Solid State Physics-II

Chapter No. 2

Energy bands

Sibt ul Hassan
Horizon college chakwal
1
Free electron gas Model:

A number of important physical properties of metals can be explained in terms of

free electron model. According to this model, the most weakly bound electrons of
constituent atoms move about freely through the volume of metal. Thus the valence
electrons of the metal become the conduction electrons. The basic assumptions of the
model are given below:
1- The metal crystals consist of positively charged ion cores called kernels and the
completely free valence electrons. In the bulk of metal, positive ion cores surround
the electron symmetrically and vice versa. The net Coulomb force experienced by
electron in the interior of metal is zero, so the electrons have freedom to move
within the interior of metal having uniform potential.
2- The mutual repulsion between the electrons is zero.
3- The constant potential inside the metal is zero, therefore energy associated with
electrons is only kinetic energy.
4- The free electrons collide with atoms and at any temperature their velocities can be
obtained by using the statistics.
5- The potential energy is neglected, hence the conduction electrons move about freely
within the metal without any collision similar to the molecules of an ideal gas.
Interpretation of metallic properties on basis of free
electron model:-
The interpretation of metallic properties in terms of free electron gas model was
developed by Drude and Lorentz on basis of classical theory applying Maxwell Boltzmann
statistics to the electron gas. This theory explains the following properties of metals:
(i)-Ohm’s Law:-
The current i is proportional to potential difference V i.e.
V
i  V or  R  constant 
i
(ii)-Electrical conductivity:-
For pure metals, electrical conductivity  is inversely proportional to absolute
temperature i.e.
I

T
For impure metals, electrical conductivity is inversely proportional to square root of
absolute temperature.
I

T
2
(iv)-Variation of electrical resistivity with pressure:-
For most of metals the resistivity  is inversely proportional to pressure P.
1
P

(v)-Wiedemann-Franz law:-
The ratio of thermal conductivity K to electrical conductivity  is proportional to
absolute temperature.
K
T

This is known as Wiedemann-Franz law.
Failures of free electron gas model:-
The classical theory could not explain the following properties of metals.
i- Specific heat
ii- Paramagnetism
iii- Long mean path at low pressure
iv- Linear variation of resistivity with temperature.
In 1928, Somerfield gave quantum treatment to free electrons and took the
electrons identical. Since electrons obey Pauli Exclusion Principle and are Fermions, so he
used Fermi-Dirac statistics instead of Maxwell-Boltzmann statistics to the electrons. Fermi-
Dirac distribution function is given by,
1
f      F
e k BT  1
This relation gives variation of f    with temperature. Here  F is Fermi energy and k B is
Boltzmann constant having value 1.38 1023 J / K .

Nearly free electron model:-

On the free electron model the allowed energy values are distributed essentially
continuously from zero to infinity. Energy levels are given by,

2
k 
2m
k 2
x  k y2  kz2            1
Where, for periodic boundary conditions over a cube of side L,

2 4
k x , k y , k z  0,  ,  ,           (1)
L L
3
The free electron wave functions are of the form
 

 k r   e ik . r


They represent travelling waves and carry momentum k .
The band structure of a crystal can often be explained by the nearly free electron
model for which the band electrons are treated as perturbed only weakly by the periodic
potential of the ion cores. This model answers almost all the qualitative questions about the
behavior of electrons in metals.

