Semiconductor electronics

Why are semiconductor electronics

important?
Semiconductor electronics has two distinct
components:
• Photovoltaics
• Processors
Understanding how semiconductor electronic
devices are made will allow you to generate
electricity and solve complex problems using
computers.
 Photovoltaics
The factor that enables photovoltaic cells to work is the photoelectric effect. The
description of the photoelectric effect won albert Einstein the Nobel Prize for physics in
1921. The basic working of the photoelectric effect is that electrons are liberated from
atoms when a photon collides with the atom.
Atoms contain different regions wherein electrons are housed. There are two distinct
bands namely:
• The valence band
• The conduction band
When electrons are bound to an individual atom they are said to be in the valence band.
Valence electrons help in the bonding of distinct atoms. The conduction band is a portion
of the band where electrons can hop between atoms, not binding with any particular one.
Valence electrons have lower energies while conduction band electrons have higher
energies, which is why they do not fall back into the valence band. The energy gap
between the conduction band and the valence band is called the band gap. Not all atoms
have identical band gaps. Electrical conductors have very small band gaps, while insulators
have very large band gaps .
Conduction band
Incoming Photon

Band gap

Valence band
Electron
In semiconductor materials there will be electrons in the conduction band
already due to thermal excitation, there will also be gaps in the valence
band where electrons have been liberated. These gaps in the valence band
are colloquially referred to as holes. These “holes” in valence band allow
other electrons to move into this gap, once again leaving a hole from
where the electron came. So holes therefore can move within the lattice
depending on whether there is an electric field present. The electrons and
holes within the matrix are called carriers, due to the fact that they are
mobile. In silicon at room temperature there is a given concentration of
holes and electrons, this is called the intrinsic concentration. This intrinsic
concentration determines whether the semiconductor will be able to
conduct a current. At very low temperatures, semiconductors act like
insulators while at room temperatures they can conduct a current .
There are other ways to increase the carrier concentration apart from
increasing the temperature. One of these ways is by introducing ions
into the semiconductor matrix that have valences different to that of
the semiconductor material. Usually group 3 and 5 elements are used
to dope silicon. Silicon has 4 valence electrons which it shares with its
surrounding atoms covalently. If a group 5 ion is introduced into the
matrix, it will bond covalently with the surrounding silicon atoms,
making bonds with 4 surrounding silicon atoms, however silicon only
needs 8 electrons in its outer orbital and the ninth electron, that is
additional due to the 5 valence electrons in group 5, will be unbonded
and will flow into the conduction band. This will increase the carrier
concentration. If a group 3 ion is implanted, there will be a hole present
in the matrix by the same logic . This arrangement can be seen below.
 Group 5
Dopant

Unbonded electron
from group 5 dopant

Group 4
semiconductor

Group 3
Dopant

Hole present due to Lack

of Electrons from Group 3
Dopant

Holes in the Matrix due to

Group 3 Doping
Implanting ions into a semiconductor is called doping. It has a dual
purpose:
• Increase the carrier concentration
• Create a built in electric field to force electrons around a circuit.

In order for a semiconductor to create useful current to be used in

external circuits, and not just non-directional electron and hole
movement, an electric field needs to be set up to control the carrier
movement. This is done by creating a positively doped silicon matrix
and a negatively doped one, and then putting these layers adjacent
to each other. The surplus electrons from the negatively doped side
will flow into the holes in the positively doped side. This will occur
until an electric field is set up to resist the further flow of electrons .
The structure can be seen below.
 Positively doped semiconductor
Holes
Direction of electric field

- - - - - - - - - - - - - - - - - - - - Negative Cores

Depleted Region
+ + + + + + + + + + + + + + + + + + + + Positive Cores

Negatively doped semiconductor Electrons

Electrons migrating into holes

The electric field is set up due to the fact that once the electrons leave the negative side
of the junction, a positive core is left behind, and what was once was neutrally charged
now has more protons than electrons. On the positive side, electrons move into holes
and shift the charge balance to negative due to the fact that there are more electrons
than protons on this side of the junction. An electric field is then set up between the
positively and negatively charged particles. The region where the electrons move into
the holes and vice versa is called the depleted region. This electric field is integral to the
operation of photovoltaic cells and without it, no useful current would be produced.

