Physics Project

Semiconductors
and
Band Theory
Akbar Hashmi
12-A

Acknowledgement
I gratefully acknowledge the kind guidance
and advice I have received from my Physics
Sir Mr Tangari

References
http://www.chemistryexplained.com/RuSp/Semiconductors.html
http://hyperphysics.phyastr.gsu.edu/hbase/solids/band.html
http://www.educypedia.be/education/sol
arcellanimations.htm
and many other linked pages associated
with this topic from Hyperphysics.
http://en.wikipedia.org/wiki/P-n_junction
In addition, for those with a sense of
humour, the somewhat eccentric
http://britneyspears.ac/physics/basics/b
asics.htm

Contents
Electrical conductivity and band theory

Summary of band theory

Intrinsic semiconductors

Extrinsic semiconductors

11

Summary of intrinsic and extrinsic semiconductors

14

pn junctions

15

Photovoltaic cells

17

Light emitting diodes

18

Why use LEDs?

21

Summary of pn junctions, LEDs and photovoltaic cells

21

SEMICONDUCTORS AND BAND THEORY

Semiconductors and band theory

The purpose of this document is to introduce the new approach that is being
brought to the Higher Physics course, in teaching about semiconductors from
the perspective of band theory.

Electrical conductivity and band theory

All solids can be classified as conductors, semiconductors or insulators
according to the availability of conduction electrons in their structures. Band
theory gives an explanation for these differences in electrical properties and
accounts for the availability, or not, of those conduction electrons.
Although individual atoms have certain permitted energy levels for their
electrons, as defined by quantum theory, when large groups of atoms are
incorporated into a solid mass those energy levels become reorganised in such
a way as to result in bands of possible energy levels (Figure 1). This is known
as the tight binding approximation.

Permitted energy
levels

Permitted energy
bands

Figure 1 Discrete energy levels within an individual atom (left) and bands of permitted
energy levels within a solid (right)

SEMICONDUCTORS AND BAND THEORY

There are such enormous numbers of electrons in a solid mass that although
the bands actually consist of very large numbers of closely packed discrete
energy levels, the bands become essentially continuous. There may be several
permitted energy level bands, but in particular we consider the two uppermost
bands. These are known as the valence band and the conduction band (Figure
2).

Conduction band
(empty)

Valence band
(full)

Figure 2 Conduction and valence bands in an insulator. These bands contain the only
permitted energy levels, and since the valence band is full and the conduction band is empty,
no net movement of electrons can occur within the material. Note the gap separating the
bands.

The electrons with lower energy levels are described as occupying the
valence band. The innermost electrons in an atom are much less influenced by
neighbouring atoms, and occupy discrete energy levels. They are sometimes
considered to be bound. At higher levels in the valence band electrons can, in
fact, move from atom to atom, but only up to the top of the valence band.
Since they are permitted only to swap places with other valence electrons in
neighbouring atoms, they are effectively unavailable for conduction.
Electrons fill the valence band from the lowest level to the highest. The top of
the valence band for a material is the highest level, which would, in theory,
be filled by all the available electrons within an atom of that material at a
temperature of 0 K. In insulators and semiconductors, the valence band is
completely filled with electrons. The conduction band is empty (Figure 2).

SEMICONDUCTORS AND BAND THEORY

The electrons fill energy levels in order because, as fermions, they must obey
the Pauli exclusion principle and cannot occupy identical energy levels. The
only way that an electron could move from one atom to another in an
insulator or semiconductor would be to occupy a slightly different energy
level in a neighbouring atom. However, all those energy levels are already
full. As discussed above, the electrons may effectively swap places, but in
order to facilitate conduction, they must leap up to the conduction band. An
energy level band must have some space within it (some vacant energy levels)
in order for there to be any net movement of electrons within the material.
For a material to be able to conduct electricity it must have electrons in its
conduction band or spaces in its valence band. There must be spaces for
charges to move into: a partially filled band.
The simulation demonstrates how energy levels differ in conductors,
semiconductors and insulators, and the impact the energy levels have on
conductivity.
Again, as a result of the wavelike behaviour of electrons within atoms,
materials may exhibit a certain range of forbidden energy levels (Figure 3).
It is simply not possible for an electron to exist with an energy level that
would place it in this range. This leaves insulators and semiconductors with a
gap between the two bands.

