Unit 5 Full Notes
Unit 5 Full Notes
• Optoelectronic integrated circuit (OEIC) involve the integration of electronic and optical components
• The monolithic integration of electronic and optical devices on the same chip will give the following
advantages.
I. High speed
II. High sensitivity
III. Compactness
IV. Reliability
V. Low cost
Lay out of optical fiber link
Hybrid Integration
• In this method, discrete devices on separate functional blocks or chips are connected using electronic
(leads) or optical (fiber) interconnects.
• Example : the integration of a junction laser with its driver circuit consisting of a bipolar transistor to
from a transmitter
• Advantage: of hybrid integration is the possibility of using high performance discrete devices as
components.
• Disadvantage are
Lack of compactness and enhanced parasitic effect in term of inter connects, bonding and lead wires.
• Parasitic are considerably reduced in ‘flip-clip’ bonding in which two chips containing component
devises and circuits are interconnected by indium bumps.
Monolithic Integration
• In this all active and passive components are fabricated on the same chip. All parts are made with
same materials and processing steps, the hetero structure and processing steps of the different
components of a OEIC can be different. This makes the realization of high performance monolithic
OEIC a real challenge.
• Advantages are size reduction, reduction of parasitic and the consequent achievements of higher-
circuit speed and bandwidth.
• Monolithic integration can be achieved by two ways
1. Vertical configuration
2. Horizontal configuration
I. Planar compatible
II. Planar regrown
1. Vertical configuration:
• In this scheme, electronic and optical device structure are epitaxial grown sequentially with
an isolation layer in between
• Disadvantage of this scheme is lake of planarity.
2. Horizontal configuration:
I. Planar compatible:
Electronic and optical devices are made from the same hetero structure
• This technique provides a large freedom in the choice of device hetero structures; the regrown
interface can have a large density of traps and other electrically active defects that can affect the
performance of the regrown device.
• OEIC is used in local area networks where the shorter distance terminals are connected.
• It is possible to use Ga-As based technology without incurring too much of losses in the optical fibers.
• Such systems are digital systems, the bit error rate (BER) must be very low ie. < 10-15) even in nosiy
environment and they should have high reliability and redundancy in critical paths.
Radar Applications
• Opto electronic integration is also important for radar applications.
• The microwave or millimetre wave phased array radar system is shown in Figure 5.7
• In principle, a phase shifted and modulated (high frequency) optical signal is injection locked to a free
running microwave or millimetre wave oscillator which forms one element of phased-array radar.
• Beam steering in such radars is achieved by incorporating a progressive phase shift ΔΦ between
successive elements
• The phase shifters using at microwave frequencies is bulkier and cumber some (weighty) so we have
to replace this by using optical phase shifters.
• An optical signal is modulated at the operation frequency by internal modulation of the laser or by an
external modulator (quantum well electro absorption modulator).
• The modulated and phase-shifted signal is coupled by injection locking to a micro wave oscillator,
which forms a single element of the phased array.
• Each element of the phased array consists of a source, a modulator, a phase shifter, a waveguide and
an oscillator
• These devices can be combined by hybrid integration, but for the sake of compactness
and ruggedness.
Image Reconstruction
Image reconstruction is the core component in DOT that involves both forward and inverse problems. The
photon diffusion/transport model establishes the mathematical relationship between the imaging parameters
(ie, optical properties) and the observable/computable photon density, hence providing a tractable basis for
image reconstruction. The forward solution to the photon diffusion/transport model and measured photon
density along the boundary are minimized through an inverse strategy that allows the reconstruction of
imaging parameters. The model is a partial differential or integral equation, which requires numerical methods
to solve, and the inverse strategies need regularization techniques to combat the inherent ill-posedness
involved in DOT
SMART PIXELARRAYS
• Smart pixel arrays (SPAs) hold great promise as an enabling technology for board to-board
interconnections in digital systems.
• For taking advantages of the high space-bandwidth product of optics, optically interconnected two
dimensional arrays of smart pixels have emerged as an attractive interconnection platform.
• Spa as an emerging optoelectronic technology is crucial to meet the demanding and flat-panel display
applications.
• The combination of photo detector and preamplifier are considered as the front end of the photo
receiver, the remaining of the circuit performs equalization, pulse shaping and gain control functions.
• To improve the gain of the signal, multiple stages of amplification are included in most of the photo
receiver circuits.
