Semiconductor Optoelectronics

Prof. M. R. Shenoy
Department of Physics
Indian Institute of Technology, Delhi

Lecture - 1
Context and Scope of the Course

(Refer Slide Time: 00:44)

Welcome to this course on semiconductor optoelectronics. What does the subject matter of
this course comprise of? Let us start with the title, semiconductor optoelectronics.
Semiconductor: what is a semiconductor? Optoelectronics: what is optoelectronics? Then we
need to understand the subject matter of this course on "semiconductor optoelectronics".

So, what is a semiconductor? As the name indicates, it’s a material which has conductivity
between those of bad conductors/insulators and good conductors/metals, to speak of in a very
elementary fashion. What kind of numbers are we talking about? Conductivity is inverse of
resistivity. Bad conductors or insulators typically have conductivity in the range 10 -12 to 10-18;
the units are Siemens per centimeter. Unit of resistivity is ohm-cm, so ohm inverse is
Siemens. For good conductors, it is typically, approximately 10 4 to 107 Siemens per
centimeter.

Semiconductors have conductivity approximately in the range 10-6 to 103 Siemens per
centimeter. But, just if a material has conductivity between those of good conductors and bad
conductors, does this qualify it to be a semiconductor? No, then it would not have had such
important applications as we have today. The primary applications of semiconductors have
come out because of the following:

(Refer Slide Time: 05:15)

that the conductivity of semiconductors varied/changed by orders of magnitude, by

 Doping the semiconductor material: We will see later that the conductivity
depends on the doping concentration. Semiconductors are normally doped: p
doped and n doped. Doping the semiconductor can change conductivity by
orders of magnitude.
 Changing the temperature of the material: Changing the temperature can
change the conductivity of the semiconductor.
 Illumination: Conductivity of semiconductor can be changed by illumination
with light of appropriate wavelength.

This is the primary reason why semiconductors have become very important, that their
conductivity can be changed by orders of magnitude. So the elementary definition of a
semiconductor is: materials which have conductivity between those of good conductors and
bad conductors, and whose conductivity can be changed by orders of magnitude, by doping,
or by changing the temperature, or by illuminating the material with light of appropriate
wavelength.
We come to the second part: Optoelectronics. What is optoelectronics? Optoelectronics
basically deals with the science and technology of electronic devices which involve
interaction with light. For example, an LED, a light emitting diode, is basically a p-n diode; a
p-n diode is an electronic device, but it involves emission. When you pass a current through a
forward biased LED, it emits light, and exactly like that, when light is incident on a photo
detector, it generates current; it is an electronic device, but involves interaction—interaction
here primarily refers to emission and/or absorption. So, semiconductor optoelectronics deals
phenomena of semiconductor electronic devices, which involve interaction with light or
emission or absorption of light.

(Refer Slide Time: 09:36)

I want to spend few more minutes on this terminology (semiconductor optoelectronics),

because there are similar terms or synonyms, like optoelectronics, optical electronics, electro-
optics, photonics, quantum electronics, and so on—a number of similar looking terms. It is
very difficult to give a precise definition of a precise domain. However, as we discussed
above, optoelectronics deals with electronic devices which involve interaction with light. In
electro-optics we deal with optical systems optical systems or devices, where the propagation
properties or characteristics of the device can be altered by application of an external electric
field.

Let me give an example of electro-optic modulator or EO modulator. Usually, these are

optical devices, maybe a crystal, through which a beam of light propagates. If you apply an
electric field to this crystal, it is possible to modify the output properties of light. For
example, the polarization can be modulated—suppose it’s an E field which leads to variation
in the refractive index of the medium; that can result in variation of the phase of the light. So
this is the basic principle of phase modulator. So, electro-optic modulator is a device which
primarily deals with optical devices and optical systems, where the application of an external
electric field modifies its properties.

Similar definition holds good for terms such as acousto-optics and magneto-optics. These
effects were discovered in the 19th century. Acousto-optics refers to optical systems where the
characteristics or propagation properties of light are altered by application of an acoustic
field. If you replace "electric field" in electro-optics by an "acoustic field", we deal with
acousto-optics. What is the acoustic field? For example, if you apply an RF to a crystal, with
a transducer and an absorber, there forms standing waves in the medium. They are
mechanical waves which can propagate through this medium, which form a standing wave,
and due to this there are compressions and rarefactions in the medium, which act like a phase
grating. An incident optical beam here can get diffracted from its original path into particular
direction/angle, satisfying certain phase matching conditions. It is called Bragg diffraction.
There are two important effects, which are Bragg diffraction and Raman-Nath diffraction. In
Raman-Nath diffraction, we get multiple diffraction orders, whereas in Bragg diffraction, if
the phase matching condition is satisfied, then the entire energy can get transferred to a
particular diffraction order. These properties/effects are discussed in detail in acousto-optics.

