Unit II - Semiconductor Physics

Syllabus: Free electron theory, opening of band gap due to internal electron diffraction from
lattice, band theory, Density of states, Fermi-Dirac distribution function, Carrier density in
intrinsic semiconductors, position of Fermi energy, Effective density of states, Carrier density
in extrinsic semiconductor, position of Fermi energy, Effective density of states, Numericals
on carrier density, Fermi Dirac Distribution, etc.
p-n junction diode. Charge density, potential and electric around the junction, and band
structure of unbiased diode. Barrier potential, Working of p-n junction diode in the forward
and reverse bias on the basis band structure, Ideal diode equation, Numericals on barrier
potential, transport, ideal diode equation, etc.

2.1 Introduction
A semiconductor and an insulator is characterized by a finite, non-zero band gap (Eg)
between the valence band (VB) and the conduction band (CB) and the VB is completely filled
and CB is completely empty at T = 0K (Fig. 2.1(a)). Typically, band gap Eg is ~ 1eV and 5 eV
for a semiconductor and an insulator, respectively. As a result of the smaller band gap, a
semiconductor has enough charge carriers for conduction as against very few in an insulator at
room temperature (T = 300K). A conductor or metal, on the other hand, has partially filled VB
or there is an overlap between the VB and the CB (Fig. 2.1(b)). The basic principle in
determining whether a material will be a conductor or an insulator/ semiconductor is the
following:

On application of an external electric field the electrons would like to gain kinetic
energy and move, leading to a current. Quantum mechanically this is equivalent to an electron
going from its present energy level to the next. This can happen only when the energy level,
next to the one in which the electron is, is allowed AND empty.

By this basic principle,

(a) for a semiconductor and an insulator at T = 0 K (Fig. 2.1(a)), electrons in the lower levels
of the VB do not have next energy level empty in the VB and next energy level for electrons
at the top of the VB are not allowed (forbidden). Hence, they are insulators at T = 0 K. At
higher temperatures (T > 0 K; (Fig. 2.1(b))), few electrons are promoted from VB to CB due
to thermal energy, thus creating empty energy states in the VB. Now few electrons in VB can
move into these empty states and the electrons in the CB have large number of empty states to
move into. The material therefore starts conducting. The conductivity of insulator/
semiconductor depends on number of electrons promoted from VB to CB. This number
depends of band gap Eg. Larger the band gap, smaller is the number of electrons and holes in
the conduction and the valence band, respectively. This will be discussed in detail later.

(b) for a conductor or a metal large number of energy levels are empty at all temperatures (Fig.
2.1(c), (d)) and hence the electrons can move, leading to a large current and hence the material
conducts.
 CB CB CB

CB

Eg

VB VB VB VB
(a) Semiconductor/ (b) Semiconductor/ (c) Conductor; (d) Conductor;
Insulator; T = 0 K Insulator; T > 0 K Partially filled VB Overlapping CB and VB
Fig. 2.1: (a) VB and CB with band gap Eg for a semiconductor or an insulator. At T = 0 K, VB is completely
filled and CB is completely empty. (b) For T > 0 K, few electrons are promoted to CB leaving behind empty
levels in VB. (c) For a conductor or metal, VB is either partially filled at all temperatures or (d) there is an overlap
between VB and CB. Shading indicates presence of electrons.

2.2 Free Electron Theory

An electron in a solid, experiences a periodic potential due to a periodic arrangement

of positive ions on the lattice points. Bloch showed that an electron, which is a quantum
mechanical particle, behave like a “free electron”, i.e., its energy is in the form of kinetic
energy:
1 2
𝑝2
𝐸 = 𝑚𝑣 =
2 2𝑚
where, m is the mass, v is the velocity and p is the momentum of the electron.

According to quantum mechanics, the deBroglie wavelength of a moving electron is

given by λ = h/p, where h is the Planck constant. By defining wave number k = 2π/λ and ħ =
h/2π (ħ is pronounced as h cross), the above equation can be rewritten as:

ħ2 𝑘 2
𝐸=
2𝑚
Diagrammatically, the E-k diagram for free electrons is a parabola as shown in Fig. 2.2 and
represents the relationship of the energy E of a given level with the wave number k. Since the
electrons have spin s = 1/2, they follow Fermi-Dirac distribution function (discussed in detail
later) and obey Pauli's exclusion principle. Each energy level can accommodate only two
electrons. The electrons start filling the energy levels from the bottom. The highest energy
level filled at T = 0 K is called the Fermi level and the corresponding energy is called Fermi
energy EF (Fig. 2.2).
 E
EF

Fig. 2.2: E-k diagram in the free electron theory for a metal. At T = 0 K, the band is filled by electrons upto a
level with energy EF, the Fermi energy.

2.3 Band Theory

Periodicity of ions on the lattice in a solid has another effect, namely, "internal electron
diffraction". In a cubic lattice, for example, electrons moving along the edge of the cubic unit
cell encounter sheets of ions (which act as diffracting planes) separated by the lattice spacing
"a" as shown in Fig. 2.3.

Crystal Planes

Electrons
a

Fig. 2.3: Electrons in a solid get diffracted internally from planes (vertical lines) of ions (open circles) separated
by the lattice spacing a.

