We have been using MicroChem SU-8 25 to create negative molds about 100 um deep
on glass wafers. We spin it on at 750 rpm. We began to have a problem several
months ago with the SU-8 shrinking back from the edge of the glass, and leaving
an irregular annular gap around the circumference of the wafer. This worsened
during the 25 minute, 95 C, pre-exposure softbake, but some gap was noticable
immediately after spin coating. The problem became so bad that only a 1.5 inch
spot in the center of the 2.5 inch wafer was covered. We finally just bought a
new batch of resist, and this seemed to solve the problem.

That was a month ago, and now it seems to be starting to happen again. A 1/8 to
1/4 inch gap is appearing around the wafer edge. Can anyone suggest why this is
occuring? I am concerned that we are storing the resist incorrectly, but we also
recently switched to a cheaper type of glass wafer. Is this a common problem?
Are there processing steps to correct it?

Thank you for any help you can offer.

Jordan Berg
Texas Tech University