Unit-4

LASERS:
Introductory Concepts
Objective
The objective of this unit is to give the student introductory but fundamental concept
regarding lasers. These concepts include the fundamental difference between spontaneous
and stimulated emission, their origins, and basic requirements of laser emission, e.g., active
medium, population inversion, and optical feedback. A brief history is also given for general
interest.

4.1 Introduction and Brief History

Laser is a device that amplifies light and produces a highly directional, high intensity beam
that typically has an almost pure frequency or wavelength. It comes in sizes ranging from
approximately one-tenth the diameter of a human hair to the size of a very large building, in
powers ranging from 10-9 to 1020 watts, and in wavelengths ranging from the microwave to
the soft x-ray spectral regions with corresponding frequencies from 10 11 to 1017 Hertz. Lasers
have pulse energies as high as 104 joules and pulse duration as short as 10-15 seconds.

Lasers are key components of some of our most modern communication systems and are the
“phonographic needle” of compact disc players. They are used for heat treatment of high-
strength materials, such as the pistons of automobile engines, and provide a special surgical
knife for many types of medical procedures. They act as target designators for military
Unit-4

weapons and are used in the checkout scanners. They can easily drill holes in the most
durable of materials and can weld detached retinas within the human eye.

The word laser is an acronym for Light Amplification by Stimulated Emission of Radiation.
One of the major processes in lasers is stimulated emission. Albert Einstein gave the idea of
stimulated emission of radiation in 1917. Until that time, physicists had believed that a
photon could interact with an atom in only two ways: it could absorbed and raise the atom to
higher energy level or be emitted as the atom is dropped to a lower energy level. Einstein
proposed a third possibility that a photon with energy corresponding to that of an energy
level transition could stimulate the atom in the upper level to drop to the lower level by
emission of another photon with the same energy and phase as the first one.

The stimulated emission is least known because at thermodynamic equilibrium more atoms
are in lower energy levels than in higher ones. Thus a photon is much more likely to
encounter an atom in a lower level and be absorbed than to encounter one in a higher level
and stimulate emission.

The first efforts to use the idea of stimulated emission were a few decades latter in the
microwave region and led to the invention of maser “microwave amplification by stimulated
emission of radiation”. The maser concept evolved nearly simultaneously in the United
States and Soviet Union. A physicist Charles H. Townes of the Colombia University is
known as the inventor of maser.

Townes thought that molecules in the excited state could be stimulated to emit microwaves
when placed in a special resonant cavity designed to enhance the emission. He outlined the
idea to post-doctoral fellow Herbert Zeiger and graduate student James P. Gordon in 1951,
and by 1953 they had a working maser. Meanwhile, Alexander M. Prokhorov and Nikolai
Basov (1954) of Lebedev Physics Institute in Moscow calculated the details of maser action
and published shortly after Townes’s results. The contributions of all these three men were
recognised and they were awarded Nobel Prize in physics in 1964.

After maser, Townes and other physicists began looking beyond the microwave region to
shorter wavelengths. They realised that at those wavelengths the physical conditions required
to produce stimulated emission would be very different. Townes and Arthur L. Schawlow (of
Bell Labs.) worked out many key parameters. Their results were published in a major paper

4.2.
 Lasers: Introductory Concepts

in 1958 and they also filed a patent application before the paper was published.

Meanwhile, a graduate student at Columbia, Gordon Gould, was working out his own
analysis of the conditions required for stimulated emission at visible wavelengths. Gould
wrote his proposals in a set of notebooks in 1957 but did not try to publish his results
promptly. He wanted to patent his work but due to some bad legal advice, he did not file a
patent application until 1959, about nine months after the Schawlow-Townes patents
application was submitted. The Schawlow-Townes patent was granted promptly, but Gould’s
application ran into a lengthy process. Finally, Gould got four patents in 1977, 1979, 1987
and 1988, based on divisions of his original application.

Schawlow and Townes have received many scientific honours for their work, but Gould
received little recognition until his patents were issued. Although Gould loses the prestige
race, yet he benefited financially. However, it was he, who first coined the word LASER in
his notebooks. Schawlow and Townes described their idea as an “optical maser”.

Publication of Schawlow-Townes paper stimulated many scientists to build lasers, and

interest spread beyond the narrow scientific community. Schawlow, Gould and most
researchers thought that gases were the best materials for lasers. However, Theodore H.
Maiman, a young scientist at Hughes Research Laboratories in Malibu, California, quietly
disagreed. He preferred synthetic ruby crystals to gases, although some theorists insisted that
ruby would not work. Maiman, who had studied energy levels in ruby extensively, proved
that the theorists were wrong. In mid-1960 he proudly demonstrated the world’s first laser,
the ruby laser. The laser era was born.

Maiman had to face a lot of difficulties; Hughes management told him to stop work on the
ruby laser, the prestigious journal, Physical Review Letters, rejected his report of ruby laser
but it was published in Nature. Today, however, Maiman is universally recognised as the
person who built first laser and has received a number of honours.

