Semiconductor

CHAPTER 1
Physics of Semiconductors
1.1 Introduction
An understanding of the electrical properties of semiconductors is
a prerequisite to the study of electronic devices, both bipolar (BJT)
and unipolar (MOS). All materials available on the earth are
grouped under three categories depending on their ability to
conduct electricity through it. They are;
Conductors (Silver, Copper, Aluminum etc.),
Semiconductors (Germanium, Silicon, Aluminum-Gallium-
Arsenide etc.), and
Insulators (Glass, Ceramic, Bakelite, Rubber, Air, element
Sulfur).
1.1.1 Conductors
What is that make a material a conductor or an insulator or a
semiconductor? We know that any solid has vast number of
electrons, but all of them do not take part in conduction of current
through them. Only the valence electrons (outer cell electrons) can
take part in conduction of current, but they are bound. They
participate in the formation of the covalent bond. These covalent
bonds of the electrons must be broken to make them free. Since,
the valence bonds are weak, because they are from the outermost
shell of the atom, they can be broken easily. In metal a large
number valence of bonds can be broken very easily with or without
even supply of external energy. So large number of electrons
become available for conduction of current at room temperature
i.e. T = 300K.
Metals such as Silver, Copper, Aluminum etc., having plenty
of free electrons, are responsible for electrical conduction through
it and are called metals. The number of available free electrons
determines the amount of conduction of electric current through it.
If the proportion of free electrons in a material is high, the material
1- 2 Electronic Devices and Linear Integrated Circuits
conducts large current. They offer very low resistively (<10
-8
m) to
electric current flow.
1.1.2 Insulators
On the contrary, the valence bonds of insulator are very strong and
require so large energy to break them that some times with that
amount of energy the solid material becomes liquid or gets
destroyed. In other words, the valence bonds of insulator is so
strong that, theoretically even, very few bonds are broken and
insignificant conduction of current takes place at T = 300K.
Materials like glass, ceramic, quartz, Bakelite, rubber, air, and the
element Sulfur are insulators having extremely high resistively
(>10
10
m), i.e., they conduct electric current rather poorly.
1.1.3 Semiconductors
The valence bonds of semiconductor can be broken very easily in
its extrinsic form and sufficient numbers of free electrons are made
available for conduction of current. In order to quantify the
concentration of carrier, we have to represent the concentration for
carrier (electrons) under equilibrium condition including different
parameters that affects its concentration by an expression as
i
n
=
kT
E
p
1
N

,
_
(1)
Where, N is the atomic concentration in Si, p is probability that
an atom can be hit by a particle, E is bond strength, kT is the
average particle energy at T under equilibrium. Here, E/kT
represents number of particle that must converge on their atom
and to provide the energy E equal to the bond strength. We know
that the probability p is always less than 1. Since, p < 1, Eqn.(1)
can be rewritten to suite the better explanation of it as
i
n ln
=
)
p
1
ln(
kT
E
N ln
(2)
The (1/p) in Eqn. (2) is more than 1 and so
p
1
ln
is a positive
quantity. Eqn.(3) can be approximated as
i
n ln

kT
E
(3)
If we try to plot Eqn.(3) for concentration of carriers n
i
on log scale
on the y-axis w.r.t. to the reciprocal of the absolute temperature on
the x-axis, we get a straight line as indicated in Fig.1. What we
Transport Phenomenon in Semiconductor 1- 3
observe from Fig.1 is that the carrier concentration is associated
with the bond strength E. As the bond strength of GaAs is 1.41eV,
Si is 1.12eV, and Ge is 0.71eV, the carrier concentration very
much depends on the theses energies. For GaAs energy
requirement is more and hence minimum number of carrier
concentration are available for it at any temperature w.r.t. the Si
and Ge.

Figure 1 Variation in intrinsic carrier concentration w.r.t.
temperature
250K 500K 400K 300K
10
6
10
8
10
7
10
9
10
10
10
11
10
12
10
13
10
14
10
16
10
15
Ge=2.5x10
13
/cm
3
Si
GaAs
1.5x10
10
/cm
3
2x10
6
/cm
3
i
n
We will be concentrating on Si semiconductor as it is presently
dominating the semiconductor manufacturing industry. Table-1
presents relevant properties of some of the important elemental
semiconductors. If we examine the table carefully, what we
observe is that as the atomic number (size of the atom) of the
element is increasing its band gap is decreasing. It indicates that
the smaller size atoms in any semiconductor are very closely
packed. The bonds of the closely packed atoms are very strong.
The carbon having very high band gap (E
G
= 5.3eV) is insulator
because its atoms are very small in size and hence very closely
packed inside the crystal structure. The atom size of silicon (Si),
germanium (Ge) and tin (Sn) are much larger than that of the
carbon and hence they are not so closely packed inside the crystal
structure have low band gaps and are called semiconductors. The
other revelation of the table-1 is that as the band gap goes on
increasing, the resistivity also goes on increasing. The resistivity of
the carbon (C) is as high 10
14
cm where the resistivity of the Sn
(semiconductor) is as low as 10
-4
cm.
Table-1.1 Properties of elemental
semiconductor
Elemen
t
Atomic
Weight
Band
gap
Resistivity
C 6 5.30eV
10
14
cm
Si 14 1.12eV
10
5
cm
Ge 32 0.67eV
10
2
cm
Sn 50 0.08eV
10
-4
cm
Advantages of semiconductor are;
1. of much greater density than a gas,
2. reduced ionisation energy (of the order of 1eV =1.6x10
-19
x1
joules to produce an electron-hole pair, compared with 5
eV to ionise a solid insulator and 30 eV to ionise a gas),
3. a much greater density of free carriers is produced in a
semiconductor,
4. ideal for making the ultra-small, high-speed transistors that
implement the modern processor,
5. can be manufactured into large pieces of uniform
composition and high quality,
6. larger the crystal (8 to 12 diameter), the more chips can
be manufactured at the same time, and the less waste
material, thus saving money,
7. of the appropriate hardness so that it can be cut into thin
slices without being so brittle that it cracks,
8. readily available and relatively inexpensive, to ensure
supply and keep costs down,
9. rapid heating at 20,000
O
C per second to unlimited
temperatures,
10. dopants moved only by 20 30 Angstroms,
11. ultra shallow junctions.
Let us concentrate our discussion on semiconductor because we
would be talking, hence forth, regarding devices made of
semiconductors and its uses. There are two types of carriers in a
semiconductor; namely
Electrons and
Holes.
We have seen in the Fig.2 that one electron has gone to the
conduction band leaving behind a vacancy. This vacancy is
conceptually a positive charge carrier having equal but opposite
charge of the electron. This carrier is called hole. Hence, the
positive charge carrier, the hole, can also move in opposite
direction of the electron can contribute to current flow.
The two types of processes occur in a pure or intrinsic
semiconductor;
generation and
recombination
An electron getting energy in the valence band moves to the
conduction band leaving behind a hole in the valence band is
called generation process. While an electron falling back from
conduction band to the valence band may though of as
recombination process. Under equilibrium, the generation is equal
to the recombination.
The process of generation is proportional to temperature i.e. G =
f(T) and the process of recombination is proportional to the product
of the electron and hole. So if we equate these two
G = f(T) = R = np =
2
i
n
It means that at any temperature, the product of electron and
hole concentration remains constant. Since, electrons and holes
are generated and recombine in pairs, the number holes must be
equal to the number of electrons. Since,
n
=
p
, then
2
n =
2
p
=
2
i
n
,
so,
n
=
p
= i
n
in an intrinsic semiconductor. The i
n
is called the
intrinsic carrier concentration. The i
n
is function of material and the
temperature both. For example; Si having 1.12eV band produces
carriers at room temperature (300K) and 500K as
) K 300 , Si ( n
i =
3 10
cm / 10 5 . 1
;
) K 500 , Si ( n
i =
3 14
cm / 10 5 . 2
Figure 2 Energy band diagram of Semiconductor
The energy band gap at 0K of Germanium and Silicon are 0.785eV
and 1.21eV respectively. The forbidden energy gap decreases with
increasing temperature leading to increasing conductivity. It is
found that E
G
decreases at a rate of 3.6x10
-4
eV/K in the case of
silicon i.e.
) T ( E
G =
T 10 x 6 . 3 21 . 1
4
Similarly in the case of Germanium, the energy gap decreases at

