HETT406: Design of

Digital & VLSI Systems

Unit 1: Semiconductor Devices and

Technology & VLSI Overview

Prof E Mashonjowa
Semiconductor Theory and Devices
◼ 1.1 Band Theory of Solids
◼ 1.2 Semiconductor Theory
◼ 1.3 Semiconductor Devices
◼ 1.4 VLSI Overview
◼ 1.5 Nanotechnology & Nanoscale Electronics
It is evident that many years of research by a great many people, both before
and after the discovery of the transistor effect, has been required to bring our
knowledge of semiconductors to its present development. We were fortunate to
be involved at a particularly opportune time and to add another small step in the
control of Nature for the benefit of mankind.
- John Bardeen, 1956 Nobel lecture

2
1.1: Band Theory of Solids

◼ Important prorties of solids:

❑ Electrical
❑ Structural,
❑ Thermal, and
❑ Magnetic

In this electronics we concentrate on electrical

conduction.

3
Categories of Solids

◼ There are three categories of solids, based on their

conducting properties:
❑ conductors

❑ semiconductors

❑ insulators

4
Electrical Resistivity
and Conductivity of
Selected Materials
at 293 K

The electrical conductivity at

room temperature is quite
different for each of these
three kinds of solids:
• Metals and alloys have
the highest conductivities,
followed by
• semiconductors, and
then by
• insulators

5
Band Theory of Solids
◼ In order to account for decreasing resistivity with
increasing temperature as well as other
properties of semiconductors, a new theory
known as the band theory is introduced.

◼ The essential feature of the band theory is that

the allowed energy states for electrons are
nearly continuous over certain ranges, called
energy bands, with forbidden energy gaps
between the bands.

6
 Band Theory and Conductivity
◼ Band theory helps us understand what makes a conductor,
insulator, or semiconductor.
1) Good conductors like copper can be understood using the free
electron
2) It is also possible to make a conductor using a material with its
highest band filled, in which case no electron in that band can be
considered free.
3) If this filled band overlaps with the next higher band, however (so
that effectively there is no gap between these two bands) then an
applied electric field can make an electron from the filled band jump
to the higher level.
◼ This allows conduction to take place, although typically with
slightly higher resistance than in normal metals. Such materials
are known as semi-metals.

7
 Valence and Conduction Bands

◼ The band structures of insulators and semiconductors

resemble each other qualitatively. Normally there exists in
both insulators and semiconductors a filled energy band
(referred to as the valence band) separated from the next
higher band (referred to as the conduction band) by an
energy gap.

◼ If this gap is at least several electron volts, the material is

an insulator. It is too difficult for an applied field to
overcome that large an energy gap, and thermal
excitations lack the energy to promote sufficient numbers
of electrons to the conduction band.

8
Smaller energy gaps create semiconductors

◼ For energy gaps smaller than about 1 electron

volt, it is possible for enough electrons to be
excited thermally into the conduction band, so
that an applied electric field can produce a
modest current.

The result is a semiconductor.

9
1.2: Semiconductor Theory

◼ At T = 0 we expect all of the atoms in a solid to

be in the ground state.
◼ The distribution of electrons (fermions) at the
various energy levels is governed by the Fermi-
Dirac distribution

β = (kT)−1 and EF is the Fermi energy.

10
 Temperature and Resistivity
◼ When the temperature is increased from T = 0, more and
more atoms are found in excited states.

◼ The increased number of electrons in excited states explains

the temperature dependence of the resistivity of
semiconductors.
✓ Only those electrons that have jumped from the valence
band to the conduction band are available to participate
in the conduction process in a semiconductor.

✓ More and more electrons are found in the conduction

band as the temperature is increased, and the resistivity
of the semiconductor therefore decreases.
11
 Some Observations
◼ Although it is not possible to use the Fermi-Dirac factor to
derive an exact expression for the resistivity of a
semiconductor as a function of temperature, some
observations follow:
1) The energy E in the exponential factor makes it clear why the band gap is
so crucial. An increase in the band gap by a factor of 10 (say from 1 eV to
10 eV) will, for a given temperature, increase the value of exp(βE) by a
factor of exp(9βE).
◼ This generally makes the factor FFD so small that the material has to
be an insulator.
2) Based on this analysis, the resistance of a semiconductor is expected to
decrease exponentially with increasing temperature.
◼ This is approximately true — although not exactly, because the
function FFD is not a simple exponential, and because the band gap
does vary somewhat with temperature.