Gap of forbidden
energies

Conduc
tor
Insulat
or

Semicondu
ctor

Figure 3: Conduction bands (blue) and valence bands (yellow) for insulators,
semiconductors and conductors. Note the energy gaps in insulators and semiconductors, and
how in a conductor there is no gap, simply a continuous, partially filled conduction band.

SEMICONDUCTORS AND BAND THEORY

In insulators this zone of forbidden energy levels is very substantial, and

separates the valence band and the conduction band significantly. The
forbidden zone is of the order of a few electron volts, and is therefore so large
that it is not normally practicable for there to be sufficient energy to move
electrons across it from the valence band to the conduction band. For
example, thermal excitations and conventional electric circuit voltages within
a material provide energies that are smaller than 1 eV on an atomic scale. It
would be necessary to expose an insulator to electric fields of the order of
1010 V m 1 in order to give the valence electrons enough energy to jump
across the gap to the conduction band, since this could provide energy in the
order of a few electron volts on an atomic scale. This is what happens when
there is dielectric breakdown.
Contrastingly, conductors only have one band, and the top of this band is only
partially filled, permitting electrical conduction. This means that there are
plenty of nearby energy levels available for electrons to move into. They can
flow easily from one atom to another when a potential difference is applied
across the conducting material.
Like insulators, semiconductors have a completely full valence band and so
electrons are not able to facilitate conduction at low temperatures. However,
for semiconductors, the forbidden energy level zone between the two bands is
sufficiently small to make it much easier for significant numbers of electrons
to move across this gap and go from the valence band to the conduction band.
This can happen if sufficient energy is supplied, for example if there is some
thermal excitation. As a result, semiconductors exhibit increased conductivity
with increasing temperatures. In many semiconductors, a temperature
increase of 10 K will permit a doubling of the numbers of electrons in the
conduction band.
In order to increase the conductivity of semiconductors, small amounts of
doping material can be used. This results in significant increases in
conductivity as a result of the narrowing of the gap between the conduction
and valence bands.

SEMICONDUCTORS AND BAND THEORY

Summary of band theory

In solids, permitted electron energy levels are organised as bands.
The valence band contains electrons that can be considered to be bound to
the atom. In insulators and semiconductors the valence band is full.
The conduction band is a region of permitted energy levels that is empty
in insulators and semiconductors, but partially filled in conductors.
Only partially filled bands may permit conduction.
There is a forbidden zone that forms an energy gap between the valence
and conduction bands in insulators and semiconductors.
That energy gap must be jumped if an electron is to move to the
conduction band, and this is not normally possible in insulators because
the gap is too large.
In semiconductors, the forbidden zone is much smaller and electrons can
jump the gap to the conduction zone as a result of thermal excitation.
Doping of semiconductors can significantly reduce the width of the energy
gap.

SEMICONDUCTORS AND BAND THEORY

Intrinsic semiconductors
Pure, undoped silicon and germanium are two simple examples of intrinsic
semiconductors (Figure 4). They are both in Group IV of the Periodic Table,
and form a tetrahedral crystalline structure, similar to diamond. Each atom of
silicon and germanium has four electrons in its outermost electron shell, and
each of these electrons is used in a covalent bond with one of the atoms four
neighbours.

Figure 4 Two-dimensional illustration of a crystal of pure undoped Si. If any individual atom
of silicon is considered, it can be seen that each of its four valence electrons is used in
maintaining covalent bonds with the atoms neighbours. These electrons are therefore
unavailable for conduction.

Since all valence electrons are involved in bonding, pure silicon and
germanium may be expected to be good insulators. However, relatively small
energies are required to move a valence electron across the energy gap to the
conduction band. This is 1.1 eV for silicon, and only 0.7 eV for germanium.
This means that a significant number of electrons are available in the
conduction band, even at room temperature (Figure 5).

SEMICONDUCTORS AND BAND THEORY

Silic
on

Germaniu
m

Conduction
band

Valence
band

Electrons
moving
across
energy gap
due to
thermal
excitation

Figure 5 It is possible for significant numbers of electrons to cross the energy gap in
semiconductors.