the rate of 2.24x10
-4
eV/K.
Example
Find the band gap of Silicon and Germanium at room
temperature(300K).
) K 300 , Si ( E
G =
300 x 10 x 6 . 3 21 . 1
4
= 1.21-0.1 = 1.1eV
) K 300 , Ge ( E
G =
300 x 10 x 24 . 2 785 . 0
4
= 0.71-0.0669 = 0.718eV
E
C
E
V
E
G
Conduction Band
Valence Band
e
h
The germanium having less band gap (E
G
= 0.72eV) has more
carriers (at least 1000 times) than that of the Si at room
temperature (300K).
The semiconductor has become very important only because its
conductivities can be changed to several orders of magnitude by
introducing controlled amount of impurities. Adding of controlled
impurities is called doping and the resulting material is called
doped semiconductor. The doped semiconductor is called the
extrinsic semiconductor. The extrinsic semiconductor is of two
types;
The type of the material depends on the type of the impurity
introduced. If pentavalent element impurity is introduced into the
tetravalent element, the resulting material is called n-type. Here,
the n stands for negative charge carriers. The pentavalent element
has five electrons in its outermost orbit. The four electrons of the
tetravalent element cover the four electrons from all around of the
fifth valent element and the fifth electron is left bound loosely to its
parent atom. This fifth electron can be detached with very small
amount of energy. Hence, each impurity atom is supplying one
loosely bound electron and was given the name n-type material.
Similarly, if controlled impurity of trivalent element is introduced
into the tetravalent element, each trivalent element introduces one
hole which is loosely bound to the parent atom. This loosely bound
hole can be made free with very small amount of energy. This is
the reason it is called p-type material. Here, the p-stands for
positive charge carriers. The group-V elements are called donor,
because it can donate one electron and group-III elements are
called acceptor because it can accept one electron.
Extrinsic (impure)
Semiconductors
Intrinsic (pure)
n-type p-type
Now again we go back to the discussion of concentration of
charge carriers as in Eqn.(3). Since, we cannot calculate the value
of probability p and hence exact numbers of carrier concentration
cannot be calculated. Only an approximate behaviour can be
predicted by the above mathematical model. In order to have
better picture of carrier concentration at any temperature, we must
consider the band model.
Band Model is more appropriately called energy band model.
What energy band model does? It tries to analyze population of
electrons in terms energy possessed by electrons. Electrons are
distributed over a range of energy at a given temperature. What we
have to do is to find out a way how electrons are distributed over
energy at a given temperature? If there is some way of finding this
distribution, we can very easily separate what part of electrons that
is free and the other part which is not free at any temperature.
Solids having conductivity between metals and insulators are
called semiconductors. Semiconductor materials, used in
manufacturing solid-state devices and ICs, are of two types,
namely,
Elemental semiconductor, and
Compound semiconductor
Important elemental semiconductor materials are from Group IV of
periodic table such as
Silicon (Si)
Germanium (Ge)
Diamond (C)
These groups IV elemental materials have diamond crystal
structures i.e. each atom is in tetrahedral configuration with four
nearest atoms. Thus they are referred to as tetrahedrally-bonded
semiconductors. Other elemental semiconductors, having various
structures differing from diamond structures, include group III
element boron, group V element phosphorus, group VI materials
such as Sulphur (S), selenium (Se), and tellurium (Te).
Germanium is one of the basic elemental semiconductors that
were used in manufacturing transistor and solid-state devices. But,
because it is difficult to process and inhibits device performance at
high temperature, it is rarely used these days.
Currently Si is the most important material used for electronic
devices manufacturing (i.e. integrated circuits). It is used in
approximately 90% of the solid-state devices and IC
manufacturing. The popularity of silicon is due to its availability in
abundance in the nature in the form SiO
2
and retention of good
electrical properties, even at high temperature.
Some of the important advantages of Si over other
semiconductors are (i) relative ease of passivating the surface by
oxidizing it in a controlled manner and forming a layer of stable
native oxide that substantially reduces the surface recombination
velocity, (ii) its hardness that allows large wafers to be handled
safely (iii) its thermal stability up to 1100
o
C, that allows high
temperature processing related to diffusion, oxidation and
annealing and (iv) its relative low cost.
The basic limitations of Si are due to magnitude and type of its
energy gap E
G
=1.12eV. In fact it is indirect energy gap material
which prohibits its use in the optoelectronic applications and (ii) its
relatively lower carrier mobility as compared to GaAs.
The semiconductor formed using two or more elements is
called compound semiconductors and they do not appear in
nature. They are synthesized using elements from groups II
through VI of the periodic table, e.g. from group III and V (III-V
compounds) or II and VI (II-VI compounds).
Important compound semiconductor materials are
The important binary compound semiconductor formed from
group-3:5 is Gallium arsenide (GaAs) and Indium phosphide (InP),
silicon germanium (SiGe) formed from group-4:4, cadmium
tellurium (CdTe) formed from group-2:6. The ternary compound
semiconductor of our interest formed using group-3:5 are (GaAsP,
AlGaAs) and the quaternary compound semiconductor is indium
gallium arsenide phosphie (InGaAsP) fromed from 3:5 group
elements.
Some of the attractive properties of compound semiconductor
such as Gallium arsenide (GaAs) are faster operating frequency
(two to three times faster than Si), low heat dissipation, resistance
to radiation, and minimum leakage between adjacent components.
These properties make GaAs an important semiconductor for use
in high performance applications. Its drawbacks are difficulty in
growing the ingots and fabricating ICs.
Materials such as Germanium and Silicon neither offer high
resistance as insulators nor offer low resistance like conductors
(metals). Their resistivity is approximately 10
2
m.
An elemental and compound semiconductor that is not
contaminated by the introduction of impurities is called an intrinsic
semiconductor. However, silicon-based semiconductors totally
dominate the present commercial market due to their advanced
fabrication technology. It has to be extracted from impure SiO
2
.
We know that rotating electrons around the nucleus in the
outermost orbit are called valence electrons. The four small lines
around the periphery of the circle inscribed Si (Figure 3) represent
the valence electrons in the outer most orbits of the tetravalent
elements (Si/ Ge/ Sn/ C).
Indirect and direct band gap semiconductors
The top of the valence band and bottom of the conduction band do
not appear at the same value of momentum of the electrons in the
indirect band gap semiconductor. Where as the top of the valence
band and the bottom of the conduction band occur at the same
value of momentum in the direct band gap semiconductors as
depicted in Fig.3.
Figure 3 Band structure of Si, GaAs and AlAs
Each photon of energy E has momentum k = E/c, where c is the
velocity of light. An optical photon has energy of the order of 10
-19
J,
and since c = 310
8
m/s. A typical photon has a very small amount
of momentum. A photon of energy = E
G
, where E
G
is the band
gap, can excite an electron producing an electron-hole pair in the
valence band to conduction band in a direct band gap
semiconductor quite easily, because the electron does require to
change the momentum. However, an electron must also undergo a
significant change in its momentum for a photon of energy E
G
to
produce an electron-hole pair in an indirect band gap
semiconductor. This is possible, but it requires such an electron to
interact not only with the photon to gain energy, but also with a
lattice vibration called a phonon in order to either gain or lose
momentum. In order to make diagonal indirect transition, both
appropriate energy photon and an appropriate momentum phonon
must be present together with the electron. This is referred to the
area of three body interaction because there are three particles
(electron, phonon, and photon) participating. Such collisions are
over 1000 times less likely than a simple electron-photon
interaction. In a direct band gap semiconductor, when an electron
in the conduction band falls to occupy the an empty state in the
valence band, the energy is given off as a photon of light.
The GaAs is a direct band gap semiconductor, which means that
the minimum of the conduction band is directly over the maximum
of the valance band as in Figure 3. Transitions between the
valance band and the conduction band require only a change in
energy, and no change in momentum, unlike indirect band-gap
semiconductors such as silicon (Si). This property makes GaAs a
very useful material for the manufacture of light emitting diodes
and semiconductor lasers, since a photon is emitted when an
electron changes energy levels from the conduction band to the
valance band.
The indirect process proceeds at a much slower rate, as it
requires three entities to intersect in order to proceed: an
electron, a photon and a phonon. Such collisions are over 1000
times less likely than a simple electron-photon interaction. This is
analogous to chemical reactions, where, in a particular reaction
step, a reaction between two molecules will proceed at a much
greater rate than a process which involves three molecules.
The same principle applies to recombination of electrons and
holes to produce photons. The recombination process is much
more efficient for a direct band gap semiconductor than for an
indirect band gap semiconductor, where the process must be
mediated by a phonon.
As a result of such considerations, gallium arsenide and other
direct band gap semiconductors are used to make optical devices
such as LEDs and semiconductor lasers, whereas silicon, which is
an indirect band gap semiconductor, is not. The important
compound semiconductors such as GaAs, InP, InGaAs, etc. are
direct bandgap semiconductors, where as important indirect
bandgap semiconductors are Si, Ge, AlAs etc. The typical
materials with their band gaps are arranged in a table-1.2.
Table-1.2 Typical materials with their band gaps
Material Direct /
Indirect
Bandgap
Band Gap
Energy at
300 K (eV)
Elements C
(diamond)
Ge
Si
Sn (grey)
Indirect
Indirect
Indirect
Direct
5.47
0.66
1.12
0.08
Groups III-V
compounds
GaAs
InAs
Direct
Direct
1.42
0.36
InSb
GaP
GaN
InN
Direct
Indirect
Direct
Direct
0.17
2.26
3.36
0.70
Groups IV-IV
compounds
-SiC Indirect 2.99
Groups II-VI
compounds
ZnO
CdSe
ZnS
Direct
Direct
Direct
3.35
1.70
3.68
1.1.4 Energy Band
An isolated atom has distinct energy levels represented by their
orbits as indicated in Fig.4 (a). All isolated atoms of the same
element have the same energy level as indicated in Fig.4 (b). As
atoms get close enough to interact, outermost orbit interact first as
demonstrated by Fig.6.
When two identical atoms are brought closer, their electrons in
two different orbits with having all the four quantum numbers
same, align themselves with different energy levels as shown in
Fig.6. In other words equal energy levels of the interacting
electrons split. For an example, atom-1 and atom-2 of the same
element are brought closer. The energy level E
11
of atom-1 and
energy level E
21
of atom-2 interact first resulting in either shifting of
the energy level E
11
up or down w.r.t. E
21
as illustrated in Fig.6.
Thus, a gap is created between the same energy levels of two-
identical isolated atoms of the same element if they are close
enough to interact.
Thus, one energy level splits into N-levels when N-atoms are
brought together, and these N-levels can accommodate at most
2N-electrons due to spin degeneracy (t spin). Fig.6 shows a Si
atom having four-electrons in its outermost orbit. When atoms are
brought close together their valence electrons get shared and form
bonds between the two-atoms called covalent bonds. A simplified
picture of the diamond-crystal bonding of silicon semiconductor is
shown in Fig.7. It is easier to analyze energy-related aspects
rather than spatial aspects such as bonds. Therefore, the concept
of energy bands is used to describe the semiconductors.
Now, we would like to describe the formation of band model in
detail. Band Model is more appropriately called energy band
model. What energy band model does? It tries to analyze
population of electrons in terms energy possessed by them.
Electrons are distributed over a range of energy at a given
temperature. What we have to do is to find out a way to show the
distribution of electrons over a range of energy at a given
temperature? If there is some way of finding this distribution, we
can very easily separate what part of electrons that is free and the
other part which is not free at any temperature.
Figure 4 (a) Representation of Tetravalent elements Si
Figure 4 (b) Two isolated atomic orbits
Si
Atom - 2 Atom - 1
E
11
E
12
E
13
E
14
E
15
E
11
E
12
E
13
E
14
E
15
Figure 5 Energy levels in two isolated Atoms
Figure 6 Shifting of energy level
Figure 7 Covalent bonding of tetravalent element
The basics of the energy band model are;
1. Allowed states
(a) Electrons are allowed only in certain energy states
(Principles of Quantum Mechanics).
(b) Electrons occupy the lowest available energy of the allowed
states.
(c) No two electrons occupy the same energy state (Paulis
exclusion principle).
E
21
E
11
E
21
E
11
Si Si
Si Si
Si
Si
Bonds
Si
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e e e e e e
e e e e e e
e e e e e e
e e
e e e e
2. Distribution of allowed states over energy E is expressed N(E).
3. What fraction of allowed state is occupied at any temperature?
Electrons will not occupy all the available states, because allowed
states are much larger than available number of electrons. At any
temperature T, under equilibrium only a fraction of allowed states
ate occupied. This fraction is expressed as f(E,T). This fraction
depends partially on bond strength E and partially on temperature
T. This function f(E,T) is given the name Fermi-Dirac function.
The Fermi-Dirac function is derived from statistical mechanics.
Since, the first two points corresponds to the quantum mechanics
and the third point corresponds to statistical mechanics, the energy
band model is the combined representation of the Quantum
mechanics and the statistical mechanics.
Let us discuss the above points one by one in more details.
What are allowed states for electrons in a crystal? What is its
distribution over a given energy and what is the fraction of the
states that are occupied? For such discussions, we start with the
picture of an isolated atom of Si with its shells and sub-shells
depicted in Fig.8. What are the allowed states in an isolated Si
atom? Then we put together N-atoms, all regularly arranged, and
start decreasing distance between atoms down to the distance
value which is actually present inside a crystal. This way we can
see how the allowed energy states of an isolated atom gets
modified by decreasing the distance between atoms to appreciate
the picture of the energy band in the crystal. We represent atomic
separation on the x-axis and the electron energy on the y-axis. We
would plot the electron energy versus atomic separation for N-
atom system. Here, all N-atoms must be regularly arranged. We
start with infinite separation of atoms. Though all atoms are in a
crystal, assumption is valid so long as the affect of one atom over
the other in not present if they are at infinitely separated
Let us draw the electron energy w.r.t. to the atomic separation
as in Fig.9. For this let us examine the atomic structure of some of
the important elements that make up various semiconductors.
IV Semiconductors III-V Semiconductors
C =

2 2 2
p 2 s 2 s 1
Ga =

1 2 10 6 2 6 2 2
p 4 s 4 d 3 p 3 s 3 p 2 s 2 s 1
Si =

2 2 6 2 2
p 3 s 3 p 2 s 2 s 1
As=

3 2 10 6 2 6 2 2
p 4 s 4 d 3 p 3 s 3 p 2 s 2 s 1
Ge =

2 2 10 6 2 6 2 2
p 4 s 4 d 3 p 3 s 3 p 2 s 2 s 1
We know that Si has atomic number 14. It means there are 14-
electrons in the silicon atom. How these 14-electrons are
distributed in the atom? They are distributed in different shells
around the nucleus. K shell has only 1s-sub shell with 2-electrons,
the L-shell has 2s and 2p sub-shells having 2+6 = 8 electrons. The
M shell has 3s, 3p and 3d sub-shell. The 3d sub-shell is completely
empty and hence its contribution in energy level will not be
considered. The arrangement of electrons in Si atom in different
shells looks like
2 2 6 2 2
p 3 s 3 p 2 s 2 s 1
. In an N-atom system of the 3p sub-
shell of Si has 6N number of states but only 2N-electrons are
present and so 2N states are occupied and 4N-states remain
unoccupied. The lowest energy correspond to 1s sub-shell and the
highest energy correspond to the upper most sub-shell i.e. 3p.
On the left hand side of Fig.9 denotes the sub-shells 1s, 2s, 2p,
3s, 3p of isolated Si atom and its occupancy. On the right hand
side of Fig.9 denotes the availability of states and the actual
occupancy of electrons in each sub-shell for a N-atom system. The
distribution of electrons in an isolated Si atom in its different sub-
shells is shown in Figs.8 and 9.
We have shown number of available states by vacant circles
and number of actually occupied states by crossed circle in each
sub-shell of Si atom in Fig.9. The 1s, 2s and 3s has 2N states and
all are field as indicate by crossed circle. The 2p has 6N states and
all of them are filled. The 3p has 6N states and only 2N states are
filed. This model is valid for an isolated atom of Si.
If spacing is reduced between atoms, the outer most orbits of
different atoms would start interacting first and its energy states
represented by single line will plit into N-number of lines because
there is N-atom system under consideration and no two electrons
can have the same energy state. This way N-number of closely
spaced lines, each representing one energy level, would form band
of energy levels with separation among them. Strictly speaking 3s
and 3p have 2n-lines of energies. As the number n is very large to
the extent of concentration of Si atoms of the order of
3 22
cm / 10
, the
separation between so many lines of energies is so low that we
consider as if a continuous band of energy is present after splitting.
So we do not show it by individual levels but a continuous band of
energy. The energy level which is very close to the nucleus would
be splitting once the atoms are brought much closer. So as the
atoms are coming closer and closer inner levels of energies also
split into bands.
Figure 8 Silicon atoms with available states and
occupancy
Figure 9 Representation of States and its
occupancy
2 states, 2e
-
2s
1s
3p
3s
2p
OOO O

