Molybdenum in the SNF

Joey Doll jcdoll at stanford.edu
Wed Nov 5 18:13:55 PST 2008


Hi SpecMat - What's the policy on molybdenum processing in the fab? It's
listed on the materials page as a case-by-case basis, however the text
refers to manganese. I'm interested in using it as an electrode for my
aluminum nitride thin films for two reasons. First, Mo yields the best
growth structure and piezoelectric performance of AlN (followed by Pt and
Ti) and more importantly, it has good etch selectivity. Etching Mo in
fluorine and aluminum nitride in chlorine is a common process. In fact, I
was speaking with an engineer from Avago today and they process Mo in their
production CMOS fab. Avago produces FBAR resonators in the ballpark of
millions per year, and use Mo/AlN/Mo stacks. I'm currently using Ti, but Mo
is a better alternative both in terms of performance and processing
capability. I'm currently etching the AlN with TMAH with the Ti as a hard
mask, which is a workable solution but the performance of AlN can be
improved by about a factor of 2 by going to Mo.

After depositing and patterning the Mo and AlN, I would need to go into
STSEtch1 for my frontside and backside cantilever releases. All metal would
be covered with photoresist, and the etch would be stopped on the buried
oxide of an SOI. The Mo patterning could be done in drytek2 and the AlN
patterning in p5000 potentially. I'm having the AlN (and potentially the Mo)
deposited by a vendor in San Jose.

Thanks!
Joey
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://snf.stanford.edu/pipermail/specmat/attachments/20081105/7878a290/attachment.html>


More information about the specmat mailing list