2 Energy Bands

Download as pdf or txt
Download as pdf or txt
You are on page 1of 113

Energy bands in solids

Some pictures are taken from Ashcroft and Mermin from Kittel from Mizutani and from several sources on the web.
we are starting to remind…

p = mv =  ⋅ k
1 2  2k 2
E = mv =
2 2m

Some pictures are taken from Ashcroft and Mermin from Kittel from Mizutani and from several sources on the web.
• The letters s, p, d, f, . . ., are often used to signify the states with , l =0, 1,
2, 3, . . ., each preceded by the principal

• n=1, l = 0, m =0, s= 1/2  (1s)2

• n=2, l = 0, m =0, s= 1/2  (2s)2

• n=2, l = 1, m = 0, 1, s= 1/2  (2p)6

• n=3, l = 0, m =0, s= 1/2  (3s)2

• etc.

• Na 11 electrons with four different orbital energy levels 1s, 2s, 2p and 3s.

• (1s)2(2s)2(2p)6(3s)1
from Alloul, Mizutani
•the energy levels for the 1023 3s electrons are split into quasi-continuously
spaced energies when the interatomic distance is reduced to a few-tenths nm.
•The quasi-continuously spaced energy levels thus formed are called an
energy band.
• Apart from metallic bonding, there are three other bonding styles:
ionic bonding, covalent bonding and van der Waals bonding (inert
gases).

• Typical examples of ionic bonding are the crystals NaCl and K Cl.
They are made up of positive and negative ions, which are
alternately arranged at the lattice points of two interpenetrating
simple cubic lattices.

• The electron configurations for both K+ and Cl- ions in a KCl crystal
are equally given as (1s)2(2s)2(2p)6(3s)2(3p)6.

See Mizutani for other examples!


Free electron under the periodic boundary condition

wigner-Seitz cell in the reciprocal space


from Kittel

from Alloul
L

• 1D
x

3D
• Free electron in a box
from Alloul
Construction of the Fermi sphere
Pauli exclusion principle
Fermi level the last occupied level = chemical potential
n = 0,±1,±2...
Mizutani
 2π 
kx =  n x 4π 3
 L  kF
N0 3
 2π 
ky =  n y
=
 L  2  2π 3
 2π 
 
kz =  n z  L 
 L 

The value of the Fermi level at absolute zero


(−273.15 C) is called the Fermi energy and
is a constant for each solid.
density of states dn = N ( E )dE
• a mole of sodium metal weighs 22.98 g with

• its density 0.97 g/cm3. Since it is a monovalent metal,


N0 (# electrons) is equal to the Avogadro number NA.
• NA=6.0231023 and V=23.6931021 nm3 results in a Fermi
radius kF=9.1 nm-1 for sodium metal.

N= number of valence electrons per atom (e/a)

bcc strucure, a= 0.422 nm, the volume per atom


Na  monovalent: 1 electron/atom-  EF= 3.2 eV

Fermi temperature: EF= KBTF if EF= 5eV  TF= 60000 K


Mizutani
Mizutani
1D free electrons
ψ( x ) = ψ( x + L)
L = na ψ ( x) ∝ e ika
E (k ) = E (k + Gi ) n = ... ± 2,±1,0
-3 -2 - 1 0 1 2 3

2π  = ... ± 2,±1,0
Gl = 
a
2 2π 2
E (k ) = (k + )
2m a empty lattice
model

l : 1 0 -1
(NEF model)
T≠0K
Kittel
NFE approxination
(perturbation theory)

Kittel
Mizutani
Fermi surface–Brillouin zone interaction

Mizutani
The Fermi surface begins to
be distorted from a sphere before making
contacts with the Brillouin zone
planes.

Mizutani
Mizutani
see Mizutani for detalis
Effective mass
 2k 2
E=
2m
dE  2 k  p
= ⇒ = ∇k E
dk m m

F = m ⋅a
*
http://people.seas.harvard.edu/~jones/ap216/lectures/ls_2/ls2_u7/sse_tut_1/solid1.html
∂2E
< 0 → m* < 0
∂k 2

∂2E
> 0 → m *
>0
∂k 2
http://people.seas.harvard.edu/~jones/ap216/lectures/ls_2/ls2_u7/sse_tut_1/solid1.html
Interpretation
• The electron is subject to internal forces from the
lattice (ions and core electrons) AND external forces
such as electric fields
• In a crystal lattice, the net force may be opposite the
external force, however:
Fext =-qE

- Fint =-dEp/dx

Ep(x)