2 states, 2e
-
6 states, 6e
-
2 states, 2e
-
6 states, 2e
-
1 atom system
2N states, 2N e
-
3s
2p
2s
1s
3p
OOO O

2N states, 2N e
-
26 states, 6N e
-
2N states, 2N e
-
6N states, 2N e
-
N atom system
Si
n = 2, 8states, 8e
-
n = 1, 2states, 2e
-
n = 3, p-shell
6-allowed states
n = 3, s-shell
2-allowed states
n = 3, 8states, 4e
-

e
e
e e
The right side of Fig.10 indicates that if the separation between
atoms is much less, inner level shell also start interacting (2p shell
has also split). The sub-shells 3p and 3s have split so much that
they have merged. The total number of states available after
merging of the 3p and 3s sub-shells is 8N and the available
numbers of electrons in these states are 4N only. Further decrease
in the separation has split the merged levels of energies of 3p and
3s sub-shells into two bands as in Fig.11 with a visible gap
between them. We know that electrons occupy the lowest energy
level. The lower band after splitting has 4N states and only 4N
electrons are available. So the lower 4N states would be occupied
by 4N numbers of electrons in that band. These electrons are
called valence electrons and the band is called the valance
band. The upper band after splitting has 4N states but no
electrons. This band is called the conduction band, because
when energy is supplied electrons jump from the valence band to
the conduction band and they become free to move. This is the
reason it called the conduction band. Our interest lies in the bands
at the atomic separation of the crystal indicated by x in Fig.10.
Hence the band picture along the x line is the interest for the
explanation of the energy band model of the crystal.
Let us draw the band picture along the x line at only one atomic
separation as in Fig.11. The lower bands of energies
corresponding to 2p, 2s, and 1s sub-shells are of no interest
presently for us because they do not take part in conduction
mechanism of electrons. More or less these conduction and
valence bands with energy gap are sufficient to describe the
behaviour of the semiconductor and hence we do not bother for
the inner bands. It is of our interest to know what the energy of
electrons is at infinity i.e. where electrons are completely
separated from the crystal. We have to indicate these energies
inside the crystal with reference to some level. This level lies at
infinity where the energy of the electrons is supposed to be zero.
This level is called the vacuum level.
3p
Electrons
energy
3p O O O O
Electrons
energy
6n states
2n e
-
O O O O
6n states
2n e
-
Figure 10 Interacting energy levels with atomic separation
The electron energy can be represented in the form cuttings
available along this line vertically from Fig.11. The energy level at
infinity is cut at only one point and hence represented by a point as
in Fig. 12, naming it vacuum level energy. The vertical line through
point x cuts the merged energy band through valence band and
conduction band. Hence, these bands are represented by vertical
lines in Fig.112 with an energy gap. The inner bands are cut at
points and hence they have been represented in Fig.12 by points
only.
Practically it is difficult to grasp this simple band diagram picture
of Fig.12 having point representation of energy. Hence, for
simplicity instead of a point we draw a line. So we represent these
points consisting of so many lines. The vacuum level is represent
by a line and designated as having the energy E
0
. The lowest line
of the conduction band is designated as E
C
and the upper most
line of the valence band is designated as E
V
with an energy gap
between them as E
G
. The energy difference labeled as
between
the bottom of the conduction band and infinity is called the
electron affinity of the semiconductor. For silicon
= 4.05eV, E
G
= 1.12eV and the width of both conduction and valence bands are
much thicker than the band gap; some thing like 5 times as
indicated in Fig.13, though it is not to scale.
3s
2p
2s
1s
Atomic separation
2s
1s
3s
2p
Atomic separation

6n states
6n e
-
2n states
2n e
-

2n states
2n e
-

2n states
2n e
-

6n states
6n e
-
2n states
2n e
-
2n states
2n e
-
It is but natural to ask a question what is the difference between
the band model and the bond model. Let us discuss how bond
model corresponds to the band model. The important events that
take place in the bond model are generation and recombination.
How do they correspond to in the band model? Electrons breaking
the bond, called generation (G) as in Fig. 15, move about the
crystal correspond to the change in the energy as shown in the
band model of Figs. 13. Similarly, the recombination (R) process
corresponds to a free electron jumping into a vacancy and getting
annihilated and breaking a bond in Fig. 14. In the band model it
looks like a hole filled with an electron in the valence band
indicated by a dot inside the circle with an arrow down from
conduction band to the valence band in Fig. 15. That is electrons
jump from the conduction band to the valence and recombines with
holes.
3s
2p
2s
1s
3p
Atomic separation
Electrons
Energy
8n states, 4ne
-
4n states, 0e
-
4n states
4ne
-
x
2n states
2n electrons
6n states
2n electrons
O O O O

2n states
2n electrons
2n states
2n electrons
6n states
6n electrons

3s
2p
2s
1s
3p
Atomic separation
Electrons
Energy

O O O O
2n states
2n e
-
2n states
2n e
-
6n states
6n e
-
2n states
2n e
-
6n states
2n e
-
Figure 11 Interacting energy levels with lesser atomic
separation
The free electrons in the bond model are gaining KE in the
crystal. This is represented in the band model by electrons moving
up in the conduction band. Since, the electrons and holes are
generated in pairs, as the electrons are gaining energy in the
conduction band of the crystal; the holes are gaining energy in the
valence band. It is shown by its movement down in the valence
band. Finally let us see the other type of transition i.e. electrons
jumping from valence band to vacuum level. This is equivalent to
the electron going out of the crystal by breaking bond. This energy
transition is shown in the band model by the arrow from valence
band to the vacuum level. It is important to know that the electrons
which have come out of the crystal whether remains at the surface
or very far away from the surface, their energies are the same,
because both electrons are out side the influence of the atoms of
the crystal. So for simplicity of explanation, electrons out side the
crystal are assumed to be at the surface. This means electrons
have moved out of the Si crystal but these electrons are localized
at the surface of the Si crystal. This is called local vacuum level. A
free electron from the conduction band is also equivalent to moving
to the local vacuum level as indicated in Fig.15.
Fig.12 Energy band at atomic separation of the crystal Fig.13
Simple band diagram of Si
Electrons
energy
Conduction Band
Valence Band
Energy gap
1s band of energy
2s band of energy
2p band of energy
Atomic Separation
Vacuum level (E
0
)
Band gap (E
G
)
= 1.12eV
Electron affinity ( = 4.05eV)
E
0
E
C
E
V
Vacuum level
G =e
-
R
E
0
E
C
E
V
h
G
e
eh
R
h
e
Holes gaining
KE
Electrons gaining
KE
Local vacuum level
Crystal
Local vacuum level
e
e
e
Fig.14 Bond Model of Si Fig.15 Band Model of Si
Let us remember that in an N-atom model, the N is a huge
number. Since, the separation between the energy levels within
the band is much smaller than the thermal energy possessed by
an electron at room temperature, the band can be viewed as
continuous. The energy E
C
is the lowest possible conduction band
energy, while the energy E
V
is the highest possible valence band
energy as depicted in Figs.16 and 17.
Figure 16 Energy bands for diamond versus lattice constant.
The band gap energy E
G
, is defined as (E
C
-E
G
). The energy gap E
V
is the energy it takes to break a bond in the spatial view of the
crystal. The band gap energies for some semiconductors at room
temperature T = 27
O
C+273 = 300K are, E
G
= 1.41eV in GaAs and
1.12eV in Si. We know that 1eV = 1.602x10
-19
J. The typical energy
band diagram of insulators, semiconductors and metals is shown
in Fig.17. The band gap in insulator is very high i.e. of the order of
8eV whereas band gap in the case of metals is either very low or
valence and conduction bands overlap.
1.1.5 Metals, insulators and semiconductors
Once we know the band structure of a given material we still need
to find out which energy levels are occupied and whether specific
bands are empty, partially filled or completely filled.
Empty bands do not contain electrons. Therefore, they are not
expected to contribute to the electrical conductivity of the material.
Partially filled bands do contain electrons as well as available
energy levels at slightly higher energies. These unoccupied energy
levels enable carriers to gain energy when moving in an applied
electric field. Electrons in a partially filled band therefore do
contribute to the electrical conductivity of the material.
Completely filled bands do contain plenty of electrons but do not
contribute to the conductivity of the material. This is because the
electrons cannot gain energy since all energy levels are already
filled.
In order to find the filled and empty bands we must find out how
many electrons can be placed in each band and how many
electrons are available. Each band is formed due to the splitting of
one or more atomic energy levels. Therefore, the minimum number
of states in a band equals twice the number of atoms in the
material. The reason for the factor of two is that every energy level
can contain two electrons with opposite spin.
To further simplify the analysis, we assume that only the
valence electrons (the electrons in the outer shell) are of interest.
The core electrons are tightly bound to the atom and are not
allowed to freely move in the material. Four different possible
scenarios are shown in Figure 17.
Figure 17 (a) shows a half filled band, (b) two overlapping bands,
(c) an almost full band separated by a small band gap from an
almost empty band and (d) a full band and an empty band
separated by a large band gap.
A half-filled band is shown in Figure 17 (a). This situation occurs
in materials consisting of atoms, which contain only one valence
electron per atom. Most highly conducting metals including copper,
gold and silver satisfy this condition. Materials consisting of atoms
that contain two valence electrons can still be highly conducting if
the resulting filled band overlaps with an empty band. This
scenario is shown in (b). No conduction is expected for scenario
(d) where a completely filled band is separated from the next
higher empty band by a larger energy gap. Such materials behave
as insulators. Finally, scenario (c) depicts the situation in a
semiconductor. The completely filled band is now close enough to
the next higher empty band that electrons can make it into the next
higher band. This yields an almost full band below an almost
empty band. We will call the almost full band the valence band
since it is occupied by valence electrons. The almost empty band
will be called the conduction band, as electrons are free to move in
this band and contribute to the conduction of the material.
Figure 17 Possible energy band diagrams of a crystal.
It is evident from Fig.17 (e) that the forbidden band is very large
i.e. 8eV in case of insulator, a moderate 1.12eV in case of a
semiconductor, and overlapping valence and conduction bands in
case of metals. As the energy band gap between the valence band
and the conduction band is very large in case of insulator, almost
no free electrons will be available in the conduction band for
conduction of current. As the energy gap between the valence
band and the conduction band is low, the probability exists that
some electrons will be lifted to the conduction band and will
become free for conduction of current in semiconductors. In metals
there is no forbidden energy gap between the valence and
conduction bands and enormous amount of free electrons are
available for the conduction of current without any external supply
of energy.
Insulators
Figure 17 (e) Energy band diagrams of Insulators,
Semiconductors, and Metals
Figure 18 Simplified energy band diagram to describe
semiconductor
We know that electrons can be emitted if light falls on certain
metals. This was observed by Einstein. However, it requires a
minimum amount of energy q
M
(eV), called the work function of
the metal. This is the amount of work done in moving the electrons
from the surface of the metal (from Fermi level of the metal) to the
vacuum. This can be explained using band diagram. The valence
band and the conduction band of most of the metals overlap. We
know that all the states below Fermi levels are occupied by
electrons. On the other hand, all the states above the Fermi level
E
G
= 8eV SiO
2
= 5eV (Diamond)
E
V
E
C
Conduction Band
Valence Band
1.12eV for Si
E
V
E
C
Semiconductors
Valence Band
Conduction Band
or
Metals
E
V
E
C
E
C
E
V
Valence Band
Conduction Band
Valence Band
Conduction Band
Valence
Band
Conduction
Band
Energy
Bandgap
Work function q q
S
E
vacuum E
E
V
E
C
E
F
E
G
E
F
q
M
E
vacuum
Metal
are empty. Hence, when electrons at Fermi level E
F
(surface of the
metal) absorbs energy equal to q
M
, it comes out of the metal to
the vacuum energy level E
vac
, called the vacuum level as shown in
Fig. 18. This electron is called the emitted electron and is free from
the bounding force of the metal. Fig. 18 depicts that q
M
= E
vac
E
F
is the minimum energy required to emit an electron since all states
all states above E
F
are empty. The work function of any metal is an
important quantity. As an instance, the work function q
M
of Al is
4.1eV and that of the Au is 5.0eV.
The Fermi level (E
F
= E
i
) of an intrinsic semiconductor lies just in
the middle of the forbidden gap of the semiconductor. There are no
electrons available at E
F
. The electrons that are emitted are usually
from the bottom of the conduction band. The energy difference
between the bottom of the conduction band (E
C
) and vacuum level
E
vac
is called the electron affinity (q in eV) of the semiconductor.
This is the minimum energy required to emit an electron from the
semiconductor as shown in Fig. 18. This is one of the important
properties of the semiconductor. The value of q for Si, as an
example, is 4.15eV. The work function of the semiconductor is not
fixed and depends on the position of the Fermi level and the
position of the Fermi level depends on the doping concentration.
The vacuum level E
vac
is taken as the reference level for drawing
the band diagram of any multi-material system.
The band diagram identifies the almost-empty conduction band
represented simply by a line that indicates the bottom of the
conduction band and is labeled as E
C
. Similarly, the top of the
valence band is indicated with a line labeled E
V
. It is important to
note that the actual band structures of semiconductors are more
complex than the reader is led to believe by the discussion. So,
semiconductors distinguish themselves from metals and insulators
by the fact that they contain almost-empty conduction band and
almost-full valence band. This also means that we will have to
deal with the transport of carriers in both bands.
To facilitate the discussion of the transport in the almost-full
valence band we will have to take into account of holes in a
semiconductor. It is important for the reader to understand that one
could deal with only electrons (since these are the only real
particles available in a semiconductor) if one is willing to keep track
of all the electrons in the almost-full valence band. The concept of
holes was introduced previously based on the notion that it is
easier to keep track of the missing particles in an almost full band,
rather than keeping track of the actual electrons in that band.
Holes are missing electrons. They have the same properties
as the electrons occupying the same states would have, except
that they carry an equal and opposite charge i.e. positive charge.
In the absence of an electric field no current can flow i.e. there
are no electrons at all in the conduction band and no holes in the
valence band to which electrons inside this band can move. A hole
was defined as an empty state in the valence band. The holes
move in the direction of the field (since they are positively charged
particles.

