Jessica,

There are a couple issues that could be involved here.  As you say, you are
using a 2-coat process for the coating of the material.  One of the issues could
be the two bake processes you use.  If your first bake is too aggressive, then
once the second bake is complete, the first layer will effectively be hard-baked
9rather than soft) and very difficult to remove regardless of exposure dose.

However, you also want to make sure that you are not under-baking the resist.
This can lead to bubbling during exposure, and film swelling and wrinkling
during develop.  Your best bet is to perform a light bake (just a few minutes)
between coats, and then a more extended bake (110-120C, for 10 - 15 minutes)
after the second coat.  Unfortunately, I cannot provide the exact parameters for
the two-coat bake, since I typically apply a 100µm layer of AZ50XT in a single
coat (with an EVG150).

A second concern is the developer that you use.  I believe that the AZ400K
material is recommended.  This chemical is always diluted before use, and the
most common dilution is 1:4 with DI water.  However, for a 100µm thick film,
this will take an extended period of time.  The time involved will also give the
material a chance to influence the un-exposed regions as well.  If you are using
a low concentration of AZ400K, you may want to consider increasing the
concentration.  Just be careful, since this will magnify any insufficiencies you
may have with the bake.

Best Regards,
Chad Brubaker

-----Original Message-----
From: Jessica Melin
Subject: [mems-talk] Thick Positive Photoresist (100um)

I am trying to find a good recipe for achieving 100um thick positive photoresist
on silicon wafers. Currently, we are trying a two layer process using AZ50XT but
there are major problems with poor adhesion of the resist after developing (even
when HMDS is used and the resist is left overnight to rehydrate). Would
appreciate any help on a successful recipe using AZ50XT or what other positive
photoresist could be used. Please contact me as soon as possible. Thank you in
advance,