Similarly, if you replace the "electric field" by "magnetic field" in the definition of electro-
optic effect, then what you have is magneto-optics. Magneto-optics deals with optical
systems where the propagation properties are modified by an applied magnetic field. A very
important effect is the Faraday Effect, where you have a linearly polarized beam travelling
through an optical medium. If you apply an axial magnetic field H, then the polarization state
gets rotated to an angle ϴ (theta), and this angle ϴ is proportional to the strength of the
magnetic field H. There are several applications of both Faraday effect and Bragg diffraction.
Acousto-optics and magneto-optics and electro-optics, all these three were known well in the
19th century.
(Refer Slide Time: 18:08)

We come to some of the other terms, what is Optical Electronics? Before that, let’s see what
is Quantum Electronics? Quantum electronics is the branch of science that primarily deals
with the study of devices and phenomena which involve interaction of radiation with matter.
The most important quantum electronic device is a laser. Non-linear optics is also a part of
quantum electronics. Now, what is Optical Electronics? In the 19th century and in the early
20th, century we had electro-optics, acousto-optics, magneto-optics, optics itself, polarization,
quantum electronics, which included lasers and non-linear optics. The subject matter which
comprises the study of all these together is optical electronics. In the 1970s, there was a
famous book by Professor Amnon Yariv: Optical Electronics. The first edition in 1970 was a
widely used text book on optical electronics, which describes all these effects, and other
propagation properties of light through a medium. In the 1980s, developments in lightwave
technology rapidly progressed due to the advent of semiconductor lasers and optical fiber
communication. So, in the 1970s with the development of semiconductor lasers,
photodetectors, optical fiber communication evolved. In both 70s and 80s, optical
communication, optical fiber communication, fiber optics, optical signal processing rapidly
developed. Rapid developments in the lightwave technology led to an entire branch
analogous to electronics, which in the 1990s was called Photonics.

So, now Photonics encompasses the earlier "optical electronics", and all the modern
communication technologies which are included in lightwave technology. One of the first
books with the title "photonics" was Fundamentals of Photonics by Saleh and Teich (1991),
and it is widely used as a text book. So, this much about the various terminologies that are
prevalent. Today photonics is a complete branch of science and technology which involves
interaction of photons and electrons, or interaction of photons and matter. Let me go a little
bit further to present the context, scope, and contents of this course.

(Refer Slide Time: 24:57)

So, let us address this question, why study semiconductor optoelectronics? This course is
generally a core course in an undergraduate or masters program on "optoelectronics and
optical communication". Therefore, let me take an example in the context of optical
communication. We take a simple optical communication link, where we have a transmitter,
and link, and a receiver. If it is optical fiber communication, and then this link is an optical
fiber. It could be a two way link, or optical fiber. Optical fiber communication is used for data
as well as voice, including all video data and digital data.

If we look at voice, i.e., the normal voice communication or telephonic conversation, then
voice is first digitized, because all communications today are digital communication. So, in
digital signal communication, voice signal is first digitized. For example, you have a NRZ
pattern, there are digital 'ones' and digital 'zeros', this is one one zero one one ( 1 1 0 1 1),
and so on; a digital sequence, the voice has been digitized, and here is the digital signal. T -
the transmitter comprises of all the electronic processing part, which converts the voice signal
into a digital signal, and the last component here is a light emitting diode, or a laser diode; the
bias current i to the diode is modulated by the same digital signal, 1 1 0 1 1 0. So, the bias
current through the LD, which is a semiconductor laser diode, is modulated, and therefore the
laser diode output yields pulses corresponding to digital ones, and no pulse corresponding to
the digital zeros. This is light pulses now, in accordance with a varying current signal. This is
a digital 1, digital 0 1 1, and 0. The light output is then coupled to the optical fiber and
transmitted towards the receiver. At the receiver end, the optical signal is incident on a photo
detector, which generates a reverse current Ip here, which again corresponds to the digital
signal that we have transmitted. Thus the photo current generated due to the incident optical
pulses, are of the same pattern as that of the original modulating current signal that was
pulsing the laser diode.

Of course, pulse shaping and other things have to be processed, which is processed in the
receiver; the receiver contains all the electronic devices and the signal processing devices, but
the first component here is the photo detector, and the last component of the transmitter is the
laser diode. So, in the context of a simple optical communication system, we need a
transmitter, which is a laser diode, and a photo detector, which is the receiver, and these are
semiconductor optoelectronic devices. It is a simple example, actually on the link there are
other optical amplifiers, which will also involve photo diodes and laser diodes, but this is
communication per se.