Maiman’s demonstration of the ruby laser opened the floodgates and was followed by the
demonstration of helium neon laser by Ali Javan, W.R. Bennett Jr., and Donald R. Harriot at
Bell Telephone Laboratories in Murray Hill, New Jersey. Their first helium-neon laser
operated at 1.15 micrometers (µm) in the near infrared. Latter, other researchers found the
632.8 nanometer (nm) red line which made the helium-neon laser most popular.

4.3
Unit-4

The laser boom really got going. In 1961, L. F. Johnson and K. Nassau demonstrated the first
solid-state neodymium laser in which the Nd ions were dopant in calcium tungstate, but the
today’s best choice of neodymium host for best commercial applications- yttrium aluminium
garnet (YAG), was demonstrated as a laser material in 1964. Three separate groups
demonstrated semiconductor diode lasers nearly simultaneously in fall 1962. All the teams
demonstrated the same gallium arsenide diodes cooled to the 77 K temperature of liquid
nitrogen and pulsed with high-current pulses lasting a few microseconds. The next few years
saw the birth of several more important lasers. W. B. Bridges (1964) observed 10 laser
transitions in the blue and green parts of the spectrum from singly ionised argon. C. Kumar
and N. Patel (1964) obtained a 10.6 µm laser emission from carbon dioxide. Sorokin and J.
R. Lankard (1966) demonstrated the first organic dye laser; hydrogen chloride emitting at 3.7
µm was demonstrated in 1965 by J. V. V. Kaspar and G. C. Pimentel.

4.2 Basic Principle of Lasers

In order for most lasers to operate, three basic conditions must be satisfied. First, there must
be an active medium, that is, a collection of atoms, molecules, or ions that emit radiation in
the optical part of the electromagnetic spectrum. Second, a condition known as population
inversion, i.e., more number of atoms available at the higher energy level as compare to
lower energy level, must exist. This condition is highly abnormal in nature. It is created in a
laser by an excitation process known as pumping. Finally, for true laser oscillation to take
place there must be some form of optical feedback present in the laser system. If this was not
present, the laser might serve as an amplifier of narrow-band light, but it could never produce
the highly collimated, monochromatic beam that makes the laser so useful.

4.2.1 Active Medium

The heart of a laser system is a material capable of emitting radiation of the required energy.
This material, known as active medium, may have any form e.g. solid, liquid, or gas but must
contain a set of energy levels in which it can absorb or emit energy in the form of optical
radiation.

4.4.
 Lasers: Introductory Concepts

4.2.1.1 Energy levels

In the macroscopic world, energy might seem to vary continuously, like the level of sand in a
pail. If we look closely at the sand, we can see that it is made up of many separate grains, and
that we can add or subtract only one grain at a time. Likewise, atoms and molecules can only
have certain amounts of energy. We call these energy states or levels.

For example, let’s look at the simplest atom (hydrogen) in which a single electron circles a
nucleus that contains a single proton. At first glance, the hydrogen atom looks like a very
simple solar system, with a single planet (the electron) orbiting a star (the proton). The force
that makes the atom stable is the attraction between the positive charge of the proton and the
negative charge of the electron.

The electron can occupy only certain orbits, as shown in Figure 4.1. We show the orbits as
circles for simplicity, but we can’t really measure exactly what the orbit looks like. If we add
energy to our simple planetary system, the planet would move farther from the star. The
same happens in the hydrogen atom. As we add more energy to the atom, the electron moves
to more and more distant orbits. However, there is crucial difference between the behaviour
of a hydrogen atom and our imaginary planetary system. The planet can be at any distance
from the sun, or could even fall into it. However, the electron can only occupy certain orbits
also called energy levels. These energy levels are plotted in Figure 4.1. Atom is, in general,
lies in the lowest possible energy level, the ground state.

The energy levels in the hydrogen atom get closer together as they get higher above the
ground state. Eventually the differences become vanishingly small. If the electron gets too
much energy, it escapes from the atom altogether, a process called ionization. If we define
the energy of the ionized hydrogen atom to be zero, we can write the energy of the atom E as
a negative number using the simple formula:

E = -R/n2 (4.1)

where, R is a constant (2.178x10-18 Joules), n is the quantum number of the orbit (counting
outwards, with one the innermost level).

4.5
Unit-4

N=6
N=5
N=6 N=4
N=3
N=5
N=4
N=3 N=2
N=2

Energy
N=1

Ground
State
N=1
Nucleus

Figure 4.1 The hydrogen atom and the corresponding energy levels.

4.2.1.2 Transitions
The process of making changes or “transitions” between energy levels is very important to
laser physics. To look at the process, let’s start again with the hydrogen atom. The electron
needs to gain energy to move from ground state to higher energy level. Conversely, it must
release energy when it drops from a higher level to lower one. One of the most convenient
ways (although by no means the only way) for an electron to absorb or release energy is as a
photon, a quantum of electromagnetic energy. The photon energy equals the difference in
energy between the two energy levels.

Suppose we start with the electron in the ground state and want to raise it up one step to the
first excited level. To do so, we must give the electron exactly as much extra energy as the
difference in energy between the ground state and the first excited level. Conversely, for the
electron to drop from the first excited level to the ground state, it must emit a photon with as
much energy as the difference in energy between the two levels. In short, the photon energy
equals the transition energy. The energy of a photon can be represented by hν; in the
frequency term, where h (= 6.626 x 10-34 J.s) is the Plank’s constant and ν is the frequency of

4.6.
 Lasers: Introductory Concepts

the transition, or in the wavelength form hc/λ, where c is the velocity of light in vacuum and
λ is the wavelength of the photon. An electron in a hydrogen atom can emit or absorb light at
only certain wavelengths.