+ + + + +
Interpretation
• electron acceleration is not equal to Fext/me, but rather…
• a = (Fext + Fint)/me = Fext/m*
• The dispersion relation E(K) compensates for the internal forces due to
the crystal and allows us to use classical concepts for the electron as
long as its mass is taken as m*

Fext =-qE

- Fint =-dEp/dx

Ep(x)

+ + + + +
Examples of Energy Bands in Solids
What we can find out:

see Mizutani for more details


Metal is a solid with a Fermi surface!
from the UvA-VU Master Course: Advanced Solid State Physics by Anne de Visser (University of Amsterdam),

BZ is half-filled

#electrons in the reciprocal unit cell

(2 atoms/u.c.)
Na metal is known to be almost a free-electron-like metal.

Mizutani

empty
lattice
calculation
Van Hove
anomaly
from the UvA-VU Master Course: Advanced Solid State Physics by Anne de Visser (University of Amsterdam),
Mizutani
In the case of copper, the bands near the Fermi level are derived
from the 4s and 3d atomic levels. The so-called 4s and 3d bands
accommodate a total of 12 electrons, while the number of available
electrons is 11. Therefore the Fermi level must cross these bands.
Consequently copper is metallic.
As another unique feature of the electronic structure of noble metals, the
parabolic E–k relation is intervened by the less dispersive E–k curves due to
3d electrons centered at about a few eV below the Fermi level.
This means that the 4s and 4p free-electron-like states are hybridized with the
3d states to form a composite band. In spite of strong hybridization, ten 3d
electrons per atom are distributed in a rather narrow energy range.

The Fermi surface of a noble metal should occupy only 50% by volume of the
first Brillouin zone of the fcc lattice.
Polyvalent Metals see Mizutani

The two outermost 4s electrons per atom


contribute to the formation of the
valence band of Ca metal.

•Since the number of electrons per


atom, e/a, is equal to 2.0, it is just high
enough to fill the first Brillouin zone.
•The simplest example of a polyvalent metal is aluminum with 3 electrons/atom and
having a 3s23p electronic configuration for the valence electrons.
•Because of the partial filling of the 3s23p6 bands, aluminum is a metal. Aluminum
crystallizes in the FCC structure
The energy bands for aluminum are very free electron-like.

from Mizutani
semimetal
•electronic band
structure intermediate
between metals and
semiconductors
Mizutani
•small Fermi surfaces
of both electrons and
holes

•a very small overlap


between the bottom of
the conduction band
and the top of the
valence band
It can be seen that the two bands slightly overlap each other so that electrons
and holes coexist at the Fermi level.
The elements C (diamond), Si, Ge and Sn( α) belong to the group IVB elements
in the periodic table and all of them are covalently bonded.
•Two interpenetrating fcc lattices and, hence, the structure is fcc but with
two atoms in each unit cell.
•In Si, two 3s and two 3p electrons per atom are hybridized
to form directionally-bonded states and the Brillouin zone is completely
filled with eight electrons per unit cell. Hence, no Fermi surface exists.

Kittel
Energy gaps of typical semiconductors are as follows:

5.3 eV for C (diamond), 1.1 eV for Si, 0.7 eV for Ge,

2.8 eV for SiC, 4.6 eV for BN and 1.4 eV for GaAs.


Is there a Fermi energy of intrinsic
Semiconductors?
•If ε F is defined as the energy separating the
highest occupied from the lowest unoccupied
level, then it is not uniquely specified in a solid
with an energy gap, since any energy in the gap
meets this test.
•People nevertheless speak of “the Fermi energy”
on an intrinsic semiconductor. What they mean is
the chemical potential, which is well defined at any
non-zero temperature. As T → 0, the chemical
potential of a solid with an energy gap approaches
the energy of the middle of the gap and one
sometimes finds it asserted that this is the “Fermi
energy”. With either the correct of colloquial
definition, ε n (k ) = ε F does not have a solution in a
solid with a gap, which therefore has no Fermi
surface!
http://hyperphysics.phy-astr.gsu.edu/hbase/Solids/Fermi.html
extrinsic semiconductors

Si semiconducting properties depend significantly on whether pentavalent or


trivalent impurities are added.

Pentavalent atoms like P, As and Sb from group VB of the periodic table will
have one more valence electron than is required for the covalent bonding.

