3/31/2016

Origin of energy bands formation in solids

An isolated atom possesses discrete energies of different electrons. Suppose two
isolated atoms are brought to very close proximity, then the electrons in the orbits
of two atoms interact with each other. So, that in the combined system, the energies
of electrons will not be in the same level but changes and the energies will be
slightly lower. So, at the place of each energy level, a closely spaced two energy
levels exists. If ‘N ’ number of atoms are brought together to form a solid and if
these atoms’ electrons interact and give ‘N ’ number of closely spaced energy levels
in the place of discrete energy levels, it is known as bands of allowed energies.
Between the bands of allowed energies, there are empty energy regions, called
forbidden band of energies.

Figure 1Schematic plot of electron

energy versus interatomic separation
for an aggregate of 12 atoms (N = 12).
Upon close approach, each of the 1s
and 2s atomic states splits to form
an electron energy band consisting of
12 states.
1

Kronig-Penney model supports the existence of these bands of energies (allowed

bands and forbidden bands). The mathematical solution for Schrödinger’s wave
equation is very tedious but it provides a clue in understanding the origin of
energy bands.

Figure 2 (a) The conventional representation of the electron energy band structure for
a solid material at the equilibrium interatomic separation. (b) Electron energy versus
interatomic separation for an aggregate of atoms, illustrating how the energy band
structure at the equilibrium separation in (a) is generated. 2

1
 3/31/2016

The conventional way of representing electron band structures in solids is shown in

Figure 2. The electrical properties of a solids are a consequence of its electron band
structure—that is, the arrangement of the outermost electron bands and the way in
which they are filled with electrons.

Four different types of band structures are possible at 0 K. In the first (Figure
3a), one outermost band is only partially filled with electrons. The energy
corresponding to the highest filled state at 0 K is called the Fermi energy Ef, as
indicated. This energy band structure is typified by some metals, in particular
those that have a single s valence electron (e.g., copper). Each copper atom has
one 4s electron; however, for a solid composed of N atoms, the 4s band is
capable of accommodating 2N electrons. Thus only half the available electron
positions within this 4s band are filled.

For the second band structure, also found in metals (Figure 3b), there is an
overlap of an empty band and a filled band. Magnesium has this band structure.
Each isolated Mg atom has two 3s electrons. However, when a solid is formed,
the 3s and 3p bands overlap. In this instance and at 0 K, the Fermi energy is
taken as that energy below which, for N atoms, N states are filled, two electrons
per state.
3

Figure 3 The various possible electron band structures in solids at 0 K. (a) The electron band structure found in
metals such as copper, in which there are available electron states above and adjacent to filled states, in the same
band. (b) The electron band structure of metals such as magnesium, wherein there is an overlap of filled and empty
outer bands. (c) The electron band structure characteristic of insulators; the filled valence band is separated from
the empty conduction band by a relatively large band gap (>2 eV). (d) The electron band structure found in the
semiconductors, which is the same as for insulators except that the band gap is relatively narrow (<2 eV).

The final two band structures are similar; one band (the valence band) that is
completely filled with electrons is separated from an empty conduction band, and
an energy band gap lies between them. For very pure materials, electrons may not
have energies within this gap. The difference between the two band structures lies
in the magnitude of the energy gap; for materials that are insulators, the band gap is
relatively wide (Figure 3c), whereas for semiconductors it is narrow (Figure 3d).
The Fermi energy for these two band structures lies within the band gap—near its
4
center.

2
 3/31/2016

Classical free electron theory of metals

The classical free electron theory was introduced by P. Drude in 1900 and
developed by Lorentz in 1909 to explain electrical conduction in metals. This
theory has some assumptions; these are:
(i) The valence electrons of metallic atoms are free to move in the spaces
between ions from one place to another place within the metallic specimen
similar to gaseous molecules so that these electrons are called free electron
gas. These electrons also participate in electrical conduction; hence they
are called conduction electrons.
(ii) There is no interaction between these conduction electrons.
(iii) The interaction of free electrons with ion cores is negligible.
(iv) The free electrons find uniform electric field of positive ions and that of
electrons in the metal.