1.2 Temperature dependence of the energy band gap
A very rough estimate of temperature dependence of the energy
gap was given in article 1.3.1. A better estimate of the temperature
dependent formula for energy gap is given here. The energy band
gap of semiconductors tends to decrease as the temperature is
increased. This behavior can be understood if one considers that
the interatomic spacing increases when the amplitude of the
atomic vibrations increases due to the increased thermal energy.
This effect is quantified by the linear expansion coefficient of a
material. An increased interatomic spacing decreases the average
potential seen by the electrons in the material, which in turn
reduces the size of the energy band gap. A direct modulation of
the interatomic distance - such as by applying compressive
(tensile) stress - also causes an increase (decrease) of the band
gap.
The temperature dependence of the energy band gap, E
G
, has
been experimentally determined yielding the following expression
for E
G
as a function of the temperature, T:
E
G
(T) =
+
T
T
) 0 ( E
2
G (2.3.5)
where E
G
(0), and are the fitting parameters. These fitting
parameters are listed for germanium, silicon and gallium arsenide
in Table-1.3.
E
Table-1.3 Parameters used to calculate the energy band gap
of germanium, silicon and gallium arsenide (GaAs) as a
function of temperature.
Germanium Silicon GaAs
E
G
(0) (eV) 0.785 1.21 1.519
(meV/K)
0.470 0.470 0.541
(K)
255 636 204
A plot of the resulting band gap versus temperature is shown in
Figure 18 (a) for germanium, silicon and gallium arsenide.
Figure 18 (a) Temperature dependence of the energy band gap
Example
Calculate the forbidden energy band gaps in Germanium, Silicon
and Gallium Arsenide at 300, 400, 500 and 600K.
Solution :
E
G
(300K FOR Si) =
+
T
T
) K 0 ( E
2
G =
636 300
) 300 ( 10 x 470 . 0
21 . 1
2 3
+
=
0452 . 0 21 . 1
= 1.1648eV 1.2eV
E
G
(300K for Ge) =
255 300
90000 x 10 x 470 . 0
785 . 0
3
+
=
0762 . 0 785 . 0
= 0.7088
0.71eV
E
G
(300K for GaAs) =
204 300
90000 x 10 x 541 . 0
519 . 1
3
+
=
0966 . 0 519 . 1
= 1.422eV
The energy band gap for Germanium, Silicon and Gallium
Arsenide at different temperatures have ben calculated and are
given in Tabl-1.4. The plot of variation in intrinsic concentration of
carriers in Ge, Si, and GaAs semiconductors are shown in Fig.1.
Table-1.4 Band gaps for germanium, Silicon and gallium
arsenide at different temperatures
Germanium Silicon Gallium
Arsenide
T = 300K 0.71eV 1.20eV 1.42eV
T = 400K 0.62eV 1.1374eV 1.3757eV
T = 500K 0.6294eV 1.107eV 1.327eV
T = 600K 0.586eV 1.0731eV 1.277eV
1.3 Free Carriers
The key parameters for Si at 300K are
intrinsic carrier concentration: i
n
=
3 10
cm / 10
and
typical doping level range: N
D
, N
A
=
15
10
to
3 20
cm / 10
If the donor concentration in an n-type material is N
D
=
3 16
cm / 10
,
then we have
3 16
cm / 10
free electrons available because almost all
electron of the donor atoms supposed to have been donated. So
the majority carrier concentration (electrons) n =
3 16
cm / 10
. Where as
the minority carrier concentration (holes) p =
n / n
2
i = ( )
16
2
10
10 / 10 =
3 4
cm / 10
only. This gives an idea of level of concentration of majority
and minority carriers in a crystal. Similarly, if the acceptor doping
concentration (holes) as majority carriers is assumed to be = N
A
=
3 16
cm / 10
, then the minority carrier (electrons) would be =
3 4
cm / 10
.
The material has to be neutral; total negative charge must be equal
to the total positive charge. So, in an n-type material electrons are
negative charges. When the free electron goes away from the
donor atom, it becomes positively charged (ionized atom). So in an
n-type material total negative charges are due to electrons only
and total positive charges are due ionized donors and holes i.e.
n
= N
D
+p. In a p-type material holes are positive charges, negative
charges are due negatively ionized acceptor atoms and electrons
i.e.
p
= V
A
+n. What we have discussed so far is under thermal
equilibrium. If we have some other type of energy sources along
with the thermal equilibrium, then
np
=
2
i
n
is no longer valid. So we
have a non-thermal equilibrium.
The non-thermal equilibrium is dictated by some other forces
also. In such circumstances, then
np
>
2
i
n
, if we have injection
(generation) of carriers. On the contrary, if we are removing
(extraction) the carrier, the
np
<
2
i
n
. The minority carrier
concentration tends to return to thermal equilibrium value always. If
we have injection, the net recombination rate of holes =
recombination rate minus the generation rate i.e. in an n-type
material R
p
-G
p
=
Lifetime Hole
Holes Excess
=
p
D
2
i
N / n p
. Where,
p
= minority
charge carriers,
D
2
i
N
n
is thermal equilibrium value of the minority
charge carriers. Hence, the excess holes =
D
2
i
N
n
p
. The life time is
another important parameter of the semiconductor material. It is
the average time before which the carriers recombine. Similarly, in
a p-type semiconductor, the net recombination rate is given as R
n
-
G
n
= n A
2
i
/ ) N / n ( n [
=
Lifetime Electron
Electrons Excess
. We are talking only about
minority carriers recombination i.e. in n-type we talk about excess
holes and in p-type we talk about excess electrons. It is because
majority carriers concentrations are so large than the minority
carrier concentration does not make any difference. Where as the
minority carrier concentration changes a lot and so we always refer
to minority carrier concentration. For the carrier concentration to be
constant, the net injection rate is equal to the net recombination
rate. The amount of excess carrier is equal to the injection rate
multiplied by the life time.
This way a covalent bond is broken and electrons move to the
conduction band whereas the holes move to the valence band as
indicated in Fig.19. This process is known as electron-hole pair
generation, sometimes, intrinsic generation. Another way to
create (almost) free charge carriers is to contaminate a material
with impurities that occupy lattice sites in place of the atoms of the
pure semiconductor. This is called doping. The amount of doping,
the doping density or concentration is usually given as
impurities/cm
3
.
An acceptor-doped material where there are more holes than
electrons are called a p-type material. The majority carrier is the
most abundant carrier in a given semiconductor sample; electrons
in n-type and holes in p-type materials.
Figure 19 Thermal excitation of a semiconductor doped with donor
1.4 Intrinsic Semiconductor
A semiconductor material without any impurity is called an
intrinsic (pure) semiconductor. In this material obviously the
number of electrons in the conduction band must be equal the
number of holes in the valence band. This is due to the fact that
any electron in the conduction band has been excited there and
left a vacant state, a hole, in the valence band (electron-hole pair
generation). We usually refer to the density of electrons and holes
in the intrinsic semiconductor as i
n
and i
p
, respectively, and give
these values as carriers/cm
3
. It is important to note that i
n
and i
p
,
depend strongly on temperature since these densities are due to
electron-hole pair generation. A sample of pure Silicon (Si) or
Germanium (Ge), by dynamic equilibrium, has intrinsic
concentration i
n
= n = p that increases rapidly with increase in
temperature. The widely accepted expression for
) T ( n
i above 50K
are
) Si , T ( n
i =
T
7000
exp T 10 88 . 3
2 / 3 16

) Ge , T ( n
i =
T
4550
exp T 10 76 . 1
2 / 3 16

where, T is in Kelvin.
Example
Calculate the intrinsic concentration of carrier in Ge and Si
semiconductors at 300K, 400K and 500K..
) Si , T ( n
i =
T
7000
exp T 10 88 . 3
2 / 3 16

) K 300 , Si ( n
i =
300
7000
exp 300 10 88 . 3
2 / 3 16

=
) 3334 . 23 exp( 15 . 5196 10 88 . 3
16

=
11 16
10 x 3525 . 7 x 15 . 5196 10 88 . 3

=
11 19
10 x 3525 . 7 x 15 . 196 . 5 10 88 . 3

=
3525 . 7 x 15 . 196 . 5 10 88 . 3
8

Increasing T
Room temperature
E
C
E
D
T0K
E
V
E
V
E
V
E
D
E
D
E
C
E
C
E
d
e e e e e e e e
e e e e e e e
=
8
10 148
=
10
10 48 . 1
/cm
3

