Dear Chad,

U are right. Though I succeeded in making 38um thick pattern with 20um feature
size, there is still some problem with undercut of the pattern and bottom
residue. I will try AZ 9260, and hoping you can send me your processing traveler
so I can use as reference and start point.

Thanks

Michael


-----Original Message-----
From: mems-talk-bounces@memsnet.org [mailto:mems-talk-bounces@memsnet.org] On
Behalf Of Brubaker Chad
Sent: 2008年6月13日 13:26
To: General MEMS discussion
Cc: Brubaker Chad
Subject: Re: [mems-talk] 20um features using AZP4620

There is actually a problem with trying to use AZP4620 to that thickness, and
one of the other responders alluded to it - transparency.

For layers thicker than about 15 - 20 µm, you'll already start to see some
issues with light penetration into an AZP4620 film (the develop behavior becomes
much less linear).  Beyond 25 µm, you begin to have extreme difficulties in
clearing the bottoms of exposed regions, because the light is just not
penetrating that far.

The suggestion of AZ9260 is a good one - this was AZ's (or Clariant's, at the
time) update to AZP4620, once it became apparent that people were using the
material for thicker-and-thicker films.  Its two primary differences from
AZP4620 is greater transparency (I've produced 5:1 aspect ratios) and an
included leveling agent to produce smoother and more uniform films.  I myself
have single coated this material to 65 µm thick, and patterned it (my minimum CD
was somewhere between 10 and 15 µm).

Best Regards,
Chad Brubaker