These free electrons participate in electrical conduction; hence they are called
as conduction electrons. The metals possess some physical properties. They
are:
(i) Due to the existence of large number of free electrons, metals possess high
electrical and thermal conductivities.
(ii) They obey Ohm’s law i.e., the current through a metal is proportional to
the applied electric field.
(iii) The resistivity (ρ) is proportional to the fifth power of absolute
temperature [i.e., ρ ∝ T5] at low temperatures, whereas resistivity is
proportional to absolute temperature (ρ ∝ T ) at high temperatures.
(iv) In metals, the ratio of thermal conductivity to electrical conductivity is
proportional to absolute temperature. This is known as Wiedmann-Franz law.
(v) Near absolute zero of temperature, the resistivity of some metals drops to
zero; hence they show superconductivity.
(vi) The resistance of a metal increases with temperature as given below:

3
 3/31/2016

Quantum Free electron theory (FET) and its success and failure

According to quantum free electron theory of metals, a conduction electron in a

metal experiences constant (or zero) potential and free to move inside the crystal
but will not come out of the metal because an infinite potential exists at the surface.
Success: This theory successfully explains electrical conductivity, specific heat,
thermionic emission and paramagnetism.
Faliure: This theory is fails to explain many other physical properties, for example: (i)
it fails to explain the difference between conductors, insulators and semiconductors, (ii)
positive Hall coefficient of metals and (iii) lower conductivities of divalent metals than
monovalent metals.
Improvements of FET: To overcome the above problems, the periodic potentials due to the
positive ions in a metal have been considered. As shown in Fig. 4(a), if an electron moves
through these ions, it experiences varying potentials. The potential of an electron at the
positive ion site is zero and is maximum in between two ions. The potential experienced by
an electron, when it passes along a line through the positive ions is as shown in Fig. 4(b).
We see that the potential experienced by an electron varies
periodically with the same period as the lattice. The potential is negative because of an
attractive force between electrons and positive ions. Along X-direction in the crystal, the
potential function V(x) has the periodicity of the lattice given by:

Figure 4:(a) Electron motion; (b) The potential

experienced by an electron in a row of ions;
(c) Rectangular potentials.

V (x) = V (x + a) (Eq. 1)
where ‘a’ is the periodicity of the
lattice. The energies of electrons can
be known by solving Schrödinger’s
wave equation in such a lattice.
The Schrödinger time-independent
wave equation for the motion of an
electron along X-direction is
given by:

(Eq. 2)

Bloch showed a type of solution for

Equation (eq. 2) given by: 8

4
 3/31/2016

(Eq. 3)
where uK(x) represents periodic function given by:

(Eq. 4)

Eq.3 is a solution to Schrödinger’s wave Eq. 2 in sinusoidal potential as shown in Fig. 4(b).
It is not easy to solve Schrödinger’s equation with these potentials. So, Kronig and Penney
approximated these potentials inside the crystal to the shape of rectangular steps as shown
in Fig. 4(c). This model is called Kronig-Penney model of potentials.

Kronig–Penney model —origin of energy bands

The rectangular potential wells and barriers, as assumed by Kronig and Penney
for one-dimensional lattice in 1931, are best suited to solve Schrödinger’s wave
equation. These potentials are shown in Fig. 5, in which the width of the
potential well and potential barrier are ‘a’ and ‘b’, respectively. The potential
energy of an electron in the well is zero; this corresponds to the potential in the
vicinity of the nucleus and in barrier it possesses a constant value represented as
V0. The periodicity of the potential is (a + b). This is an approximate model but
close to reality. The energies and wave functions of electrons associated with
this model can be calculated by solving time-independent one-dimensional
Schrödinger’s wave equations for the two regions I and II as shown in Fig. 5.