This neat ordering of energy levels is evident only in hydrogen atom, where there is just a
single electron. Add more electrons, and the energy level picture quickly becomes more
complicated. Electrons interact with each other, and with the nucleus, shifting energy levels
slightly. Electrons can occupy subshells within each shell. The more complex the energy
level structure, the more transitions between energy levels is possible. The more transitions,
the larger number of possible spectral lines. Superimpose them all on a single spectrum and
look almost like a set of random lines.

Another complication is that all transitions are not equally likely. One reason is that more
atoms are in some states than in others. For example, under normal circumstances, more
atoms are in ground state than in excited levels. Another is that quantum mechanical rules
make some transitions much more likely than others. That means that an atomic or molecular
species will absorb or emit some wavelengths much more strongly than others. This effect
shows up both in absorption spectra (which shows that wavelengths the material absorbs
when light from another source passes through it) and emission spectra (the wavelengths the
material emits when it is itself de-excited).

4.2.1.3 Types of Transitions

So far we have concentrated on electronic transitions, partly because we picked the hydrogen
atom as our introductory example. Electronic transitions can cover a wide range of
wavelengths. These occur at ultraviolet, visible, or infrared wavelengths from 100 nm in the
ultraviolet through to near infrared wavelengths.

The shortest wavelengths come from inner-shell electronic transitions in heavy elements,
which involve much more energy than outer-shell transitions. These short wavelengths are
considered to be X-rays. On the other hand, transitions between high lying electronic energy
levels (say, level 18 and 19 of hydrogen) involve very little energy, putting them deep in the
infrared, microwave, or even radio-frequency range. Because these are qualitatively different
than higher-energy transitions of outer electrons, they are put into special class called

4.7
Unit-4

Rydberg transitions.

Neither Rydberg transitions nor X-ray emission falls into the optical region, which is a very
small portion of the electromagnetic spectrum, therefore, these are not likely events under
normal laser conditions. Transitions between nuclear energy levels can produce even higher
energy photon, called gamma rays.

On the other end of the wavelength spectrum are transitions between vibrational and
rotational energy levels of molecules. Vibrational transition energies typically correspond to
wavelengths of a few to tens of micrometers; rotational transitions have less energy, typically
corresponding to wavelengths of at least 100 micrometers.

Transitions in two or more types of energy levels can occur simultaneously. For example, a
molecule can undergo a vibrational and rotational transition simultaneously, with the
resulting wavelength close to that of more energetic vibrational transition. Many infrared
lasers emit families of closely spaced wavelengths on such vibrational-rotational transitions.

Remember in considering transitions that longer wavelengths correspond to lower energy,

and shorter wavelengths correspond to higher energy (E = hc/λ). Thus, the energy of a
vibrational transition is much larger than that of a rotational transition, even though the
rotational wavelength is much larger. A combination of a rotational and vibrational transition
thus has only slightly different energy than the original vibrational transition. Transition
energies or frequencies add together in a straightforward manner:

E1+2 = E1 + E2 (4.2)

where, E1+2 is the combined transition energy, E 1 and E2 are the energies of the separate
transitions. The same rule holds for frequencies, with ν substituted for the energy. However,
wavelength (λ) of combined transitions add by an inverse rule:

1/λ1+2 = 1/λ1 + 1/λ2 (4.3a)

or λ1+2 = 1/(1/λ1 + 1/λ2 ) (4.3b)

4.2.1.4 Spontaneous Emission

We know that atoms have well defined energy levels and they can be pushed to the higher

4.8.
 Lasers: Introductory Concepts

energy states (excited states) by absorption of a photon or by some other means e.g. electric
discharge. Atoms spent some time in the excited state and then decay to the lower energy
state. The average time required to de-exite the 1/e number of atoms from the upper level to
the lower level is called lifetime of the state. This lifetime can be very small e.g. in nano-
seconds and up to a few seconds. Typically, for electronic transitions it is in the order of ten
nano-seconds. After spending this period in the excited state, atoms come down to the lower
energy state by emitting a photon. This emission of photon from an excited atom is called
spontaneous emission. This phenomenon one sees in daily life. Sun, bulb, tubelight and all
the fluorescent devices emit photons spontaneously in the visible region. These photons are
emitted in all the direction and illuminate the whole area, probably, this is the blessing of
spontaneous emission.

Example 4.1: Describe spontaneous emission from excited helium

When we see helium discharge lamp directly, it appears to emit pinkish white light as
shown in Figure 4.2. In fact in the discharge, helium atoms are being excited in the
upper energy levels. They soon give up their energies by dropping down to lower
energy level with emission of photons. This spontaneous emission involves quite a
number of different energy levels and thus produces the photons of different colours.
When viewed the same discharge through a diffraction grating (an optical element to

He spectrum
7065
6678

5876

5015
Wavelength in Angstrum

4922 Transparent Eye

4713 grating
4471

Helium
discharge
lamp

Figure 4.2 Helium discharge spectrum observed through a transmission grating.