According to Bragg's law, diffraction condition is:

2𝑑𝑠𝑖𝑛𝜃 = 𝑛𝜆

where d is the separation between the crystal planes, 𝜃 is the angle between the incident beam
and the diffracting crystal planes, n is the order of diffraction and λ is the wavelength of X-ray
or deBroglie wavelength of the incident particle.

In the case illustrated in Fig. 2.3, d = a (the lattice parameter of the cube) and 𝜃 = 90°.
2𝜋 2𝜋 𝑘
Therefore, 2a = nλ. This can be rewritten as = = , where, k = 2π/λ. The condition
2𝑎 𝑛𝜆 𝑛
for internal diffraction is therefore
 𝜋
𝑘 = 𝑛 (𝑎 ), n = ±1, ±2, ±3, ±4, .....

This equation represents vertical lines which cut the E-k diagram and modify it.
𝝅
Particular case of 𝒌 = (𝒂) i.e., n = 1 is shown in Fig. 2.4. The different pieces of this E-k
diagram can be translated through k values which are multiples of 2π/a without loss of its
meaning. This leads to a VB which is approximately an inverted parabola near its top and a
CB represented by an erect parabola with a band gap Eg (Fig. 2.4). The top of the VB and
bottom of the CB have energies denoted by Ev and Ec, respectively. The band gap is then Eg =
Ec - Ev.

Fig. 2.4: Development of a band gap due to internal electron diffraction. The E-k diagram can be arranged by
shifting pieces to the right by k = 2π/a to obtain CB and VB and a band gap of Eg.
Example: Lattice spacing of Si is 𝑎 = 5.431Å. Calculate the value of k and the average E for
which the first band gap will appear.

Solution:
𝜋 𝜋
𝑘= = = 5.785 × 109 𝑚−1
𝑎 5.431 × 10−10 𝑚
ℎ 𝜋 ℎ 6.63 × 10−34 𝐽𝑠
𝑝 = ħ𝑘 = = = = 6.1 × 10−25 𝑘𝑔𝑚⁄𝑠
2𝜋 𝑎 2𝑎 2 × 5.431 × 10−10 𝑚
ħ2 𝑘 2 ℎ2 ℎ 𝜋 ℎ
𝐸= = 2
, 𝑏𝑒𝑐𝑎𝑢𝑠𝑒 ħ𝑘 = =
2𝑚 8𝑚𝑎 2𝜋 𝑎 2𝑎

ℎ2 (6.63 × 10−34 𝐽𝑠)2

𝐸= 2
= −31 −10 2
= 2.047 × 10−19 𝐽 ≡ 1.28𝑒𝑉
8𝑚𝑎 8 × 9.1 × 10 𝑘𝑔(5.431 × 10 𝑚)

2.3.1 Relation between covalent bond strength and band gap energy

The electronic configuration for silicon is 1s22s22p63s23p2. The 3s23p2 orbitals undergo
sp3 hybridizations. These sp3 hybridized orbitals form covalent bonds with those of
neighbouring atoms in a 3-dimensional diamond structure. For simplicity, it is drawn as 2-
dimensional network (Fig. 2.5). Each bond in the solid has slightly different energy.
Enumeration of these energies and corresponding wave numbers is nothing but the VB of the
solid. Each bond thus corresponds to one energy level in the band. When a bond breaks an
electron is no longer restricted to the covalent bond around a Si+4 ion but can move around
throughout the solid and now is called a free electron. The energy of the free electron also is
quantized. Enumeration of the energies of the free electrons and corresponding wave numbers
is nothing but the CB of the solid. The energy required to break the weakest bond and set the
electron free corresponds to the band gap energy Eg. As the energy of bottom of the conduction
band and top of the valence band is denoted by Ec and Ev, respectively, Eg = Ec - Ev.

free electron

Hole with +1 charge

Si+4 Si+4
CB
Ec Hole
Eg
Si+4 Si+4
Ev

VB

Fig. 2.5: Schematic diagram illustration covalent bonds in a Si crystal (in 2D for simplicity) and its equivalence
to the band diagram.

2.3.2 Concept of hole

When an electron is promoted from the VB to CB it leaves behind an empty state in the
VB (Fig. 2.5). This vacancy in the VB is called a hole. In the physical bond picture, it is
equivalent to a vacancy in the bond and has a positive charge (Fig. 2.5). The hole has a positive
charge and seems to apparently move due to electrons from other bonds jumping into this
vacancy and thus creating a vacancy in the bond from which the electron has jumped.

2.3.3 Effective mass

As shown in figure 2.4, the E-k diagram deviates from the ideal parabolic behaviour.
However, mathematically it is convenient to keep using the parabolic equation for free
electrons for electrons in the band. The deviation from the parabolic E-k diagram is taken care
of by introducing the concept of effective mass m* instead of the rest mass m and is given by

ħ2
𝑚∗ = 𝜕2𝐸
(𝜕𝑘 2 )

The effective masses for the electrons in the CB and the holes in VB are denoted by m e* and
mh*, respectively.