Figure 5. Kronig–Penney model of potentials. 10

5
 3/31/2016

The Schrödinger’s equations are:

(Eq. 5)

(Eq. 6)

(Eq. 7)

Hence, Eq.5 and 6 becomes:

(Eq. 8)

(Eq. 9)

The solution that will be appropriate for both the regions will be of the form of
a plane wave eiKx modulated with a periodic function uK(x): 11

(Eq. 10)

Differentiating Eq. 10 twice, we get:

Substituting these values in Eq. 8 and 9, we get:

(Eq. 11)

12

6
 3/31/2016

Similarly,

(Eq. 12)

Now to slove these differential equations, assume the solution of the form:

which on substitution in Eq.11 gives:

(Eq. 13) 13

Similarly from Eq. 12,

(Eq. 14)

Here, A, B, C and D are constants. These constants may be obtained by applying the
following boundary conditions.

(Eq. 15)

(Eq. 16)

The boundary conditions represented in Eq. 15 show that the wave functions and their
first derivatives are equal at x = 0 because of the continuity of wave functions at that
point. The boundary conditions represented in Eq. 16 show that the wave function and its
first derivative at x = a is equal to that at x = -b, because of the periodicity of wave
functions.
Applying boundary conditions Eq. 15 and 16 on Eq. 14 gives four equations involving the
constants A, B, C and D.

14

7
 3/31/2016

(Eq. 17)

(Eq. 18)

(Eq. 19)

(Eq. 20)

The constants A, B, C and D can be determined by solving Equations 17–20,

and thus the wave function u1 and u2 can be obtained. The four Equations 17–20
have solution only if the determinant of the coefficients of A, B, C and D
vanishes. The above condition leads to the following equation.

(Eq. 21)
15

This equation is quite complicated. To express it in a more simplified form, Kronig and
Penney suggested delta function such as V0 → ∞ and b → 0 but the product V0b remains
finite. Under these circumstances, sin hβb → βb and cos hβb → 1 as b → 0. Hence, Eq. 21
becomes:

(Eq. 22) Since V0>>E

22 becomes,

(Eq. 23)

Equation 23 represents a condition for the wave function. P is referred to as

the scattering power of the potential barrier. It represents the strength with which
electrons in a crystal are attracted to the ions. The term V0B is called barrier
strength. With an increase of P, an electron is bound more strongly to a potential
well, when P→0, the electrons are free. The right-hand side of Equation 23 can
take values between 1, but the left-hand side exceeds this value. By plotting the
left-hand side of Equation 23 for a finite value of P ( say 3π/2) against αa, it is
possible to determine the allowed values of αa. To find the allowed parts of the
curve, horizontal lines are drawn at +1 on vertical line as shown
in Fig. 6.
16

8
 3/31/2016

The parts of the curve, which lies between the 1 horizontal line, are acceptable
to the left-hand side of Equation 23. From the graph, the following conclusions
may be drawn.

(i) The allowed values of αa and hence energy, for which wave mechanical
solutions exist are shown by the shadow portions. From the graph, we know
that the conduction electrons in periodic potentials of lattice ions possess the
bands of allowed energy (shaded region) separated by forbidden regions
[unshaded region].

Figure 6. A graph plotted . 17

(ii) As the value of αa increases, the width of allowed energy bands increases and the width
of forbidden bands decreases.
(iii) Now, we will see the effect of varying ‘P ’. If P is large, then curve crosses  1 line
steeply so that the allowed bands are narrower and forbidden bands are wider as shown in
Fig. 7. In the limit P → ∞, the allowed energy bands reduce to single energy levels as
shown in Fig. 8. The energy levels in this case are discrete and similar to the energy levels
of a particle in a constant potential box of atomic dimensions. When P → 0, then the left-
hand side of Equation 23 will not cross  1 line as shown in Fig. 8. Hence, all the energies
are allowed to the electrons. Thus, by varying P from zero to infinity, the energies of
electrons will vary from continuous to bound, i.e., free electrons to bound electrons.

Figure 7. Shows the decrease of allowed energy bands width for large value of P.
18

9
 3/31/2016

Figure 8 Shows that all energies are allowed to the electrons as P → 0.

(iv) E–K diagram: The free electrons moving in periodic potentials of lattice can
have energy values only in the allowed regions or zones. It is possible to plot the
total energies of free electrons versus their wave number or propagation vector
K as shown in Fig. 9.

