Chapter 3.

Optical Sources :

Critical Parameters:
• Emission wavelength
• Spectral width
• Modulation speed, amplitude and phase modulation capability
• Divergence/ability to couple into fiber
• Power/energy consumption/efficiency
• Compactness, reliability, cost-effectiveness
• Possible sources: LEDs, LASER Diodes

1
Light Production:
• Taking the most general view, there is only one way that light can be
produced: that is, through the rapid change of state of an electron from
a state of relatively high energy to a (more stable) state of lower
energy.
• When this happens, the energy has to go somewhere, and it is often
emitted in the form of light. Of course, this almost always takes place in
the context of a particular material and structure.
• When light is not emitted, energy is given up in the form of “phonons”.
Phonons are discrete quantities of kinetic energy which cause
mechanical vibrations of the atom concerned.
• The electron concerned could be bound within a molecule (albeit
loosely) or it could be relatively free within the material.
• Emission of light, (in the form of a photon) can take place either
spontaneously or it can be “stimulated” by the presence of another
photon of the right energy level.

2
Light Production:
• Spontaneous emission is really the normal case.
• When an electron is elevated to a high energy state this state is usually
unstable and the electron will spontaneously return to a more stable
state very quickly (within a few picoseconds) emitting a photon as it
does so.
• When light is emitted spontaneously its direction and phase will be
random but the wavelength will be determined by the amount of energy
that the emitting electron must give up.

3
 Light Production:
• Stimulated emission is what happens in the operation of a laser. In
some situations when an electron enters a high energy (excited) state
it is able to stay there for a relatively long time (a few microseconds)
before it changes state spontaneously.
• When an electron is in this semi-stable (metastable) high energy state it
can be “stimulated” by the presence of a photon of light to emit its
energy in the form of another photon. In this case the incident photon
must have the right energy (wavelength) within quite small limits.
• When stimulated emission takes place the emitted photon has
exactly the same wavelength, phase and direction as that of the
photon which stimulated it!
• LASER is an acronym for “Light Amplification by the Stimulated
Emission of Radiation”.
Source of energy:
✓Heat
✓Electrical Discharge
✓Electric current
✓Chemical reaction
✓Biological reactions
✓Absorption of light
4
✓Nuclear reaction
The Semiconductor Junction Diode:

• Figure 53 shows a p-n junction with an electrical potential applied

across it. When the field is applied in one direction the device
conducts electricity (called the forward direction), but when the field
is applied in the opposite direction (the reverse direction) no current
can flow.
• The quantum of energy is radiated as electromagnetic energy with
the wavelength depending on the size of the energy “gap” that the
free electron crosses when it fills the hole. This phenomenon is
called Injection Luminescence.
• If we choose the materials correctly this emits visible light and we
have built an LED.
5
 Construction and Operation of LEDs :

• The wavelength of light emitted by the LED is inversely proportional to

the bandgap energy. The higher the energy the shorter the wavelength.
The formula relating electron energy to wavelength is given below:

• This means that the materials of which the LED is made determine the
wavelength of light emitted.

6
Construction and Operation of LEDs:
• The following table shows energies and wavelengths for commonly
used materials in semiconductor LEDs and lasers:

• Some materials ranges of energies and wavelengths are given. This is

because we can mix the materials in different proportions and get
different bandgap energies.
• The level of dopant used is very important in determining the amount of
power that can be produced and also has some affect on the
wavelength. Typically, the junction area has quite a high level of doping7
Construction and Operation of LEDs:

• Every time an electron recombines with a hole one photon is emitted.

• The output power is directly proportional to the drive current
• Transitions can take place from any energy state in either band to any
state in the other band.
• Because of the range of states possible in both bands there is a range
of different energy transitions possible. This results in a range of
different wavelengths produced in this spontaneous emission.
• So, LEDs produce a range of wavelengths. Typically the range is about
80 nm or so.
• Indirect bandgap materials (silicon, germanium and most alloys of
8
aluminium) – not suitable
Indirect Bandgap Materials:
• Some of the energy possessed by an electron in either the conduction
band or the valence band takes the form of “lattice momentum”.
• Some materials (silicon for example) have different amounts of lattice
momentum in each of these bands.
• In order to emit a photon, lattice momentum must first be given up (in the
form of a phonon). This requirement for both phonon and photon
emission simultaneously means that photon emission is quite an unlikely
event.
• What tends to happen in these materials is that impurity sites in the
material offer intermediate states between the conduction and valence
bands. Electrons are able to jump between bands without radiating a
photon by transiting these intermediate states.
• Light emission from indirect bandgap materials has a very low “quantum
efficiency”.
• Indirect bandgap materials: silicon, germanium and most alloys of
aluminium.