10
10 5 . 1
/cm
3

) K 398 , Si ( n
i =
398
7000
exp 398 10 88 . 3
2 / 3 16

=
8 3 16
10 x 2995 . 2 x 10 x 94 . 7 10 88 . 3

=
2995 . 2 x 10 x 94 . 7 10 88 . 3
3 11

=
2995 . 2 x 10 x 94 . 7 88 . 3
11
=
11
10 84 . 70
/cm
3
=
12
10 084 . 7
/cm
3
) K 300 , Ge ( n
i =
300
4550
exp 300 10 76 . 1
2 / 3 16

=
) 1667 . 15 exp( 10 x 19615 . 5 10 76 . 1
3 16

=
7 3 16
10 x 589 . 2 x 10 x 19615 . 5 10 76 . 1

=
589 . 2 x 19615 . 5 10 76 . 1
12

=
13
10 367 . 2
/cm
3

13
10 5 . 2
/cm
3

Intrinsic carrier concentration for most common semiconductors in
use is
) K 300 , Si ( n
) K 400 , Si ( n
i
i
=
)
300
1
400
1
( 7000 exp{ )
300
400
(
2 / 3

=
)
300 x 400
400 300
( 7000 exp{ )
300
400
(
2 / 3

=
)
3 x 4
70
exp( 54 . 1
=
) 833 . 5 exp( 54 . 1
=
38 . 341 x 54 . 1
= 525.727
) K 400 , Si ( n
i =
2 10
10 x 2573 . 5 x 10 5 . 1
=
12
10 886 . 7
/cm
3

) K 500 , Si ( n
i =
)}
300 x 500
500 300
( 7000 exp{ )
300
500
( 10 x 5 . 1
2 / 3 10

=
)
3 x 5
2 x 70
exp( 152 . 2 x 10 x 5 . 1
10
=
) 333 . 9 exp( 152 . 2 x 10 x 5 . 1
10
=
4 10
10 x 13 . 1 x 152 . 2 x 10 x 5 . 1
=
14
10 x 65 . 3
/cm
3

) K 300 , Ge ( n
) K 400 , Ge ( n
i
i
=
300
)} 1
400
1
( 4550 exp{ )
300
400
(
2 / 3

=
)
3 x 40
455
exp( 54 . 1
=
) 792 . 3 exp( 54 . 1
=
345 . 44 x 54 . 1
= 68.29
) K 400 , Ge ( n
i =
29 . 68 x ) K 300 , Ge ( n
i =
29 . 68 x 10 5 . 2
13
=
15
10 707 . 1
/cm
3

) K 500 , Ge ( n
i =
)}
300
1
500
1
( 4550 exp{ )
300
500
( x ) K 300 , Ge ( n
2 / 3
i

=
)
3 x 50
2 x 455
exp( 152 . 2 x 10 x 5 . 2
13
=
) 066 . 6 exp( 152 . 2 x 10 x 5 . 2
13
=
95 . 430 x 152 . 2 x 10 x 5 . 2
13
=
16
10 x 39 . 1
/cm
3

Material E
G
[eV] n
i
[cm-3]
Si 1.12 1.5x10
10
Ge 0.71 2.5x10
13
GaAs 1.42 1.79x10
6
Typical values of i
n
are:
3 6
cm 10 2
in GaAs,
3 10
cm / 10 5 . 1
in Si and
3 13
cm / 10 5 . 2
in Ge at room temperature i.e. T = 27+273 = 300K. Let
us compare Si with, 5x10
22
atoms/cm
3
and with 4-bonds (valence
electrons) per atom which yields a total of 5x10
22
x4 =
2x10
23
valence electrons/cm
3
. How many covalent bonds can be
broken with a i
n
of
3 10
cm / 10 5 . 1
? It is equal to 2x10
23
/1.5x10
10
=
1.33x10
13
. It means less than one (1/1.33 = 0.75) bond in 10
13
can
be broken in Si at room temperature. When impurities are added,
by doping, to a semiconductor it is said to be an extrinsic
semiconductor. At equilibrium the extrinsic semiconductor is said
to have carrier concentrations i
n
and o
p
, both different from i
n
. At
T = 0K, o
n

= o
p
= i
n
= 0. This state is sometimes called freeze-out.
We know that at room temperature pure silicon and germanium
behave as an insulator. Thus, the material to be used in fabrication
of semiconductor devices must be tailored electrically to produce
enough charge carriers (electrons or holes). The crystalline piece
of semiconductor is doped with atoms of neighboring group of
elements. Generally the pure crystalline structure has poor current
carrying capabilities. A very small amount of impurity (
13
10 : 1
impurity/ host) is added with an atomic structure that is different
than that of the host atom to obtain free charge carriers (electrons
or holes). This impurity will modify the crystal lattice such that
current carriers are developed. Usually the amount of impurities
added are in very small proportion to that of the host, but even this
small amount is sufficient to unbalance the number of holes and
electrons in the host materials.
The acceptor atom when mixed as impurity accepts electrons
and creates holes. A semiconductor material made after adding
the acceptor atoms is called p-type material. Similarly, each donor
atom donates one free electron. A semiconductor material made
after adding the donor impurity is called n-type material. Each
donor atom introduces one electron in the conduction band without
introducing hole in the valence band. The approximate
conductivity, concentration and resistance of conductors,
semiconductor, and insulator are shown in Fig.20.
The conductivities of a few commonly used conductors,
semiconductors, and insulators are shown in Table-1.5.
Table-1.5 Classification of materials based on conductivity,
resistivity, and bandgap
Material Conductivity
(S/m)
Resistivities
(m)
Band
gap
Classifica
tion
Silver
Copper
Aluminum
7
10 25 . 6
7
10 88 . 5
7
10 82 . 3
8
10 6 . 1

8
10 7 . 1

8
10 7 . 2

Overla
ps
Conducto
r
Germaniu
m (Ge)
Silicon (Si)
1.5
4
10 5

0.60
3
10 2
0.7eV
1.2eV
Semicon
ductor
Porcelain
Glass
Hard
Rubber
10
10 33 . 3

12
10 88 . 5

16
10
9
10 3
11
10 7 . 1
16
10
Very
large
Insulators
On one hand good conductors have high conductivity while on
the other, good insulators have high resistivity. The word
semiconductor has been derived from two woods
Semiconductor = Semi + conductor
Semi means half and conductor means metal. Thus,
semiconductor is a material whose behaviour is halfway between
that of metal and insulator. So its resistivity and the resistance fall
half way between metal and insulator. Important physical
properties of both Si and Ge semiconductors are given in Table
1.6.
Increasing resistively
3
10 2
11
10
8
10 7 . 1

Increasing conductivity
Glass Si
Ge
Cu
0.6
(m)
10
-3
10
5
Metal Semiconductor Insulator
Resistance
10
21
/cm
3
10
6
/cm
3
Metal Semiconductor Insulator
Concentration

Figure 20 Resistivity and conductivity of few conductors,
semiconductor, and insulator
Table 1.6 Physical properties of Silicon and Germanium
Ge Si
Atomic Number 32 14
Density gm/m
3
6
10 32 . 5
6
10 33 . 2
Atomic Weight gm/gm.atom 72.6 28.09
Relative Dielectric Constant 16 12
Atomic Density (atoms/m
3
)
28
10 41 . 4
28
10 5
E
G
(eV) at 0K 0.785 1.21
E
G
(eV) at 300K 0.72 1.1
Intrinsic Concentration
3
i
m / n
at 300K
19
10 5 . 2
16
10 5 . 1

Intrinsic Resistivity at
300K(m)
0.45 2300
n
(m
2
/V.s) at 300K
0.38 0.135
p
(m
2
/V.s) at 300K
0.18 0.048
D
n
(m
2
/s) at 300K 0.01 0.0035
D
p
(m
2
/s) at 300K 0.0045 0.0012
Electronic Charge (q)
19
10 6 . 1

Coulomb
Electronic Mass (m
eo
) at rest
kg 10 108 . 9
31
Effective mass of electron

eo
*
e
m / m
in Si
0.92
Effective mass of hole eo
*
h
m / m

in Si
0.49
Proton Mass (m
po
) at rest
kg 10 672 . 1
27
Neutron Mass (m
n
) at rest
kg 10 672 . 1
27
Permittivity of free space (

o
)
m / F 10 854 . 8
12
Permeability of free space

(
o
)
m / H 10 4
7

Boltzman Constant (
k
)
K / J 10 38 . 1
23
=
K / eV 10 62 . 8
5
Plank's Constant (h)

Js 10 625 . 6
34
.
Avogadro Number
23
10 025 . 6
per gm mole
1 =
m 10
10
, 1mil =
inch 10
3
,
1micron=
m 10
6
, Si atom
size = 0.24nm
1eV =
J 10 6 . 1
19
.
Velocity of Light =
s / m 10 3
8
Universal Gas Constant (R)

= 8.314J/K.mol
Thermal voltage = KT/q =
T/11600 =25.9mV
The Ge/Si element lies in the fourth group of periodic table and
hence is known as tetravalent element with 4-valence electrons in
the outermost orbit. These valence electrons are responsible for
the chemical behavior of the elements. There is a general
tendency of element to enter into chemical reaction in such a way
as to acquire a stable state. According to the rule of octet, the
most stable state of matter contains 8-electrons in the outermost
orbit. Hence, each atom of Si/ Ge crystal has 8-electrons, 4-of its
own and the rest 4-shared by neighbouring atoms in its outermost
orbit. Useful assorted semiconductor elements are shown in the
Table-1.4 with atomic numbers and Weights.
Dopants
The materials chosen as suitable dopants depend on the atomic
properties of both the dopant and the material to be doped. In
general, dopants that produce the desired controlled changes are
classified as either electron acceptors or donors. A donor atom
that activates (that is, becomes incorporated into the crystal lattice)
donates weakly-bound valence electrons to the material, creating
excess negative charge carriers. These weakly-bound electrons
can move about in the crystal lattice relatively freely and can
facilitate conduction in the presence of an electric field. (The donor
atoms introduce some states under, but very close to the
conduction band edge. Electrons at these states can be easily
excited to the conduction band, becoming free electrons, at room
temperature.) Conversely, an activated acceptor produces a hole.
Semiconductors doped with donor impurities are called n-type,
while those doped with acceptor impurities are known as p-type.
The n and p type designations indicate which charge carrier acts
as the material's majority carrier. The opposite carrier is called the
minority carrier, which exists due to thermal excitation at a much
lower concentration compared to the majority carrier.
For example, the pure semiconductor silicon has four valence
electrons. In silicon, the most common dopants are commonly
known as Group-III and Group-V elements. Group-3 elements all
contain three valence electrons, causing them to function as
acceptors when used to dope silicon. Group 5 elements have five
valence electrons, which allow them to act as a donor. Therefore, a
silicon crystal doped with boron creates a p-type semiconductor
whereas one doped with phosphorus results in an n-type material.
The useful dopants are arranged in Table-1.7.
Table-1.7 Periodic Table of useful Dopants
Atomic III-
group Atomic
Number
Weight
Atomic IV-
group Atomic
Number
Weight
Atomic V-
group Atomic
Number
Weight
5 Boron (B
10.82
6 Carbon
(C)
12.01
7 Nitrogen
(N)
14.008
13 Aluminum
(Al)
26.97
14 Silicon
(Si)
28.09
15
Phosphorus (P)
31.02
31 Gallium
(Ga)
69.72
32 Germanium
(Ge)
72.6
33 Arsenic
(As)

74.91
49 Indium
(In)
114.8
50 Tin (Ti)

118.7
51 Antimony
(Sb)
121.
8
Mobilities of electron and hole in intrinsic silicon and germanium
at room temperature having the dimensions in m
2
/V.s are
Mobility
) s . V / m (
2
n
) s . V / m (
2
p
Si 0.13 0.048
Ge 0.38 0.18
Why n-type material is mainly used in manufacturing the
semiconductor devices? The answer to this question becomes
clear from the mobilities of electrons and holes in Si. The mobility
of electrons (
n
= 0.13) in Si is roughly two and half times more
than that of holes (
p
= 0.048). This is the main reason why n-type
semiconductor is preferred over p-type for manufacturing the
semiconductor devices is high mobility of the electron w.r.t. hole.
The atoms of semiconductors like Ge and Si are arranged in an
orderly manner known as crystal lattice. We know that the typical
value of concentration of charge carriers in intrinsic Si and Ge at
room temperature (300K) are i
n

