durusmail: mems-talk: AZ4620 processing
AZ4620 processing
2009-09-02
2009-09-03
AZ4620 processing
Y.Tian
2009-09-03
Hi Anirban,

>From my experience, if you want to get a perfect square pattern through the AZ
+ UV process, your mask should be very good and your UV should be very good. I
has this kind of problem before, the square patterns always looked like oval or
nearly circles.

I guess your small patterns were not washed away by the electroplating solution.
The resist is good enough to stand in most of the electroplating solutions for
hours. I guess it was caused by the development step. Smaller patterns got over-
developed. Also, rehydration is a good way to improve the quality. I always
leave it for half day or overnight before exposure.

If you want hard bake it after exposure, be careful of the temperature you use.
I would recommend below 100 C.

Hope this helps.

Yingtao

-----Original Message-----
From: mems-talk-bounces@memsnet.org [mailto:mems-talk-bounces@memsnet.org] On
Behalf Of ANIRBAN SARKAR
Sent: 02 September 2009 20:02
To: General MEMS discussion
Subject: [mems-talk] AZ4620 processing

Hi all,

I am trying to develop a suitable recipe for obtaining around 20 microns
thick photoresist mold by double coating of AZ4620.

Till now I have encountered  two problems and I would be obliged if you can
help me in solving the problem.

I am using the following recipe for double coating of AZP4620:

Si wafers with Ti and Copper deposited on it by thermal evaporation
Spincoating of HMDS  for proper adhesion
1st layer Spincoating: 500 rpm for 5 secs 1000 rpm for 10 secs 2100rpm for
20 secs
Soft bake:110'c for 4 minutes hot plate bake
2nd layer Spincoating: 500 rpm for 5 secs 1000 rpm for 10 secs 2100rpm for
20 secs
Soft bake:110'c for 4 minutes hot plate bake

**** Is it necessary to provide some sort of rehydration time before exposure?

Exposure: Quintel UV exposure station 1600 mJ/cm2
Development: AZ400K: DI water=1:4 for 4-5 minutes.

Problem #1: I have varying linewidths in my patterns ranging from 10um (min)
to 100um(max). I have no problems in getting the bigger patterns OK but the
smaller patterns of 10um linewidth are not following the mask.  Most of the
square patterns came out closely to circle as well ovals mostly.

Problem#2:  I have to use this thick photoresist for Copper
electroplating. From my previous experiments I can see that the smaller
patterns are getting washed away after a certain period of time in the
electroplating solution.  The bigger patterns are holding up well and I am
able to get reasonably good metal deposition.

****Do I need to hard bake the resist before performing the electroplating?
If so what is the temperature which has to be used for this purpose?

Also if I am aiming for thick PR mold is double coating with AZP4620 is a
safe option or should I switch to  another positive resist ( I cannot use
negative resist like Su8) like AZ92XX for the purpose.

I am looking forward for any help in this regard.

Thanks
Anirban

Electrical Engineering
Louisiana State University
Baton Rouge
70802,USA
reply