We've also seen the ashy residue when etching with O2 plasma (barrel
etcher, 60W forward, 5W reflected, 2x10-1 mbar O2, crossed plates). I
put it down to non-volatile antimony oxide residue from the
photoinitiator in the SU-8. I didn't analyse the residue for heavy
metals, but it seems plausible.

If I do a "back of the envelope" calculation, I get something like this:
1. Assume a mixture of 5wt% (C6H5)3SSbF6 + 95wt% SU-8. The photoacid
generator (PAG) is more complex than this, but it's close enough for
this estimate.
2. Assume the reaction gives stoichiometric antimony(III) oxide:
    2(C6H5)3SSbF6 + xs O2 --> Sb2O3
    2 x 499 g/mol         --> 291.5g/mol
    That is, for every gram of PAG one starts with, there'd be 0.292g
solid oxide left.
3. So for 1g SU-8 + PAG, there'd be 0.015g Sb2O3
4. Antimony(III) oxide is around 4.3 times more dense than SU-8
(5.2/1.2), so for a given volume (or thickness) of SU-8, I'd expect to
be left with a residue with 0.34% of the original volume or thickness.
The Sb2O3 would be powdery, however.

Chris


On Thursday, March 13, 2003, at 06:15  am, Greg Reimann wrote:

> Hi,
>
> May I add a question? I've tried to ash SU-8 with O2 plasma in
> ICP/RIE system. But there is a film of residue on wafer. Have you seen
> such problem?
>
> Isaac
--
Christopher F. Blanford
Inorganic Chemistry Laboratory, South Parks Road, Oxford, OX1 3QR, UK
Phone: +44 (0)1865 282603; Fax: +44 (0)1865 272690
PGP keyID: 8D830BC9  http://pgp.mit.edu/