3 10
cm / 10 5 . 1
and
3 13
cm / 10 5 . 2

respectively.
The difference in concentration of Ge and Si materials is
primarily due to energy gap. Band gap for germanium is
approximately 0.72eV while for Silicon is 1.12eV. The factor of
energy gaps of Ge is almost 2-times that of the Si that results into
factor of almost
10 13
10 5 . 1 / 10 5 . 2
=
3
10 7 . 1
of the free carrier
concentration between them. Hence, less number of electron hole
pairs will be generated in Si than in Ge at any temperature. This
way the conductivity of Si will be lesser than that of Ge at room
temperature.
1.5 Tetravalent Elements
We know that valence electrons provide the cohesive force of
binding among different atoms of Si/ Ge crystal. Since the bond is
shared by valence electrons, it is named covalent bond as shown
in Fig.21. The covalent bond results from the sharing of a pair of
the valence electrons of anti parallel spin between neighboring
atoms (similar or dissimilar).
Electrons in the outermost orbit (valence shell) are bound by
valence bond having no free charge carriers either in Si or Ge at
absolute zero temperature (0K= -273
o
C). So electrons of Si/ Ge
are unable to take part in the conduction process and hence
behave as insulators. The energy levels of these valence electrons
are called valence band and in this band electrons cannot conduct
any current. If sufficient amount of energy is supplied under the
influence of external factors (such as temperature, light etc.), the
covalent bond may break open and electron-hole pair may be
generated as shown in Fig.22.
The band diagram representation of the intrinsic Si/ Ge is
illustrated in Fig.23. These free electrons under the influence of
electric field move in a particular direction and conduct current.
The band in which these free elections move is known as
conduction band. Different semiconductor elements require
different amounts of energy to break open the covalent bond. The
same situation can be explained using Figs.23.
The Si and Ge atoms approximately require 1.12eV and 0.72eV
energy to break open their covalent bond and elevate electrons
from valence band to conduction band. An electron in the valence
band (filled band) jumps to the conduction band (empty band) as
soon as it receives the amount of energy required overcoming the
energy gap. Each electron transferred from valence band to
conduction band leaves behind a vacancy similar to a physical
hole. This physical hole is associated with a positive charge equal
but opposite to the charge of an electron. This comes from the
principle that electrons and holes are generated in pairs as shown
in Fig.22. Therefore the concentration of free electrons and holes
will always be equal in an intrinsic semiconductor.
Figure 21 Covalent bonds in intrinsic Si
Figure 22 Thermal generations of electron-hole pair

Figure 23 Intrinsic semiconductor energy bands
1.6 Effect of Temperature on Intrinsic Conductivity
The conductivity of intrinsic semiconductor increases with
increasing temperature. So resistivity decreases with increase in
the temperature. Thus, semiconductors have negative temperature
coefficient of resistance. Electrons experience a force of attraction
towards the positive terminal and holes experience a force of
attraction towards negative terminal of the battery as shown in
Fig.24. Thus, there is net movement of particular type of charge
carriers in a particular direction. This is called drift. The electrons
drift in conduction band where as the holes drift in the valence
band. Hence, both electrons and holes have drift velocities.
The process of covalent bond rupture caused either by
increasing temperature or supply of light energy accompanied by
Conduction band
Valence band
Inner bands
E
G
= Band gap
Mobile Electrons
Mobile Holes
E
C
E
V
E
Fi
formation of a free electron and a free hole is called electron-hole
pair generation. The number of free electrons is equal to the
number of free holes and the semiconductor in which number of
free electrons is equal to the number of free holes is called an
intrinsic semiconductor. Intrinsic semiconductor is the trade name
of pure semiconductor. The whole crystal structure of the intrinsic
semiconductor is just the replica of one section of the structure.
So, in an intrinsic semiconductor, conduction is due to the two
separate and independent particles carrying equal and opposite
charges and drifting in opposite directions under the influence of
applied electric field. Thus, in an intrinsic semiconductor, the
probability of getting electrons and holes is equal i.e. 50:50.
Hence, Fermi-level in intrinsic semiconductor lies in the middle of
energy gap ( G
E
) as depicted in Fig.24.
Figure 24 Forces due to drift of holes and electrons
1.7 Conduction of Current
The conduction of current in a semiconductor is controlled by 2-
methods, namely, Drift, and Diffusion.
1.7.1 Drift Current
Fig. 25 is the sample of a crystal lattice wherein the free electrons
move randomly in Brownian motion. They collide with the
stationary atoms and get reflected in a different direction. The
average distance traveled between 2-collisions is called the mean
free path. In absence of any electric field, no net movement in any
direction takes place as shown by solid arrows in Fig.25.
With application of the electric field, electrons get accelerated in
opposite direction of the electric field and the net movement takes
place in a particular direction as indicated by arrow from left to
right. This movement is called drift. Similarly, holes drift in the
I
V
e
e
h
h
same direction of the electric field. The drift velocity of the carriers
is proportional to the electric field and related as
v E
or, v = E
(1.7.1)
Here, the proportionality factor is the mobility of carriers =
m
V
/
s
m
=
s . V
m
2
(1.7.2)
The motion of charge particles constitutes electric current. Total
charge of moving electrons in a volume (Ax) of the semiconductor
is
Q = (nqAx) (1.7.3)
where, A = cross sectional area perpendicular to the flow of
electrons
n = number of electrons/volume
x = length of semiconductor in the direction of flow of
electrons
q = electric charge of an electron =
C 10 6 . 1
19
Thus, current due to drift of electron as shown in Fig.1.18a can be

expressed as
n
I

=
dt
dQ
=
dt
) Axnq ( d
=
,
_
dt
dx
) nq ( A
= n
Aqnv
=
E Aqn
n
(1.7.4)
Electron drift current density = n
J

=
A
I
=
A
E Anq
n
=
E nq
n
=
E
n

(1.7.5)
where, n
= conductivity = n
nq
(1.7.6)
Direction of electric field E
e
e
e
e
e
e
e
e
e
e
e
Figure 25 Brownian motion of electrons
In a semiconductor both electrons and holes act as charge
carriers and hence current density of drifting holes as shown in
Fig.26 can be expressed as
p
J
=
E nq
p
=
E
p
(1.7.7)
where,
p
=
p
nq
= conductivity of p-type semiconductor (1.7.8)
Figure 26 Drift of electrons and holes
Now total current densities due to drifting electrons and holes as
indicated in Fig.27 is
J
= n
J
+
p
J
=
E nq
n
+
E nq
p
=
E
n
+
E
p
=
E
(1.7.9)
Hence conductivity of the material = n
+
p
=
) p n ( q
p n
+

(1.7.10)
Figure 27 Drift of holes and electrons
The intrinsic conductivity of pure Ge at room temperature is
approximately 2.2mho/m, while the pure Si may approach as low
as
m / mho 10 4
4
. The resistivity is
=
1
=
) p n ( q
1
p n
+
(1.7.11)
I
V
I
V
h h
h
e
e
e
I
V
h e
e h
Resistance of the material R =
A
x
=
)
p
p
n
n ( Aq
x
+
(1.7.12)\
Example
Find the resistivity of intrinsic silicon of length 3mm and rectangular
cross section of 50x100m. When it is doped with a pentavalent
impurity of one impurity atom for each 60-million silicon atoms,
calculate the resistivity of the doped silicon. Given that number of
silicon atoms =
3 28
m / 10 5 . 4
, intrinsic carrier concentration = i
n
=
3 16
m / 10 5 . 1
, electron mobility
n
=
Vs / m 135 . 0
2
and hole mobility
p
=
Vs / m 048 . 0
2
.
What would be the intrinsic and extrinsic resistances offered by the
silicon bar?
Solution:
The conductivity of the intrinsic silicon =
i
= (
n
+
p
)n
i
q
i
=
( ) 048 . 0 135 . 0 10 6 . 1 10 5 . 1
19 16
+

=
183 . 0 10 4 . 2
3
=
m / mho 10 439 . 0
3

Intrinsic resistivity =
i
=
i
1
=
392 . 4
10
4
=
m 10 3 . 2
3

R
i
=
A
L
i
=
12
3
3
10 x 100 x 50
10 x 3
10 3 . 2

=
5
10 x 3 x 3 . 2
9
=
9
10 x 38 . 1
If N
D
= number of donor atoms =
6
i
10 60
n
=
6
28
10 60
10 5 . 4

=
3 20
m / 10 5 . 7
Hence, number of free electrons (each atom gives one electron) =
n
n
= D
N
=
3 20
m / 10 5 . 7
Number of hole =
n
2
i
n
n

=
20
2 16
10 5 . 7
) 10 5 . 1 (
=
3 11
m / 10 3
Thus, the number of holes (=
3 11
m / 10 3
) in the doped semiconductor
are much smaller the numbers of electrons (=
3 20
m / 10 5 . 7
). Hence,
the semiconductor is n-type and its conductivity will be
n
= qn
n
n
=
135 . 0 10 5 . 7 10 6 . 1
20 19

= 0.162/m
Hence, resistivity of the doped semiconductor =
n
=
n
1
=
162
1000
=
6.192m
R
n
=
A
L
n
=
12
3
10 x 100 x 50
10 x 3
192 . 6

=
6
10 x 5
192 . 6 x 3
6
10 x 7 . 3
1.7.2 Recombination/ Generation and Lifetime
Carriers (electrons and holes) are generated in pairs due to
thermal agitation or absorption of radiation. Some of the free
electrons come near the empty covalent bond (hole) and fall into it.
For such recombination momentum of electrons and holes must be
conserved. Since the momentum after collisions is zero, it is
imperative that electrons and holes must be moving with equal and
opposite velocities before collisions. Such direct method of
recombination is a rare possibility and hence very less amount of
recombination takes place. Normally recombination takes place
through traps or recombination centers. There are atoms having
energy states in the forbidden band. Such condition exists due to
imperfection. This condition can also be created artificially by
doping the material with impurity atoms. These traps act as third
body in collision and absorb the residual momentum. Thus the rate
of recombination increases. The Lifetime of carriers denoted by
n
(tau) is defined as the average time elapsed between two
generations or recombination. In fast recovery diodes, the lifetime
of the carrier is desired to be very low and this is achieved by
doping with gold.
Force on an electron = -qE
The minus sign is due to negative charge of electron.
F =
q
KT
=
f m
e (1.7.13)
Acceleration = f =
dt
dv
n
=
2
2
dt
x d
=
e
m
qE
(1.7.14)
where, e
m
= effective mass of electron.
Hence, on integrating, electron velocity comes out to be
n
v
=
dt
dt
dv
n
=
dt
e
m
qE
=
n
e
m
qE
=
E
n
(1.7.15)
n
=
q
m
n n
(1.7.16)
Eqn.(1.7.16 ) indicates that the mobility of electron is related
with its lifetime (
n
). The lifetime of the carrier is defined as the time
taken by the carrier to transfer the energy from excitation to
recombination. The motion of the carriers in semiconductor is
retarded by lattice vibration, impurities and crystal defects. At
higher temperature carriers are scattered randomly by lattice
vibrations, while at lower temperature Coulomb interaction
between carriers and impurity ions dominates over scattering
mechanism. When an electric field is applied across it, electrons
and holes are accelerated in opposite directions till they are
scattered by impurities or phonons. Special feature of this
interaction or collision is that the velocity of carriers after each
encounter becomes essentially random once again and their
acceleration in the direction of the field starts afresh. Thus for
moderate field intensities carriers never go far off from the thermal
equilibrium. In spite of the presence of electric field, the carrier
energy remains essentially thermal.
Example
If the electron mobility
n
=
Vs / m 135 . 0
2
, m
n
= 9.108x10
-31
kg, what
would be duration of the life of electron?
n
=
q
m
n n
=
19
31
10 x 6 . 1
135 . 0 x 10 x 108 . 9
= 0.768x10
-12
secons = 0.768psecons
1.7.3 Diffusion Current
Let us consider a sample semiconductor material in Fig.28 wherein
concentration of carriers (holes) is much high on the left side of the
surface YY. Since the electrons (holes) are in Brownian motion,
they cross over the surface YY randomly from left to right and right
to left. But as the concentration of electrons (holes) on the left side
is much higher than on the right side, the average number
electrons (holes) going from left to right is more than number of
electrons (holes) going from right to left of the surface YY. This
constitutes a net flow of electrons (carriers) from left to right i.e. a
net current flow due to flow of holes from left to right. This is called
diffusion current because it results from diffusion of carriers. Thus
diffusion current is due to flow of charge carriers from higher
concentration to lower concentration region.
In order to understand the diffusion process of carriers in
semiconductors, let us consider one-dimensional variation of
carrier concentration shown in Figure 1.19. Here, the temperature
has been assumed to be constant so that the thermal velocity of
carriers remains independent of the distance x. If we assume that
the length (L) in Figure 1.19 is the mean free path (L = n n
v