Typical band gap and effective mass at room temperature for common semiconductors
and insulators is tabulated in Table 2.1. It is noted that Eg ~ 1eV for semiconductors and ~ 5eV
for insulators. Further, the effective masses also deviate quite significantly from the rest mass
of an electron.

Table 2.1: Band gap and ratio of effective mass m* and rest mass m 0 of an electron for typical semiconductors
and insulators

Material Eg(eV) me*/m0 mh*/m0

Si 1.12 1.18 0.8

Ge 0.66 0.55 0.36
GaAs 1.42 0.066 0.52
SiO2 8
Diamond (C) 5

Physically, the meaning of effective mass is as follows. When an electric field E is

applied, the force experienced by a charge carrier with a charge q is F = q E = m*(dv/dt).

2.3.4 Density of states

A band in a solid has closely spaced energy levels or states. The Density of States
(DOS) at a given energy E in the conduction band (gc(E)) and valence band (gv(E)) is given
by:
3⁄
4 𝑚𝑒∗ 2 1⁄
gc(E) = [ ] (𝐸 − 𝐸𝑐 ) 2 for E ≥ Ec
√𝜋 2𝜋ħ2

and
3⁄
4 𝑚∗ 2 1⁄
gv(E) = [ ℎ] (𝐸𝑣 − 𝐸) 2 for E ≤ Ev
√ 𝜋 2𝜋ħ2

The number of states in an energy interval of E to E+dE is then given as g(E)dE.

2.3.5 Fermi-Dirac Distribution Function

Electrons are particles with spin s = 1/2. All particles which have half integral spin are
called Fermions and follow Fermi-Dirac distribution function given by
 1
𝑓(𝐸) =
1+ 𝑒 (𝐸−𝐸𝐹)⁄𝑘𝑇
where, E is the energy of the electron, EF is the Fermi energy, k is Boltzmann constant, T is
temperature in Kelvin and f(E) represents the probability of finding an electron with an energy
E.

Fig. 2.6: Fermi-Dirac distribution function.

At T = 0 K,

f(E) = 1/(1+e-∞) = 1 for E < EF,

f(E) = 1/(1+e∞) = 0 for E > EF,

f(E) = 1/2 for E = EF

For T > 0 K,

f(E) < 1 for E < EF,

f(E) > 0 for E > EF,

f(E) = 1/2 for E = EF

As the temperature increases more electrons gain energy and therefore the probability of
finding them with higher energy increases. This is represented in Fig. 2.6.

Further, the probability of finding an electron with energy EF +E is equal to

probability of absence of an electron with energy EF -E.

1 1
𝑓(𝐸𝐹 + ∆𝐸) = =
1+ 𝑒 (𝐸𝐹+∆𝐸−𝐸𝐹)⁄𝑘𝑇 1 + 𝑒 (∆𝐸)⁄𝑘𝑇
1
1 − 𝑓(𝐸𝐹 − ∆𝐸) = 1 −
1+ 𝑒 (𝐸𝐹−∆𝐸−𝐸𝐹)⁄𝑘𝑇
 1 + 𝑒 (−∆𝐸)⁄𝑘𝑇 − 1 𝑒 (−∆𝐸)⁄𝑘𝑇
1 − 𝑓(𝐸𝐹 − ∆𝐸) = =
1 + 𝑒 (−∆𝐸)⁄𝑘𝑇 1 + 𝑒 (−∆𝐸)⁄𝑘𝑇

𝑒 (−∆𝐸)⁄𝑘𝑇 𝑒 (∆𝐸)⁄𝑘𝑇 1
1 − 𝑓(𝐸𝐹 − ∆𝐸) = × =
1 + 𝑒 (−∆𝐸)⁄𝑘𝑇 𝑒 (∆𝐸)⁄𝑘𝑇 1 + 𝑒 (∆𝐸)⁄𝑘𝑇
∴ 𝑓(𝐸𝐹 + ∆𝐸) = 1 − 𝑓(𝐸𝐹 − ∆𝐸)

Example: Find the probability of finding an electron at an energy 0.2 eV above and 0.2 eV
below the Fermi energy at 300K. k = 8.6 × 10-5 eV/K

Solution:

Given: E = 0.2 eV, T = 300 K, k = 8.6 × 10-5 eV/K

Probability of finding the electron at an energy E above the Fermi energy is

1 1
𝑓(𝐸𝐹 + ∆𝐸) = =
1 + 𝑒 (𝐸𝐹+∆𝐸−𝐸𝐹)⁄𝑘𝑇 1 + 𝑒 (∆𝐸)⁄𝑘𝑇
kT = 8.6 × 10-5 eV/K × 300 K = 0.0258 eV

1
𝑓(𝐸𝐹 + 0.2) = ~0.00043
1+ 𝑒 (0.2𝑒𝑉)⁄0.0258𝑒𝑉
𝑓(𝐸𝐹 + ∆𝐸) = 1 − 𝑓(𝐸𝐹 − ∆𝐸)

Rearranging the terms, we get

𝑓(𝐸𝐹 − ∆𝐸) = 1 − 𝑓(𝐸𝐹 + ∆𝐸)

Therefore, probability of finding the electron at an energy E below the Fermi energy is

𝑓(𝐸𝐹 − 0.2) ~ 1 − 0.00043 = 0.99957

Example: At what temperature can we expect 10% probability that electrons in Silver have an
energy which is 1% above the Fermi level. The Fermi energy in Silver is 5.5 eV.