4.9
Unit-4

separate different wavelengths of light) multiple images of different colours of the

lamp appear at different angles. A strong yellow line at 588 nm is prominent but
violet, green, blue, red and deep red lines can also be seen.

To understand the spontaneous emission a little more, let us consider two energy levels, 1
and 2, of some given material, their energies being E 1 and E2 (E1<E2) and population
densities N1 and N2 as shown in Figure 4.3. It is convenient, however, to take level 1 to be
the ground level. Let us now assume that an atom (or molecule) of the material is initially in
level 2. Since E2>E1, the atom will tend to decay to level 1. The atom must therefore release
the energy, corresponding energy difference (E2-E1). This energy is delivered in the form of a
photon of frequency ν given by

(E 2 − E1 )
ν= (4.4)
h

N 2
E 2
Energy

N 1
E 1

Figure 4.3 Two level energy system.

Spontaneous emission is therefore characterized by the emission of a photon of energy hν =

E2-E1, when the atom decays from level 2 to level 1 (Figure. 4a). Note that radiative emission
is just one of the two possible ways for the atom to decay. The decay can also occur in a non-
radiative way. In this case the energy difference E 2-E1 is delivered in some form other than a
photon (e.g., it may go into kinetic energy of the surrounding molecules).

The probability of spontaneous emission can be characterised in the following way: Let us
suppose that, at time t, there are N2 atoms (per unit volume) in level 2. The rate of decay of
these atoms due to spontaneous emission, i.e., (dN2/dt)sp, will obviously be proportional to
N2. We can therefore write

4.10.
 Lasers: Introductory Concepts

 dN 2 
  = − A21 N 2 (4.5)
 dt  sp

The coefficient A21 is called the spontaneous emission probability or the Einstein coefficient.
The quantity τsp=1/A21 is called the spontaneous emission lifetime. The numerical value of
A21 (and τsp) depends on the particular transition involved.

Initial state Final state

(a)
Spontaneous
emission

(b)
Stimulated
emission

(c)
Stimulated
absorbtion

Figure 4.4 Energy level diagram illustrating (a) spontaneous emission, (b) stimulated
emission and (c) stimulated absorption.

In spontaneous emission each individual atom acts like a small randomly oscillating antenna
emitting at the transition frequency. Therefore, the total emission from a collection of
spontaneously emitting atoms exhibit noise-like character (Figure 4.5).

4.11
Unit-4

incoherent radiations

Figure 4.5 Spontaneous emission is incoherent or noise-like character, emerging randomly in

all directions.

4.2.1.5 Stimulated Emission

In stimulated emission, the atom is triggered to undergo the transition in the presence of
photons of energy (E2-E1). In these transitions each individual atom acts like a passive
resonant antenna that is set oscillating by the applied signal. Therefore, the internal motion or
oscillation in the atom is not random, but is driven by an applied signal. Let us again assume
that the atom is found initially in level 2 and a photon of frequency ν as of equation (1) is
incident on the material. Since this wave has the same frequency as the atomic frequency,
there is a finite probability that this wave will force the atom to undergo the transition 2 → 1.
In this case the energy difference E2-E1 is delivered in the form of a photon that adds to the
incident one as shown in Figure 4b. There is, however, a fundamental distinction between the
spontaneous and stimulated processes. In the case of spontaneous emission, the atom emits a
photon that has no definite phase relation with that emitted by another atom. Furthermore, the
photon can be emitted in any direction. In the case of stimulated emission, since the process
is forced by the incident photon, the emission of any photon adds in phase to that of the
incoming wave. This wave also determines the direction of the emitted wave. In this case,
too, we can characterise the process by means of the equation

dN 2  ρ
 = −ν
B21 N 2
dt  st (4.6)

4.12.
 Lasers: Introductory Concepts

where (dN2/dt)st is the rate at which transitions 2 → 1 occur as a result of stimulated emission,
B21 is called the Einstein coefficient for stimulated emission, the energy density ρν = Nhν,
where N is the number of photons per unit volume having frequency ν.

4.2.1.6 Absorption
All the atoms and molecules have discrete energy levels. An atom can absorb a photon and
become excited. Let us now assume that the atom is initially in level 1. If this is the ground
level, the atom will remain in this level unless some external stimulus is applied to it. We
shall assume that a photon of frequency ν given by equation (1) is incident on the material. In
this case there is a finite probability that the atom will be raised to level 2 (Figure 4.4c). The
energy difference E2-E1 required by the atom to undergo the transition be obtained from the
energy of the incident photon. This is the absorption process.

In a similar fashion to equations 2 & 3, we can write an equation for absorption rate,

dN1
=
−ρ
νB12 N1
dt (4.7)

where N1 is the number of atoms per unit volume that at the given time are lying in level 1
and B12 is called the Einstein coefficient for stimulated absorption.

