This is challenging since the resist are designed to have sharp
profiles, however things can be done. I am assuming that you are talking
about positive resist based on a novolak resin.

These are some methods.

1. Bake the resist at a temperature where it flows and the edge melts.
This tends to give an edge that is an arc, but you don't have the sharp
edge. Since resists batches vary in the chemistry of the resin,
specifically the ratio of o-, p-, and m-creosote isomers which go into
the novolak polymer, the flow behavior can vary, so the temperature
needs to be high enough that you get flow from batch to batch.

This is crude but it will work where you just need some sloping.

2. For a more sloped or really tapered resist profile you will need to
defeat the mechanisms of the designers of the resist, who designed the
resist to have a sharp resist edge, and high resolution. That is they
designed the resist to have a non-linear response to exposure.

One mechanism is the formulations to have a sort of crust in the very
surface of the resist. To defeat this you might have to try a very light
flood exposure.

The resolution will need to be defeated by running the process in ways
contrary to high resolution. That is have the develop time too long, or
maybe way too short. Have the concentration of the developer too high
which lowers selectivity.

Finally, I would expose out of focus.

You will need to do some considerable process optimization of flood
exposure, develop conditions, and out-of-focus exposure.

Process control will be adversely effected, since the high contrast
mechanisms of resist contribute to process control.

3. Multiple masks. Have differently sized masks and do partia exposures.
The problem is that the alignment of the masks need to be close to dead
on. I don't think this will work well.

4. Multiple exposures: If you need to taper it along one axis or maybe
two axis, but not in every direction, you can align the substrate, and
then do partial exposures and move the substrate slight amounts.

Best of luck,

Ed



-----Original Message-----
From: mems-talk-bounces@memsnet.org
[mailto:mems-talk-bounces@memsnet.org] On Behalf Of QiuWeibin
Sent: Friday, October 30, 2009 4:16 PM
To: mems-talk@memsnet.org
Subject: [mems-talk] vertical taper Photoresist profile



 Dear All,

    We are trying to get the vertical taper PR profile with 8mm size and
0-1.5um thickness. Can anyone give us some suggestion?  We want to avoid
a grey scale mask, since they are so costly.

Thanks a lot

Qiu