This sounds like incomplete developer liquid exchange in the narrow
features.  Longer time in the developer and/or more agitation
(ultrasound) may help clear the narrower channels. Also, we find such
residue can be removed with a low power, short O2/Ar plasma ash. You
will need to experiment to find a compromise that minimizes damage to
the pattern.

Good luck,

Don Friedrich
JDSU
Santa Rosa

-----Original Message-----
From: Jan Newman [mailto:newman@optimim.demon.co.uk]
Sent: Thursday, December 18, 2008 2:07 PM
To: 'General MEMS discussion'
Subject: Re: [mems-talk] Problem with photolithography

Hello Laura
             Certainly could be a cause, what proximity gap between
wafer
and mask  are you using?
Also what aligner ?

Jan

-----Original Message-----
From: mems-talk-bounces@memsnet.org
[mailto:mems-talk-bounces@memsnet.org]
On Behalf Of Yinyan Gong
Sent: 18 December 2008 15:16
To: mems-talk@memsnet.org
Subject: [mems-talk] Problem with photolithography

I used shipley 1813 and the spin speed is 4000rpm (thickness of
photoresist is about 1.3um). After spin coating the wafer is softbaked
in oven at 90C for 30min. But after the wafer is developed in MIF 319,
there is always some photoresist left in the exposed area with 10um in
width but the photoresist was removed in the exposed area of 100um by
100um. I tried to increase the exposure time, but the problem still
exist? Could that possibly because the mask and wafer are not contacted?

Thanks.

Laura