Hello mems-talkers,

In the archive of the mems-talk forum bubbles in the SU-8 are often
mentioned, but I would like to reopen the discussion.

I work with SU-8 25 (the "old" version of SU-8).
I make 10 to 20 um SU-8 layers upon glass wafers. The biggest problem
was to obtain a resolution of 2,5 um with near UV lithography. Now a
have a recipe to do so; the drawback is that the adhesion of my SU-8 on
glass is less.
Especially after rinsing with IPA.I think that this is related with the
little bubbles I have in the resist.

I started to control the SU-8 layer after every process step with a
microscope. After the prebake, after the lithography, after the PEB I
could not observe any bubbles (maybe because of lack of contrast of the
layer under the microscope?). But after development the bubbles are
clearly visible and are about 1 um in diameter; at the surface they form
those "pivots" (a topic I found at the mems-talk as well !!) and they
seem related.

Important to know is that I dispense the resist manually on the wafer.
The spinners we use do not allow me to dispense SU-8 on a spinner that
is already rotating. The bubbles are homogeneous distributed, so I think
it can not be a caused by dispensing the resist. Neither have I degassed
the resist before use.

I am not a chemist, but some MSDSheets mention that the PAG SbF6 can
react to produce H2 or F2 gas. Can this be the case and causing the
bubbles? If so, is there any knowledge about this? Or should degassing
solve the problem? Can someone advise me a way of degassing? I am afraid
that a degas step will change the solvent concentration of the SU-8 ,
and so, I will need to change/optimize my recipe again.

Regards,

Herman Walgraeve

Delft University of Technology

Department of Biotechnology
Analytical Biotechnology Section
Julianalaan 67
2628 BC Delft
The Netherlands