Brillouin Zone
19
Figure 9 E–K diagram

where n = 1, 2, 3 … . The dotted parabolic curve shows E-K relation for

completely free electrons.
From the graph, we see that the electrons have allowed energy values in the
region or zone extending from

This is called the first Brillouin zone. After a break in the energy values, called
the forbidden region or band, the electrons have another allowed zone of energy

values in the region extended from

This zone is called the second Brillouin zone. Similarly, the higher order Brillouin
zones can be defined.
The discontinuities occur at the boundaries of the Brillouin zones. Each portion
of the curve gives a number of allowed energies called allowed band of energies.
The curves are horizontal at the top and bottom and they are parabolic near the
top and bottom with curvatures in opposite directions. As ‘P’ decreases, the
discontinuous E-K graph will reduce to a continuous parabolic graph as shown by
the dotted lines and the forbidden bands disappear. Then, the energy values are
practically continuous.
20

10
 3/31/2016

Sources of electrical resistance

In a metallic crystal, the electrons move through periodically varying electric

potentials (or fields) of positive ions of the lattice. The cause for electrical
resistance is the electron scattering and the cause for electron scattering is the
non-periodicity of the lattice potentials. The causes for non-periodicity of
lattice potentials are: (i) impurities in crystals, (ii) crystal defects (or
imperfections) and (iii) thermal vibrations.

Freely moving electrons are associated with de Broglie [or matter] waves. In a
perfect crystal, an electron moves similar to an electromagnetic wave without
attenuation so that the mean free paths of electrons will be of several hundred
Angstroms. This is in agreement with the observed values of mean free paths.
According to classical free electron theory, the electrical resistance is due to
collisions of free electrons with the positive ion cores of the crystal. From this
concept, the mean free path will be a few Angstroms only. This is contradictory
to the observed mean free paths of electrons, so the scattering of electrons by
positive ions is ignored. Now, we will see in detail the causes of non-periodic
potentials of lattice.
21

(i) Presence of impurities in crystals: The presence of impurities introduces electrical

resistance in a metallic substance. With an increase of impurity concentration, the
resistivity of a metal increases as shown in Fig. 10. The presence of impurity atomic
site changes the periodicity of electrical potential. If the impurity concentration is
more, then at large number of atomic sites non-periodic potentials exist and they cause
large electron scattering and hence resistivity.

Figure 10. Resistivity variation

of impure metal with temperature

(ii) Crystal defects: The presence of imperfections such as vacancies and grain boundaries
changes the periodic potentials. So, larger the number of defects, larger will be the non-
periodic potentials and hence larger the electron scattering and resistivity. 22

11
 3/31/2016

Figure 11 The electrical resistivity versus temperature for copper and three copper–nickel alloys, one
of which has been deformed. Thermal, impurity, and deformation contributions to the resistivity are
indicated at -100C. [Adapted from J. O. Linde, Ann. Physik, 5, 219 (1932); and C. A. Wert and R. M.
Thomson, Physics of Solids, 2nd edition, McGraw-Hill Book Company, New York, 1970.]
23

(iii) Thermal vibrations: At high temperatures, the resistivity is linearly

proportional to absolute temperature, whereas at low temperatures, the
resistivity is proportional to the fifth power of temperature. Due to thermal
vibrations, the positive ions move away from their ideal positions so that
periodicity of potential changes. With an increase of temperature, the amplitude
of thermal vibrations of positive ions will be more, so non-periodicity increases.
Hence, resistivity increases. Near absolute zero of temperature, some materials
show zero resistivity or superconductivity. For some materials, the resistance is
not equal to zero even at 0 K. The total resistivity (ρ) of a metal is due to
impurities and defects and also due to thermal vibrations.
∴ ρ = ρ i + ρ T + ρd
where ρi is the resistivity due to impurities, ρ(T ) = resistivity due to thermal
vibrations and ρd is deformation resistivity. Above equation is sometimes called
Matthiessen’s rule.
24

12
 3/31/2016

Influence of Temperature
For the pure metal and all the copper–nickel alloys shown in Figure 11, the
resistivity rises linearly with temperature above about -200C. Thus,
ρ(T) = ρ0 + aT
where ρ0 and a are constants for each particular metal. This dependence of the
thermal resistivity component on temperature is due to the increase with
temperature in thermal vibrations and other lattice irregularities (e.g., vacancies),
which serve as electron-scattering centers.

