Optical Amplifiers

Continued

1
Photonic Network
By Dr. M H Zaidi
Review of Previous Lecture
Optical amplifiers are a key component in long
haul optical communication
Amplifiers are majorly of four types
Electro optic regenerators
Erbium-doped optical amplifiers (EDFA)
Raman Amplifiers
Semiconductor optical amplifiers (SOA)
Pump powers required for L-band EDFAs are
much higher than their C-band counterparts.

Photonic Network 2
By Dr. M H Zaidi
EDFA Multi Stage Designs

1st Active Stage 2nd Active Stage

Co-pumped Counter-pumped

Er3+ Er3+
Input Doped Fiber Doped Fiber Output
Signal Signal

Optical Optical
Optical
Isolator Isolator
Isolator

PUMP PUMP

Photonic Network 3
By Dr. M H Zaidi
EDFA Multi Stage Designs
1st stage gives high gain, low noise
2nd stage give high output power
Combined both, we get a high gain and high power
output signal with low noise.
Also multiple pumps give failure recovery benefits.
Optical isolator/filter can be placed in between to flatten
the gain or remove pump signal.

Photonic Network 4
By Dr. M H Zaidi
EDFA - Advantages
EDFAs have high pump power utilization (>50%)
Directly and simultaneously amplify a wide wavelength
band (> 80 nm) in the 1550 nm region, with a relatively
flat gain.
Flatness can be improved by gain-flattening optical filters
Gain in access of 50 dB
Low noise figure
Transparent to optical modulation formats
Suitable for long haul applications

Demerit
EDFAs are not small and cannot be integrated with other
semiconductor devices
Photonic Network 5
By Dr. M H Zaidi
Stimulated Raman Scattering
If two or more signals at different wavelengths
are injected into a fiber, SRS causes power to be
transferred from the lower-wavelength channels
to the higher-wavelength channels.
This coupling of energy from a lower
wavelength signal to a higher-wavelength signal
is a fundamental effect that is also the basis of
optical amplification and lasers.

Photonic Network 6
By Dr. M H Zaidi
Stimulated Raman Scattering

Fig: The effect of SRS. Power from lower-wavelength channels is transferred to the higher-
wavelength channels

Fig: SRS gain coefficient as a function of channel separation

Photonic Network 7
By Dr. M H Zaidi
Raman Amplifiers
Raman Amplification is based on the Stimulated Raman
Scattering (SRS) phenomenon.
Raman gain spectrum is fairly broad and the peak of the gain is
centered about 13 THz below the frequency of the pump signal
used.
Unlike EDFAs, we can use the Raman effect to provide gain at
any wavelength.
Raman amplification can potentially open up other bands for
WDM, such as the 1310 nm window, or the so-called S-band
lying just below 1528 nm.
Also, we can use multiple pumps at different wavelengths and
different powers simultaneously to tailor the overall Raman gain
shape.
Another major concern with Raman amplifiers is crosstalk
between the WDM signals due to Raman amplification.

Photonic Network 8
By Dr. M H Zaidi
Raman Amplifier

Fig: Distributed Raman amplifier using a backward propagating pump, shown

operating along with discrete erbium-doped fiber amplifiers

Photonic Network 9
By Dr. M H Zaidi
Raman Amplifier Vs EDFA (1/2)
Unlike EDFAs, Raman effect to provide gain at
any wavelength. An EDFA provides gain in the
C- and L-bands (1528-1605 nm).
Thus Raman amplification can potentially open
up other bands for WDM, such as the 1310 nm
window, or the so-called S-band lying just below
1528 nm.
Also, multiple pumps at different wavelengths
and different powers can be used simultaneously
to tailor the overall Raman gain shape.
Photonic Network 10
By Dr. M H Zaidi
Raman Amplifier Vs EDFA (2/2)
Raman amplification relies on simply pumping
the same silica fiber used for transmitting the
data signals, so it can be used to produce a
lumped or discrete amplifier, as well as a
distributed amplifier.
Today the most popular use of Raman amplifiers
is to complement EDFAs by providing
additional gain in a distributed manner in ultra-
long-haul systems.