(Refer Slide Time: 31:20)

But optical communication is not the only application, because we know that semiconductor
lasers, laser diodes, photo detectors, and LEDs are widely used in consumer electronic and
consumer optoelectronic products, such as laser printers, CD drives, barcode scanners, and
we have a number of other applications: laser pointers, DVDs, CCD cameras, LED lightings
etc; you need optoelectronic components everywhere. A large consumer electronic market
exists where optoelectronic components are widely used, optical communication being just
one. Historically, one of the earliest optoelectronic component is the opto-coupler.

(Refer Slide Time: 32:54)

Let me discuss the opto-coupler; this was developed in the early 1970s. Opto-coupler
basically comprised of LEDs and photo detectors. For example, applications wherein two or
more electronic circuit boards had to be connected, but had to be completely electrically
isolated. It is an electronic circuit board, say a PCB (printed circuit board) and there is
another electronic circuit board, which requires complete electrical isolation—there are
certain processing circuits which have ground as the common terminal, whereas some other
circuits have the positive or negative supply line as the common terminal; in such cases, you
need complete electrical isolation between the two circuit boards.

And one of the component chosen was the opto-coupler; it was very easy. It was based on
LED (light emitting diode) and photo detector. The LED would give out light, depending on
the current i through the LED, and the photo detector would generate the same current i or a
current proportional to i; it would generate photo current, which would then go to the next
circuit board. So this was one of the earliest optoelectronic components: the opto-couplers,
widely used in electronic circuits. Today, a similar thing, on a much more different scale is
used for high speed, high density optical interconnects. It is for this reason I got off
mentioning this earliest simple device, the opto-coupler; that there is a renewed interest in
this kind of coupling between two circuit boards, which handle very large speeds with high
density optical interconnects.

(Refer Slide Time: 36:28)

You have a MOS device here, you have C-MOS device here, a C-MOS circuit board, and
there is another C-MOS circuit board: very high speed electronics, gigabit speed. You need
communication between them; one of way is of course, you could have a large number of
electrical wires, but the electrical wires will have their own losses and inductance; at very
high speeds and very high data rates, these losses become prohibitive. One of the alternatives
which is being developed is high-speed high-density optical interconnects, where you have an
optoelectronic chip. The electronic circuit board is connected to an optoelectronic (OE) chip
here, and there is another OE chip here on the other side. What is this OE chip? OE chip is a
chip which contains optoelectronic components, such as laser diodes and photo detectors, and
you can have high-speed communication between these two OE chips in the optical domain,
so this is optical communication.

It could be free space optical communication, or it could be this OE chip is connected to the
other OE chip through optical fibers. Such chips, such high density optical interconnects, are
being developed. The advantage is that at very high speeds, there is no deterioration of the
data if you handle it in the optical domain. A higher version of this includes chip to chip
communication: the OE chips talk to each other directly. The sources here are monolithic
VCSELS, and the detectors are photo detectors. VCSEL is short for Vertical Cavity Surface
Emitting Lasers. You can draw an array of VCSELS on the chip, which emit vertically
upwards, or emit vertically, and you can have an array of photo detectors. With both the array
of VCSELS and photo detectors, one can have free space chip to chip communication at very
high data rates, i.e., tens of gigabits, and several of them, is what is being developed now.
Therefore, it’s a highly improved version of the original opto-coupler.

(Refer Slide Time: 40:39)

Let me now come to the context and scope of this course. It is a first course; by first course I
mean that most of the times we start with fundamentals and elementary concepts and build on
it. So, it is a first course, which would require some background of optics, EM theory,
electronics or semiconductors, little bit of quantum mechanics, or simply a first course for
under graduates who have done first year undergraduate level of optics, EM theory,
semiconductors, and quantum mechanics. It is a first course, which could be for senior
undergrads or for first semester masters. That is the level of course, and the students who take
this course are from various backgrounds: they are from electrical engineering background,
optics, physics background, applied physics background, and also some people from
mechanical engineering background. So we really do not require any specific prerequisite
other than some basics concepts in the area of optics, electromagnetic theory, and quantum
mechanics at the first year level.
(Refer Slide Time: 43:07)

So, let me give you the contents of this course. The course will have four parts. In the first
part, we review essential semiconductor physics for optoelectronics. We start with energy
bands in solids, the E-k diagram, the band structure, the density of states; the density of states
in quantum well structures, a little bit of quantum mechanics would be required here, but we
will discuss from the basics. Then we will go over to occupation probability of carriers, and
carrier concentration in semiconductors, the carrier concentration and its dependence on the
Fermi level; the quasi Fermi levels, which are very important in optoelectronic devices.
Semiconductor materials: we briefly discuss essential semiconductor materials for
optoelectronics. We will not be able to go into details of all semiconductors, but those which
are widely used in optoelectronics; just to name a few, silicon, germanium, gallium arsenide,
indium phosphide, and so on.