Donor

As5+
Mizutani
Acceptor

B3+

•trivalent atoms like B, Al, Ga and In from group IIIB in the periodic
table possess one electron less than the number required for covalent
bonding.
Hence, an electron from one of the adjacent Si atoms will enter the
orbit about the trivalent impurity atom. This causes the trivalent atom to be
negatively charged and the host Si atoms to lose an electron in this process.
Mizutani
Mizutani
Culity, Graham
Introduction to
magnetic materials
Wiley 2009.
Pauli Paramagnetism
W. Pauli
Nobel Price 1945

Spin - Magnetization
for small fields B (T=0)

 
EB = −µ B ⋅ B

n = g ( E F ) ⋅ ∆E

The small size of the paramagnetic ∆E = 2 µ B ⋅ B


susceptibility of most metals was a
puzzle until Pauli pointed out that is
M = µ B (n↑ − n↓ ) = g (E F )µ B B
2
was a consequence of the fact that
electrons obey Fermi Dirac rather
than classical statistics.
Pauli paramagnetism is a weak effect compared to paramagnetism in
insulators (in insulators one electron at each ion contributes, in metals
only the electrons at the Fermi level contribute).

Magnetic Spin - Susceptibility

M µ0 M
χP = = = µ 0µ B g (EF )
2

H B
3N 3N
g ( EF ) = =
2 EF 2k BTF
M µ0 M 3µ 0µ B N
2
χP = = =
H B 2 EF

(Pauli Paramagnetism)
Culity, Graham
Introduction to
magnetic materials
Wiley 2009.
- Copper is diamagnetic, since (a) its Fermi energy is close to the band
edge [i.e.g(EF) is small] and thus χp is relatively small, and (b) copper has
ten d-electrons (i.e. Z ≈ 10) orbiting on a fairly large radius, which
results in a large χdia.
- Also in bivalent metals (e.g. Be) g(EF) is small → they are diamagnetic
- In semiconductors, the valence bands are filled, i.e. g(EF) = 0
→ semiconductors have no paramagnetic susceptibility and are
diamagnetic
Direct Exchange between delocalized Electrons

Itinerant ferromagnetism

Spontaneously Split bands: e.g. Fe M = 2.2μB/f.u. is non integer


.... this is strong evidence for band ferromagnetism

Mean field Model: all spins feel the same exchange field λM
produced by all their neighbors, this exchange field can
magnetize the electron gas spontaneously via the Pauli
Paramagnetism, if λ and χP are large enough.

Quantitative estimation: what is the condition that the system


as a whole can save energy by becoming ferromagnetic ?
M = µ B (n↑ − n↓ )

moving De(EF)δE/2 electrons from


spin down to spin up band

kinetic energy change: ∆EK .E . = De ( EF )(δE )2


1
2
exchange energy change:
M
∆E P. E . = − ∫ µ 0 (λM ')dM ' = − µ 0 λM 2 = − µ 0 λµ B2 (n↑ − n↓ ) 2
1 1
0
2 2
see Mizutani
total energy change:

U ≡ µ 0 µ B2 λ
n↑ − n↓ = De ( EF )δE

⇒ ∆E = ∆EK . E . + ∆EP. E . = De ( EF )(δE ) (1 − UDe ( EF ) )


1 2

2
there is an energy gain by spontaneous magnetization, if
Stoner Criterion
UDe ( EF ) ≥ 1
... Coulomb Effects must Edmund C.
be strong and density of Stoner
states at the Fermi energy (1899-1968)
must be large in order to
get sponatneous
ferromagnetism in
metals.
from Mizutani
kittel
kittel
Figure shows the calculated valence band structure for the spin-up and
spin-down electrons in ferromagnetic Ni metal. It can be seen that the spin-up
band is shifted to higher binding energies relative to the spin-down band
because of the exchange energy. The Fermi level is determined by fitting a
total of 10 electrons per atom into the hybridized 3d and 4s bands. It is clear
that the 3d spin-up band is fully filled by electrons while the Fermi level falls in
the middle of the 3d spin-down band, leaving holes in this sub-band. The
number of holes per atom is calculated to be ~ 0.6. This is obviously
proportional to the magnetization of pure Ni metal and is consistent with the
observed non-integer value
(Cu:[Ar] 3d10 4s1)
Ni1-xCux Ni1-xZnx (Zn: [Ar] 3d10 4s2)
average electron concentration

M=0, for x = 0.6 X= 0.3


the Ni 3d spin-down band possessing 0.6 holes per atom Hence, the Ni 3d spin-down band will be filled at x= 0.3
will be filled when x=0.6

Mizutani
Mizutani
from
Mizutani
see Mizutani

You might also like