It must be noted at this stage that although most thermal excitation involves
energies much less than even 0.7 eV, quantum mechanics clearly shows that
there is a small but significant probability of an electron being able to jump
the energy level gap, even at relatively low temperatures. As previously
discussed, this probability increases rapidly with temperature.
Once an electron jumps up to the conduction band in the crystal lattice, it
leaves behind a hole in the covalent bond. This hole can enable another
neighbouring valence band electron to move into it. As such, a hole behaves
rather like a positive charge carrier, even though it is actually a vacancy for
an electron. A hole can travel through the crystal lattice of the semiconductor.
A helpful analogy might be to consider a queue of cars on a road. If a space
appears at the front of the queue, cars may move forward in turn. Each time a
car moves forward, it leaves a space behind it, into which the next car may
now move. An observer from above might consider that the cars are moving
forwards or that the space is moving backwards.
Some semiconductors, like pure silicon or germanium, are known as intrinsic
semiconductors. Intrinsic semiconductors must always contain equal numbers
of conduction electrons and holes. If an electron can move from its place then
it must leave behind a hole (Figure 6).

SEMICONDUCTORS AND BAND THEORY

electro
ns
hole
s

Figure 6 In intrinsic semiconductors like pure silicon or germanium, every electron that
moves up to the conduction band must leave a hole in the valence band. Electrons and holes
exist in equal numbers and both contribute to conduction. There are no majority charge
carriers in intrinsic semiconductors.

Extrinsic semiconductors
Often, it is more useful to control the properties of a Group IV semiconductor
by deliberately introducing very small proportions of a Group III or Group V
element. This is known as doping and results in what is known as an extrinsic
semiconductor. Extrinsic semiconductors have majority charge carriers that
may be either electrons or holes.
Consider a semiconductor that is doped with a Group III element (Figure 7).
Each atom of the doping agent will have only three electrons in its outer
shell. This is insufficient to form the four covalent bonds with its Group IV
neighbours and therefore results in a hole. Countless holes are now built into
the semiconductors crystal lattice. It may be referred to as a p-type
semiconductor as the majority charge carriers are positively charged holes. As
a result of the doping process, it will require much less energy to allow
charge to flow through the semiconductor and so its conductivity is greatly
enhanced. Unlike metals, a p-type semiconductors conduction occurs in the
valence band. In effect, the doping agent adds an extra energy level just
above the valence band, sometimes called an acceptor band.

SEMICONDUCTORS AND BAND THEORY

Electro
n
Hole

Silicon

Group
III

Figure 7 Introducing small quantities of Group III atoms into a silicon lattice (in practice
only around one part in a million) leaves holes built into the valence band.

Technically, there will also still be a small degree of intrinsic behaviour, as

electrons leave behind holes, but this is not considered to be significant in
comparison with the overwhelming number of majority charge carriers.
A similar process is involved if a Group V element is used for doping. This
gives an extra electron, surplus to covalent bonding requirements, for each
atom of the doping agent. These electrons are negatively charged and so an
n-type semiconductor has been produced. In an n-type semiconductor, the
majority charge carriers are electrons. The conductivity has been greatly
enhanced as before, but this time conduction occurs in an extra energy level
just below the conduction band, which is sometimes called the donor band.
Therefore, in p-type and n-type semiconductors conduction can occur easily
because there is effectively unfilled space within either the valence or the
conduction band, respectively.
Semiconductors are crucial to modern life. According to estimates (Sheffield
University) 43% of all semiconductor production goes into computers, 23%
into consumer products, 13% into communication and 12% into
manufacturing.

SEMICONDUCTORS AND BAND THEORY

Summary of intrinsic and extrinsic semiconductors

Semiconductors allow conduction by means of negative charge carriers,
which are electrons, or positive charge carriers, which are holes.
The energy band gap in semiconductors is small enough that thermal
excitation is sufficient for significant numbers of electrons to be able to
move up from the valence to the conduction band.
Intrinsic semiconductors, such as pure silicon, will always have equal
numbers of holes and electrons since each conduction electron will leave
behind a hole.
Semiconductors may be doped with impurities that add either extra
electrons or holes to the lattice.
These doped semiconductors now have a majority charge carrier present
and are known as extrinsic semiconductors.
Group III doping agents result in p-type extrinsic semiconductors, which
contain extra holes.
Group V doping agents result in n-type extrinsic semiconductors, which
contain extra electrons.

