Anibar,

Thanks for the details.

The first observation is that your bake is not long enough.
Typically, a film of 12µm thickness requires 5 minutes bake on a hot
plate at 110-120˚C.  If you used a convection oven, then not very much
of the solvent was removed.  I also think that 120˚C in a convection
oven is too hot.  Try 90˚-100˚C for at least 45 minutes, even an
hour.  This could explain the rounded edges of the pattern.  The
exposure dose is probably reasonable (I am assuming that you are using
a contact printing tool like a Suss).  For a film that thick, you
should require a development between 2-4 minutes.  The fact that you
need only 80 seconds development is consistent with high level of
solvent remaining in the resist from insufficient bake.  Try that and
post your results on the forum.

Brad Cantos
brad.cantos@holage.com
http://holage.com


On Jul 17, 2009, at 2:06 PM, ANIRBAN SARKAR wrote:

> Hey,
>
> Let me describe the process conditions and I hope that will discuss
> the
> problems in more detail.
>
> The substrate is Silicon nitride deposited on Si by LPCVD.
>
> Spin coating: AZP 4620 for 2000rpm for 30 secs The  thickness of the
> resist
> came out to be 12um.
>
> Pre baking for 5 mins at 120'c in convection oven
>
> Exposure: 400mJ/cm2
>
> Development time: 80 secs in AZ400k:Di water= 1:4
>
> The patterns in the mask ranges from 10 microns upto 100 microns
> maximum.
>
> When I  developed according to the above conditions, I saw that the
> patterns
> did not follow the mask patterns at all.Most of the square patterns
> came out
> closely to circle as well ovals mostly.Also some of the areas were
> very
> poorly developed.
>
> The thick PR mold will be used for electroplating of Copper.
> Also in the other substrate(GLASS)  the PR  will be used for lift
> off after
> a thick metal deposition (evaporation) of around 5 um.
>
> Any suggestions will be highly appreciated.
>
> Best wishes
>
> Anirban