Solution:
1
𝑓(𝐸) =
1+ 𝑒 (𝐸−𝐸𝐹 )⁄𝑘𝑇
∆𝐸 0.055𝑒𝑉
∴𝑇= 1
= 1
= 291𝐾
𝑘 × ln (𝑓(𝐸) − 1) 8.6 × 10−5 𝑒𝑉⁄𝐾 × ln (0.1 − 1)

2.3.6 Charge carrier density

Number of electrons per unit volume in the conduction band is given by
 𝐸𝑡𝑜𝑝
𝑛=∫ 𝑔𝑐 (𝐸)𝑓(𝐸)𝑑𝐸
𝐸𝑐

Where, 𝐸𝑐 is bottom of the conduction band, 𝐸𝑡𝑜𝑝 is top of the conduction band, 𝑔𝑐 (𝐸) is the
density of states and f(E) is the Fermi-Dirac probability of presence of an electron at E.

Similarly, number of holes per unit volume in the valence band is given by
𝐸𝑣
𝑝=∫ 𝑔𝑣 (𝐸)(1 − 𝑓(𝐸))𝑑𝐸
𝐸𝑏𝑜𝑡𝑡𝑜𝑚

Where, 𝐸𝑣 is the top of the valence band, 𝐸𝑏𝑜𝑡𝑡𝑜𝑚 is the bottom of the valence band, (1-f(E)) is
the probability of absence of an electron, i.e. presence of a hole.

It can be seen from Fig. 2.7 that the area under the curve of 𝑔𝑐 (𝐸)𝑓(𝐸) and that for
𝑔𝑣 (𝐸)(1 − 𝑓(𝐸)) is the same for an intrinsic semiconductor so that number of electrons is
equal to number of holes. For n-type and p-types semiconductors, electrons and holes are the
majority charge carriers, respectively (Fig. 2.7).

(a)

(b)

(c)

Fig.2.7 Density of states, Fermi-Dirac distribution function and their product for (a) n-type (b) p-type and (c)
intrinsic semiconductor

For all practical purpose, 𝐸𝑡𝑜𝑝 and 𝐸𝑏𝑜𝑡𝑡𝑜𝑚 can be taken as ∞ and -∞, respectively.

Number of electrons per unit volume in the conduction band is given by

 3⁄ 1
𝑚𝑒∗4 2 ∞(𝐸 − 𝐸𝑐 ) ⁄2
∴𝑛= [ ] ∫ 𝑑𝐸
√𝜋 2𝜋ħ2 𝐸𝑐 1 + 𝑒
(𝐸−𝐸𝐹 )⁄𝑘𝑇

(𝐸−𝐸𝑐 ) (𝐸𝐹 −𝐸𝑐 )

To solve the integral, let 𝜂 = and 𝜂𝑐 = .
𝑘𝑇 𝑘𝑇

Therefore,
𝑑𝐸
(a) 𝑑𝜂 = 𝑜𝑟 𝑑𝐸 = (𝑘𝑇)𝑑𝜂
𝑘𝑇

1⁄ 1
1⁄ (𝐸− 𝐸𝑐 )×𝑘𝑇 2 1⁄ (𝐸− 𝐸𝑐 ) ⁄2 1⁄ 1
(b) (𝐸 − 𝐸𝑐 ) 2 =( ) = (𝑘𝑇) 2( ) = (𝑘𝑇) 2 (𝜂) ⁄2
𝑘𝑇 𝑘𝑇

(𝐸−𝐸𝐹 ) (𝐸−𝐸𝑐 −𝐸𝐹 +𝐸𝑐 ) (𝐸−𝐸𝑐 ) (𝐸𝐹 −𝐸𝑐 )

(c) = = − = 𝜂 − 𝜂𝑐
𝑘𝑇 𝑘𝑇 𝑘𝑇 𝑘𝑇

(d) 𝐸 → 𝐸𝑐 𝑖𝑚𝑝𝑙𝑖𝑒𝑠 𝜂 → 0
3⁄ 1 1
𝑚𝑒∗ 4 2 ∞
𝜂 ⁄2 (𝑘𝑇) ⁄2
∴𝑛= [ ] ∫ (𝑘𝑇)𝑑𝜂
√𝜋 2𝜋ħ2 0 1+𝑒
(𝜂−𝜂𝑐 )

3⁄ 1
𝑚𝑒∗ 4 2 3
∞
𝜂 ⁄2
∴𝑛= [ ] (𝑘𝑇) ⁄2 ∫ 𝑑𝜂
√𝜋 2𝜋ħ2 0 1 + 𝑒 (𝜂−𝜂𝑐)
1
𝜂 ⁄2
2 ∞
∴ 𝑛 = 𝑁𝑐 ∫ 𝑑𝜂
√𝜋 0 1 + 𝑒 (𝜂−𝜂𝑐)

3⁄
𝑚𝑒∗ 𝑘𝑇 2
where, 𝑁𝑐 = 2 [ 2𝜋ħ2 ]

For low doping concentrations, i.e. if Ec - EF > 3kT, 𝑒 (𝜂−𝜂𝑐) ≫ 1.