• The overall performance of the circuit is mostly decided by the front end
•
Photo detector:
Features of photo detectors are
(i) High quantum efficiency
(i) Low capacitance
(iii) Small response time and
(iv) Low dark current.
Pre-amplifier:
• Low noise with high output gain are very important for preamplifier circuit.
• the input capacitance, parasitic resistances and leakage currents of the amplifier should be minimized:
• Field Effect Transistors (FETs) and Bipolar transistors have been used in the preamplifier circuit.
Integration of photo detector and pre-amplifier:
• The integration for the photo detector and Low-noise pre-amplifier circuit is also very important in
determining the overall performance of the photo receivers.
• Three types of integration are commonly used.
(i) The Low-input impedance design
(ii) The high-input impedance design and
(iii) The transimpedance design
• High-input impedance design and transimpedance designs are mostly used.
• The high-input impedance design requires the additional equalization circuit to extend the
value of the bandwidth due to the large RC time constant.
• The transimpedance design is most popular because no equalization is required. AFET can
also be used as an active feedback element, hence increasing the impedance value to as high
as 80KΩ. This significantly reduces the overall thermal noise.
4K a Tθm 4K Tθ
+ ( 2πCr ) f0 If B2 + B m ( 2πCr ) ID3 B3
2 2
gm gm
1444424444 3 1444 424444 3
IV V
4K a Tθm 2
n
IA
+ ( 2πCr ) r r B2
gm
144444 r =1 τr 3
244444
VI
Where
Iph → photo current
Ig → Leakage current in the gate
ID → Dark current in the photodiode
F0 → Noise corner frequency.
IP1.IP2,IP3→ personick integrals
IF→1/f noise integral
IT→ trap integral
Θm→ materials related to parameter
gm → FET trans conductance
τr→ trap emission time constant
Ar, → constant dependent on Trap density and transistor parameters
• The first term is the signal value, associated with the photo current IPH.
• The second term is the 'thermal noise' due to load resistance in the high and low input impedance
designs or feedback resistance in the transimpedance design.
• The Third term represents 'short noise, due to leakage current in the gate, land dark
/ current in the photo diode, ID
• . The fourth term is the '1/f noise' in the FET device
• The fifth term represents the 'noise' associated with the channel conductance of the FET
• The sixth term is due to noise resulting from traps in the channel or buffer regions.
• To minimize the noise, low Ig, ID, low total input capacitance and high trans-conductance are
needed.
• At higher frequencies and bit rates the noise is dominated by the noise associated with the
channel conductance. It is proportional to the cube of the bandwidth of the circuit and square of
the total capacitance, CT, given by
CT = Cgs + CJ + CP
Where
CGS→ grate source capacitance in the FET
CJ,- PIN diode capacitance
CP, → parasitic capacitance.
1 − r q
Where
hv
→ photon energy
q
Where
m→ modulation index
S m 2 I0
=
N 2 i 2N
Where
• The SNR equation can be used to evaluate the noise performance of practical photo receiver circuits.
• Transit time of the generated carriers in the diode and RC time constant of the circuit are the most important
parameters to determine the bandwidth of the receiver.
• The frequency response of the PIN photodiode J(w), limited by transit time and the electrical frequency
response H(w), of the overall circuit includes the diode capacitance and resistances.
• The frequency response for the complete receiver can be expressed as
J0(w) = J(w)H(w)
• In many designs, the noise is the main concern because equalization can be used in later stages of the
photoreceiver, thereby increasing the effective bandwidth of the photo-receiver.
• The calculated and measured electrical frequency response of a PIN-MODFET front end photo receiver
circuit is shown in figure 5.20.
Figure 5.19 Equivalent circuit model of a 10W input impedance PIN-MODFET front end photoreceiver
Figure 5.20 Calculated and measured frequency (GHz) response of the circuit
6) Draw the diagram of PIN-HBT integrated front end photo receiver and explain its operation?
• The other possible combinations of photodetectors and amplifiers can be formed as a monolithically
integrated front end photoreceiver circuit based on the compatibility of material used and desired device and
circuit characteristics.
• To form the integrated controller modulator, we use Heterojunction bipolar transistor (HBT) can be integrated
with PIN diode.
• Reasons for using a PIN-HBT combinations are
(i) Conveniently used to realize a front-end photo receiver ie PIN photodiode serves as front-end photo
detector and HBT is the preamplifier.