Photonic Network 11
By Dr. M H Zaidi
Semiconductor Optical Amplifiers
They are not a good as EDFAs for use as amplifiers.
Used for other applications: in switches and wavelength
converter devices.
First, the populations are not those of ions in various energy
states but of carriers-electrons or holes.
Semiconductor consists of two bands of electron energy levels: a
band of low mobility levels called the valence band and a band of
high mobility levels called the conduction band.
At thermal equilibrium, there is only a very small concentration
of electrons in the conduction band of the material,

Photonic Network 12
By Dr. M H Zaidi
Semiconductor Optical Amplifiers
Population inversion condition, the electron concentration in the
conduction band is much higher.
Population inversion in an SOA is achieved by forward-biasing a
pn-junction.
Nevertheless, EDFAs are widely preferred to SOAs for several
reasons.
Main reason is that SOAs introduce severe crosstalk when they
are used in WDM systems.
Gains and output powers achievable with EDFAs are higher.
Coupling losses and the polarization-dependent losses are also
lower with EDFAs since the amplifier is also a fiber. Due to the
higher input coupling loss, SOAs have higher noise figures
relative to EDFAs.

Photonic Network 13
By Dr. M H Zaidi
Semiconductor Optical Amplifiers

Block diagram of a semiconductor optical amplifier. Amplification occurs when light

propagates through the active region. The facets are given an antireflective coating to
prevent undesirable reflections, which cause ripple in the amplifier gain.

Photonic Network 14
By Dr. M H Zaidi
Semiconductor Optical Amplifiers

The energy bands in a p-type semiconductor and the electron

concentration at
(a) thermal equilibrium and
(b) population inversion

Photonic Network 15
By Dr. M H Zaidi
Semiconductor Optical Amplifiers
Finally, the SOA requires very high-quality
antireflective coatings on its facets (reflectivity
of less than 10-4) which is not easy to achieve.
Higher values of reflectivity create ripples in the
gain spectrum and cause gain variations due to
temperature fluctuations.

Photonic Network 16
By Dr. M H Zaidi
Crosstalk in SOAs
Consider an SOA to which is input the sum of two optical
signals at different wavelengths.
Assume that both wavelength are within the bandwidth of the
SOA.
Presence of one signal will deplete the minority carrier
concentration by the stimulated emission process so that the
population inversion seen by the other signal is reduced.
Thus the other signal will not be amplified to the same extent
and, if the minority carrier concentrations are not very large,
may even be absorbed.

Photonic Network 17
By Dr. M H Zaidi
Crosstalk in SOAs
Thus, for WDM networks, the gain seen by the signal in one
channel varies with the presence or absence of signals in the
other channels.
This phenomenon is called crosstalk, and it has a detrimental
effect on the system performance.
This crosstalk phenomenon depends on the spontaneous
emission lifetime from the high-energy to the low-energy state.
The spontaneous emission lifetime in an EDFA is about 10ms.
Therefore lifetime is large enough compared to the rate of
fluctuations of power in the input signals.
Elecrtons cannot make the transition from the high-energy state
to the lower-energy state in response to these fluctuations.
Thus there is no crosstalk whatsoever in EDFAs.

Photonic Network 18
By Dr. M H Zaidi
Crosstalk in SOAs
In the case of SOAs, this lifetime is on the order of
nanoseconds.
Thus the electrons can easily respond to fluctuations
in power of signals modulated at gigabit/second rates,
resulting in a major system impairment due to
crosstalk.
Thus crosstalk is introduced only if the modulation
rates of the input signals are less than a few kilohertz,
which is not usually the case.
EDFAs are better suited for use in WDM systems
than SOAs.
Crosstalk effect is not without its uses.

Photonic Network 19
By Dr. M H Zaidi
EDFA - Applications

Photonic Network 20
By Dr. M H Zaidi
Optical Amplifiers - Applications
In line amplifier
-30-70 km
-To increase transmission link

Pre-amplifier
- Low noise
-To improve receiver sensitivity

Booster amplifier
- 17 dBm
- TV

LAN booster
amplifier

Photonic Network 21
By Dr. M H Zaidi
Front-End Amplifiers
Two Front-End amplifier types:
High-impedance Amplifier
Trans-impedance Amplifier
Thermal noise current that arises due to the random
motion of electrons and contaminates the photocurrent
is inversely proportional to the load resistance.
Thus, to minimized the thermal noise, we must make
RL large.
Thus there is a trade-off between the bandwidth of the
photodiode and its noise performance.
Thus the transimpedance front end is chosen over the
high-impedance one for most optical communication
systems.