∞ 1⁄
2 𝜂 2
∴ 𝑛 = 𝑁𝑐 ∫ 𝑑𝜂
√𝜋 0 𝑒 (𝜂−𝜂𝑐)

As 𝑒 −𝜂𝑐 is a constant, it can be taken out of the integral,

 1
𝑁𝑐 2 ∞ 𝜂 ⁄2
∴ 𝑛 = −𝜂 ∫ 𝑑𝜂
𝑒 𝑐 √𝜋 0 𝑒 𝜂
1
∞ 𝜂 ⁄2 ∞ 1⁄ −𝜂
Integral ∫0 𝑑𝜂 = ∫0 𝜂 2 𝑒 𝑑𝜂 is called a gamma integral.
𝑒𝜂

∞
Gamma integral in general has a form 𝛤(𝛼) = ∫0 𝑥 𝛼−1 𝑒 −𝑥 𝑑𝑥 , where α is a fraction.
∞ 1
Therefore, our integral ∫0 𝜂 ⁄2 𝑒 −𝜂 𝑑𝜂 = 𝛤(3⁄2).

The properties of gamma integral relevant to our derivation are: 𝛤(𝛼 + 1) = 𝛼𝛤(𝛼)and
√𝜋
𝛤(1⁄2) = √𝜋. Therefore, 𝛤(3⁄2) = 𝛤(1⁄2 + 1) = 2

Therefore,

2 3 2 √𝜋 𝜂
𝑛 = 𝑁𝑐 𝑒 𝜂𝑐 𝛤 ( ) = 𝑁𝑐 𝑒 𝑐
√𝜋 2 √𝜋 2

(𝐸𝐹 −𝐸𝑐 )
∴ 𝑛 = 𝑁𝑐 𝑒 𝑘𝑇

Here Nc is called the effective density of states of conduction band.

Similarly, the number of holes per unit volume in the valence band is given by
𝐸𝑣 (𝐸𝑣 −𝐸𝐹 )
𝑝=∫ 𝑔𝑣 (𝐸)(1 − 𝑓(𝐸))𝑑𝐸 = 𝑁𝑣 𝑒 𝑘𝑇
𝐸𝑏𝑜𝑡𝑡𝑜𝑚

∗
3⁄
𝑚ℎ 𝑘𝑇 2
Here 𝑁𝑣 = 2 [ 2𝜋ħ2 ] is called the effective density of states of valence band.

2.3.6.1 Intrinsic Semiconductor

In an intrinsic semiconductor the number of electrons is equal to the number of holes.
Therefore,

n = p = ni
(𝐸𝐹𝑖 −𝐸𝑐 ) (𝐸𝑣 −𝐸𝐹𝑖 )
∴ 𝑛𝑖 = 𝑁𝑐 𝑒 𝑘𝑇 = 𝑁𝑣 𝑒 𝑘𝑇

Where EFi is the Fermi energy of an intrinsic semiconductor

𝑁𝑣 (𝐸𝐹𝑖 −𝐸𝑐 −𝐸𝑣 +𝐸𝐹𝑖 )

=𝑒 𝑘𝑇
𝑁𝑐

𝑁𝑣 (𝐸𝐹𝑖 − 𝐸𝑐 − 𝐸𝑣 + 𝐸𝐹𝑖 )
𝑙𝑛 ( ) =
𝑁𝑐 𝑘𝑇
 𝑁𝑣
2𝐸𝐹𝑖 − (𝐸𝑐 + 𝐸𝑣 ) = 𝑘𝑇 𝑙𝑛 ( )
𝑁𝑐

𝐸𝑐 + 𝐸𝑣 𝑘𝑇 𝑁𝑣
𝐸𝐹𝑖 = + 𝑙𝑛 ( )
2 2 𝑁𝑐

3⁄ 3⁄
𝑚∗ 𝑘𝑇
𝑒 2 ℎ 𝑚∗ 𝑘𝑇 2
Putting 𝑁𝑐 = 2 [ 2𝜋ħ2
] and 𝑁𝑣 = 2 [ 2𝜋ħ2
] in the above equation

𝐸𝑐 + 𝐸𝑣 3𝑘𝑇 𝑚ℎ∗
𝐸𝐹𝑖 = + 𝑙𝑛 ( ∗ )
2 4 𝑚𝑒

If 𝑚𝑒∗ = 𝑚ℎ∗ , then

𝐸𝑐 + 𝐸𝑣
𝐸𝐹𝑖 =
2
The Fermi level of an intrinsic semiconductor thus lies at the centre of the band gap.

2.3.6.2 Extrinsic Semiconductor

n-type Semiconductor
A n-type semiconductor is obtained by doping penta-valent impurities in tetra-valent
Silicon. These dopants act as donors of electrons to the conduction band. Let ND be the number
of dopant atoms per unit volume. At room temperature, all of them ionize and contribute one
electron each thus creating ND free electrons in the conduction band. Apart from this, thermal
energy promotes few electrons from the valence band to conduction band thus producing
electrons and holes. But the number of electrons and holes produced by thermal energy is
usually much less than ND. Therefore, electrons are the majority charge carriers (~ND) and holes
are the minority charge carriers.

