing the resist vendor seemed to work when nothing else did. You
change resist vendors as follows:

        A. Email/Phone/write the resist vendor describing your problem.
Since you have  Shipley you will find it impossible to contact Rhom &
Hass on the web. You would      have to contact the resist reseller by
getting the number from purchasing.

        B. If you get no response, follow up again so you can say they
didn't respond to       repeated inquiries and then change the resist
vendor.

        C. If you do get a response Test the suggestions of the resist
vendor, and then        write up the report of the testing and the
failed results. Theorectically one of   these suggestions of the resist
vendor might work, however, I have no experience        regarding
successful resist vendor suggestions. Make sure it really works for more
than one lot.

        I have listed the usual things already, so if you do them, you
can say, "I have        done that, I have done that," and you can get on
to changing the resist.

        Also, don't accept that the problem is due to something else in
your fab.

        D. Then change resist.

The above exercise will be necessary since the vendor will otherwise
claim that they can fix it and your manager will insist that you try
their suggestions.

Ed


-----Original Message-----
From: mems-talk-bounces@memsnet.org
[mailto:mems-talk-bounces@memsnet.org] On Behalf Of Sadhana Patil
Sent: Friday, October 05, 2007 12:45 PM
To: mems-talk@memsnet.org
Subject: [mems-talk] fudging/spreading around the PR features

I am new to Photolithography. I am doing Photolithography with S1813.
Many a times I get fudging/spreading of photoresist around the edge of
the photoresist features after developing. They look more or less like
threads spreading outward from the feature.
Does anyone know what am I doing wrong?