In the absorption process, the incident photon is simply absorbed to produce the 1 → 2
transition. In the beginning of the century, Einstein also showed that the coefficients of
stimulated emission and absorption are equal, i.e. B 21=B12 (assuming that degeneracy in both
the upper and lower level is same)

4.2.1.7 The Einstein Relations

Einstein showed that the parameters describing the above three processes are related through
the requirement that for a system in thermal equilibrium, the rate of upward transitions (E 2 to
E1) must be equal to the rate of downward transition processes.

We can write the upward transition rate as N1ρνB12. The total downward transition rate is the
sum of the induced and spontaneous contributions i.e. N 2ρνB21+N2A21. In the preceding
discussions A21, B21 and B12 are called the Einstein coefficients. The relationship between

4.13
Unit-4

them can be established as follows.

For a system in equilibrium, the upward and downward transition rates must be equal and
hence we have

N1ρνB12 = N2ρνB21+N2A21 (4.8)

Thus

N 2 A 21
ρν = (4.9a)
N1B12 − N 2 B21

or

A 21 B21
ρν = (4.9b)
N1 N 2 − 1

The populations of various energy levels of a system in thermal equilibrium are given by
Boltzmann statistics to be:

N 0 exp(− E j kT)
Nj =
∑ exp(− E i kT) (4.10)
i

where Nj is the number of atoms in the jth level with energy E j, N0 is the total number of
atoms, k is Boltzmann constant, and T is temperature in Kelvin.

N1
= exp(( E 2 − E1 ) kT)
Hence N 2 (4.11)
= exp( hν kT)

From equations 4.9b and 4.11 we get ratio of spontaneous emission to stimulated emission in
thermodynamic equilibrium as

A 21
= exp( hν / kT) − 1 (4.12)
ρν B21

Equation (4.9) shows that whenever there is emission of photons, each in thermal
equilibrium, stimulated emission is always present. ;However, the ratio of stimulated
emission to spontaneous emission depends on the temperature (in thermal equilibrium)
frequency of the photons and the number of photons available for stimulation.

4.14.
 Lasers: Introductory Concepts

Example 4.2: Calculate the ratio of spontaneous emission to stimulated emission for
tungsten filament lamp operating at a temperature of 1500 K (assume the average frequency
ν to be 5x1014 Hz).

A 21
Solution: The ratio R of spontaneous emission to stimulated emission ( R = ) is as
ρν B21
follows

R = exp( hν / kT) − 1

Taking k=1.38x10-23 JK-1, h=6.6x10-34J.s, ν =5x1014 Hz.

 6.6x10−34.5x1014 
R = exp − 23
 − 1
 1.38x10 .1500 
≈ e16 ≈ 8.4x106

This confirms that under conditions of thermal equilibrium stimulated emission is not very
significant. For sources operating at lower temperatures and higher frequencies stimulated
emission is even less likely.

Example 4.3: At what temperature are the rates of spontaneous and stimulated emission
equal for λ=550 nm radiation? At what wavelength are they equal at room temperature
(T=300 K)?

Solution: For spontaneous and stimulated emission rates to be equal at wavelength of 550
nm, we have

 h.c 
R = exp  −1 = 1
 λkT 
h.c 1
⇒T=
λk ln(1 + R )

Substituting values of h=6.6x10-34 J.s, k=1.38x10-23 JK-1, c=3x108 m/s, λ=550 nm, and R=1,
we get T=37635 K. Hence to achieve the equal spontaneous and stimulated emission
temperature of more that thirty seven thousand Kelvin is required.

Similarly for equal spontaneous and stimulated emission rates at room temperature we can

4.15
Unit-4

find the required wavelength, λ, as

h.c 1
λ=
k.T ln(1 + R )

Again substituting values of all the constants and temperature T=300 K, we found λ~69 µm
which lies in the microwave region. Thus radiation of 69 µm wavelength have equal
spontaneous and stimulated emission rates at room temperature.

The above discussion indicates that the process of stimulated emission competes with the
processes of spontaneous emission and absorption. Clearly if we wish to amplify a beam of
light by stimulated emission then we must increase the rate of this process in relation to the
other two processes. To achieve this for a given pair of energy levels we must increase i)
radiation density and ii) the population density N2 of the upper level in relation to the
population density N1 of the lower level. We shall show that to produce laser action we must
create a condition in which N2>N1, even though E2>E1 that is we must create a so-called
population inversion.

4.2.2 Laser Pumping

The population inversion required for light amplification constitutes an abnormal distribution
of atoms among the various available energy levels. To understand how light amplification
can be achieved in a medium, it is necessary to consider the Boltzmann distribution and then
pumping mechanism to achieve the population inversion.

4.2.2.1 Population Inversion

The population inversion condition required for light amplification is a non-equilibrium
distribution of atoms among the various energy levels of the atomic system. The Boltzmann
distribution, which applies to a system in thermal equilibrium, is given by equation 7, it is
obvious that as Ej increases Nj decreases for a constant temperature. We note that if the
energy difference between E1 and E2 is nearly equal to kT (~0.025 eV at room temperature)
then the population of the upper level would be 1/e or 0.37 times of the lower level. For an

4.16.
 Lasers: Introductory Concepts

energy difference large enough to give visible radiation (~2.0 eV), however, the population
of the upper level is almost negligible.