Fermi energy
As derived earlier,
(𝐸𝐹𝑖 −𝐸𝑐 ) (𝐸𝐹 −𝐸𝑐 )
𝑛𝑖 = 𝑁𝑐 𝑒 𝑘𝑇 𝑎𝑛𝑑 𝑛 = 𝑁𝑐 𝑒 𝑘𝑇

From the first equation,

(𝐸𝑐 −𝐸𝐹𝑖 )
∴ 𝑁𝑐 = 𝑛𝑖 𝑒 𝑘𝑇

Putting this in the second equation, we get

(𝐸𝐹 −𝐸𝑐 ) (𝐸𝑐 −𝐸𝐹𝑖 ) (𝐸𝐹 −𝐸𝑐 ) (𝐸𝐹 −𝐸𝐹𝑖 )
𝑛 = 𝑁𝑐 𝑒 𝑘𝑇 = 𝑛𝑖 𝑒 𝑘𝑇 𝑒 𝑘𝑇 = 𝑛𝑖 𝑒 𝑘𝑇
 𝑛
𝐸𝐹 = 𝐸𝐹𝑖 + 𝑘𝑇𝑙𝑛 ( )
𝑛𝑖

For n-type semiconductors, n = ND

Therefore,

𝑁𝐷
𝐸𝐹 = 𝐸𝐹𝑖 + 𝑘𝑇𝑙𝑛 ( )
𝑛𝑖

Fermi energy for n-type semiconductor is near Ec

p-type semiconductor

Fermi energy
For p-type semiconductor, using
(𝐸𝑣 −𝐸𝐹𝑖 ) (𝐸𝑣 −𝐸𝐹 )
𝑛𝑖 = 𝑁𝑣 𝑒 𝑘𝑇 𝑎𝑛𝑑 𝑝 = 𝑁𝑣 𝑒 𝑘𝑇

Fermi energy for p-type semiconductor can be derived as

𝑁𝐴
𝐸𝐹 = 𝐸𝐹𝑖 − 𝑘𝑇𝑙𝑛 ( )
𝑛𝑖

Fermi energy for p-type semiconductor is near Ev

Further, using equations for ni, n and p we can prove that

𝑛𝑖2 = 𝑛𝑝

Example: The forbidden gap in pure Si is 1.1 eV. Compare the number of electrons at
temperature 37°C and 7°C.

Solution:

𝑛𝑖 = 𝐶𝑒 −𝐸𝑔 ⁄2𝑘𝑇
3⁄ −𝐸𝑔
3⁄
𝑛𝑖 (𝑇2 ) 𝑇2 2 (
1 1
− ) 310 2 −1.1𝑒𝑉
(
1
−
1
)
=( ) 𝑒 2𝑘 𝑇2 𝑇1 =( ) 𝑒 2×8.6×10−5𝑒𝑉/𝐾 310𝐾 280𝐾
= 10.6
𝑛𝑖 (𝑇1 ) 𝑇1 280

Example: (a) Find EF - EFi for an n-type semiconductor if the doping concentration is ND= 1.8
×1015 cm-3. Given, ni = 1 × 1010 cm-3 at T = 300K. Boltzmann's constant = 8.6 × 10-5 eV/K.

(b) How far is EF from Ec if the band gap is 1.1 eV for Si?

(c) What is the concentration of minority charge carriers?

Solution: EF - EFi = kTln(ND/ni)

Given: ND= 1.8 ×1015 cm-3

ni = 1 × 1010 cm-3

T = 300K

k = 8.6 × 10-5 eV/K

Eg = 1.1 eV

(a) EF - EFi = 8.6 × 10-5 × 300 × ln(1.8 ×1015 / 1 × 1010 ) = 0.0258 ln(1.8 × 105) = 0.31 eV

(b) Since EFi is at the centre of band gap, Ec- EFi = Eg /2 = 1.1/2 = 0.55 eV

Therefore,

Ec - EF =Ec - EFi - (EF - EFi ) = 0.55 - 0.31 = 0.24 eV

(c) 𝑛𝑖2 = 𝑛𝑝

𝑛𝑖2 𝑛𝑖2 (1 × 1010 𝑐𝑚−3 )2

∴𝑝= = = = 0.56 × 105 𝑐𝑚−3
𝑛 𝑁𝐷 1.8 × 1015 𝑐𝑚−3

Example: The band gap for Silicon is Eg = 1.12 eV at T = 300K. Find the number of intrinsic
charge carriers at T = 300K. Given √𝑁𝑐 𝑁𝑣 = 9.5 × 1019 𝑐𝑚−3, 𝑘 = 8.6 × 10−5 𝑒𝑉/𝐾.
−𝐸𝑔
⁄
Solution: 𝑛𝑖 = √𝑁𝑐 𝑁𝑣 𝑒 2𝑘𝑇