12
 Holes and Intrinsic Semiconductors
◼ When electrons move into the conduction band, they leave
behind vacancies in the valence band. These vacancies are
called holes. Because holes represent the absence of negative
charges, it is useful to think of them as positive charges.
◼ Whereas the electrons move in a direction opposite to the
applied electric field, the holes move in the direction of the
electric field.
◼ A semiconductor in which there is a balance between the
number of electrons in the conduction band and the number of
holes in the valence band is called an intrinsic
semiconductor.
◼ Examples of intrinsic semiconductors: pure carbon and Ge

13
 Impurity Semiconductor
◼ It is possible to fine-tune a semiconductor’s properties by
adding a small amount of another material, called a dopant, to
the semiconductor creating what is a called an impurity
semiconductor.

◼ As an example, silicon has four electrons in its outermost shell

(this corresponds to the valence band) and arsenic has five.
❑ Thus while four of arsenic’s outer-shell electrons participate in
covalent bonding with its nearest neighbors (just as another
silicon atom would), the fifth electron is very weakly bound.
❑ It takes only about 0.05 eV to move this extra electron into the
conduction band.
◼ The effect is that adding only a small amount of arsenic to
silicon greatly increases the electrical conductivity.

14
n-type Semiconductor

◼ The addition of arsenic to silicon creates what is

known as an n-type semiconductor (n for
negative), because it is the electrons close to the
conduction band that will eventually carry
electrical current.

The new arsenic energy levels just below the

conduction band are called donor levels
because an electron there is easily donated to
the conduction band.

15
Acceptor Levels & p-type semiconductors
◼ Consider what happens when indium is added to silicon.
❑ Indium has one less electron in its outer shell than silicon. The result is
one extra hole per indium atom. The existence of these holes creates
extra energy levels just above the valence band, because it takes
relatively little energy to move another electron into a hole
❑ Those new indium levels are called acceptor levels because they can
easily accept an electron from the valence band. Again, the result is an
increased flow of current (or, equivalently, lower electrical resistance)
as the electrons move to fill holes under an applied electric field
◼ It is always easier to think in terms of the flow of positive charges
(holes) in the direction of the applied field, so we call this a p-type
semiconductor (p for positive).
❑ acceptor levels p-Type semiconductors
◼ In addition to intrinsic and impurity semiconductors, there are many
compound semiconductors, which consist of equal numbers of
two kinds of atoms.

16
1.3: Semiconductor Devices

pn-junction Diodes
◼ Here p-type and n-type semiconductors are
joined together.

◼ The principal characteristic of a pn-junction

diode is that it allows current to flow easily in
one direction but hardly at all in the other
direction.

We call these situations forward bias and

reverse bias, respectively.
17
 Operation of a pn-junction Diode
(a) This is the no-bias case. The small
thermal electron current (It) is offset by
the electron recombination current (Ir).
The net positive current (Inet) is zero.

(b) The diode is in reverse bias. (c) Here the diode is in forward bias.
Now Ir is slightly less than It. & there Because current can readily flow from p
is a small net flow of electrons from
to n, Ir can be much greater than It.
p to n & positive current from n to p.

Figure 11.12: The operation of a pn-junction diode. 18

 Bridge Rectifiers
◼ The diode is an important tool in many kinds of electrical circuits (Fig 1.14).
◼ The bridge rectifier is set up so that it allows current to flow in only one
direction through the resistor R when an alternating current supply is placed
across the bridge.
◼ This is the first step in changing alternating current to direct current. The
design of a power supply can be completed by adding capacitors and
resistors in appropriate proportions.
◼ This is an important application, because direct current is needed in many
devices and the current that we get from our wall sockets is alternating
current.

Figure 1.14: Circuit diagram for a diode bridge rectifier. 19

 Zener Diodes
◼ The Zener diode is made to operate under reverse bias once a sufficiently high
voltage has been reached.
◼ The I-V curve of a Zener diode is shown in Figure 1.15. Notice that under
reverse bias and low voltage the current assumes a low negative value, just as
in a normal pn-junction diode. But when a sufficiently large reverse bias voltage
is reached, the current increases at a very high rate.

