A colleague of mine ran into a similar issue when sputtering NiTi.  It turns
out he needed a NiTi target and a separate Ti target to achieve the desired
stoichiometry in the sample film, either because some elemental Ti was
reacting with residual oxygen in the chamber or Ti had a lower sputtering
efficiency than Ni.  If your deposition tool can accommodate another target,
a similar approach may work well in your case - you could vary the power to
the additional target while keeping the IO target power constant to achieve
your desired stoichiometry.

Good luck,

Brian C. Stahl
Graduate Student Researcher
UCSB Materials Research Laboratory
brian.stahl@gmail.com / bstahl@mrl.ucsb.edu
Cell: (805) 748-5839
Office: MRL 3117A


On Wed, Jan 26, 2011 at 2:37 PM, Ruiz, Marcos Daniel (SENCOE) <
Dan.Ruiz@honeywell.com> wrote:

> Thanks for the clear replies to my questions.  Based on your answers, it
> is safe to say that you are doing a good job of eliminating oxygen and
> moisture from the chamber before deposition.
>
> The only other source of oxygen is your "IO" sputter target.  Although I
> have no experience sputtering this alloy, I suspect that the oxygen in
> the "IO" target is sputtered at a greater rate (maybe 4X faster) than
> the other element.  Thus, the resulting film on your samples is oxygen
> rich.
>
> This supports the advice you've already received about getting a new
> target.  I think the other piece of advice you got about wafer
> temperature probably serves to keep free oxygen out of the solid phase.
>
> Your sputter target vendor would probably be a good resource here.  They
> may be able to tell you what target you need to buy (percent oxygen) to
> achieve the oxygen level you require in your samples.
>
> Dan Ruiz