−1.12⁄ −1.12⁄
𝑛𝑖 = 9.5 × 1019 𝑒 2×8.6×10−5 ×300 = (9.5 × 1019 𝑒 0.0516 ) 𝑐𝑚−3

𝑛𝑖 = 9.15 × 1019 × 3.75 × 10−10 = 35.58 × 109 𝑐𝑚−3

2.4.1 p-n junction diode

A p-n junction is a basic element in electronics which can be designed to rectify, emit
light, detect light, convert solar energy into electrical energy, be a voltage reference, etc. Let
NA and ND be the doping concentration of the acceptor impurities on the p-side and donor
impurities on the n-side, respectively. The difference ND - NA and the energy levels Ev, Ec, EF
and EFi for n and p type of semiconductors are shown in Fig. 2.8(a). When a junction is formed
the Fermi energy EF is the same throughout the solid (Fig. 2.8(b)). The levels Ev, Ec and EFi
therefore shift as shown in Fig. 2.8(b). The valence band and conduction band bend in the
region depletion region around the junction as shown in Fig. 2.8(c).

Fig.2.8 The positions of Fermi energy for intrinsic, n-type and p-type and bending of bands in a p-n junction diode

The magnitude of the difference in the value of Ec, Ev and EFi on the n side and the p
side is eVbi where Vbi is called built-in voltage or barrier potential (Fig. 2.9(a)). The variation
of potential as a function of space around the junction is shown in Fig. 2.9(b). The value of
-xp denotes the start of depletion layer on the p-side and xn denotes the end of the depletion
layer on the n-side. The potential reaches a value of Vbi at xn and remains at that value beyond.
The electric field ℰ , which is defined as derivative of potential then looks as shown in Fig.
2.9(c). The point of inflection in V versus x (Fig. 2.9(b)) occurs at the junction and therefore
the electric field has maximum value at the junction. Differentiating electric field gives the
𝑑ℰ 𝜌
Poisson's equation, = where, ρ is the charge density and ϵ is the permittivity of the
𝑑𝑥 𝜖
material. The charge density shown in Fig. 2.9(d) is due to negatively charged immobile ions
on the p-side and positively charged immobile ions on the n-side of the depletion region.
 xp xn

Fig. 2.9(a) illustrates bending of bands by an energy eVbi. The density ρ (in number/volume) of immovable charges
in a p-n junction diode is shown in (d). Electric field around the junction is shown in (c) Potential around the
junction, Vbi is the barrier potential, xn-xp is the barrier width is shown in (b).

The value of Vbi can be easily obtained by observing that the difference in EFi on the
two sides of the junction is equal to eVbi. The value of Fermi level on the p side EF w.r.t. (EFi)p
can be defined as

𝑁𝐴
𝐸𝐹 = (𝐸𝐹𝑖 )𝑝 − 𝑘𝑇𝑙𝑛 ( )
𝑛𝑖

or

𝑁𝐴
(𝐸𝐹𝑖 )𝑝 = 𝐸𝐹 + 𝑘𝑇𝑙𝑛 ( )
𝑛𝑖

Similarly, on the n-side,

𝑁𝐷
𝐸𝐹 = (𝐸𝐹𝑖 )𝑛 + 𝑘𝑇𝑙𝑛 ( )
𝑛𝑖

or

𝑁𝐷
(𝐸𝐹𝑖 )𝑛 = 𝐸𝐹 − 𝑘𝑇𝑙𝑛 ( )
𝑛𝑖
 Therefore,

𝑁𝐴 𝑁𝐷 𝑁𝐴 𝑁𝐷
(𝐸𝐹𝑖 )𝑝 − (𝐸𝐹𝑖 )𝑛 = 𝑘𝑇𝑙𝑛 ( ) + 𝑘𝑇𝑙𝑛 ( ) = 𝑘𝑇𝑙𝑛 ( 2 )
𝑛𝑖 𝑛𝑖 𝑛𝑖

If the energy is given in electron-volts (eV) then the energy for one electron moving
through a potential of V is equal to 1×V in electron-volts.

𝑁𝐴 𝑁𝐷
∴ 𝑉𝑏𝑖 = (𝐸𝐹𝑖 )𝑝 − (𝐸𝐹𝑖 )𝑛 = 𝑘𝑇𝑙𝑛 ( 2 )
𝑛𝑖

where k = 8.6×10-5 eV/K.

Example: Calculate the built in potential Vbi for a diode which has a doping of 2.5×1014cm-3
on the n-side and 9.6 × 1016 cm-3 on the p-side. The intrinsic charge carrier density at room
temperature (T = 300K) for Silicon is 1 × 1010 cm-3.

Given, k = 8.6×10-5 eV/K.

𝑁𝐴 𝑁𝐷
Solution: 𝑉𝑏𝑖 = 𝑘𝑇𝑙𝑛 ( )
𝑛𝑖2

Given: ND = 2.5 × 1014 cm-3

NA = 9.6 × 1016 cm-3

ni = 1 × 1010 cm-3

k = 8.6 × 10-5 eV/K

T = 300K

2.5 × 1014 × 9.6 × 1016

𝑉𝑏𝑖 = 0.0258 𝑙𝑛 ( )
1020

24 × 1030
𝑉𝑏𝑖 = 0.0258 𝑙𝑛 ( ) = 0.0258 × 26.2 = 0.68 𝑒𝑉
1020

2.4.2 Diode characteristics

2.4.2.1 Un-biased diode

In an unbiased diode there is an equilibrium of charge carriers moving from n to p and
p to n sides of the diode (Fig. 2.10(a)) and hence the net current is zero. It is noted that the
distribution of electrons in the conduction band and holes in the valence band is such that their
number decreases as one goes away from Ec and Ev as shown in Fig. 2.10(a). This decrease is
exponential.