(ii) Stems from noise and sensitivity considerations.
• Comparison of calculated sensitivity of PIN-FET and PIN-HBT photo receivers are shown in figure 5.21
Figure 5.21
• Sensitivity of HBT based photo receiver is better than a FET-based photo receiver at high frequencies.
• The integration involved a single step epitaxy of the HBT, from which the PIN modulator was selectively
defined by processing.
• The collector region of the HBT also serves as the i-region of the diode. The two devices were then
monolithically integrated with the addition of the required passive elements.
Equivalent circuit:
• Inductive peaking is the technique to enhance photo receiver response at high frequencies which means an
inductor is placed in series after the photodiode at the input of the amplifier.
• The resistors and inductors are usually formed with evaporated single or multi-layered metals in monolithic
circuits.
• Typical values of peaking inductors range from 2-5nH.
• Measured frequency response of monolithic PIN-HBT photo receiver is shown in figure 5.24.
Figure 5.24
• A -3dB modulation bandwidth of 19.5GHz is recorded which makes the circuit good for 24 Gbits/s
bitrates in a digital application.
7)Explain the of eye pattern of a photoreceiver?
• Eye diagram measurements are useful for estimating the noise performance and sensitivity of the photo
receiver.
• To characterize the system pulse performance, digital optical communications systems need optical time
domain measurements.
• Such measurements include both single and multi-valued waveforms.
• Multi-valued waveform measurements, are commonly called 'eye diagrams' which includes pulse parameter,
mask, extinction ratio and jitter.
• Pulse parameter measurements include rise time, fall time, overshoot, undershoot, preshoot, pulsewidth, duty
cycle, period and setting time.
• Eye diagrams are formed by overlaying multiple single-valued pseudo-random binary sequence (PRBS)
waveforms on the display of the oscilloscope or eye diagram analyzer.
• Generally, more open the eye is, lower the likelihood that the system may mistake a “1”bit for a "0" bit, or
vice versa.
• The eye diagram opening width, the time between the Zero to one/one to Zero crossings shows the time
interval of sampled signals without error due to inter symbol interference.
• The rise and fall times, generally measured from 10% to 90% point and the slope of the rise and fall times are
important for estimating the system's sensitivity to sample timing.
• Masks are used to standardize the minimum quality for eye opening.
Block diagram:
• The optical transmitter sends a pseudo-random sequence of binary digits to the receiver through optical link.
• The received signal, and a clock waveform generated by the receiver from the received signal are fed to the
sampling oscilloscope.
• Two parameters ie 'eye closure' and 'jitter' are defined by the eye diagrams given by
V
Percentage eye closure = 1 − 1 × 100
V2
Δt
and percentage Jitter = × 100
tc
• The central open region of the eye diagram is a measure of the BER of the circuit. It gives a probability of the
circuit in making correct decisions regarding incoming zeros and ones.
• Another important characteristics of photo receivers is ‘Dynamic range’. It is defined as the ability of the
circuit to respond to a wide range of optical input powers and this response must be linear.
• A typical dynamic range shows a linear relation between input optical power and measured electrical
output. The dynamic range in this figure is 25dB.
OEIC Transmitters
• A light source such as a high power LED or a laser are included in the transmitter.
• Integration of laser with the associated electronics such as driver circuits is more complicated than the
fabrication of a photo receiver.
• Because the laser has following disadvantages.
1. Laser structure is nearly 4 um high, which makes the processing steps for integration with an
incompatible hetero structure for the electronic device very difficult.
2. The optical cavity in an edge emitting laser needs to be defined by two end mirrors.
3. Electrical and optical confinement needs to be achieved in the lateral dimension.
4. The operation of the laser necessitates efficient heat sinking of the whole chip.
• Inspite of these disadvantages it is worthwhile to a integrate the component devices to ensure a higher
modulation bandwidth.
Figure 5.27 Cross sectional diagram of a OEIC transmitter using the hetero structure for the laser and
driver transistor
• For optimization of the performance of both devices, separate optimized structures are more desirable.
• Early transmitter circuits consisted of a single edge-emitting laser, whose facet was created by
clearing, integrated with a single transistor.
• Example GaAs-based Laser-MESFET OEIC
• Fabrication process of GaAs based Laser-MESFET OEIC with 3 MBE steps.
