> I need to etch both shallow and deep features using Deep Reactive Ion
Etching.
> The first etch step is fine. However, during the second etch step, deep
> trenches develop at the edges of the features produced during the first
etch.
> I am not sure what this is due to but I think the photoresist (Shipley
9260)
> spin-coated after the first DRIE etch coats the edges only very thinly and
is
> removed during the second etch step.
>
> Any suggestions on how to overcome this will be greatly appreciated.
Thanks.
>
> Regards,
> Melissa

Spinning on photoresist over existing trenches is always tough.
There is a trick we have used successfully to get around this:

Start by depositing oxide on the wafer.
It should be thick enough to endure the second etch (figure on a selectivity
of Si:SiO2 of better than 100:1).
Pattern the oxide with the *second* mask and etch the oxide.
Strip that resist.

Now, pattern photoresist with the *first* mask (figure on a selectivity of
Si:PR of better than 50:1).
Do the first DRIE.
Strip the resist.

Your second mask is already on the wafer in the form of oxide.
Do the second DRIE.
Strip the oxide.

    --Kirt Williams, Ph.D.    consultant