=
distance between two collisions), then on the average one half of
electrons moving from right to left at x = L and one half of
electrons moving from left to right at x = +L will cross the x = 0
plane at any instant of time. To calculate current, we determine the
net flow of electrons per unit time per unit area crossing the plane
at x = 0.Then, net flow of electrons passing x = 0 plane from left to
right in one mean free path per unit area is given as
n
(x) = n n
v ) L x ( n
2
1
v ) L x ( n
2
1
+
=
)} L x ( n ) L x ( n { v
2
1
n
+
(1.7.17)
Expanding Eqn. (1.7.17) using Taylor series keeping only first two
terms gives electron concentration in the neighborhood of x = 0
plane as
n
(x) =
}] L
dx
) x ( dn
) x ( n { } L
dx
) x ( dn
) x ( n [{ v
2
1
n
+
=
L
dx
) x ( dn
v
n
=
L
dx
) x ( dn path free one
n

=
L
dx
) x ( dn L
n
=
dx
) x ( dn L
n
2
(1.7.18)
The quantity
n
2
L
having dimension of cm
2
/s is called diffusion
constant D
n
. Negative sign indicates net flow of electrons due to
diffusion is in the direction of decreasing electron concentration.
Thus, electron flux density of Eqn. (1.7.18) can be simplified as
) x (
n
=
dx
) x ( dn
D
n
(1.7.19)
Similarly, the holes flux density due to concentration gradient can
be shown as
) x (
p
=
dx
) x ( dp
D
p
(1.7.20)
Now the diffusion current crossing a unit area (current density) is
the electron flux density multiplied by unit charge of electron (-q) as
) x ( J
n =
) q (
dx
) x ( dn
D
n

=
dx
) x ( dn
qD
n (1.7.21)
Similarly, the hole current density is
) x ( J
p
=
) q (
dx
) x ( dp
D
p
+
=
dx
) x ( dp
qD
p
(1.7.22)
Hence, total electron current density due to drift and diffusion is
J
= ) Diffusion ( n ) Drift ( n
J J +

E qn
n
+

,
_
dx
dn
qD
n =
'
,
_
,
_
+
dx
dn D
nE q
n
n
n (1.7.22)
Figure 28 Electron concentrations versus distance
These charges can be assumed as gas molecules. For gas in
steady state
P = nKT (1.7.23)
where, P = pressure
n = concentration or number of gas molecules at STP=
4 . 22
10 023 . 6
26
=
25
10 689 . 2
molecules/m
3
The pressure gradient that provide force to expand the gas is
dx
dP
=
,
_
dt
dn
KT
= n
KTv
(1.7.24)
The excess density of electrons in a semiconductor with non-
uniform excess electrons at a certain distance from the injecting
point is n/m
3
. The internal electric field at the same point E
provides the force on one electron = qE and hence force on n
electrons
= nqE =
n
n
v
nq
(1.7.25)
Dividing n
J
=
'

dx
) n ( d
qD
n by n
yields
n
n
J
,
_
n
n
qD
'

dx
) n ( d
= n
v
qD
n
n
,
_
(1.7.26)
This force corresponds to the pressure gradient of gas = force of
the charge gradient and hence Eqns.(1.7.22) and (1.7.26) are
similar. Thus,
Y
Y

Slope =
dx
dp
=
dx
dn
KT =
n
n
qD
(1.7.27)
n
n
D
=
q
KT
= T
V
(1.7.28)
Similarly, for holes
p
p
D
=
q
KT
n
n
D
=
q
KT
= T
V
=
p
p
D
(1.7.29)
At room temperature (300K) T
V
=
q
KT
=
11600
T
= 0.0259V (1.7.30)
Hence, total hole current due drift and diffusion of holes is
I
p
=
'
,
_
,
_

dx
dp
D
pE Aq
p
p
p =
'
,
_
,
_

dx
dp
q
KT
pE Aq
p (1.7.31)
Similarly, total electron current due to drift and diffusion is
I
n
=

,
_
+
dx
dn D
nE Aq
n
n
n =
'
,
_
,
_
+
dx
dn
q
KT
nE Aq
n (1.7.32)
Another definition required for the explanation of energy band
diagram of solids is Fermi level that is indicated in Fig.29. Fermi
level (intrinsic) of any solid may be regarded as the reference
energy level from which all other energies may be measured
conveniently.
Example
Obtain the diffusion current density in a piece of Germanium
having concentration gradient of 1.5x10
22
electrons/m
2
and D
n
=
0.0012m
2
/s.
J
n
=
dx
dn
qD
n =
22 19
10 x 5 . 1 x 0012 . 0 x 10 x 6 . 1

=
3
10 x 5 . 1 x 0012 . 0 x 6 . 1
=2.88A/m
2

Figure 29 Intrinsic semiconductor energy band
Inner bands
Conduction band
Valence band
E
C
E
F
E
V
E
G
1.8 Extrinsic Semiconductor
The extrinsic semiconductor is the trade name of impure
semiconductor and they are of two- types; namely, n-type, and p-
type. After mixing very small amount (say 1 part in
8
10 ) of impurity
atoms either from pentavalent element or from trivalent elements in
pure Silicon or Germanium atoms, the resulting Si or Ge crystal is
said to be doped semiconductor and the process of making a pure
semiconductor an impure one is known as doping.
1.8.1 n-type material
If extremely small amount of impurity from pentavalent elements
(Antimony, Phosphorus or Arsenic) is added to intrinsic Si or Ge at
the time of crystal growing and the semiconductor is allowed to
crystallize so that the lattice remain intact, then each Arsenic atom
displaces one of the Silicon atoms from its regular site and forms
covalent bonds with the neighboring silicon atoms as shown in
Fig.30, using 4-of its 5-valence electrons.
The fifth electron of the pentavalent element does not fit into 4-
bonds and so left loose dangle to the arsenic atom with relatively
low binding energy. In fact, with the addition of just little thermal
energy these electrons can break to become free to move about in
the crystal. This excess electron constitutes a mobile, localized
negative charge. The Arsenic atom, on the other hand, is an
immobile, localized positive charge. This way the Arsenic impurity
atom adds an excess electron to the crystal but does not disturb
the overall electrical neutrality of the crystal.
In addition to the existing large number of free electrons, a few
thermally generated electron-hole pairs also exist. The thermally
generated electron joins the large numbers of free electrons
forming the majority carriers whereas the thermally generated
holes, here, becomes the minority charge carrier. Thus, it can
be inferred that conduction can be controlled by proper addition of
impurity. As the conduction is now mainly due to negative charge
carrier, the semiconductor is said to be n-type.
In an n-type semiconductor, the conduction is primarily due to
electrons and hence electrons are in majority and holes can be
thought of in minority.
The n-type semiconductor is neutral, because by way of doping
(pentavalent atoms) equal amount of negative charge (in the form
of electrons) and positive charge (in the form of protons) have
been made available as illustrated in Fig.30. In Fig.31 only one
hole and four electrons are deliberately shown to indicate minority
and majority carriers respectively, though this proportion of
majority and minority is much smaller here than the exact position.
Figure 30 Tetravalent crystal with one As atom
The presence of donor impurity from the point of view of energy
level diagram creates extra-localized energy levels just below the
conduction band as shown in Fig.32. Such level is termed as donor
level and the gap between this level and conduction band is
approximately 0.01eV and 0.05eV for Ge and Si respectively. The
estimation of D
E
can be done similar to the hydrogen atom. The
5th electron is bound to the parent atom of the impurity only
without forming any covalent bond. Here slight difference does
exist as the electrons move in the solid. This suggests the effective
mass of the electron (
*
e
m
) should be used in place of the rest mass.
The relative permittivity of the semiconductors must be taken into
consideration. Under such conditions, the excitation energy D
E
is
expressed as
E
D
=
,
_
m
m
*
n
6 . 13 eV
1
2

,
_
(1.8.1)
Very small amount of Arsenic/ Antimony greatly increases the
number of conduction electrons in addition to intrinsically available
electrons and holes. Hence, concentration of electrons in
conduction band exceeds the concentration of holes in the valence
band. This is the reason why Fermi level of n-type material shifts
upward (towards the bottom of conduction band) as demonstrated
in Fig.32.
The experimental values of donor levels created by various donors
in Ge are given as
As = 0.0127eV, P =
eV 012 . 0
, Sb = 0.0097eV

Figure 31 Current due to electron (majority carriers)
Example
Calculate excitation energy required to excite the electrons from
donor levels to the conduction band in a Si crystal having effective
mass
*
e
m
= 0.26m
e
and relative permittivity
r
= 12.
D
E
=
2
) 12 ( 26 . 0 6 . 13

= 0.025eV
Conduction band
E
d
= Binding energy
Valence band

E
C
E
D
E
F
E
Fi
E
V
E
G
e e e e e
I
n
V
I
P

O
n-type
e
h
e
e
h
e
e
Figure 32 Band diagram in n-type
Representation of n-type material
Fig.33 shows an n-type semiconductor with large number of
electrons (majority charge carriers), a few holes (minority charge
carriers) and sufficiently large number of immobile positive ions.
These types of charges are represented with the symbols shown in
legends and it will be followed in the discussion. Electrons, holes
and immobile ions are represented by black circle, white circle and
encircled positive sign respectively as shown in Fig.33.