Figure 5.28
• Recently, laser facets are formed ‘on-chip’ by micro leaving or by using dry-etching techniques such
as ion beam etching.
Figure 5.29 Integrated transmitter circuit with monitoring photo diode formed b 2-step MBE
• Figure 5.29 shows the cross section of a simple transmitter circuit fabricated by using ion beam
etching technique and micro leaving of laser facets. The laser structure was grown in the first epitaxial
step.
• The circuit consists of a photodiode for monitoring laser output power. The laser is a SQW GRIN-
SCH device.
Equivalent Circuit
• Figure 5.30 shows the equivalent circuit for single channel laser – MESFET transmitter. It includes
three FETs and a 50Ω input load resistance.
• The identical FETs labeled T1 and T2 form a differential amplifier or current source that provides the
advantages of common mode rejection and noise reduction.
• Inputs are applied at VG1 and VG2 are small signal modulated signals. FET (T3) acts as a constant
current source and provides the drive current.
• The current source IB is a DC source to bias the laser at threshold.
• The photodiode PD acts as a monitoring device. If the transconductance of two FETS is equal to ‘gm’
and the slope in the lasing portion of the light-current characteristics is 'S', the output power of laser,
Pout is expressed as
gms
Pout = ( VG1 − VG 2 )
2
• The modulation bandwidth limitation of a laser is set by the relaxation oscillation frequency, fr.
• The modulation bandwidth can be increased by a large photon density in the cavity and a short photon
lifetime.
• Large photon density is achieved by driving the laser well above threshold which can become a
critical issue due to the problem of heat dissipation.
• The power dissipation in a laser can be expressed as
Pd = Pin − Pout
hv hv
Pd = I B + I B R S − ( I B − I th ) ηp
q q
Where
ηp → power efficiency
• Power dissipation should be kept small, even for a large bias current, if the series resistance and
threshold current are small and ηp is large.
• The intrinsic model of a laser diode can be represented by an RLC circuit that has a resonance
1
frequency f =
2π LC
• In practice, the combined impedance of this RLC circuit is smaller than parasitic circuit elements.
• The most important parasitic elements are the series resistance of the diode RS, the bond wire
induction Lb, and parasitic capacitance Cp between bonding pad and ground plane.
• The equivalent circuit of a laser monolithically integrated with a FET is shown in figure 5.31.
Figure 5.31 Equivalent circuit of a laser monolithically integrated with FET
9) With necessary diagram , Explain the concept of optical control in microwave oscillator ?
• Optoelectronic phased army antenna is an example of the application of optical control of high speed
electronic circuits.
• Due to wide B. W of the optical control signal and their inherent isolation from radio frequency
signes, the use of optical signals to control high speed electronic circuit is advantageous.
• Optical signals can be routed through light weight fibers or monolithically integrated optical
waveguides without affecting signals transmitted on microwave guides.
• Microwave-optical links also allow remote control antennas, which allows moving personnel and
expensive control equipment to be located at the command center rather than at antenna.
• It will be useful in cellular telephone transmission, cable television where a transmitting antenna is
located in remote locations.
• Similarly fiber optic links have been used in microwave measurements and instrumentation where
optical fiber over come the limitations on phase stability imposed by coaxial cables.
• To realize the optical control of microwave oscillators three basic forms are used.
Optical switching:
• With optical switching, the intensity of the input light controls the output power of the oscillator.
• Applied in a non-linear fashion for turn ‘on’ and ‘off’ the controlled oscillator.
• The characteristic of optical switching is shown in figure 5.3:2
Optical tuning
• With this method, the intensity of the input light controls the output frequency of the oscillator.
• The characteristics is shown in figure 5.33.
Figure 5.33
• Optical Injection locking refers to the use of a high – frequency modulated optical control signal to fix
the frequency of free running oscillator.
• When in locked condition, the oscillator will oscillate at the same frequency as the injection locking
signal. There will be a phase offset ‘φ’ between the oscillator’s output signal and the injected signal.
The phase offset varies by ±90° overs the injection locking bandwidth Wmax.
• The injection locking bandwidth and phase error are given by
w0 Pinj
Bandwidth Wmax =
2Q Po
w − w0
and Phase Error φ = sin −1 lock
w max
Where
w 0 → free running oscillation frequency
• ‘Q’ provides to a measure for loss in the resonant circuit and is defined as
• The characteristics is shown in Figure 5.34
Figure 5.34
• Tuning and injection locking characteristics of an GaAs HBT oscillator circuit is shown below.