Semiconductivity
The electrical conductivity of the semiconducting materials is not as high as that
of the metals; nevertheless, they have some unique electrical characteristics that
render them especially useful. The electrical properties of these materials are
extremely sensitive to the presence of even minute concentrations of impurities.

Intrinsic semiconductors are those in which the electrical behavior is based on

the electronic structure inherent in the pure material.

When the electrical characteristics are dictated by impurity atoms, the

semiconductor is said to be extrinsic.
25

INTRINSIC SEMICONDUCTION
Intrinsic semiconductors are characterized by the electron band structure shown in
Figure 12: at 0 K, a completely filled valence band, separated from an empty conduction
band by a relatively narrow forbidden band gap, generally less than 2 eV.

Fig. 12: Possible band structure of intrinsic

semiconductor at 0K

The two elemental semiconductors are silicon (Si) and germanium (Ge), having
band gap energies of approximately 1.1 and 0.7 eV, respectively. Both are found in
Group IVA of the periodic table and are covalently bonded. In addition, a host of compound
semiconducting materials also display intrinsic behavior. One such group is formed
between elements of Groups IIIA and VA, for example, gallium arsenide (GaAs) and
indium antimonide (InSb); these are frequently called III–V compounds. The compounds
composed of elements of Groups IIB and VIA also display semiconducting behavior; these
include cadmium sulfide (CdS) and zinc telluride (ZnTe).
26

13
 3/31/2016

Energy Band Structures and Conductivity

• Fermi Energy (EF) - highest filled state at 0 K

• Conduction band - a partially filled or empty energy band
• Valence band – the highest partially or completely filled band

In semiconductors and insulators, the valence band is filled, and no more electrons
can be added (Pauli‘s principle).

Electrical conduction requires that electrons be able to gain energy in an electric field.
This is not possible in these materials because that would imply that the electrons are
promoted into the forbidden band gap.

27

Energy States: Insulators & Semiconductors

• Insulators: • Semiconductors:
-- Higher energy states not -- Higher energy states separated
accessible due to gap (> 2 eV). by smaller gap (< 2 eV).
Energy Energy
empty
band empty
? band
GAP GAP

filled filled
filled states
filled states

valence valence
band band

filled filled
band band
28

14
 3/31/2016

Charge Carriers
• Two charge carrying mechanisms

– Electron – negative charge

– Hole – equal & opposite
positive charge
• Move at different speeds - drift
velocity

Higher temp. promotes more electrons into the conduction band

 s as T
Electrons scattered by impurities, grain boundaries, etc.

29

Energy Band Structures and Conductivity

(semiconductors and insulators
• In semiconductors and insulators, electrons have to jump across the band
gap into conduction band to find conducting states above Ef
• The energy needed for the jump may come from heat, or from irradiation
at sufficiently small wavelength.
• The difference between semiconductors and insulators is that in
semiconductors electrons can reach the conduction band at ordinary
temperatures, where in insulators they cannot.
• The probability that an electron reaches the conduction band is about
exp(-Eg/2kBT), where Eg is the band gap. If this probability is < 10-24 one
would not find a single electron in the conduction band in a solid of 1 cm3.
This requires Eg/2kBT > 55. At room temperature, 2kBT = 0.05 eV →Eg >
2.8eV corresponds to an insulator.
• An electron promoted into the conduction band leaves a hole (positive
charge) in the valence band, that can also participate in conduction. Holes
exist in metals as well, but are more important in semiconductors and
insulators.

30

15
 3/31/2016

Energy Band Structures and Conductivity (metals)

In metals, highest occupied band is

partially filled or bands overlap.