Fig (a): shows plot of energy  versus wave vector k for a free electron. Fig (b) shows plot of
energy versus wave vector for an electron in a monatomic linear lattice of lattice constant
a. We know that Bragg reflection is a characteristic feature of wave propagation in crystals.
We explain physically the origin of energy gaps in the simple problem of a linear solid of
lattice constant a. The low energy portions of the band structure are shown in Fig, in (a) for
entirely free electrons and in (b) for electrons that are nearly free, but with an energy gap
  
at k   .The region in k space between  and  is the first Brillouin zone of this
a a a
 
 i x i x
lattice. The wave functions at k   are not the traveling waves e a or e a
of free
a
electrons but are standing waves.
The time-independent state is represented by standing waves. We can form two different

i x
standing waves from two travelling waves e a
, as
 
i  
   e
i x x
a
e a
 2 cos x 
a 
            (2)
i  
    e
i x x
a
e a
 2i sin x
a 


4
Origin of energy gap:-
The two standing waves  () and () pile up electrons at different regions, and
therefore the two waves have different values of the potential energy in the field of the ions
of the lattice. This is the origin of the energy gap. The probability density  of a particle is
 *   . For a pure traveling wave e ikx , we have    *  eikx eikx  1 , so that the charge
2

density is constant. The charge density is not constant for linear combinations of plane
waves. Consider the standing wave     in (2) ;for this we have

 ()   ()  cos 2 x
2

a
This function piles up electrons (negative charge) on the positive ions centered at
x  0, a, 2a,       in Fig., where the potential energy is lowest
.
Figure (c) shows the variation of the electrostatic potential energy of a conduction electron
in the field of the positive ion cores.

The ion cores bear a net positive charge because the atoms are ionized in the metal, with
the valence electrons taken off to form the conduction band. The potential energy of an
5
electron in the field of a positive ion is negative, so that the force between them is
attractive.
For the other standing wave  () the probability density is

 ()   ()  sin 2 x
2

a
This concentrates electrons away from the ion cores. In Fig.(d) we show the electron
concentration for the standing waves  () , () and for a traveling wave.
When we calculate the average or expectation values of the potential energy over these
three charge distributions, we find that the potential energy of  ( ) is lower than that of
the traveling wave, whereas the potential energy of  () is higher than the traveling wave.
We have an energy gap of width Eg if the energies of  ( ) and  () differ by Eg .

Magnitude of the Energy Gap :-

 2 2
The wave functions at the Brillouin zone boundary k  are x and 2 sin
2 cos x,
a a a
normalized over unit length or line. We write the potential energy of an electron in the
crystal at point x is
2
U ( x)  U cos x
a
The first-order energy difference between the two standing wave states is

 
1
Eg   U (x)  ( )   () dx
2 2

0
1
    2
 2  U cos 2 x  sin 2 x  cos xdx
0  a a  a
Eg  U
We see that the gap is equal to the Fourier component of the crystal potential.

Bloch Theorem:
In case of free electron theory we assumed that the electron movie in a region of constant
potential inside a one and three dimensional potential well. Although the free electron theory is
able to explain the phenomena such as electrical conductivity, thermonic emission etc, it fails to
explain why some metals are good electrical conductor and some other are semiconductors.

In order to explain the difference between the conductors and insulators, it is necessary to
incorporate the variation of potential inside the crystal due to presence of positive ion cores in
the free electron model as shown. It appears more realistic to assume the potential inside a

6
metallic crystal to be periodic with the periodicity of lattice. The potential is minimum at ion
cores and maximum between two ion cores.

The corresponding one dimensional Schrodinger wave equation can be written as,

Where the periodic potential V(x) may be defined by means of lattice constant ‘a’ as

Bloch has shown that the one dimensional solution of Schrodinger wave equation is,

In case of three dimensions,

The equation (1) and (2) are known as Bloch functions in one and three dimensions respectively.

They represent free electron wave is modulated by periodic function or , where

or is periodic with the periodicity of the lattice in one and three dimensions

7
respectively. Therefore, considering only one dimensional case and suppose we have N(even)
number of atoms in a linear chain of atoms of length L, then we can write

And depends on the exact nature of potential field. From the equations (1) and (3)

This is known as Bloch condition. The complex conjugate of above equation is

So

This indicates that the probability of finding electron is same everywhere in the whole chain of
atoms. It is not localized around any particular atom but it is shared by all atoms in the chain.
Now we have,

This will be true only if

Where n= and L is the length of chain of atoms. When , we have .