Example 4: An atom has two energy levels with a transition wavelength of 694.3 nm.
Assuming that all of the atoms in an assembly are in one or other of these levels, calculate
the percentage of the atoms in the upper level at room temperature (T=300 K) and at T= 500
K.

Solution: The energy of the radiation of wavelength 694.3 nm,

i.e. E=hc/λ=6.6x10-34.3x108/694.3x10-9=2.85x10-19 Joules

We have

N2 −E  − 2.85x10 −19 

= exp  = exp − 23

N1  kT   1 . 38 x10 .T 

At room temperature i.e. T=300 K, N2/N1=1.2x10-30, i.e. population of the upper level is
1.2x10-28% of the lower level in thermal equilibrium.

At 500 K, N2/N1=1.12x10-17, i.e., population of the upper level is 1.12x10-15% of the lower
level in thermal equilibrium.

This shows that population of the upper level increases about 10 13 times by increasing the
temperature up to 500 K from room temperature, but in both cases N2<<N1.

Example 4.5: The relative number of atoms N1 and N2 in two energy levels E1 and E2
separated by an energy gap E2-E1 are given at thermal equilibrium by Boltzmann ratio.
Evaluate the ratio N2/N1 for the following cases:

an optical transition, λ=550 nm, at room temperature, 300 K;

a microwave transition,  = 3 GHz, at room temperature;

A 10 GHz transition at liquid-helium temperature, 4.2 K.

For an optical transition at λ =550 nm to have N2/N1 = 0.1, what temperature would be

4.17
Unit-4

required?

Solution:

The energy difference for λ = 550 nm is E2-E1 = hc/550x10-9 = 3.6x10-19 J

N2  − 3.6 x10−19 
= exp − 23

N1  1 . 38 x10 × 300 
≈ e −87 ≈ 10− 37

The number of atoms present in the upper energy level for a visible transition at room
temperature is negligibly small.

The energy difference for  = 3 GHz, hν = 6.6x10-34.3x109 = 1.98x10-24 J

N2  − 1.98x10−24 
= exp − 23

N1  1.38x10 × 300 
≈ 0.99952

In the microwave region, number of atoms present in the upper energy level is a little less
than the lower energy level at room temperature. Therefore to create a population inversion
in microwave region is easier than in optical region.

The energy difference for  = 10 GHz, hν = 6.6x10-24 J

N2  − 6.6x10−24 
= exp − 23

N1  1.38x10 × 4.2 
≈ 0.89237

At low temperature the number of atoms present in the upper energy level decreases.

To have N2/N1 = 0.1 at λ = 550 nm, the temperature required is 11329 K.

This shows that to increase the number of atoms in the upper energy level at thermal
equilibrium requires very high temperatures.

From the above example it is obvious that population can not be achieved by increasing the
temperature of the medium. To create a population inversion in the optical region, we must

4.18.
 Lasers: Introductory Concepts

supply energy selectively to the lower energy level, to excite atoms into the upper energy
level. This excitation process is called pumping. Pumping produces a non-thermal
equilibrium situation.

One of the methods used for pumping is stimulated absorption, that is the energy levels
which one hopes to use for laser action are pumped by intense irradiation of the system. Now
as B12 and B21 are equal once atoms are excited into upper level the probabilities of further
stimulated absorption or emission are equal so that even with very intense pumping we can
not get more atoms in the excited state. The best we can achieved is the equal population in
both the levels, therefore, to make a laser amplifier with just two energy levels is not
possible, i.e. a population inversion in a two level system can never be achieved by optical
pumping.

Therefore, we must look for materials with either three or four energy levels system. This is
not a problem as atomic systems generally have a large number of energy levels. A three
level system is illustrated in Figure 4.6. Initially the distribution obeys the Boltzmann law. If
the collection of atoms is intensely illuminated by photons, they can be excited into the level
E3 from the ground level E1. From the level E3, the atoms decay by non-radiative process to
the level E2 and a population inversion may be created between E 2 and E1. Ideally the
transition from level E3 to E2 should be very rapid to keep E3 level almost empty. The
transition from E2 to E1 should be slow, that is E2 should have relatively longer lifetime. This
allows a large build-up in the number of atoms in level E2. Hence N2 may become greater
than N1 and then population inversion will be achieved.

The level E3 should preferably consist of a large number of closely spaced levels so that
pumping uses a wide range of the radiation. This increases the pumping efficiency. Three
level lasers, for example ruby, require very high pumping powers because the terminal level
of the laser transition is the ground state. This means that more than half of the ground state
atoms has to be pumped to the upper state to achieve population inversion.

4.19
Unit-4

E E

E 3 E 3
Rapid decay

E 2 E 2

Pumping
Laser
transition

E 1 E 1

N N
(a) (b)

Figure 4.6 Population of energy levels by pumping in a three level system: (a) Boltzmann
distribution before pumping and (b) distribution after pumping and the transitions involved.

The four-level system shown in Figure 4.7 has much lower pumping requirements. The
populations of the levels E4, E3, and E2 are all very close in conditions of thermal
equilibrium. Thus, if atoms are pumped from the ground state to the level E 4 from which they
decay very rapidly to the level E 3. The population inversion is quickly created between levels
E3 and E2. The energy level schemes of the media used in lasers are often complex but they
can usually be approximated by either three or four level schemes.