Figure 33 Representation of n-type material
1.8.3 p-type semiconductor
The opposite effect to that of the pentavalent element occurs if
trivalent impurity such as Boron, Aluminum, Gallium, and Indium
(having 3-valence electrons per atom) is added to an intrinsic Ge
or Si. Since the impurity ratio is 1 part in 10
6
, the host Silicon or
Germanium atoms surround each impurity atom, all around. The
trivalent atom has only 3-valence electrons. These valence
electrons from Boron (trivalent impurity) atom does not have the
fourth electron, the fourth bond cannot be completed as depicted in
Fig.34.
The thermal energy at room temperature is sufficient to provide
this energy so as to complete the covalent bonds around the
trivalent atoms available. This process of filling a vacancy around
the trivalent atom creates a vacancy of one electron in that
Legends
Electron
Hole
Ionized donor
e
h
Thermally generated electron-hole

e e e e e e e

e e e e e e e
e
h
particular position from where the electron must have jumped as
shown in Figs. 34 and 35. The vacancy of electron must have
opposite charge i.e. positive and since it goes on moving from one
atom to another, there is creation of a movable (mobile) positive
charge named hole.
The Boron atom becomes an immobile, localized negative
charge as in Fig.35. Since the holes are created in the valence
band and in p-type material concentration of holes in valence band
are much more than concentration of electrons in conduction band,
the Fermi level lies immediately above the binding energy E
A
as
indicated in Fig.36.
The binding energies E
A
for various acceptors in Ge are
B = 0.0104eV, Al = 0.102eV, Ga = 0.0108eV, In =
0.0112eV

Figure 34 Si crystals with one B-atom

Figure 35 Incomplete bond representation and movement of hole
In Silicon these values are of the order of 0.046eV.
An extra-localized energy level of acceptor element on the
energy level diagram is termed acceptor levels in the forbidden
gap just above the top of the valence band as shown in Fig.36.
Electrons that occupy these levels leave behind then, in the
valence band, vacancies (holes) that permit the electric current to
flow. The estimation of excitation energy can be done using Eqn.
(1.8.1) (assuming the effective mass of hole in the Si crystal to be
0.33m) as
E
A
=
2
) 12 ( 26 . 0 6 . 13

= 0.0312eV
Figure 36 Band diagram of p-type material
Once again we note that though p-type semiconductor has
excess of holes for conduction processes, as a whole it is
electrically neutral. Only one electron in Fig.37 has been shown
deliberately to indicate the minority carrier and 4-holes to represent
the majority carriers, though this proportion of majority and minority
is much smaller than the exact position.
Conduction band
Ionized acceptor
Holes
E
a
=E
A
-E
V
Valence band
E
C
E
V
E
F
E
G
E
A
E
Fi

h
h h h h
I
n
V
I
P

O
p-type
h
e
h
e
e
e
e
Figure 37 Drift current due to holes (majority carriers)
1.8.4 Representation of p-type material
Fig.38 shows p-type semiconductor with white circles representing
holes, black dot representing electrons and encircled negative sign
representing immobile ions. In a p-type semiconductor holes are
majority charge carriers and there are a few thermally generated
electron-hole pairs. Hence, the thermally generated holes join with
the already present large number of free holes giving the picture of
majority carriers as holes and few thermally generated minority
charge carriers as electrons as depicted in Fig.38.
Figure 38 Two-dimensional view of p-type semiconductor
1.8.5 Carrier Concentration
The concentration of one type of carriers can be increased by
mixing desired type of impurity. This phenomenon gives the idea of
majority and minority charge carriers. A typical dose of phosphorus
might be N
D
= 10
16
atoms/cm
3
in a Si crystal with 10
23
atoms/cm
3
.
Thus, N
D
augments n by 10
16
atoms/cm
3
but i
n
is only 10
10
approximately. This gives a million to one increase! The new value
of n describes excellent conducting properties. The phosphorus
atoms are referred to as donors since each phosphorus atom
donates one electron to the conduction band. The large increase in
the value of n gives a boost to recombination so that p falls to
Legends
Electron
Hole
Ionized acceptor
e
h
Thermally generated electron-hole

h h h h h h h

h h h h h h h
e
h
quite a low level. Eventually, the equilibrium is reached in a
manner such that
) T ( n
2
i = np (1.8.2)
It is true for both doped as well as intrinsic semiconductors. This
is called law of mass action. In this situation, electrons are the
majority charge carriers and holes are the minority charge
carriers. The latter contributes very little to the current flow. The
approximation that is valid is n = N
D
and p =
/n n
2
i = D
2
i
N / n
. Since
the majority carriers carry negative charge, the conduction
mechanism and the material is called n-type.
Doping with trivalent atoms (Boron) to tetravalent atoms result
into acceptor material. Since, a trivalent element accepts an
electron in order to form a stable bond, it is called acceptor. The
concentration N
A
of boron acceptor contributes directly to p to a
very large amount. The large hole population leads to short lived
increase in recombination so that electron population falls and
equilibrium is reached when p = N
A
and n =
p / n
2
i = A
2
i
N / n
.
If an n-type sample is subsequently doped with an equal
concentration of acceptor atoms i.e. A
N
= N
D
=
3 16
cm / atoms 10
, their
effects cancel and the sample becomes intrinsic again. As the
electrical conductivity must be conserved,
n N p N
A D
+ +
, so
p n
= A D
N N
(1.8.3)
This is true for any doping condition. The left hand side represents
the mobile charge and the right hand side the immobile charges.
Thus, in an n-type semiconductor majority charge carriers = n
n
=
N
D
and in p-type semiconductor majority charge carriers =
p
p
= N
A
.
1.9 Energy Distribution
The number of free electrons E
dn
per unit volume within energy
range E to E+dE at temperature T can be obtained using Fermi-
Dirac statistics. The number of free electrons per cubic meter of a
metal whose energies lie in the interval E to E+dE is given by
E
dn
= (E)dE (1.9.1)
Here (E) represents the density of electron in the energy
interval dE. We can safely assume in the case of free electrons
that there is no potential variation within the metal. Each cubic
meter of the metal must contain the same number of electrons.
Although electron density (electrons per cubic meter) remains
constant within the metal, yet electrons in each unit volume have
all possible energies. The function N(E) represents this distribution
in energy. This represents the number of electrons per electron
volt per cubic meter of the metal. The function (E) may be
expressed as
(E) = N(E)f(E) (1.9.2)
Here, N(E) is the density of states (number of states per electron
volt per cubic meter) in the conduction band and f(E) is probability
that an electron occupies a quantum state with energy N(E)f(E) is
called as Fermi-Dirac probability function.
Example
Estimate the density of Si.
Solution:
The density of states in Si can be expressed as
(Si) =
cell unit of volume
cell unit in ofatoms mass
=
3
constant attice l
mass) on wt.)(prot cell)(at. atoms/unit of (No.
=
3 9
27
) 10 543 . 0 (
10 672 . 1 1 . 28 8

=
3 8
27
) 10 43 . 5 (
10 672 . 1 1 . 28 8

=
1 . 160
10 672 . 1 1 . 28 8
3

=
2.35x10
-3
kg/m
3

The density of states N(E) is found to be proportional to the half
power of energy E, i.e.
NE)
2 / 1
E =
2 / 1
E
(1.9.3)
where, =
2 / 3
2
*
e
h
m 2
4

,
_
=
2 / 3
2 34
31
) 10 625 . 6 (
10 1 . 9 1 . 1 2
4

,
_
= 1.2210
56
(1.9.4)
*
e
m
= effective mass of electron in kg,
h = Planks constant in Joules-sec (
Js 10 62 . 6
34
)
The Fermi-Dirac probability function is expressed as
f(E) =
)
KT
E E
1
1
F
exp(

+
(1.9.5)
Let us now examine the Fermi-Dirac probability function at T = 0K
f(E) =
)] E E (
F
exp[ 1
1
+
(1.9.6)
Three possibilities results from Eqn.(1.9.6);
(a) E< E
F
. In this case exponential term of Eqn.1.9.6 becomes
zero and hence
f(E) =
1
.
It suggests that all quantum (energy) levels with energies less than
E
F
will be occupied at absolute zero.
(b) E > E
F
. In this case exponential term of Eqn.1.9.6 becomes
infinite and hence
f(E) = 0.
This reveals that there is no probability of finding an occupied
quantum state for energy greater than E
F
at absolute zero.
(c) For E = E
F
, the exponential term in Eqn.(1.9.5) become unity
and Eqn. (1.9.5) reduces to
f(E) = 1/2 (at all temperatures), (1.9.7)
Combining Eqns.(1.9.3) and (1.9.5) results as
(E) = N(E)f(E) =
2 / 1
E
)
KT
E E
1
1
F
exp(

+
=
)
KT
E E
F
exp( 1
E
2 1
and E
dn
=
dE
1
E
)
KT
E E
F
exp(
2 1
(1.9.8)
From Eqn.(1.9.8), (E) =
2 / 1
E
for E < E
F
= 0 for E > E
F
. (1.9.9)
Eqn.(1.9.9) reveals that there are no electrons having energies
more than E
F
. The Fermi level energy E
F
may be defined as the
maximum energy that electrons may possess at absolute zero.
Eqn.(1.9.8) is referred to as the completely degenerate distribution.
The variation of f(E) with energy (E-E
F
) at absolute temperatures
0K, 300K and 2500K is shown in Fig.39. Fig.40 shows the plot of
E-E
F
versus f(E) at absolute temperatures.
The distribution function changes only very slightly with
temperature inspire of large temperature changes from 0K to
2500K as shown in Fig.41. The effect of high temperature is to give
still higher energies to those electrons that are having high
energies at absolute zero (in the neighborhood of E
F
) i.e. only
outermost or high-energy valence electrons are affected by
increased temperature. The lower energy electrons are left
practically undisturbed. Further the curve for T = 2500K is
asymptotic with the energy axis and Hence, only few electrons
have large value of energy.
Figure 39 f(E) versus E-E
F

1.0 1.0 0
A
300K
0K
2500K
0.5
0
1.0
f(E)
(EE
F
)eV
T=0K
A
f(E)
1.0 0.5 0
1
0
1.0
300K 2500K
(EE
F
)eV

Figure 40 (E-E
F
) versus f(E) (b) Energy distribution
Classically all particles should have zero energy at 0K. It can be
observed from Fig.40 that the electrons actually have energies
extending from 0 to E
F
at absolute zero. This is due to Pauli
Exclusion Principle which states that no two electrons may have
the same set of quantum numbers i.e. all the electrons cannot
have the same energy at 0K. Thus, absolute zero is the condition
of lowest possible energy but not the zero energy. As the
temperature is raised, more thermal energy is stored in electrons
and their average kinetic energy is increased. Now some of the
electrons acquire energies higher than E
F.

1.9.1 Fermi level
The expression for E
F
can be obtained from energy distribution
curve in Fig.40 at T = 0K. The area under the curve represents the
total number of free electrons.
n =
dE
F
E
0
2 1
E .

=
2 3
F
E
3
2
(1.9.10)
E
F
=
3 2
2
n 3
,
_
=
3 2
56
10 22 . 1 2
n 3
,
_

=
( )
3 / 2
54
10 0123 . 0

3 / 2
n =
3 / 2 36
n 10 05 . 0

(1.9.11)
We know that the density n varies from metal to metal and
hence E
F
also varies as per Eqn.(1.9.11). The density of electrons
n may be calculated by knowing specific gravity, atomic weight,
and the number of electrons per atom. The Fermi energy E
F
for
most of the metals is numerically less than 10eV. Fermi level is a
characteristic energy of the material. We know that number of
electrons and holes are equal in an intrinsic semiconductor. The
concentration of electrons increases above the bottom of
conduction band. Similarly, the concentration of holes increases
below the top of the valence band as indicated in Fig.41.
E
C
E
F
E
V
E
G
E
C
E
F
E
V
E
G
E
C
E
F
E
V
E
G
O O O O
O O
O O
O O O
O
O
O O O
H
o
l
e

E
n
e
r
g
y
(
e
V
)

E
l
e
c
t
r
o
n

E
n
e
r
g
y

(
e
V
)
e e
e e e e
h h
h h h h
h h h h
h h h h
e e e e

e e e e
e e e e
e e e
e e e e
h h h h
h h h h
h h h h
Figure 41 (a) intrinsic semiconductor (b) n-type, (c) p-
type
The center of gravity of electrons and holes in intrinsic
semiconductor lies exactly at the middle of forbidden gap. This
central level is known as Fermi level. The Fermi level is the
energy that corresponds to the center of gravity of the conduction
electrons and holes weighted according to their energies. Fermi
level is a concept like a hollow body having a center of gravity
where there is no matter. When a donor impurity is added to an
intrinsic semiconductor, it becomes n-type and it has more free
electrons than holes. Thus moves the center of gravity up i.e.,
Fermi level shifts towards the conduction band. Similarly, when an
acceptor impurity is added to an intrinsic semiconductor, it
becomes p-type and it has more holes than electrons. This shifts
the Fermi level towards the valence band as shown in Fig.41.

Semiconductor

Caricato da

Informazioni sul documento

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Semiconductor

Caricato da

Copyright:

Formati disponibili

CHAPTER 1

Similarly in the case of Germanium, the energy gap decreases at

Effective mass of electron

Permittivity of free space (

Permeability of free space

Plank's Constant (h)

Universal Gas Constant (R)

Thus, current due to drift of electron as shown in Fig.1.18a can be

Potrebbero piacerti anche