Dear All,

I am having quite a lot of difficulty developping the recipe for 150microns
thick layers of SU8 2075 on fused silica substrated sputtered with gold
(4inch wafers). My main problems are:

(a) Sticking of the wafer to the mask.
(b) Some non-developped stuff stack to my patterns and I cannot get rid of
that only if I reduce the dose and then, I begin havng some initiation of
adhesion problems.

 My process is as follows:

1. Spin Omnicoat: I spin 5 layers of Omnicoat to try to get a thick layer so
that it is easier to remove the SU8 after my process. Each layer is spin at
3000rpm (as described in the Omnicoat data sheet) and baked for 30sec at
200C. After the 5 layers are spun, the total stack is baked for 1min at
200C.

2. Spin of SU8 2075: I spin it to a 150microns thickness. The coatings look
normally pretty nice and uniform.

3. Soft-bake: I place my samples on the hotplate at 40C and ramp it up
slowly to 95C. Keep it at 95C for 45min and then ramp it back down to 40C.
The ramps are the key for us to avoid inducing cracking and lifting of the
SU8 during developing.

4. EBR using acetone at 500rpm/sec for 70sec

5. Rest time: 3h

6. Exposure: 30sec (12mW/cm2). This dose is lower than the recommneded one
but when we went to higher doses I was getting non developped stuff on the
corners of my patterns and I could not get rid of that). As the wafers are
gold coated, I thought that maybe this dose should be enough? I have here my
first problem: THE WAFER STICKS NORMALLY TO THE MASK!

7. PEB: using the same ramps as for the soft-bake, I ramp it up to 95C, keep
it for 25min and ramp it back down. Let it rest for 5 min or so to let it go
down to RT before development.

8. Development: 8min in PGMEA. Rinsing in IPA.

Anybody can detect something wrong in my process??

Any suggestion will be greatly appreciated!

Best regards,

Sonia.