I would use a Novolak positive photoresist, since they are easy to use,
and are resistance to plasma compared to many negative resists.

You need a carbon rich plasma mixture to etch SiO2 to balance the
stoichometry. So SF6 would be out. Unfortunately CF4 is still rather
flourine rich and carbon poor relative to what you need, people usually
use C2F6 or CHF3 to make the plasma more "reducing".

However, I am not sure that you can't somehow make it work. It might be
a slow etch and you might have to watch closely your resist loss. If you
are just doing contacts, your resist might provide enough reducing
component, but I think you could end up with a very unreliable and
marginal process.

You need to think about what conditions of power, pump speed (time
constant of the gases in the chamber), pressure, temperature, would tend
to favor the fragmentation of CF4 to carbon compounds that deposit on
the SiO2 and react with the substrate SiO2 extra O.

There well may be no process space in which the process works. However,
I am thinking there might be a condition where etch will occur slowly in
which the SiO2 is oxygen rich due to the Si being removed by the plasma
flourine and so the flourine part of the etch is limited and what carbon
arrives will react with the surplus oxygen and balancing the reaction
stoichometrically after all.

I would think you would need to have low pressure high power to fragment
the CF4 to create some C, CF, CF2 fragments and avoid flourine
recombination. But I am just speculating. I would suggest doing a
fractional factorial DOE which you are going to have to do anyways
regardless of our advice.

Also, there are probably much more informed plasma experts than me.

Ed Sebesta
EVG620 User Group Newsletter Editor.



-----Original Message-----
From: mems-talk-bounces@memsnet.org
[mailto:mems-talk-bounces@memsnet.org] On Behalf Of alim polat
Sent: Tuesday, February 05, 2008 3:17 AM
To: MEMS-talk@memsnet.org
Subject: [mems-talk] Dry etch to get patterned SiO2 hard mask

I have 200-400 nm thick SiO2 thin film grown using PECVD-RIE machine.
Now I need to etch the SiO2 thin film at the same machine to get
patterned SiO2 hard mask. I only have CF4 and SF6 gases in the machine.
Could I get some advices on dry etching SiO2 to get patterned hard mask.
Also I need what kind pf PR I should use on top of SiO2.