Conduction occurs by promoting

electrons into conducting states, that
starts right above the Fermi level. The
conducting states are separated from
the valence band by an infinitesimal
amount.

Energy provided by an electric field is

sufficient to excite many electrons
into conducting states. → High
conductivity.

31

Energy Band Structures and Bonding

(metals, semiconductors, insulators)

Relation to atomic bonding:

☼ Insulators – valence electrons are tightly bound to (or shared
with) the individual atoms – strongest ionic (partially covalent)
bonding.
☼ Semiconductors - mostly covalent bonding somewhat weaker
bonding.
☼ Metals – valence electrons form an “electron gas” that are not
bound to any particular ion.

32

16
 3/31/2016

Energy Band Structures and Conductivity

33

Semiconductors
• Intrinsic semiconductors – A Semiconductor in its pure form; electrical conductivity
is defined by the electronic structure of pure material.
• Extrinsic semiconductors – A semiconductor with added impurities; electrical
conductivity is defined by impurity atoms.

Intrinsic semiconductors
Since both electrons and holes conduct the conductivity of an intrinsic
semiconductor is
σ = n|e|μe + p|e|μh
where p is the hole concentration and μh the hole mobility.
Electrons are more mobile than holes, μe > μh

In an intrinsic semiconductor, a hole is produced by the promotion of each

electron to the conduction band. Therefore, n = p and
σ = n|e|(μe + μh) = p|e|(μe + μh)
(only for intrinsic semiconductors)

n (and p) increase exponentially with temperature, whereas μe and μh decrease

(about linearly) with temperature.
⇒ The conductivity of intrinsic semiconductors is increasing with
temperature (different from metals!) 34

17
 3/31/2016

Conduction in Terms of Electron and Hole Migration

• Concept of electrons and holes:

valence electron hole electron hole
electron Si atom
pair creation pair migration

- + - +

no applied applied applied

electric field electric field electric field
• Electrical Conductivity given by:
# holes/m 3
s  n e e  p e h
hole mobility
# electrons/m3 electron mobility

35

Intrinsic vs Extrinsic Conduction

• Intrinsic:
Number of electrons = Number of holes (n = p)
--case for pure Si
• Extrinsic:
--n ≠ p
--occurs when impurities are added with a different
number of valence electrons than the host (e.g., Si atoms)
• n-type Extrinsic: (n >> p) • p-type Extrinsic: (p >> n)
Phosphorus atom Boron atom
hole
4+ 4+ 4+ 4+ conduction 4+ 4+ 4+ 4+
s  n e e 4+ 5+ 4+ 4+
electron
4+ 3+ 4+ 4+
s  p e h
valence
4+ 4+ 4+ 4+ electron 4+ 4+ 4+ 4+
no applied Si atom no applied
electric field electric field 36

18
 3/31/2016

Extrinsic semiconductors
Si is considered to be extrinsic at room T if impurity concentration is one atom per 1012
(remember our estimation of the number of electrons promoted to the conduction band by
thermal fluctuations at 300 K)

Unlike intrinsic semiconductors, an extrinsic semiconductor may have different

concentrations of holes and electrons. It is called p-type if p > n and n-type if n > p.

One can engineer conductivity of extrinsic semiconductors by controlled addition of impurity

atoms – doping (addition of a very small concentration of impurity atoms). Two common
methods of doping are diffusion and ion implantation.

37

n-type semiconductor

n-type extrinsic semiconductors

Excess electron carriers are produced by substitutional impurities that have
more valence electron per atom than the semiconductor matrix

Example: phosphorus (or As, Sb etc.) with 5 valence electrons, is an

electron donor in Si since only 4 electrons are used to bond to the Si lattice
when it substitutes for a Si atom. Fifth outer electron of P atom is weakly
bound in a donor state (~ 0.01 eV) and can be easily promoted to the
conduction band.

Impurities which produce extra conduction electrons are called donors,

ND = NPhosphorus ~ n

Elements in columns V and VI of the periodic table are donors for

semiconductors in the IV column, Si and Ge.