Once a photon has created an electron hole pair either in the positive or negatively
doped silicon, the minority carrier (Hole in N-type semiconductors and electrons in P-
type semiconductors) will be swept across the depleted zone by the electric field. Once
the minority carrier has been swept across the depleted region, it will once again be a
majority carrier. If there is a front and rear charge collection plate that is connected to a
circuit, the electron or hole will flow through this pathway, expending its energy and
eventually recombining when it reaches the side where it is once again a minority
carrier. This can be seen in the figure below.

The top plate that collects either holes or electrons, is either made up of metallic fingers
that collect the charge while letting photons into the semiconductive layer, or it is made
of a transparent coating oxide which collects the majority carriers and allows light to
pass through unimpeded.
 Hole migrating to the N-type
side
Photon used to create the
electron hole pair

Transparent collector plate

P-type
Electron hole pair

N-type Electron and hole merging

after their energy has been
dissipated
Back collector plate

Electron migrating to the P-type

side
 Manufacturing solar cells
Monocrystalline solar cells are cells that have a highly ordered crystal lattice and are doped
positively and negatively in order to transport holes and electrons. There are several
manufacturing techniques including the Czochralski process and Float-zone technique. Their
manufacturing techniques will be discussed below.
In 1918 J Czochralski submitted a paper outlining the method to create a monocrystalline ingot of
silicon. He showed that if a single seed of monocrystalline silicon was lowered into a bath of
molten polycrystalline silicon, a monocrystalline silicon ingot forms on the seed as the seed is
pulled out. It is important to ensure that the seed is raised very slowly out of the bath so that no
dislocations form in the crystal matrix . This ingot of monocrystalline silicon is then sawed into
wafers using a band saw. Unfortunately during this process a significant amount of the material is
lost due to the sawing. These wafers are then used in the monocrystalline solar panels used
widely today. This process is very energy intensive and requires very pure silicon. The process
itself does however contribute contaminants in the form of carbon from the bath walls. Oxygen is
also a contaminant in this process .
The majority of monocrystalline silicon is produced in this way, due to the fact that you can
produce large diameter ingots with a highly ordered crystal structure. This process is currently the
most suited to mass producing highly efficient solar panels, however there is definitely scope to
find new processes that are cheaper and less energy intensive.
Stage 1

Stage 2

Stage 3

Stage 4

Stage 5

Czochralski process
A technique has also been developed that works on the same
principles as the Czochralski process but is slightly modified. A seed
is placed beneath a column of silicon, a mobile heating element then
slowly moves downward, heating a small region of the silicon and
allowing it to form a monocrystalline structure. This is called the
float zone technique. This process is less energy intensive than the
original czochralski process, however it is still very energy intensive.
Far less oxygen contaminant is present in this silicon due to the fact
that there is less movement of the silicon and hence it is more
difficult for impurities to diffuse into the structure. This process is
mostly used in laboratory grown silicon wafers.
Multicrystalline solar cells still have a highly ordered crystal
structure, however there are individual grains of monocrystalline
silicon within the wafer. Grain sizes are typically in the range of a few
millimeters.
 Processors
• Processors are made up of a combination of
billions of transistors. Transistors consist of 3
layers of doped silicon. Either NPN or PNP
 How transistors work
• Transistors work as switches. A sandwich of two
negatively doped silicon layers envelops a positively
doped layer. At the interface between the positively
doped and negatively doped layer a depletion zone
exists. This prevents electrons from flowing between the
layers. When a current is supplied to the base(usually
positive) it causes electrons in the p type layer to be
attracted to the base. This causes the depletion zone to
be minimised and a conducting channel to be created,
which allows electrons to flow across the transistor.
 How transistors compute
• When transistors are arranged in specific ways
they produce structures called logic gates.
These logic gates allow calculations to be done
when they are arranged in certain
combinations. In the next slide some logic
gates are shown
Transistor arrangement for OR gate
Logic gate arrangement for full adder
How microprocessors are manufactured
1. Monocrystalline ingots of silicon are created
2. These ingots are sliced into wafers and are polished.
3. A photoresitive layer is deposited using spin deposition.
4. The photo resistive layer is exposed to UV light in the pattern of the final circuit
5. The exposed photoresistive layer is removed by chemical washes.
6. The silicon is doped via ion implantation
7. A second photoresist layer is applied. The exposed areas are removed by chemical
washes as before.
8. The silicon wafer is then etched to remove unwanted silicon.
9. A insulative layer is then deposited onto the surface along with electroplated copper
directly above that. The copper is then ground off untill it is only in the channels
10.Interconnects are formed
11.Heat sinks are applied
12.Processors are tested and binned according to their performance