2.4.2.2 Forward biased diode

In the forward bias (Fig. 2.10(b)), the barrier decreases to Vbi-VF where VF is the applied
voltage VA> 0. As a result, there are more number of electrons on the n-side than those on the
p-side in the same energy level, leading to electron current from n- to p- side. Similarly, there
is a hole current from p- to n- side. The current increases exponentially (Fig. 2.10(d)) with
applied voltage because the electron and hole distributions (majority charge carriers) are
exponential.

2.4.2.3 Reverse biased diode

In the reverse bias (Fig. 2.10(c)), the barrier increases to Vbi+ VR where VR is the applied
voltage VA< 0. As a result, there are less number of electrons on the n-side than those on the
p-side in the same energy level, leading to electron current from p- to n- side. Similarly, there
is a hole current from n- to p- side. Since this current is due to minority charge carriers, its
magnitude is small and saturates (Fig. 2.10(d)) to a value called reverse saturation current I0.

Fig. 2.10 (a) Unbiased (VA = 0) p-n junction diode. Fermi energy EF is the same on both the sides. Ev, Ec and EFi
have bent. The pile of electrons (majority charge carriers) in the conduction band represent exponentially
decreasing number as one goes to higer energy levels. Similar situiation is for holes also. Net current is zero. (b)
In forward bias (VF = VA> 0), the electrons from higher levels in conduction band and holes from lower levels in
valence band constitute electron and hole currents, respectively. (c) In reverse bias (VR = VA< 0), minority charge
carriers are able to go to the opposite side leading to reverse caturation current. (d) Ideal I-V characteristics for a
p-n junction diode.

2.4.3 I-V characteristics of a diode

The I-V characteristics for both forward and reverse bias is combined into one equation
which is
𝑉𝐴⁄
𝐼 = 𝐼0 (𝑒 𝑘𝑇 − 1)

where k is in eV/K.
The reverse saturation current I0 is determined by the doping concentrations on the n- and the
p- sides and the intrinsic charge carrier density. For large reverse voltage (VA< 0), the term
𝑉𝐴⁄
𝑒 𝑘𝑇 becomes small compared to 1 and thus can be neglected. Then, 𝐼 = −𝐼0 . For VA = 0, I
𝑉𝐴⁄
= 0. For large forward voltage (VA> 0) the term 𝑒 𝑘𝑇 becomes larger than 1 and therefore, 1
𝑉𝐴
can be neglected. ∴ 𝐼 = 𝐼0 𝑒 ⁄𝑘𝑇 . The current increases exponentially. This behaviour is
shown in Fig. 2.10 (d).

Example: If the reverse saturation current I0 = 1nA for a diode, find the forward current at T
= 300K for an applied voltage of 0.5V and 0.3V.

Solution: Diode equation is

𝑉𝐴⁄
𝐼 = 𝐼0 (𝑒 𝑘𝑇 − 1)

Where, 𝐼0 is reverse saturation current, 𝑉𝐴 is applied voltage, I is the total current through the
diode. I is positive if 𝑉𝐴 is positive and corresponds to forward current. I is negative if 𝑉𝐴 is
negative and corresponds to reverse current.

For VA=0.5V, the energy in electron-volts is 1×VA=VA.

𝑉𝐴⁄ 𝑉𝐴⁄
𝐼 = 𝐼0 (𝑒 𝑘𝑇 − 1) ~𝐼0 (𝑒 𝑘𝑇 ) = 1 × 10−9 (𝑒 0.5⁄0.0258 )~0.261 𝐴 = 261𝑚𝐴

For VA=0.3V,
𝑉𝐴⁄
𝐼 = 𝐼0 (𝑒 𝑘𝑇 − 1) ~1 × 10−9 (𝑒 0.3⁄0.0258 )~0.000112 𝐴 = 0.112𝑚𝐴

Note the vast difference in the forward current due to the fact that 0.5V is above the cut-in
voltage and 0.3V is below the cut-in voltage.
2.5 Avalanche/ Zener diode (For extra reading, Not in syllabus)
When the reverse bias voltage becomes too large there is a breakdown of the diode.
There are two types of breakdowns:

1) Zener breakdown: Here the high electric field ruptures the covalent bond and gives rise to
free electrons

2) Avalanche breakdown: Here the slope of the Ec and Ev levels (Fig. (c)) in the depletion
region is so steep that the electrons gain large energy and in the process break covalent bonds.

After the bonds are broken by the above processes, free electrons accelerate and break more
bonds thus creating an avalanche of electrons. The current thus increases rapidly at the
breakdown voltage.

The breakdown voltage can be decided doping level. A high doping level decreases the
depletion region and hence increases the electric field around the junction. The enhance
electric field leads to Zener and avalanche effects.

Zener diodes are used as voltage reference.