Dear Yilei,

I have several questions about your case. Which method did you use to tell
the your SiO2 mask is still left on wafer after HF etching? And how thick
was your SiO2 mask before RIE? I assume you know that CF4/O2 plasma etches
SiO2 as well, although slower than etching Si with your gas ratio. CFx
polymer layer is not significantly formed on SiO2 surface because SiO2 is
a source of O atoms to gasify CFx polymer by combustion reactions. HF
should have a high chance of penetrating this very thin layer and remove
both SiO2 and the polymer layer together (something like lift-off). If it
is a purely CFx polymer layer, O2 plasma should remove it by combustion
reactions. By the way, how did you form your SiO2 mask? Was it from a
photoresist mask? Could your SiO2 have a scum resist layer that you didn't
strip properly? Anyway, you can still use O2 plasma to remove this scum
layer.

What is your chamber and electrode material? Metal contaminants may turn
your polymer to organometallic residues. In that case, O2 plasma or
piranha etch won't remove the residues in a great amount.  You will need
special RIE residue remover solvent to remove them. Please contact me
directly if you have further questions.

Yours sincerely,

Isaac Chan, Ph.D.
University of Waterloo


On Wed, 25 May 2005, yilei zhang wrote:

> Yes, we do have oxygen in the receipt and I also found it was said that
> oxygen will avoid the formation of the polymer layer. I am not sure whether
> the polymer layer is the reason that sio2 cannot be etched with HF after
> RIE. Maybe it is fluorinated silicon? And if it is caused by "sputtered"
> polymer, it should be removed by pure o2 etching, and I will do it and let
> you know the result.