Figure 1.15: Typical I-V curve for a Zener diode. Figure 1.16: A Zener diode reference circuit.

20
 Light Emitting Diodes
◼ Another important kind of diode is the light-emitting diode (LED). Whenever
an electron makes a transition from the conduction band to the valence band
(effectively recombining the electron and hole) there is a release of energy in
the form of a photon (Figure 1.17).
◼ In some materials the energy levels are spaced so that the photon is in the
visible part of the spectrum. In that case, the continuous flow of current through
the LED results in a continuous stream of nearly monochromatic light.

Figure 1.17: Schematic of an LED. A photon is released as an

electron falls from the conduction band to the valence band. 21
 Photovoltaic Cells
◼ In a photovoltaic cell, a solar cell takes incoming light energy and
turns it into electrical energy.
◼ The solar cell can be thought of as an LED in reverse (Figure 1.18).
◼ Even though the efficiency of most solar cells is low, their
widespread use could potentially generate significant amounts of
electricity.

Figure 1.18: (a) Schematic of a photovoltaic cell. (b) A schematic showing more of the working
parts of a real photovoltaic cell. From H. M. Hubbard, Science 244, 297-303 (21 April 1989). 22
Transistors

◼ Another use of semiconductor technology is in

the fabrication of transistors, devices that
amplify voltages or currents in many kinds of
circuits.

◼ The first transistor was developed in 1948 by

John Bardeen, William Shockley, and Walter
Brattain (Nobel Prize, 1956).

23
 Transistors
24

Transistors are categorized by:

 Semiconductor material:

germanium, silicon, gallium arsenide, silicon carbide

 Structure:
BJT, JFET, MOSFET, IGBT, "other types"
 Polarity:
NPN, PNP (BJTs);
N-channel, P-channel (FETs)
 The Bipolar Junction Transistor (BJT)
25

 A bipolar junction transistor consists of three regions of

doped semiconductors.
 emitter
 base
 collector
 p-n-p transistor or n-p-n transistor

Figure 1.22: (a) In the npn

transistor, the base is a p-type
material, and the emitter and
collector are n-type. (b) The two-
diode model of the npn
transistor.
 Bipolar Junction Transistor

 the arrow on the emitter lead specifies the direction of current

flow (conventional) when the emitter-base junction is forward
biased.
 IE, IB, and IC are assumed positive when the currents flow into the
transistor.
26
Bipolar Junction Transistor

27
 The transistor as a current amplifier
A small current in the base region can be used to control a larger
current flowing between the emitter and collector.

The larger collector current IC is related to the base current IB by:

IC =  IB
 is the current gain (also called hFE).  ~ 50 –250; typically 100.
28
 Types of BJT amplifiers
29

 Since a junction transistor is a three-terminal device and

there are four input-output terminals, one of the transistor
terminals must be common to the input and output circuits.
 This leads to the names for the three basic types of
amplifiers:
 "common emitter",
 “common collector”, and

 “common base”.
 BJT configurations

Common base Common collector Common emitter

30
 BJT Currents
31

IE = IB + IC = ( + 1) IB  IC (for large  or  >> 1)

◆ Because  >> 1, the main utility of a transistor is evident:

• we are able to control a large current IC  IE with a small
current IB, thus achieving current amplification.
 Field Effect Transistors (FETs)
32

 The JFET is a long channel of semiconductor

material, doped to contain an abundance of positive
charge carriers (p-type), or of negative carriers (n-
type).
 Contacts at each end form the source(S) and drain
(D).
 The gate (G) (control) terminal has doping opposite
to that of the channel, which surrounds it, so that
there is a P-N junction at the interface.
 Types of Field Effect Transistors
(The Classification)
33

n-Channel JFET
FET JFET
p-Channel JFET

MOSFET (IGFET)

Enhancement Depletion
MOSFET MOSFET

n-Channel p-Channel n-Channel p-Channel

EMOSFET EMOSFET DMOSFET DMOSFET

• JFET (Junction Field-Effect Transistor): Applications in analog and RF circuit design

• MOSFET (Metal-Oxide Semiconductor Field-Effect Transistor): Primary component
in high-density VLSI chips such as memories and microprocessors
 Field Effect Transistors (FETs)
34

n-Channel JFET

p-Channel JFET

The gate, drain, and source correspond roughly to the

base, collector, and emitter of BJTs.
 Field Effect Transistors (FETs)
35

 The names of the terminals refer to their functions.