This is the edge of the first Brillouin zone. When L is large, the allowed values of k would come
closer together and their distribution along k-axis becomes quasi continuous. The total number of
allowed k values in first zone is,

=N

8
This is equal to the total number of atoms in the chain of atoms (or in the unit cell in three
dimensions).

The Kronig-Penny Model:

In order to know the exact E-k relationship for electron it is necessary to consider a well defined
one dimensional periodic potential. For the purpose, we shell consider Kronig-Penny model.
They suggested a simpler potential in the form of an array of square wells as shown.

The corresponding Schrodinger equations for two regions I and II are of form,

Where

And

9
Where

Since the expected solutions of Schrodinger equations should have the form of Bloch function,
this requires both to be continuous throughout the crystal. Therefore, let us suppose
that the solutions of above equations are of form,

For

Where A, B and C, D are constants in the region I and II, respectively. Their values can be
obtained by applying the following boundary conditions,

Since, for a periodic lattice with V(x+a) =V(x), it is expected that the wave function will also
exhibit the same periodicity. Therefore, the expected solutions of the above Schrodinger equation
must have the same form as that of the Bloch function. Making use of Bloch functions,

Using , the equations (7) and (8) become

Now applying boundary conditions in (5), (6), (9) and (10), we obtain
10
Or

Above equations can be written as,

We have non-vanishing solutions if and only if the determinant of the coefficients A, B, C and D
vanishes that is,

On simplifying this determinant, we have

In order to simplify equation (15), Kronig and Penny assume that the potential energy is zero at
the lattice sites and equals Vo in between them. They further assume that as the height of the
potential barrier V tends to infinity and the width of the barrier b approaches zero in such a way
that the product Vob remains finite. This assumption is equivalent to the Dirac δ-function type

11
potential energies Vo separated by distance a, the potential energy being zero in between the δ-
function spikes. Under these assumptions,

Hence equation (15) becomes

Where

Where

12
Equation (16) is schematically represented in figure. The quantity on the right hand side is
plotted as a function of . The cosine term on the left hand side of the equation can only have
values between -1 and +1 as indicated by horizontal lines in the figure. A consequence of this
limitation is that only certain values of α (and hence E) are allowed. Further, from the left side of
equation (16), it is clear that for a specific value of energy E, can have only one value.
Moreover, since is an even periodic function, it will have the same values whether ka is
positive, negative or it is increased by integral multiple of 2п. Accordingly, the total energy E of
the electron is an even periodic function of k with period of . Figure below shows a plot of
energy as a function of k. if k is to be real, the magnitude of should be less than one,
which corresponds to the allowed energy band. On the other hand, those value of energy E for
which . Only the imaginary values of k are possible which correspond to the forbidden
bans. The allowed and the forbidden energy bands schematically shown in figure below.

13
Wave equation of electron in periodic potential:
Let u(x) denotes the potential energy of an electron in a linear lattice of lattice constant ‘a’. We
know that potential energy is invariant under a crystal lattice translation,

A function invariant under crystal translation may be expanded as Fourier series in a reciprocal
lattice vector G. We write Fourier series for potential energy,

This is Fourier representation of u(x) in terms of reciprocal lattice vector.

We can write equation (1) as

For convenience, we have assumed that crystal is symmetric about x=0 and

So above equation becomes,

We have assumed that crystal is symmetric,

14
The wave function 𝜓(x) may be expressed as Fourier series summed over all values of the wave
vector permitted by boundary conditions, so that

The wave equation of electron in periodic potential is,

Where H is Hamiltonian, E is energy eigen value and 𝜓 is called wave function.

The wave equation is,

Making use of equations (1),(2) in (3)

Put

Using

The above equation is known as Central equation and is S.W.E in periodic potential.