Photonic Network 22
By Dr. M H Zaidi
Front-End Amplifiers
Amplifier
Photodiode +
RL C A
Ip -

High-impedance Front End Amplifier Circuit

RL
+
A
-
Photodiode C Amplifier
Ip

Trans-impedance Front End Amplifier Circuit

Photonic Network 23
By Dr. M H Zaidi
Front-End Amplifiers
There is another consideration in the choice of a front-
end amplifier: dynamic range. This is the difference
between the largest and smallest signal levels that the
front-end amplifier can handle.
However, dynamic range of the receivers is very
important consideration in the case of networks where
the received signal level can vary by a few orders of
magnitude, depending on the location of the source in
the network.
Transimpedance amplifier has significantly higher
dynamic range than the high-impedance one, and this is
another factor in favor of choosing the transimpedance
amplifier.

Photonic Network 24
By Dr. M H Zaidi
 Optical Photodetectors

Photonic Network 25
By Dr. M H Zaidi
Optical photodetectors (PDs)
PDs convert photons
to electrons
Two photodiode types q
N electrons
N photons
PIN
APD
For a photodiode
it is reqd that it is
sensitive at the used l
small noise
long life span
small rise-time (large BW,
small capacitance)
low temperature sensitivity
quality/price ratio

26
Photonic Network
By Dr. M H Zaidi
PhotoDetectors
Basic principle is electron hole pair generation.
Photo-detectors made up of S.C materials
Photons incident on S.C absorbed by electrons in
valence band
These electrons acquire higher energy and are excited
into the conduction band, leaving behind a hole in the
valence band.
When an external voltage is applied to the
semiconductor, these electron-hole pairs give rise to an
electrical current, termed the photocurrent.
Principle of quantum mechanics is that each electron
can absorb only one photon to transit between energy
levels.
Photonic Network 27
By Dr. M H Zaidi
Photodetection Principle
Conduction Band
Electron Energy (eV)

Electron

hV/e Eg

Valence Band

Photon Hole

The basic principle of photo-detection using a semiconductor. Incident photons are

absorbed by electrons in the valence band, creating a free or mobile electron-hole pair.
Electron-hole pair gives rise to a photocurrent when an external voltage is applied.

Photonic Network 28
By Dr. M H Zaidi
Materials commonly used to produce photodiodes:

MaterialWavelength range (nm)

Silicon 1901100

Germanium 8001700

Indium gallium arsenide 8002600

lead sulfide <1000-3500

Photonic Network 29
By Dr. M H Zaidi
PhotoDetectors
Energy of the incident photon must be at least equal to the
band gap energy in order for a photocurrent to be generated.
This gives us the following constraint on the frequency fc or
the wavelength at which a semiconductor material with
band gap Eg can be used as a photodetector
hfc = hc/ > eEg
The Largest value of for which this equation is satisfied is
called the cutoff wavelength and is denoted by cutoff
We see from the table 3.2 that the well-known
semiconductors silicon (Si) and gallium arsenide (GaAs)
cannot be used in the 1.3 and 1.55 m bands.

Photonic Network 30
By Dr. M H Zaidi
PhotoDetectors
Although germanium (Ge) can be used in both these bands, it has
some disadvantages that reduce its effectiveness for this purpose.
New compounds indium gallium arsenide phosphide (InGaAsP) are
commonly used to make photodetectors in the 1.3 and 1.55 m bands.
Silicon photodetectors are widely used in the 0.8 m bands.
Fraction of the energy of the optical signal that is absorbed and gives
rise to a photocurrent is called the efficiency of the photodetector.
For transmission at high bit rates over long distances, optical energy is
scarce, and thus it is important to design the photodetector to achieve
an efficiency as close to 1 as possible.
The power absorbed by a semiconductor slab of thickness L m can be
written as
Pabs = (1- e-L) Pin
Also,
= Pabs / Pin = 1- e-L

Photonic Network 31
By Dr. M H Zaidi
PhotoDetectors
Where Pin is the incident optical signal power, and is the
absorption coefficient of the material.
depends on the wavelength and is zero for wavelengths >
cutoff
Typical values of are on the order of 104 /cm, so to achieve
an efficiency > 0.99, a slab of thickness on the order of
10m is needed.
Area of the photodetector is usually chosen to be sufficently
large so that all the incident optical power can be captured by
it.
Photodectectors have very wide operating bandwidth since a
photodetector at some wavelength can also serve as a
photodetector at all smaller wavelengths.
Thus a photodetector designed for the 1.55m band can also
be used in the 1.3m band.
Photonic Network 32
By Dr. M H Zaidi
Band gap Energies and Cutoff wavelengths Table 3.2

