Hi Prakash,

I've tried AP300 (Silicon Resources -
http://www.siliconresources.com/apromoter.html) but have seen limited
improvements to SU-8/substrate adhesion relative to other forms of
chemical adhesion promoter. If you are using HMDS, try to repeat your
process but ensuring the wafer is dehydration-baked at 150 deg C for
20-30 mins after HMDS application and prior to SU-8 spinning. If there
is any moisture on the wafer pre- or post- HMDS application, this will
hydrolyse the trimethyl substituted groups from the HMDS, reversing
the process of silylation. In fact, removal of the HMDS substrate
monolayer is achieved by rinsing in water. Applying HMDS in a humid
environment worse for adhesion than not using any form of (chemical)
adhesion promoter. Blowing nitrogen over your wafer surface (post
HMDS) just before applying SU-8 would also be beneficial.

You mentioned that you need to achieve robust adhesion between your
SU-8 structures and the substrate. If the above does not yield the
improvements you seek, you could try to adopt a strategy of physically
attaching the SU-8 to the substrate by using mechanical interlocking.
If this is appropriate for your application, have a look at the
following paper on this technique:
http://www.iop.org/EJ/abstract/0960-1317/15/11/012
(J. Micromech. Microeng. 15 2074-2082, 2005).

Best of luck and let us know how things work out.

Regards,

Michael


> ---------- Forwarded message ----------
> From: "Prakash Mukre" 
> To: mems-talk@memsnet.org
> Date: Wed, 20 Dec 2006 17:28:10 -0500
> Subject: [mems-talk] SU-8 Adhesion with SiO2
> Hi All,
>
> I am trying to deposit SU-8 over thermal oxide. SU-8 is part of the final
> device I am fabricating, so I want it to adhere to the substrate strongly.
> But with my current recipe SU-8 delaminates after development. Following are
> the some process specs I am following.
>
> SU-8 2100 Thickness: 200-250um
> Softbake: 15min @65C,  100min @95C.
> UV Dose: I have tried 400mJ/cm2 to 1000mJ/cm2
> Hard Bake: 5min @65C, 20min @ 95C.
> Development time: 10-20min.
>
> I have tried using primers like HMDS, P20. Also I have tried using Omicoat.
> What could be the possible reason for this? or SU-8 does not adhere to
> thermal oxide at all?
>
> Thanks,
> Prakash