Solution of central equation:

Let g be the shortest reciprocal lattice vector such that

Then

Central equation can be written as,

15
Also it can be expanded interms of k. So put in equation (1).

Put

Put

Put

In order to solve these equations for E, the determinant of coefficients must vanish.

k  2 g   U 0 0 0
U k  g   U 0 0
0 U k   U 0 0
0 0 U k  g   U
0 0 0 U k  2 g  

Since we have five simultaneous equations so there are five values of E in this case.

Solution of central equation at zone boundary:

The central equation is gives as,

Where

We have considered, also

16
For first Brillouin zone substitute

Where and

There is no difference in both values, so

Now equations (3) and (4) can be written as,

To solve these equations make determinant of these two equations,

Equation (7) is solution of central equation at the boundary of first Brillouin zone.

17
This equation tells us that factor ‘u’ is added or subtracted from free electron energy. when factor
‘u’ is added, the energy of electron becomes greater than free electron energy which is shown by
upper part in the figure, when factor ‘u’ is subtracted, the energy of electron becomes smaller
than free electron energy which is shown by the lower part in the figure.

So the band gap is produced whose width is equal to ‘2u’. Now from equation (5)

We have found the value of energy of electron at zone boundary, now we will see what will be
the wave vector of that electron.

As we know that

Substitute

Using equation (8) in above,

In above equation wave function is for electron whose energy value is and one is for
electron whose energy value is .

18
Solution of central equation Near zone
boundary:
The central equation is gives as,

Where

We have considered, also

Since k can increase or decrease in steps of G so putting in 2.

Since k in nearly to zone boundary so C(k) and C(k-G) are important terms and we neglect other
terms. So equation (2) and (3) becomes,

To solve these equations we make determinant as

This is quadratic equation in E and it will have two roots,

19
Here and

This is solution of central equation near zone boundary.

Now put

This is energy of electron near zone boundary interms of E.

20
We can also write above equations in the form,

Each root is potted in the figure.

1
These are roots for energy when wave vector is very close to zone boundary G . For U
2
negative, the solution     corresponds to upper of two bands and     to the lower of
two bands. The two C’s are plotted in fig.

21
Fig: Ratio of the coefficients as calculated near the boundary of the first Brillouin zone. One
component dominates as we move away from the boundary.

Restatement of the Bloch theorem:-

(general proof of Bloch theorem whether state is degenerate
or non-degenerate):-
The wave function  ( x ) may be expressed as Fourier series summed over all values
of wave vector permitted by boundary conditions, so that,
 ( x)   C  k  eikx          1
k

Central equation is,

 k    C  k   UGC  k  G   0        (2)
G

Once we determines C’s from equation (2), the wave function given in equation (1) is
written as,
 k ( x )   C  k  G  ei  k G  x
k

 
 k ( x)    C  k  G  e  iGx  eikx
 k 
 k ( x)  uk  x  eikx
With
uk  x    C  k  G  e iGx
k

Because uk  x  is a Fourier series over the reciprocal lattice vectors, it is invariant under a
crystal lattice translation T, so that uk  x   uk  x  T  .We verifies this result directly by
calculating uk  x  T  :

22
  
uk  x  T    C  k  G  e  e iGT  C  k  G  e iGx 
 iG  x T 

k k 
uk  x  T   eiGT uk  x 
Since e  iGT  1 , so
uk  x  T   uk  x 
There by establishing the periodicity of uk . This completes the proof.

Crystal Momentum of an Electron:-

Q: What is the significance of the wave vector k used to label the Bloch function?
It has several properties:
  
1- Under a crystal lattice translation which carries r to r  T we have
 k  r  T   eik .T .ei k . r uk  r  T   eik .T k  r 
          

     
  
    

Because uk  r  T   uk  r  . Thus eik .T is the phase factor by which a Bloch function

   
is multiplied when we make a crystal lattice translation T.

2- If the lattice potential vanishes, the central equation reduces to  k    C  k   0 so
 
  
  

 

that all C  k  G  are zero except C  k  , and thus uk  r  is constant. We
     
 
  

have k  r   ei k . r , just as for a free electron.