The hole created in donor state is far from the valence band and is
immobile. Conduction occurs mainly by the donated electrons (thus n-
type).
σ ~ n|e|μe ~ ND |e|μe
(for extrinsic n-type semiconductors)
38

19
 3/31/2016

Band diagram for n-type semiconductors

39

p-type extrinsic semiconductors

• Excess holes are produced by substitutional impurities that have fewer valence
electrons per atom than the matrix.
• A bond with the neighbors is incomplete and can be viewed as a hole weakly bound
to the impurity atom.
• Elements in columns III of the periodic table (B, Al, Ga) are donors for
semiconductors in the IV column, Si and Ge.
• Impurities of this type are called acceptors, NA = NBoron ~p
• The energy state that corresponds to the hole (acceptor state) is close to the top of the
valence band. An electron may easily hop from the valence band to complete the bond
leaving a hole behind. Conduction occurs mainly by the holes (thus p-type).
σ ~ p|e|μp ~ NA |e|μp
(for extrinsic p-type semiconductors)

40

20
 3/31/2016

Band diagram for p-type semiconductors

41

Conductivity: Comparison
• Room T values (Ohm-m)-1 = ( - m)-1
METALS conductors CERAMICS
7 -10
Silver 6.8 x 10 Soda-lime glass 10 -10-11
Copper 6.0 x 10 7 Concrete 10 -9
Iron 1.0 x 10 7 Aluminum oxide <10-13

SEMICONDUCTORS POLYMERS
Silicon 4 x 10 -4 Polystyrene <10 -14
Germanium 2 x 10 0 Polyethylene 10 -15-10-17
-6
GaAs 10
semiconductors insulators
42

21
 3/31/2016

Pure Semiconductors:
Conductivity vs T
• Data for Pure Silicon:  E gap / kT
-- s increases with T
sundoped  e
-- opposite to metals Energy
electrical conductivity, s empty
? band
(Ohm-m) -1 GAP
10 4 electrons
10 3 filled can cross

filled states
valence gap at
10 2 band
higher T
10 1
filled
10 0 pure band
(undoped)
10 -1
material band gap (eV)
10 -2 Si 1.11
50 10 0 1000 Ge 0.67
T(K)
GaP 2.25
CdS 2.40
43

Doped Semiconductor: Conductivity vs. T

• Data for Doped Silicon: • Comparison: intrinsic vs
-- s increases doping extrinsic conduction...
-- reason: imperfection sites -- extrinsic doping level:
lower the activation energy to 1021/m3 of a n-type donor
produce mobile electrons. impurity (such as P).
-- for T < 100 K: "freeze-out“,
thermal energy insufficient to
10 4 0.0052at%B
excite electrons.
electrical conductivity, s

-- for 150 K < T < 450 K: "extrinsic"

10 3
-- for T >> 450 K: "intrinsic"
(Ohm-m) -1

10 2 doped
0.0013at%B
10 1 doped
concentration (1021/m3)

undoped
10 0
conduction electron

pure 3
freeze-out

extrinsic

intrinsic

(undoped)
10 -1 2
10 -2 1
50 100 1000
T(K) 0
0 200 400 600 T(K)
44

22
 3/31/2016

p-n Rectifying Junction

• Allows flow of electrons in one direction only (e.g., useful
to convert alternating current to direct current.
• Processing: diffuse P into one side of a B-doped crystal.
• Results: p-type n-type
+ + + - -
--No applied potential:
+ + - - -
no net current flow.
--Forward bias: carrier
flow through p-type and p-type + - n-type
+ + -
n-type regions; holes and ++- - -
electrons recombine at +-
p-n junction; current flows.

--Reverse bias: carrier n-type -

+ p-type
flow away from p-n junction; - + + - - +
carrier conc. greatly reduced + + - -
at junction; little current flow.
45

Properties of Rectifying Junction

46

23
 3/31/2016

Transistor MOSFET
• MOSFET (metal oxide semiconductor field effect
transistor)

47

Integrated Circuit Devices

• Integrated circuits - 50 nm line width

– > 100,000,000 components on chip

– chip formed layer by layer

48

24