We then discuss semiconductor heterostructures, lattice matched layers; the need for lattice
matching, strained layer epitaxy, which usually have certain advantages, and which also deal
with quantum well structures, or when the layer thicknesses are small enough that we discuss
quantum well structures. We also discuss band gap engineering, i.e., various techniques for
tailoring of band gap. And then finally, in a device we have p-n junctions, Schottky junctions,
i.e. junction between metal and semiconductors, and ohmic contacts. We will discuss the last
part here, namely p-n junctions, schottky junctions, and ohmic contacts, from a point of view
of the final device, and briefly overview fabrication of heterostructure devices; there are
several techniques which are used. We will not again go into the details, because there are full
courses on the fabrication technology of semiconductor devices. This is Part I of the course,
so approximately the first fifteen lectures will cover most of the essential semiconductor
physics, or device physics.

(Refer Slide Time: 46:02)

In Part II of the course, we will discuss light emission, absorption, amplification and
modulation, in approximately ten lectures. We start with the interaction of photons with
electrons and holes in a semiconductor. We will discuss optical joint density of states,
probabilities of emission and absorption, rates of emission and absorption, amplification by
stimulated emission—which is the basic principle for laser action—and we discuss the laser
amplifier, absorption spectrum of semiconductors, gain and absorption in quantum well
structures; one of the important devices based on the absorption spectrum of quantum well
structures immediately follows, namely the electro-absorption modulator. We will also
discuss the principle of operation and device configuration here, and in this part we finally
conclude with semiconductor laser amplifiers, or SOAs.
(Refer Slide Time: 47:11)

Part III of the course is Semiconductor light sources. It is Part III, or third module; these can
be treated as different modules. We start with injection electroluminescence, light emitting
diodes, device structure and output characteristics, modulation bandwidth, and materials for
light emitting diodes and their applications. Then we come to laser basics, in particular
semiconductor laser, the device structure and output characteristics; we will discuss singe
frequency lasers, the distributed feedback and distributed Bragg reflector lasers. Also, the
vertical cavity surface emitting laser (VCSEL), about which I mentioned above. We will also
discuss quantum well lasers and practical laser diodes, laser packages, and handling of these
devices.
(Refer Slide Time: 48:13)

In the last part of the course, we will discuss about semiconductor photo detectors. This is a
small part, because most of the characteristics and device structures have already been
discussed in the earlier parts. So, directly we go over to the general characteristics of photo
detectors, the impulse response of photo detectors, and one of the simplest photo detector:
photoconductor, with specific applications; we go to semiconductor photodiodes, two
important configurations being pin diodes and avalanche photodiodes (APDs), their structure,
materials, characteristics and device performance. We will briefly discuss other photo
detectors, such as photo-transistors, solar cells, and CCDs. As an epilogue to these four parts,
finally I would like to summarize and give a talk on photonic integrated circuits, or PICs.
(Refer Slide Time: 49:19)

Here are some of the references which are listed. There are large number of books in the area
of optoelectronics, so I have listed some of the important text books, namely Fundamentals
of Photonics by Saleh and Teich: This book covers most of the photonic sciences and
technologies, but chapters 16 to 18 primarily deal with the semiconductor optoelectronics;
Yariv and Yeh, Photonics: Optical Electronics in Modern Communication, the 6th edition of
the original book Optical Electronics of 1970. Here again, chapters 15 to 17 would be useful;
more recent book by Ghione, Semiconductor Devices for High Speed Optoelectronics, and
two classic books on Semiconductor Optoelectronics—the early books exclusively on
semiconductor optoelectronics—by Pallabh Bhattacharya and Jaspreet Singh. I have listed
them here, and although they are dated, they are still very useful in terms of fundamentals and
the early developments in optoelectronics. So I would strongly recommend you to go through
these two books as well for the basics in semiconductor optoelectronics.

So these are the course contents, and as I said the course will have four modules, which I
have written as Part I, Part II, Part III and Part IV, the first part being the review of essential
semiconductor physics for optoelectronics. Those of you who have had a first course already
in semiconductor physics, can even go over directly to part two of the course. I think with
this I will stop this first lecture. Almost all lectures will follow as per the course contents
listed here. We will start with part one of the course, i.e., review of semiconductor physics. In
the next lecture, we start with energy bands in solids: the E-k diagram.