Oh, as a follow-up.  I just realized you're implying that your UV source isn't
even collimated.  You can get a rough estimate of geometric beam divergence
(neglecting diffraction effects) by simply assuming your light acts like a point
source, so that the emitted light is roughly spherical.  That's a reasonable
engineering esimate for "bad" divergence.  So if your optical throw distance is,
say, 0.5 m and your substrate diameter is 150 mm, then the worst divergence
angle will be roughly arctan(75/500) or about 37 degrees, which might be enough
to cause problems near the edge of the wafer.  On the other hand, if the
substrate is only 20mm wide for the same throw, then the worst divergence angle
will only be about 1.1 degree.

If geometric divergence is a problem, the easiest thing you can do is just move
your light source farther from the substrate.  That will reduce divergence at
the expense of exposure time.

Your other option would be to try adding in a UV lens to improve the collimation
a bit, again assuming the light is kind of like a point source.  You can look
through an optical supply catalog like Edmund for all kinds of UV lenses with
various coatings - something with NR coating in the 350-360 nm range is a good
bet.  If you can find a short focal length lens with sufficient diameter for
your illumination, you can just mount it in the optical path so that your lamp
is near the lens's focal plane.  (You don't want it right at the focal plane,
since at that point you'll be focusing a sharp image of the filament or LED or
whatever onto your substrate - unless you also add a diffuser to the beam path.)

Looks like you can get a near-UV achromatic lens designed for source focusing
for around $100 (though note that I'm not claiming to have evaluated any of
these for suitability for your use!):
        http://www.edmundoptics.com/optics/optical-lenses/achromatic-lenses
/near-uv-nuv-achromatic-lenses/3189

Best,
Mike

On Dec 6, 2013, at 1:25, SHANE GUO  wrote:

> Hi all,
>
> We have a UV light source which is being used for exposing SU-8. It is not a
mask aligner with an advanced UV source but it still works for patterning SU-8
features that are more than 100um wide with an aspect ratio of 0.5. However, I
found that the side walls are not very vertical for narrow(smaller than 30um)
and tall (over 200um) features.
>
> I believe the UV light is the culprit, which is not very directional so the
side walls are tapered, especially for those tall and narrow features.
>
> Does anyone know if there is a way to improve the directionality of the light
source?
>
>
> Best
> _______________________________________________
> Hosted by the MEMS and Nanotechnology Exchange, the country's leading
> provider of MEMS and Nanotechnology design and fabrication services.
> Visit us at http://www.mems-exchange.org
>
> Want to advertise to this community?  See http://www.memsnet.org
>
> To unsubscribe:
> http://mail.mems-exchange.org/mailman/listinfo/mems-talk

_______________________________________________
Hosted by the MEMS and Nanotechnology Exchange, the country's leading
provider of MEMS and Nanotechnology design and fabrication services.
Visit us at http://www.mems-exchange.org

Want to advertise to this community?  See http://www.memsnet.org

To unsubscribe:
http://mail.mems-exchange.org/mailman/listinfo/mems-talk