4.2.3 Optical Resonator

The population inversion is not all it takes to make a laser. A material with an inverted
population can emit light in every direction. The light may be due to stimulated emission,
and it may be at a single wavelength, but is not converted into a laser beam. To extract
energy efficiently from a medium with a population inversion and make a laser beam, we

4.20.
 Lasers: Introductory Concepts

need a resonant cavity that helps building (or amplifying) stimulated emission by feedback,
i.e., reflecting some of the light back into the laser medium.

Before discussing the feedback, let us start by looking at the process of amplification.

E E

E 4
E 4
Rapid decay
E 3
E 3

Laser
transition
E 2
E 2
Pumping

Rapid decay

E 1
E 1

N N
(a) (b)

Figure 4.7 Population of the energy levels in a four level system: (a) before pumping and (b)
after pumping.

4.2.3.1 Amplification of Light

In lasers, light is amplified. To understand it let us consider a collimated beam of perfectly
monochromatic radiation passing through an absorbing medium (Figure 4.8). We assume for
simplicity that there is only one relevant electron transition, which occurs between the energy
levels E1 and E2. For homogeneous medium, the change in irradiance of the beam dI is
proportional to the distance travelled dx and to the incident intensity I, i.e. dI= -αI.dx. Here
the constant of proportionality, α, is the absorption coefficient. The negative sign indicates
the reduction in beam irradiance due to absorption. Writing this equation in differential form
we have

4.21
Unit-4

dI
= −αI (4.13)
dx

By integrating Eq. 4.10 we have

I = Ioe-αx (4.14)

where Io is the incident intensity.

I(x+ dx)
I(x)

I0

x+dx
x
0

Light

Figure 4.8 Collimated beam of light traversing an absorbing medium. The change in the
irradiance across a small slab of the medium is proportional to the irradiance at the slab and
to the thickness of the slab dx.

Example 4.6: If 1% of the light incident into a medium is absorbed per millimetre, what
fraction is transmitted if the medium is 0.1 m long?

Solution: Transmission of light passing through an absorbing medium of length x is given by

I=I0 e-αx

For α=0.01 mm-1, x=0.1 m=100 mm, we have

I=I0 e-0.01x100 =I0.e-1 =0.368xI0

Therefore, 36.8% of the incident light will transmit through the medium and rest will be
absorbed.

Let us consider the absorption coefficient in more detail. Clearly the degree of absorption of
the beam will depend upon N 1, number of atoms with electrons in the lower energy level E 1,

4.22.
 Lasers: Introductory Concepts

and N2, number of atoms with electrons in upper energy level E 2. If N2 is zero then the
absorption would be maximum, while if all of the atoms are in the upper level the absorption
would be zero and the probability of stimulated emission would be large.

When a beam having N number of photons per unit volume pass through a medium then net
rate of loss of photons can be determined from the stimulated transitions, i.e. absorption and
emission.

dN
=
N2ρ
B21 −
ν N 1ρ
B12
ν
dt (4.15)

For the same degeneracy of both levels (i.e. B12=B21), we can write

dN
=−
(N2 N 1 )ρ
B21
ν
dt (4.16)

In this discussion we have ignored photons generated by spontaneous emission as these are
emitted randomly in all direction and do not contribute to the initial collimated beam.
Similarly we have ignored scattering losses.

Irradiance of a beam is the energy crossing a unit area in unit time. In mathematical form it
can be written as

I=N.hν21.u (4.17)

where u is the velocity of light in the medium. If n is the refractive index of the medium then
u=c/n, where c is the velocity of light in free space. The rate of change of irradiance when
passing through a medium is given by

dI dN
= hν 21 . u (4.18)
dx dx

since

u 1 1
= = , we can write equation 4.18 as
dx dx u dt

dI dN
= hν 21 (4.19)
dx dt

Combining equations 4.16 and 4.19 we get

4.23
Unit-4

dI
=
( N2 −
N 1 )ρ
.B 21 .hν
ν 21
dx (4.20)

Now from equations 4.13, 4.20, ρν = Nhν, and equation (4.17)

1
( N1 − N 2 ).ρν . B21 = α.ρ ν . u (4.21)
hν 21

Re-arranging for absorption coefficient α, and substituting the value of u = c/n, we have

B21 . hν 21 . n
α = ( N1 − N 2 ) (4.22)
c

The absorption coefficient, α, depends on the difference in the populations of the two energy
levels E1 and E2. For a collection of atoms in thermal equilibrium, N1 will always be greater
than N2. If, however, we can create a situation in which N2 becomes greater than N1 then α is
negative and the quantity (-αx) in the exponent of equation 11 becomes positive. Thus the
irradiance of the beam grows as it propagates through the medium in accordance with the
equation:

I = Ioekx (4.23)

where k is referred to as the small signal gain coefficient and is given by

B21 . hν 21 . n
k = ( N 2 − N1 ) (4.24)
c

In this situation the medium acts as an amplifier and may be called active medium.

Example 4.7: If the irradiance of light doubles after passing once through a laser amplifier
0.5 m long, calculate the small signal gain coefficient assuming no losses in the medium. If
the increase in irradiance were only 5% what would be k?

