Re: effective masses in silicon

[Vern Lindberg <vwlsps@RITVAX.ISC.RIT.EDU>]

I don't know what the differences between all the following are, or when to
use them, but

 From Handbook Series on Semiconductor Parameters, vol 1, M. Levinshtein, S.
Rumyantsev, M. Shur, World Scientific, Singapore, 1996

For Silicon (As multiples of true electron mass, m 0)

Effective electron mass longitudinal  0.98   transverse  0.19
Effective hole mass heavy, 0.49    light, 0.16

Effective density of states in conduction band, M = 6 equivalent valleys,
each m c = 0.36, for a total of m cd = 1.18
     (N conduction (cm^-3) = 4.82E15  M (m c)^3/2 T^3/2  or
     (N conduction (cm^-3) = 4.82E15  (m cd)^3/2 T^3/2)

Electrons surfaces of equal energy are ellipsoids.
Effective mass of conductivity = m cc = 0.26

Holes in split-off band, m so = 0.24
Effective mass of density of states at 300 K  m v = 0.81




> Is there a semiconductor expert out there?  I'm wondering what numbers
> to use for the density-of-states effective masses of electrons and
> holes in silicon.  Kittel and Kroemer (chapter 13) give 1.06 and 0.58
> times the bare electron mass for electrons and holes, respectively.
> But my local solid-state physicist says 1.06 is way too high, referring
> me to the book by Sze, which says the d-of-s effective mass is the
> geometric mean of the effective masses in the three directions,
> which it gives as .98, .19, and .19, respectively.  The geometric
> mean would then be only .33.  (For holes there's hardly any
> discrepancy.)  Can anyone tell me why there might be such a
> discrepancy, and which number I should use, say, in computing the
> chemical potential of pure silicon at room temperature?
>
> With many thanks,
>
> Dan Schroeder
> Weber State University
> dschroeder@cc.weber.edu


Dr. Vern Lindberg                       716-475-2546
Department of Physics                   Fax 475-5766
85 Lomb Memorial Drive
Rochester Institute of Technology       Computer Haiku
Rochester, NY 14623
                                         A file that big?
                                      It might be very useful.
                                        But now it is gone.