[mems-talk] SU-8 and S1813

[Joseph Grogan]

Good timing with this question as I was just trying a new process this
week where I pattern 1813 as a sacrificial layer and then encapsulate it
with SU8 2005 with the intention of removing the 1813 to leave me with
enclosed channels (and inlets patterned in the SU8). I was running into
a problem with alignment because after I coated the SU8 2005, I couldn't
see the 1813 beneath it anymore which prevented me from patterning my
inlets in the SU8. I thought maybe it was just a contrast problem and 
bad illumination. I did a couple tricks to get around the alignment 
problem and eventually patterned the SU8. But then after development of 
the SU8, it appeared that the channels left over by the stripped 1813 
were not clearly defined anymore. It occurred to me that maybe the 
solvent in the freshly spun SU8 was doing something to the 1813 right 
from the start.

So I coated a slide with 1813 and baked it for an hour at 125C to make
sure it was super tough. Then I put the slide in Cyclopentanone (SU8 
thinner/solvent) and the 1813 was stripped in about 15 seconds. This 
leads me to believe that it wasn't a contrast issue that was preventing 
me from seeing the underlying 1813, but rather, the underlying 1813 had 
been attacked by the SU8.

So my question is: has anyone successfully encapsulated 1813 with SU8
and then removed the 1813 to be left with a channel? I'm suspicious that
it may not be possible because the solvent in the SU8 attacks the
patterned 1813 and destroys the sacrificial layer. If this is the case, 
are there any other sacrificial positive resists anyone could suggest to 
use with SU8?

any input would be greatly appreciated, thanks.
-Joe Grogan

Brubaker Chad wrote:
> Oh yes.  It would most likely strip it, since SU-8 developer is the same
> solvent as the casting solvent for S1813 (PGMEA - propylene glycol
> monomethyl ether acetate, or 1-methoxy-2-propanol acetate)