(i)waveguide coupler
(ii)waveguide interferometer
• Guided wave components are required for routing optical signals on a chip and also for the functions
of directional coupling, filtering and modulation.
Couplers
• The guided wave integrated optical circuit element is called directional coupler.
• Directional coupler is used for transferring optical energy from one wave guide to another (or) from
one region of an OEIC chip to another.
• It consists of two parallel waveguides between which the transfer of optical energy occurs due to the
overlapping of waveguide modes.
• This energy exchange requires that the light propagating in both guides have same velocity and
propagation vector.
• If these parameters in the two channels are exactly same, then the power propagating in the two guides
is given by
P1 ( Z ) = cos 2 ( KZ ) e − γz
P2 ( Z ) = sin 2 ( KZ ) e− γz
Where
Z → direction of propagation
K → coupling constant
Coupling constant (K) is given by
2β2y be − bd
K=
βz w ( β2γ + β2z )
Where
b → extinction coefficient
d → separation between the guides
w → width of each guide
βγ & βz → mode propagation constants in the transverse and propagation directions respectively.
• 'Coupling length (lc)' of a directional coupler is defined as the length at which the total transfer of
power takes place is given by
1π
Ic = m + , m = 0,1, 2...
2k
• But in real couplers, the two guides may not be same and hence the propagation constants may differ
by a small amount ∆β z
∆βz
2
K 2r = K 2 +
2
Dual-Channel Coupler
Figure 5.41 GaAs Dual channel single mode ridge waveguide coupler
• Curved sections of guides are important elements of integrated optics.
• Sharp bends constitute a large radiation loss.
• For the bends to loss-free, the radius of curvature at the bend must be larger than ‘λ’, the wavelength
of light propagating in guide.
Branching Networks
• ‘Branching networks’ divide optical power among two or more outputs or combine power from two
or more inputs.
• The parameters of branching networks are
(i) ‘Power division ratio’ - reflects the loss of input power at the output ports.
(ii) ‘Excess or insertion loss’ – reflects the loss of input power due to material losses
• Commonly used Branching Network is the waveguide 'Y' structure. It can be used as a symmetric
power divider or combiner shown in Figure 5.42.
• The modes at various points of propagation are also illustrated. At point ‘I1’, the two arms are
uncoupled and behave as independent single mode guides, supporting the lower order mode. Near the
taper, the structure transitions from a single wave guide of width 2W (at I2) to a waveguide of width
W.
• At ‘I2’, both the symmetric and antisymmetric modes are supported. As these propagate towards point
I3, the antisymmetric mode is cut off and its energy is radiated into the substrate.
• At ‘I3’ single mode characteristics are again restored. For equal input intesities in the two arms, the Y
combiner essentially behaves like 3-dB directional couplers.
Figure 5.43 (a) Power splitters based on directional coupler (b) Power splitter based on Y structure
• Active guided wave components can be integrated in OEICs with active optoelectronic devices.
1. Laser
2. Electro optic modulator
1. Modulators
2. Interferometers and
3. Filters
Mach-Zehnder Interferometer
• A simple guided wave modulation/switching device based on the electro-optic effect is the mach-
zehnder interferometer
Figure 5.43 (a) Power splitters based on directional coupler (b) Power splitter based on Y structure
• Figure 5.44 shows the schematic diagram of guided wave mach-zehnder interferometer with input and
output 3-dB couplers.
• The incoming optical beam is split equally between the two branches of the input coupler.
nλ 02 q 2
∆n r =
8λ 2 ε 0 n g mC 2
Where
Disordered MQW:
Low-Loss waveguides:
• If the guides are made of high-quality, defect free epitaxial layers, then the major sources of loss are
surface scattering and absorption.
• Therefore, etching and formation technique become critical in the fabrication of low-loss waveguides.
• Figure of merit is its loss coefficient (γ), which determines the insertion loss of a waveguide, is mainly
determine by free-carrier absorption.
• Therefore, material quality and processing become important in determining loss coefficient (γ).
• Transmission of optical power in the guide is given by
P (z) = P (0) eγz
from which the guide loss is given by
L = 4.3γ (dB/cm)
• The range of Low-loss waveguides is from ‘0.1 to 1.0 dB/ Cm’.