Steven,
           A problem a colleague of mine had was similar.  He cured it
with a combination of image reversal and a Suss holographic 3
dimensional aligner.  He achieved sub micron images on the top, down the
side walls, and at the bottom of a trench.  When I get back in the
office I will a) contact him for more details and b) send you all the
details I have.  Bill Moffat

________________________________

From: mems-talk-bounces@memsnet.org on behalf of Steven Yang
Sent: Mon 4/7/2008 7:45 AM
Subject: [mems-talk] How to pattern photoresist in channel bottom?

Hi, all

Got a photoresist pattern problem to pattern electrode from channel
surface to channel bottom. My channel is DRIE etched into Si substrate
with depth around 25-40 microns. My electrode need to fall on the
bottom of the channel, while its connection will extend from bottom
along the side wall to the up surface of the Si sub.

Now, I use AZ4620 with typical thickness around 6.5um on the plane
surface to pattern the electrode and its connection. However, it
turns out that the pattern on the up surface getting over-developed
while the one on the bottom of the channel still under develop, which
totally destroy the whole pattern. and I also suspect that the pattern
on the sidewall of the channel actually did not formed, as I think the
photoresist on the sidewall and corner of the bottom will be thicker
than elsewhere.

So, do you guys have any suggestion or standard procedure to pattern
electrode cover up and down surface of the microchannel?