Material Eg (eV) cutoff (m)

Si 1.17 1.06
Ge 0.775 1.6
GaAs 1.424 0.87
InP 1.35 0.92
In0.55Ga0.45As 0.75 1.65
In1-0.45yGa0.45yAsyP1-y 0.75-1.35 1.65-0.92

Photonic Network 33
By Dr. M H Zaidi
PhotoDetectors
Photodetectors are commonly characterized by their
responsivity R.
If a Photodetector produces an average current of Ip
amperes when the incident optical power is Pin watts,
then the responsivity is
R = IP/Pin A/W
Since an incident optical power Pin corresponds to an
incidence of Pin/hfc photons/s on the average, and a
fraction of these incident photons are absorbed and
generate an electron in the external circuit, we can write
R = e/hfc A/W.
Responsivity is commonly expressed in terms of ; thus
R = e / hc = /1.24 A/W
Photonic Network 34
By Dr. M H Zaidi
PhotoDetectors
Since can be made quite close to 1 in practice, the
responsivities achieved are on the order of 1 A/W in the
1.3m band 1.2 A/W in the 1.55m band.
Using slab of semiconductor does not give high
efficiencies.
A photodetector is called a photodiode when a reverse
bias voltage is applied to a semiconductor instead of
using homogenous slab of it.
Depletion region in a pn-junction creates a built-in
electric field.

Photonic Network 35
By Dr. M H Zaidi
PIN Photodiodes
To improve the efficiency of the photodetector, a very lightly
doped intrinsic semiconductor is introduced between the p-
type and n-type semiconductors.
Such photodiodes are called pin photodiodes, where the I in
PIN is for Intrinsic.
Width of the p-type and n-type semiconductors is small
compared to the intrinsic region so that much of the light
absorption takes place in this region.
This increases the efficiency and thus the responsivities of the
photodiode.
From table 3.2, we see that the cutoff wavelength for InP is
0.92 m, and that for In GaAs is 1.65 m.
Thus the p-type and n-type regions are transparent in the 1.3-
1.6m range.

Photonic Network 36
By Dr. M H Zaidi
PIN Photodiodes
A reverse-biased pn-junction used as a
photodetector.
(a) A pn-junction photodiode.
(b) Depletion region with no bias voltage
applied.
(c) Depletion region with
a reverse bias voltage, Va.
(d) Built-in electric field on reverse bias.

Photonic Network 37
By Dr. M H Zaidi
PIN Photodiodes
A pin photodiode based on a heterostructure.
The p-type and n-type regions are made of InP,
which is transparent in the 1.3 and 1.55 m
wavelength bands.
The intrinsic region is made of InGaAs, which
strongly absorbs in both these bands.

Photonic Network 38
By Dr. M H Zaidi
PIN Photodiodes
PIN diode is a variation on standard pn-diode
An intrinsic (pure) layer of semiconductor is fabricated between the p and
n-types
Depletion layer widens
Internal electric field is maintained over a wider layer
Because very few electrons and holes are in this region
Its resistivity is low
Only a small reverse bias is needed to increase the depletion region
Stretches almost entire way between the terminals

Very fast response times

A few nanoseconds or less

Photonic Network 39
By Dr. M H Zaidi
PinFET
A field-effect transistor (FET) has very high
input impendence and for this reason is often
used as the amplifier in the front end.
A pin photodiode and an FET are often
integrated on the same semiconductor substrate,
and the combined device is called pinFET.

Photonic Network 40
By Dr. M H Zaidi
Concluding Remarks
In multi-staged EDFA, 1st stage gives high gain, low
noise whereas 2nd stage give high output power.
The demerit of EDFAs is that they are not small and
cannot be integrated with other devices.
Raman Amplification is based on the Stimulated Raman
Scattering (SRS) phenomenon.
Population inversion in an SOA is achieved by forward-
biasing a pn-junction.
The biggest drawback of SOAs is that they introduce
severe crosstalk when they are used in WDM systems.
Basic principle of photo-detectors is electron hole pair
generation.
Photonic Network 41
By Dr. M H Zaidi