SEMICONDUCTORS AND BAND THEORY

pn junctions
If a single semiconducting crystal is doped in such a way that one end is
p-type and the other n-type, then some very useful properties come into play.
The interface between the p-type and n-type sections is known as a pn
junction. In this boundary region, electrons from the n-type material may
diffuse across the boundary and combine with holes from the p-type material,
and vice versa. This results in a lack of majority charge carriers in the
immediate vicinity of the junction and as such the region is known as the
depletion zone. The pn junction greatly affects the conductivity of the
semiconductor as a whole. When electrons from the n-type material diffuse
into the p-type material, they form negative ions as they combine with holes.
Positive ions are also left behind in the n-type material. Eventually, this
process results in there being no further diffusion of electrons or holes as a
result of Coulomb attraction and repulsion (Figures 8a and 8b).

Depletion
zone

--ptype

---------

++
+
++
++

ntype

Junctio
++
n
+
Figure 8a In a pn junction a depletion zone is formed by the diffusion of electrons from the
n-type material into the p-type material. As the electrons combine with holes, ions are
formed in the depletion zone.

SEMICONDUCTORS AND BAND THEORY

p-type

n-type

Conduction
band

Depletion
zone
Valence band

Figure 8b Band theory gives us a model for explaining what happens at a pn junction.
Holes and electrons diffuse towards the junction in their different bands. At the junction they
combine, producing the depletion zone. This pn junction is forward biased.

The pn junction may only allow current to flow if it is forward biased. By

connecting the negative terminal of a power supply to the n-type material, the
junction becomes forward biased. Electrons may be pushed across the
depletion zone if the supply has a sufficient potential difference to overcome
the Coulomb repulsion discussed above. This is typically of the order of 0.7
V. Once the depletion zone is crossed, conduction is easily facilitated by the
majority charge carriers in each of the semiconducting materials.
If the pn junction is reverse biased, ie the n-type material is connected to the
positive terminal of a power supply, the depletion zone effectively becomes a
greater and greater barrier to conduction. One can imagine the depletion zone
illustrated in Figures 8a and 8b becoming a higher and higher barrier to
conduction as electrons are driven further and further back from the depletion
zone. The junction can only allow a tiny leakage current to flow because of
the intrinsic semiconductors electrons and holes. Since this is usually
undesirable, silicon is preferable to germanium because its leakage current is
so much smaller as a result of it having a larger energy gap between its
valence and conduction bands (Figure 5).
Eventually, if the reverse voltage continues to increase, the semiconductor
will break down and may result in damage to the junction.

SEMICONDUCTORS AND BAND THEORY

Photovoltaic cells
A photovoltaic cell consists of a very thin layer of p-type semiconductor that
is in contact with a layer of n-type material. The conduction electrons are
freed through the action of photons of light. The photons provide sufficient
energy to the electrons to enable them to jump up across the energy gap to the
conduction band, leaving behind a hole. The band gap energy for silicon is of
the order of 1.1 eV, and so only photons with at least 1.1 eV of energy can
cause the release of conduction electrons. The wafer of semiconductor is very
thin and so there is a good chance that this process will happen at or very
close to the pn junction. The electric field produced by the depletion layer at
this junction forces the electron and hole apart, creating a potential
difference, and so a current can flow if the cell is connected to a circuit. The
p-type layer must be very thin, perhaps 1 m thick, to prevent conduction
electrons from being captured and immobilised by holes.

Photons interact
with electrons close
to pn junction

n-type

Contac
ts
Figure 9 Simplified photovoltaic cell in cross-section.

ptype

SEMICONDUCTORS AND BAND THEORY

A solar cell must also have a layer of antireflective coating (not shown in
Figure 9) to improve efficiency because a silicon crystal is so shiny that
without this layer many of the photons would be reflected before they could
cause the release of an electron. Even so, typical efficiencies of solar cells
stand close to 15% and the greatest efficiencies rarely exceed 25%. Finally,
the cell must be coated in glass to reduce damage from the elements.
A compromise must be reached in the choice of materials used to optimise the
performance of the solar cell. By reducing the band gap energy in the
semiconducting material, photons with longer wavelengths and lower
frequencies may be harnessed to free electrons and holes. Although this may
seem desirable and will release more charges, it has the effect of reducing the
strength of the electric field across the junction. It turns out that a band gap
energy of about 1.4 eV is close to ideal, maximising the current and voltage,
and therefore the power, of the cell.

