3.7 The kp method
Using the kp method [Bir74,Yu03] one can obtain analytical expressions
for the band dispersion and the effective masses. It allows the extrapolation
of the band structure over the entire Brillouin zone from the energy gaps and
optical matrix elements at the zone center. While the kp theory has been
frequently used to model the valence band of semiconductors, we will
additionally apply it to model the impact of strain on the conduction band
minimum.
The kp method can be derived from the one-electron Schrödinger equation
|
(3.58) |
Here
denotes the one-electron Hamilton operator and
the periodic lattice potential. The wavefunction of an electron in an
eigenstate labeled and its energy are denoted by
and
, respectively. In a periodic potential Bloch's theorem
applies, and the solutions of (3.58) can be expressed as
|
(3.59) |
where is the band index,
a wave vector, and
has
the periodicity of the lattice. Assuming that the potential
is
local3.1, one can
substitute
into (3.58) to obtain an
equation for
|
(3.60) |
Considering any fixed wavevector
, the above equation
yields a complete set of eigenfunctions
, which completely span
the space of lattice periodic functions in the real space. Hence, the
wavefunction
at
can be expanded in terms of
|
(3.61) |
Once,
and
are known, the functions
and the eigenenergies
at any
vector
in the vicinity of
can be
obtained by
treating the term
in (3.60) as a perturbation. Either degenerate or nondegenerate
perturbation theory has to be used. The method has been first applied by
Seitz [Seitz35] and was later extended to study the band structure of
semiconductors [Luttinger55,Kane56,Cardona66].
This method for calculating the band structure is known as the kp method. It works
best for small
and can be applied to calculate the band
structure near any given point
provided that the matrix elements of
between the wavefunctions (or the wavefunctions themselves) and the
energies at
are known. When using a sufficiently large number of
to approximate a complete set of basis functions, the band
structure over the entire first Brillouin zone can be calculated by
diagonalizing (3.60) numerically [Cardona66].
In the following, a nondegenerate kp theory will be used to derive the band
dispersion and the effective masses for the nondegenerate conduction band of
Si. To analyze the effect of shear strain on the
lowest conduction band , a degenerate kp method is adopted in
Section 3.7.2, since the conduction band is expanded
around the symmetry point (zone boundary) where it touches the
conduction band.
Subsections
E. Ungersboeck: Advanced Modelling Aspects of Modern Strained CMOS Technology