You shouldn't get any polymer formation etching with SF6.  I have etched
Nb with no observable residue using pure SF6 RIE.  However my experience
is limited to etching Nb on SiO2 or fused silica.  SF6 tends to etch Si
very fast.  Perhaps you're getting some sort of reaction between the Nb
and Si etch products.  A solution might be to oxidize the Si before
depositing and etching the Nb, then if necessary remove the oxide after
etching the Nb.

Roger Shile

-----Original Message-----

Hi,
I am trying to etch Nb with SF6. I noticed that there is lot of polymer
formation on the Si Substrate surrounding my desired Nb feature. I was
thinking of using SF6+O2 to avoid polymer formation. Adding O2 to SF6
means photoresist will be etched at a greater rate. It also means that
polymer which is most likely acting as a etch stopper for Si will in its
absence cause Si to etch faster, which is even more detrimental because
it will end up in making a big trench next to Nb feature.(I already
noticed the Si etch of 500nm in 4 min)
Any alternative that anybody knows to avoid Si etch?
Thanks!
~T