Light-emitting diodes
When a diode is forward biased, electrons from the n-type semiconductor may
move across the junction and combine with holes in the p-type material. The
electrons in the n-type semiconductor move within the higher energy

SEMICONDUCTORS AND BAND THEORY

conduction band, and as they cross the junction they move briefly into the
empty conduction band of the p-type material. Since the lower energy p-type
valence band is only partially filled, however, the electrons rapidly fall into
an energy level within that valence band. In effect, electrons fall into holes,
and as this happens energy is released in the form of emitted photons (Figure
10). For ordinary diodes, these photons have a relatively low frequency and
long wavelength, which means that they fall outside the visible spectrum.
However, in the construction of LEDs the semiconducting materials may be
engineered in such a way as to result in the photons having sufficiently high
frequency that they fall within the range of visible light.
The frequency of the light emitted from LEDs is controlled by the size of the
energy gap between the conduction and valence bands. A bigger gap will
result in a larger energy change and, in accordance with the relationship E =
hf, a higher frequency of light will be emitted. So, a small energy gap will
result in red light and a much larger energy gap is required for green or blue
light.

ntype

Junctio
n

ptype
Photons
emitted as
electrons drop
down to
valence band

Conduction band
(partially filled)

Valence
band
(partially
filled)
Figure 10 In an LED electrons cross the junction from n to p in the conduction band. Once
on the p side of the junction, they fall back across the energy gap to the valence band. This
releases photons of light.

SEMICONDUCTORS AND BAND THEORY

Since electrons usually drop from the bottom of the conduction band into the
top of the valence band, light from LEDs tends to be nearly (although not
completely) monochromatic, with a narrow emission spectrum.
To alter the energy gap in an LED, different doping agents are used. They
may typically include indium, gallium and nitrogen to produce blue light,
gallium and phosphorous for green light, and gallium, phosphorous and
arsenic for red light. By using combinations of red, green and blue it is
possible to produce any colour of light and this has led to the advent of LED
televisions. Furthermore, by varying the proportions of the doping agents,
single intermediate colours may also be produced.

SEMICONDUCTORS AND BAND THEORY

Why use LEDs?

LEDs have many advantages over other light sources.
An LED typically has an efficiency of around 80%.
This is far superior even to energy-saving compact
lights because of the way in which light is produced.
This clearly has major implications for reducing
carbon dioxide emissions and fuel bills. Because of the
very low levels of undesirable thermal energy
produced, LEDs may also be expected to have much
greater longevity than conventional light sources. LED
and OLED TVs are expensive but with low power
usage and high reliability they are set to become more
common in future.
Infrared LEDs are used in remote controls because of
their reliability and low power consumption.
Ultraviolet LEDs can be used for detecting counterfeit
notes and even for sterilisation procedures.
LEDs have extremely fast switching speeds, which
means that they are particularly useful for applications
where light sources must be pulsed, strobed or simply
switched on and off rapidly with reliability.
Additionally, they work well at low temperatures, are
shock resistant and contain fewer hazardous materials
than energy-saving light bulbs.

SEMICONDUCTORS AND BAND THEORY

Summary of pn junctions,
LEDs and photovoltaic
cells
The interface between p-type and ntype material is called the pn
junction.
Majority charge carriers diffuse
towards the junction and electrons
combine with holes, forming ions.
This lack of charge carriers results in a
depletion zone across the pn
junction, with positive ions on the ntype side and negative ions on the ptype side.
If the p-type material is connected to
the positive terminal of a supply and
the n-type to the negative terminal,
then the junction is forward biased.
If the terminals are reversed, the junction is
reverse biased and cannot conduct.
LEDs emit photons of light as electrons fall
from the conduction band of the n-type material
into holes in the valence band of the p-type
material.

SEMICONDUCTORS AND BAND THEORY

The bigger the energy gap between the bands,

the greater the energy, and therefore the
frequency, of the emitted photons.
Photovoltaic cells use the energy of absorbed
photons to separate electrons and holes and
thus produce a potential difference.

Thank you!!