Hi Michael,

300nm of thermal SiO2 is reported to be a good diffusion barrier and to prevent
silicide formation even at temperatures as high as 1000°C. Actually, Pt films
(evaporated & sputtered) adhere on thermal SiO2 without any adhesion layer and
it survives also high-temperature annealing. Metallic adhasion layers such as Ti
or Cr rather cause problem at high temperatures due to diffusion of adhesion
layer itself which results in degraded adhesion and drift of electric resistance
of Pt. I have read that metal oxide layers such as Al2O3 should work as a good
adhesion layer & diffusion barrier in one.

See e.g.

"High temperature materials for thin-film thermocouples on silicon wafers", K.G.
Kreider, G.Gillen, Thin Solid Films 376 (2000) 32-37

"Adhesion layer for platinum based sensors", Robert G et al., US patent no
4952904


Regards,

Natsuki


-----Ursprüngliche Nachricht-----
Von: mems-talk-bounces@memsnet.org [mailto:mems-talk-bounces@memsnet.org] Im
Auftrag von Michael D Martin
Gesendet: Mittwoch, 5. März 2008 15:58
An: General MEMS discussion
Betreff: Re: [mems-talk] Anneal recipe for Pt thin film on Si wafer

I've had this problem in the past, i.e. formation of Pt-Si alloys during an
anneal with (I think) a Cr adhesion layer.  Could anyone recommend a material
stack Pt/ adhesion layer/ diffusion barrier?

Thanks,
   Michael
   U. of Louisville

>>> Oliver Horn  3/3/2008 4:38 AM >>>
Hello Natsuki,

you will get PtSi or PtSi2 if you heat up Pt on Si without diffusion
barrier, even at 260°C.  The adhesion is only a bit improved.

Oliver