9
 Heterojunctions (Practical LEDs):

• It's easy enough to construct a p-n junction that will emit light of the
required wavelength. What isn't easy is getting the light out of the junction
and into a fibre.
• p-n junctions are necessarily very thin, flat and need to cover a relatively
large area if they are to produce any meaningful amount of light.
• Light is spontaneously emitted in all directions and since the
semiconductor material is transparent over the band of wavelengths
produced, the light will disperse in all directions. It is very difficult to get
any meaningful amount of light into a fibre from a regular p-n junction.
• What is needed is a way of producing light in a more localised area, with
a greater intensity and with some way of confining the light produced
such that we can get it (or a lot of it), into a fibre. The heterojunction is the
answer to this problem.
10
 Heterojunctions (Practical LEDs):

• A heterojunction is a junction between two different semiconductors with

different bandgap energies.
• The difference in bandgap energies creates a one-way barrier.
• Charge carriers (electrons or holes) are attracted over the barrier from the
material of higher bandgap energy to the one of lower bandgap energy.
• The double heterojunction forms a barrier which restricts the region of
electron-hole recombination to the lower bandgap material. This region is
then called the “active” region.

11
 Heterojunctions (Practical LEDs):

• The heterojunction allows us to have a small active region where the light
is produced.
• The material in the active region usually has a higher refractive index than
that of the material surrounding it. This means that a mirror surface effect
is created at the junction which helps to confine and direct the light
emitted.
12
 Four design challenges:
• Getting the Light into a Fibre
✓ This can be accomplished in two ways: by emitting the light on
the surface (the Surface Emitting LED or SLED) or by directing
the light out the side of the device (the Edge Emitting LED or
ELED).
• Confining and Guiding the Light within the Device
✓ Within the device the light must be confined and directed to the
exit aperture so that it can be directed into the fibre.
✓ It's almost a lucky accident here that the active layer in a
heterostructure almost always has a higher refractive index than
the adjacent (higher bandgap) material. This junction forms a
mirror layer and helps to confine the light to the active layer. For
this reason the outer layers are often called “confinement
layers”.
• Getting Power to the Active Region
• Power in the form of electrons and holes must be delivered to
13
the active
 Four design challenges:
• Getting Power to the Active Region
✓ Power in the form of electrons and holes must be delivered to
the active region in sufficient quantity to produce the desired
amount of light.
✓ This is done primarily by using three different techniques:
1. Careful positioning of the electrical contacts where power is
supplied. If you provide a low resistance path through the
device to the active region then a large proportion of the
available current will follow that path. Emission will take
place in the part of the active region where current flows.
2. Using different levels of dopant in the host material.
3. Using insulating materials to confine the active region and
the current path. One method is to use layers of SiO2 (good
old sand) within the device to form the needed barriers.
However, SiO2 is not a good conductor of heat. A better
(albeit more expensive) method is to use proton
bombardment of localised areas.
14
 Four design challenges:

• Getting Rid of the Heat

✓ During operation the active region produces a considerable
amount of heat.
✓ This must be conducted away and dispersed with some form of
“heat sink”.

15
Conceptual structure of a double heterojunction:

edge emitting
LED??

• Note the convention of using the capital letters N and P to denote high
levels of dopant and lowercase n and p to indicate lower levels of
doping.
• The insulating material could be SiO2 as shown in the picture or a
proton bombarded semiconductor material (proton bombardment
renders the material insulating).
• The active region is typically only about 40 microns across
16
Two configurations of practical LEDs:

• This Surface Emitting LED (SLED) operates at 850 nm wavelength and

the edge emitter could typically operate in the 1310 nm region.
• In both types of LED a combination of insulating materials and junctions
is used to:
1. Guide the current flow to a small “active region” and
2. Guide the light produced out of the device and into an easy position
for coupling to a fibre.
17
Coupling to a Fibre:

(a) Use of a Graded Index Lens (GRIN lens) is fairly common. A GRIN lens
is very similar to just a short length of graded index fibre (albeit with a
much larger diameter). The lens collects and focuses the light onto the end
of the fibre.
(b) A Ball lens is also often used. This is bonded to the surface of the LED
with an epoxy resin that has a specific refractive index. However, the RI of
the epoxy can't match to both the RI of the fibre and the RI of the
semiconductor since the semiconductor will have an RI of around 3.5 and
the fibre of around 1.45.
18
Coupling to a Fibre:

(c) The Direct Coupling method is becoming increasingly popular. Just

mount the fibre end so that it touches the LED directly. A common way to
do this is to mount the LED inside a connector so that when a fibre is
plugged in (mounted in the other half of the connector) you get firm
mounting in good position. This has the advantage of low cost and low
complexity.
(d) Another common way is to fix a ball lens to the end of the fibre as
shown in the diagram.