 The gate terminal may be thought of as controlling the
opening and closing of a physical gate.
 This gate allows the flow of electrons through or blocks
their passage by constricting the current-carrying channel..
 Electrons flow from the source terminal towards the drain
terminal if influenced by an applied voltage.
 JFET operation
36

 In normal usage, a voltage is applied across

the channel, with the drain being made
positive with respect to the source.
 Thus, electrons will move through the n-type
channel from source to drain, with the
applied voltage and the resistance of the
channel determining the magnitude of
current flowing, in accordance with Ohm's
Law.
 JFET operation
37

 However, the depletion region in the

middle of the channel will not carry
current.
 Therefore, the effective width of the
channel is restricted by the depletion
region, and its effective resistance is
higher in this part of the channel, than
in any other part of the channel.
 JFET operation
38

 If the gate becomes forward biased with

respect to the channel, of course, the depletion
region shrinks away and no longer controls the
flow of current through the channel.

 This is not generally useful in practical

electronic circuits.
 JFET operation
39

 However, if the gate becomes

reverse biased with respect to
the channel, the applied electric
field will enlarge the depletion
region, as shown.

 As a result, the working channel

width is reduced, and the
effective channel resistance
is increased significantly.
 JFET operation
40

 In this manner, a small voltage applied

to the gate can have a profound effect
on the current flowing through the
channel.
 If the applied reverse bias becomes high
enough, the depletion region can cover
the entire width of the channel, and cut
off current flow completely.
• Because the electric field produced by the applied
gate voltage controls the resistance of the channel
(and therefore the current flowing through the
channel), this device is known as a field-effect
 Schottky Barriers
◼ Here a direct contact is made between a metal and a
semiconductor.
◼ If the semiconductor is n-type, electrons from it tend to migrate
into the metal, leaving a depleted region within the semiconductor.
This will happen as long as the work function of the metal is
higher (or lower, in the case of a p-type semiconductor) than that
of the semiconductor.
◼ The width of the depleted region depends on the properties of the
particular metal and semiconductor being used, but it is typically
on the order of microns.
◼ The I-V characteristics of the Schottky barrier are similar to those
of the pn-junction diode.
◼ When a p-type semiconductor is used, the behavior is similar but
the depletion region has a deficit of holes.
41
Schottky Barriers

42
 Schottky Barriers

Figure 1.26: (a) Schematic drawing of a typical Schottky-barrier FET. (b)

Gain versus frequency for two different substrate materials, Si and GaAs.
From D. A. Fraser, Physics of Semiconductor Devices, Oxford: Clarendon
Press (1979).

43
Schottky Diodes

urrent
chott y
diode
◼ Because of their majority-
carrier conduction
diode
mechanism, Schottky diodes
can achieve greater
olta e
switching speeds than p–n
junction diodes, making them
appropriate to rectify high-
frequency signals.
◼ voltage is always implied)
due to direct tunneling.

44
 Semiconductor Lasers
◼ Semiconductor lasers operate using population inversion —
an artificially high number of electrons in excited states

◼ In a semiconductor laser, the band gap determines the

energy difference between the excited state and the ground
state

◼ Semiconductor lasers use injection pumping, where a large

forward current is passed through a diode creating electron-
hole pairs, with electrons in the conduction band and holes in
the valence band.

A photon is emitted when an electron falls back to the

valence band to recombine with the hole.
45
Semiconductor Lasers

◼ Since their development, semiconductor lasers have

been used in a number of applications, most notably in
fiber-optics communication.

◼ One advantage of using semiconductor lasers in this

application is their small size with dimensions typically
on the order of 10−4 m.

◼ Being solid-state devices, they are more robust than

gas-filled tubes.

46
 Integrated Circuits

◼ The most important use of all these semiconductor

devices today is not in discrete components, but
rather in integrated circuits called chips.

◼ Some integrated circuits contain a million or more

components such as resistors, capacitors, and
transistors.

◼ Two benefits: miniaturization and processing

speed.

