Dear Suzanne

Many thanks for your help and sorry for the delayed response.
Im not sure I understand about using proximity flags for WEC. Could
you elaborate a bit on this.

As for using chrome masks I will try this once I get hold of some mask
blanks. However, I am attaching the film mask to a glass plate and am
reasonably sure I am doing it in such a way to ensure it is flat when
making contact with my substrate (and making sure the printed side
contacts the wafer).

One possible source of error may depend on how the WEC is performed
and which points on the mask as used as reference, i.e. if a point is
used that is referncing the glass plate rather than the mask surface.
In this past I have had good results using film masks with a fairly
well collimated UV lamp. The flexibility of the film can actually aid
conformal contact with the SU-8 surface despite the edge bead effect.
However, I need to accurately align mulitple layers, so can't do this
in this case.

N.B We do have an appropriate filter fitted so the spectrum shouldn't
be the cause.

Best regards

Gareth

On Sat, Aug 9, 2008 at 00:46, Suzanne Scullen Ericson
 wrote:
> Gareth,
>
> I recognize a few problems with your process. 200um features in 100um of SU8
> resist can be done very reliably on a contact printer such as the SUSS.
>
> You can expect a profile in thick SU8 that will have a slight hourglass
> shape but one can easily match bottom to top CD with optimization of
> exposure and bakes (soft and PEB).
>
> Using a SUSS tool you will need to filter out the wavelengths below 350nm.
> SU-8 is highly absorbing below 350nm and to expose thick SU8 you need to get
> light to the bottom without overexposing at the surface.  I-line band pass
> filter will do this or a UV400 cut-off filter (if you want to have broadband
> available for other resists used on same tool).
>
> Uniform gap is especially important to prevent proximity effect which is
> what I believe you are seeing.  First try a profilometer of your wafer
> surface.  I know unless you've done multiple coats with edge strips in
> between you will have height variation from edge to center.  The 'bead' can
> be millimeters wide and microns high even though it appears relaxed.
> Typically there will also be a high spot or 'button' in the center.
>
> If doing contact WEC on a non-uniform coating the surface leveling will not
> be accurate and therefore your proximity or contact gap will not be
> repeatable or reliable.  It would be best to use proximity flags for WEC as
> long as they can swing out beyond the edge bead.  Minimizing contact to
> resist surface as ProxWEC does will also reduce resist contamination and
> build up between mask and wafer which can very dramatically affect gap
> accuracy at exposure.
>
> You should also be using a chrome mask or at the very least place your
> acetate film between two glass plates to hold flat. On the SUSS tools the
> WEC leveling, whether done in proximity or contact, will reference the mask
> and the film is too flexible to endure the WEC pressure and you will not get
> an accurate gap.
>
> Lastly, with 200um features you could expose SU8 at proximity exposure gaps
> of 50-200um which minimizes contamination induced proximity effects as long
> as you can keep the WEC head finely tuned.