19
 Characteristics of LEDs:

• Low Cost
• Low Power
• Relatively Wide Spectrum Produced
• Incoherent Light
• Digital Modulation
• Analogue Modulation

20
 LASERs:

• LASER is an acronym for “Light Amplification by the Stimulated

Emission of Radiation”.
• Ideal laser light is single-wavelength only.
• Lasers can be modulated (controlled) very precisely (the record is a
pulse length of 0.5 femto seconds).
• Lasers can produce relatively high power.
• Because laser light is produced in parallel beams, a high percentage
(50% to 80%) can be transferred into the fibre.
• Lasers have been quite expensive by comparison with LEDs.
• The wavelength that a laser produces is a characteristic of the material
used to build it and of its physical construction.
• Amplitude modulation using an analogue signal is difficult with most
lasers because laser output signal power is generally non-linear with
input signal power.

21
Principle of the LASER:

• The critical characteristic here is that when a new photon is emitted it

has identical wavelength, phase and direction characteristics as the
exciting photon.
22
Principle of the LASER:

23
 Lasing:

• We need a “population inversion” to take place. A population Inversion occurs

when there are more electrons in the higher energy state than there are in the
lower energy state. Without this condition stimulated emission (lasing) cannot
occur.
• Materials that have this “4-level” energy state system are much more efficient
at lasing than 3-level materials because population inversion is relatively easy
to attain.
• Because the lower energy state (after the radiative transition) is itself a
transient state it is easier to get a population inversion (more electrons in the
high energy state than the low energy one) than if the lower energy state was
itself the ground state.
• There are typically a lot of electrons in the ground state.
24
 Need for Population Inversion:
• The requirement for a population inversion to be present as a precondition for
stimulated emission is not at all an obvious one. Electrons in the high energy
state will undergo stimulated emission regardless of how many electrons are
in the ground state. The condition seems irrelevant.
• The problem is that an electron in the ground state will absorb photons at
exactly the wavelength at which electrons in the higher energy state will
undergo stimulated emission! You must have a greater probability of
stimulated emission than absorption for lasing to occur.
• It happens that the probability that an electron in the ground state will absorb
an incoming photon is usually different from the probability that an electron in
the excited state will undergo stimulated emission. So what you really need is
not an inversion in the numbers of electrons in each state. Rather you need
the probability that an incoming photon will encounter an excited electron and
stimulate emission to be greater than the probability that it will encounter an
electron in the ground state and be absorbed.
• So an inversion takes place when the number of electrons in the excited state
multiplied by the probability of stimulation by an incoming photon exceeds the
number of electrons in the ground state multiplied by the probability of
absorption of an incoming photon. 25
 Semiconductor Laser Diodes:

• The most common communication laser is called the “Fabry-Perot”

laser.
• The Fabry-Perot laser can be modified by placing something in the
cavity that will disperse unwanted frequencies before they reach the
lasing threshold.
• There are a number of alternatives, but a common way is to place a
diffraction grating within the cavity. When this is done, the laser can
produce a very narrow spectral linewidth (typically today .2 to .3 nm).
Lasers using this principle are called Distributed Feedback (DFB) or
Distributed Bragg Reflector (DBR) lasers.

26
Fabry-Perot Lasers: Fabry-Perot resonator

27
• On the left we have solved the equation above for a cavity 100 microns
long, a wavelength of 1500 nm and a refractive index of 3.45 (InP). We
can see that there are 7 wavelengths within 10 nm of 1500 nm where
resonance may occur.
• On the right of the figure we can see the same solution but for a cavity
200 microns long. Here there are 13 possible resonant wavelengths
• The longer the cavity (and the shorter the wavelength) the more
resonant wavelengths we can find within the vicinity of our centre
wavelength.

28
Construction of a Fabry-Perot Laser:

29