Assignment 6: Solution: K V N K
Assignment 6: Solution: K V N K
Assignment 6: Solution: K V N K
Assignment 6: Solution
Vk
1. In class, we derived the relationship vk = N
where vk is the volume of a point in
k-space, Vk is the volume of the first Brillouin zone and N is the total number of unit
(2π)3
cells in the solid’s volume. Show that vk is also equal to Ω
where Ω is the volume
of the unit cell in direct space. Hence the number of k points in the first Brillouin
Solution:
2π~
~a∗ = b × ~c
Ω
~b∗ = 2π ~c × ~a (2)
Ω
2π
~c∗ = ~a × ~b
Ω
where, Ω = ~a · ~b × ~c.
Solving for Vk
(2π)3 ~
~a∗ · ~b∗ × ~c∗ = (b × ~c) · ((~c × ~a) × (~a × ~b)) (3)
Ω3
to show
= ~a(~a · ~b × ~c)
(5)
= ~a((~c × ~a).~b)
= Ω~a
(2π)3 ~
Vk = (b × ~c) · (Ω~a)
Ω3
(2π)3 (6)
= Ω
Ω2
(2π)3
=
Ω
Now Vk is equal to value of the first BZ and Ω is equal to volume of the unit cell in
Vk
vk =
N (7)
(2π)3
=
NΩ
(2π)3 (8)
vk =
Vsolid
The volume of a k-point is (2π)3 /Vsolid . The periodic potential does not change the
~ 2 d2 ~2 k 2
~
− + kpx + V (x) ukn (x) = εn (k) − ukn (x). (10)
2m dx2 m 2m
Solution:
~2 d2
− ψkn (x) + V (x)ψkn (x) = εn (k)ψkn (x). (11)
2m dx2
~2 d2 ikx
εn (k)eikx ukn = − 2
(e ukn ) + V (eikx ukn ),
2m dx
~2 d
ikx d
=− e ukn + ike ukn + V (eikx ukn ),
ikx
2m dx dx
~2 2
ikx ikx d ikx d ikx d 2 ikx
εn (k)e ukn =− e ukn + ike ukn + ike ukn − k e ukn
2m dx2 dx dx
+ V (eikx ukn ),
~2 2
ikx d ikx d
=− e ukn + i2ke ukn − k e ukn + V eikx ukn ,
2 ikx
2m dx2 dx
~2 d2 ~2 k 2
~ d
= − ukn + k(−i~ )ukn + ukn + V ukn eikx .
2m dx2 m dx 2m
d
Using definition of momentum operator along x, px = −i~ dx , we obtain
~2 d2 ~2 k 2
~
εn (k)ukn = − ukn + kpx ukn + ukn + V ukn (12)
2m dx2 m 2m
Upon the arranging this given the desired form of the Schrodinger equation for the
~ 2 d2 ~2 k 2
~
− + kpx + V (x) ukn (x) = εn (k) − ukn (x). (13)
2m dx2 m 2m
the Fermi sphere kF relative to the size of the first Brillouin zone (FBZ), ki ? Will
Solution:
The reciprocal lattice of a simple cubic lattice is also cubic with edge length 2π/a.
The minimum distance from the origin (kx , ky , kz ) = 0 to the nearest zone boundary
1/3
3π 2
kF a
=
ki a3 π
1/3 (14)
3
= = 0.9848.
π
Yes, distortion of the spherical Fermi surface occur because the Fermi surface is close
upto 98% to the boundaries. The diagram below shows the relation of the FBZ
a
~a = (1, 1, −1)
2
~b = a (−1, 1, 1) (15)
2
a
~c = (1, −1, 1)
2
2π
a~∗ = (1, 1, 0)
a
2π
b~∗ = (0, 1, 1) (16)
a
2π
c~∗ = (1, 0, 1).
a
Solution:
Take ~a, ~b, ~c, as the direct lattice primitive basis vectors of the face centered cubic
a
~a = (1, 1, −1)
2
~b = a (−1, 1, 1) (17)
2
a
~c = (1, −1, 1)
2
Take ~a∗ , ~b∗ , ~c∗ , as the basis vectors in reciprocal space. To find these basis vectors
~b × ~c
~a∗ = 2π
Ω
~b∗ = 2π ~c × ~a (18)
Ω
~a × ~b
~c∗ = 2π
Ω
Now,
1
1 −1
3 a3
a
~a.(~b × ~c) = −1 1 1 = . (19)
8
4
1 −1 1
î k̂ k̂
2
8π a 2π
~a∗ = 3 . −1 1 1 = (1, 1, 0) (20)
a 4 a
1 −1 1
Similarly,
~b∗ = 2π (0, 1, 1)
a (21)
2π
~c∗ = (1, 0, 1)
a
These vectors define a primitive cell in the reciprocal space, indicating an FCC
reciprocal lattice.
(b) Construct the first Brillouin zone and show that a dodecahedron (with 12 sides)
Solution:
We construct a Wigner Seitz cell. The boundary planes bisect ±~a∗ , ±~b∗ , ±~c∗ at
(kx , ky , kz ) = (±1, ±1, 0), (±1, 0, ±1), (0, ±1, ±1). These are 12 planes.
(c) What is the minimum distance from the origin of the k-space to the nearest
√
2π
zone boundary? Show that the distance is ki = a
.
Solution:
The minimum distance of any one such plane (e.g. (±1, ±1, 0) = (kx , ky , kz ))
√
πp 2 2
π 2 (22)
ki = (±1) + (±1) = .
a a
1
(d) What is the number of electrons per unit volume (n)? Find kF = (3π 2 n) 3 where
Solution:
For body centered cubic lattice, we have two atoms per unit cell (one electron
1
kF = (3π 2 n) 3
13 1 (24)
(6π 2 ) 3
2 2
= 3π 3 =
a a
kF
(e) Find ki
. How far or close to the zone boundaries does the Fermi surface extend?
Solution:
1
kF (6π 2 ) 3 /a
= √
ki π 2/a
1
(6π 2 ) 3
= √
π 2
2 13 (25)
6π 1
= 3
√
π 2
13
6 1
= √ = 0.88
π 2
So, the Fermi surface extents to about 88% of the zone boundaries.
(f) Repeat all the above calculations for a monovalent FCC metal. Which of the
Solution:
a
~a = (0, 1, 1)
2
~b = a (1, 0, 1) (26)
2
a
~c = (1, 1, 0)
2
2π
a~∗ = (−1, 1, 1)
a
2π
b~∗ = (1, −1, 1) (27)
a
2π
c~∗ = (1, 1, −1).
a
The boundary planes bisect ±~a∗ , ±~b∗ , ±~c∗ at (kx , ky , kz ) = (±1, ±1, ±1).
The Minimum distance of any one such plane from the origin is
√
πp π 3 (28)
ki = (±1)2 + (±1)2 + (±1)2 = .
a a
For face centered cubic lattice, we have four monovalent atoms per unit cell
4
n= . (29)
a3
1
kF = (3π 2 n) 3
13 1 (30)
(12π 2 ) 3
2 4
= 3π 3 =
a a
1
kF (12π 2 ) 3 /a
= √
ki π 3/a
1
(12π 2 ) 3
= √
π 3
1 (31)
12π 2 3 1
= √
π3 3
13
12 1
= √ = 0.903
π 3
So, the Fermi surface extents to about 90.3% of the zone boundaries.
Fermi surface for FCC monovalent extends more toward boundaries than BCC