 

3- The quantity k enters in the conservation laws that govern collision processesin

crystals. Thus k is called the crystal momentum of an electron. If an
 
electron k absorbs in a collision a phonon of wave vector q , the selection rule
     
is k  q  k  G . In this process the electron is scattered from a state k to a state k  ,

with G reciprocal lattice vector.

Zone Schemes::
With zone schemes we mean the graphical representation of wave vector k and energy eigen
value E. There are three kinds of zone schemes.

1. Extended Zone Scheme:

23
In the treatment of crystal we usually work in reciprocal space. In reciprocal space we divide the
crystal into Brillouin zones, such that as real crystal is made up of unit cells. In one dimensional
case the boundaries of first Brillouin zone are at . So the width of first BZ

Consider the wave function of free electron,

The energy associated with this wave function is given by,

So k increases as increases hence we get a graph of parabolic shape.

In extended zone scheme we travel in a car of wave vector from first BZ to the nearest and so on.

For nearly free electron the extended zone scheme is shown in figure.

2. Reduced Zone Scheme:

As it is clear from its name in this zone scheme we reduce the graph between wave vector and
energy into first Brillouin zone. Consider the wave function of free electron,

The energy associated with this wave function is given by,

24
So k increases as increases hence we get a graph of parabolic shape. The vertex of parabola is
at k=0.

The vertex of parabola for is at .

Although every electron state is characterized by the wave vector, but it is not unique. They
differ from one another by reciprocal lattice vector. This non-uniqueness allows us to bring all
wave vectors into the first Brillouin zone.

25
 3. Periodic Zone Scheme:
We can translate the desired portion of first zone to any or every zone boundary. We can obtain a
periodically repeated zone scheme. This construction is known as periodic or repeated zone
scheme. It a reverse process of reduced zone scheme.

26
NUMBER OF ORBITALS IN A BAND:-
Consider a linear crystal constructed of an even number N of primitive cells of
lattice constant a. In order to count states we apply periodic boundary conditions to the
wave functions over the length of the crystal. The allowed values of the electron wave
vector k in the first Brillouin zone are given by:
2 4 N
k  0,  , ,         ,
L L L
N  N 
We cut the series off  , for this is the zone boundary. The point    is
L a L a
not to be counted as an independent point because it is connected by a reciprocal lattice

vector with .The total number of points is exactly N, the number of primitive cells.
a
Each primitive cell contributes exactly one independent value of k to each energy band.
This result carries over into three dimensions. With account taken of the two independent
orientations of the electron spin, there are 2N independent orbital’s in each energy band.

Metals and Insulators:-

If the valence electrons exactly fill one or more bands, leaving others empty, the
crystal will be an insulator. An external electric field will not cause current flow in an
insulator. Provided that a filled band is separated by an energy gap from the next higher
band, there is no continuous way to change the total momentum of the electrons if every
accessible state is filled. Nothing changes when the field is applied. This is quite unlike the
situation for free electrons for which k increases uniformly in a field.
A crystal can be an insulator only if the number of valence electrons in a primitive
cell of the crystal is an even integer. If a crystal has an even number of valence electrons
per primitive cell it is necessary to consider whether or not the bands overlap in energy. If
the bands overlap in energy, then instead of one filled band giving an insulator, we can
have two partly filled bands giving a metal.

Fig.(a) an insulator, fig. (b) a metal or a semi metal because of band overlap, and fig.(c) a
metal.
27
 The alkali metals and the noble metals have one valence electron per primitive cell,
so that they have to be metals. The alkaline earth metals have two valence electrons per
primitive cell; they could be insulators, but the bands overlap in energy to give metals, but
not very good metals. Diamond, silicon, and germanium each have two atoms of valence
four, so that there are eight valence electrons per primitive cell; the bands do not overlap,
and the pure crystals are insulators at absolute zero.

28