47
 Why Make ICs
48

 Integration improves
 size

 speed

 power

 Integration reduces manufacturing costs

 (almost) no manual assembly
 IC Evolution (1/3)
49

 SSI: Small Scale Integration (early 1970s)

 contained 1 -10 logic gates

 MSI: Medium Scale Integration

 contained 100 logic gates capable of logic functions, counters

 LSI: Large Scale Integration

 contained 1000 logic gates: first microprocessors on the chip

 VLSI: Very Large Scale Integration

 the process of creating an integrated circuit (IC) by combining
millions of MOS transistors onto a single chip.
 now offers 64-bit microprocessors, complete with L1 & L2 cache
memory, floating-point arithmetic unit(s), etc.
 L1 (level-1) cache memory is built onto the microprocessor chip itself
 L2 (level-2) on separate chip (on expansion card) accessed more quickly)
 IC Density of Integration
50
 IC Evolution (2/3)
51

 Bipolar technology
 TTL (transistor-transistor logic)
 ECL (emitter-coupled logic)
 MOS (Metal-oxide-silicon)
 although invented before bipolar transistor, was initially
difficult to manufacture
 nMOS (n-channel MOS) technology developed in 1970s
required fewer masking steps, was denser, and consumed
less power than equivalent bipolar ICs
 MOS IC was cheaper than a bipolar IC and led to
investment and growth of the MOS IC market.
 IC Evolution (3/3)
52

 Aluminum gates replaced by polysilicon by early 1980

 CMOS (Complementary MOS) n-channel and p-channel
MOS transistors; lower power consumption, simplified
fabrication process
 Bi-CMOS - hybrid Bipolar, CMOS (for high speed)
 GaAs - Gallium Arsenide (for high speed)
 Si-Ge - Silicon Germanium (for RF)
Nanotechnology & Nanoscale Electronics

◼ Nanotechnology is generally defined as the scientific

study and manufacture of materials on a submicron
scale.

◼ These scales range from single atoms (on the order of

0.1 nm up to 1 micron (1 m = 1000 nm).

◼ This technology has applications in engineering,

chemistry, and the life sciences and, as such, is
interdisciplinary.

53
 Carbon Nanotubes

◼ In 1991, Japanese physicist

Sumio Iijima discovered a new
geometric arrangement of pure
carbon into large molecules.

◼ In this arrangement, known as a

carbon nanotube, hexagonal
arrays of carbon atoms lie along a
cylindrical tube instead of a
spherical ball.

54
 Structure of a Carbon Nanotube
Figure 1.30: Model of a carbon
nanotube, illustrating the
hexagonal carbon pattern
superimposed on a tubelike
structure. There is virtually no
limit to the length of the tube.

http://www.hpc.susx.ac.uk/
~ewels/img/science/nanotubes/.

55
Carbon Nanotubes

◼ The basic structure shown in Figure 1.30 leads to

two types of nanotubes. A single-walled nanotube
has just the single shell of hexagons as shown.

◼ In a multi-walled nanotube, multiple layers are

nested like the rings in a tree trunk.

◼ Single-walled nanotubes tend to have fewer defects,

and they are therefore stronger structurally but they
are also more expensive and difficult to make.

56
Applications of Nanotubes

◼ By their strength they are used as structural

reinforcements in the manufacture of composite
materials
❑ (batteries in cell-phones use nanotubes in this way)

◼ Nanotubes have very high electrical and thermal

conductivities, and as such lead to high current
densities in high-temperature superconductors.

57
Nanoscale Electronics
◼ One problem in the development of truly small-scale
electronic devices is that the connecting wires in any
circuit need to be as small as possible, so that they do
not overwhelm the nanoscale components they connect.

◼ In addition to the nanotubes already described,

semiconductor wires (for example indium phosphide)
have been fabricated with diameters as small as 5 nm.

◼ These nanowires have been shown useful in connecting

nanoscale transistors and memory circuits.

These are referred to as nanotransistors.

58
Information Science

◼ It is possible that current photolithographic

techniques for making computer chips could be
extended into the hard-UV or soft x-ray range,
with wavelengths on the order of 1 nm, to
fabricate silicon-based chips on that scale.

◼ Possible quantum effects as devices become

smaller, specifically the superposition of
quantum states possibly leading to quantum
computing.

59
Moore’s Law and omputin ower

Figure 1.29: Moore’s law - the theory that the processing

60
power of computers doubles every two years or so