Hmm.  You run into the material saturation limits here, which become
relevant.  Given that boron is an interstital impurity, the
concentration can by necessity be higher than that of phosphorus,
which is substitutional (in Si only).  In SiGe, this becomes a bit
more complicated, depending on the alloy fraction of Si-x-Ge-(1-x)
you're using.  The best numbers (based on papers that I have seen) to
use for estimates here are to use the same ones you have get for
bulk-Si, they will get you within a ballpark (~75%) of the practical
measurements.

For bulk Si, max concentrations of phosphorus achievable even with
in-situ doping during growth (epi or poly) I believe does not exceed
~10^20-10^21, that is, approximately 1e-3 ohms/sq.  For boron, I
believe you can get it down a bit more, 2e-4 ohm/sq is the best I've
seen.  I work with sheet resistivity as that's easier, since bulk
wafers are measured in it.

Contact me for references, I can cite some to you, there's a good
discussion of this is in the 2nd edition of Microelectronics
Fabrication (forget the author right now, it's the green-colored
edition, not the orange-colored first one).

-Eric

> ---------- Forwarded message ----------
> From: "Wang Ziyang" 
> Does any one know the possible lowest electronic resistivity of boron doped
and phosphorus doped SiGe after annealing? And I also found the non-uniformity
of electronic resistivity on the deposited SiGe layer by LPCVD, even after
annealing. I am wondering how I can do to render the distribution of electronic
resistivity more uniform.