Solution: In an optical amplifier, the light intensity increases as

1  I 
I = I 0 e kx ⇒ k = ln 
x  I 0 

(a) for I/I0=2 and x=0.5 meter; k=1.386 m-1

4.24.
 Lasers: Introductory Concepts

The value of signal gain coefficient is 1.386/m

(b) for I/I0=1.05 and x=0.5 meter; k=0.098 m-1

The value of signal gain coefficient is 0.098/m.

4.2.3.2 Optical Feedback

The gain (amplification) per unit length of most active media is so small that very little
amplification of a beam of light results from a single pass. In the multiple passes through the
same active media may result a large amplification.

To make an oscillator from an amplifier, it is necessary to introduce a suitable positive

feedback. In the laser, the positive feedback may be obtained by placing the active material
between two highly reflecting mirrors (e.g., plain-parallel mirrors as shown in Figure 4.9).
The initial stimulus is provided by any spontaneous transitions between appropriate energy
levels in which the emitted photon travels along the axis of the system. The signal is
amplified as it passes through the medium and fedback by mirrors. Saturation is reached
when the gain provided by the medium exactly matches the losses incurred during a complete
round trip.

Light reflector Partially transmitting mirror

R 1 =100% R 2 ~90%

L
M1 Active medium M2
Laser light

Figure 4.9. Scheme of laser oscillations.

If one of the two mirrors is made partially transparent, a useful output beam can be extracted.
It is important to note that for laser, a certain threshold condition must be fulfilled. For laser,
the oscillation will start when the gain of the active medium compensates the losses. Thus,

4.25
Unit-4

while a population inversion is a necessary condition for laser action it is not the sufficient
one because the minimum or threshold value of gain coefficient must be large enough to
overcome the losses and sustain oscillations. The threshold gain specifies the minimum
population inversion required.

The total loss of the system is due to a number of different processes, the most important
ones include:

transmission at the mirrors, in fact transmission from one of the mirrors usually provides the
useful output, the other mirror is made as reflective as possible to minimise losses;

absorption and scattering at mirrors;

absorption in the laser medium due to transmissions other than the desired transitions;

scattering at optical inhomogeneities in the laser medium;

diffraction losses at the mirrors.

To simplify, let us include all the losses except those due to transmission at the mirrors in a
single effective loss coefficient γ, which reduces the effective gain coefficient to (k-γ). We
can determine the threshold gain by considering the change in irradiance of a beam of light
undergoing a round trip within the laser cavity. We assume that the laser medium fills the
space between the two mirrors M1 and M2 which have reflectance R1 and R2 and a separation
L. In travelling from mirror M1 to M2 the beam irradiance increases from I0 to I according to
equation 4.23 as,

I = I 0 .e ( k −γ).L
(4.25)

After reflection at M2, the beam irradiance will be R2.I0.exp[(k-γ).L] and after a complete
round trip the final irradiance will be such that the round trip gain G is

G = R 1 .R 2 .e 2 ( k − γ ) L (4.26)

If G is greater than unity then there will be a net amplification and the oscillation will grow;
if G is less than unity, the oscillation will die out. Therefore, we can write the threshold
condition as

4.26.
 Lasers: Introductory Concepts

G = R 1 .R 2 .e 2 ( k th − γ ) L = 1 (4.27)

where kth is the threshold gain. It is important to realise that the threshold gain is equal to the
steady-state gain in continuous output lasers. This equality is due to gain saturation, which
can be explained as follows. Initially, when laser action starts the gain may be well above the
threshold value. The effect of stimulated emission reduces the population of the upper level
of the laser transition so that the degree of population inversion and consequently the gain
will decrease. Thus the round trip gain may vary. It is only when G has been equal to unity
for a period of time that the laser output power settles down to a steady-state value, that is
when the gain just balances the losses in the medium. In terms of the population inversion
there will be a threshold value

Nth= (N2-N1)th (4.28)

In the steady state situation (N2-N1) remains equal to Nth regardless of the amount by which
the threshold-pumping rate is exceeded. The small signal gain required to support steady
state operation depends on the laser medium through k and γ, and on the laser construction
through R1, R2 and L.

1  1 
k th = γ + ln  (4.29)
2L  R 1 .R 2 

The above equation shows that k can have a wide range of values, depending not only (N 2-
N1) but also on the intrinsic properties and design parameters. If k is high then it is relatively
easy to achieve laser action. With low gain media, mirrors must have high reflectivities, be
very clean and carefully aligned.

Example 8: Calculate the mirror reflectance required to sustain laser oscillations in a laser
which is 0.1 m long given that the small signal gain coefficient is 1 m -1 (assume one mirror
has 100% reflectance).

Solution: Gain of a laser amplifier is given by G=R 1R2e2kL and for threshold it should be
unity, therefore,

R1R2 e-2kthL =1

4.27
Unit-4

or

R1R2= e-2kthL

It is given that kth=1 m-1, R1=1, and L=0.1 m. The value of R2 is:

R2=e-0.2=0.82

The reflectance of the second mirror (i.e. output coupler) is 82 %.

The efficiency of laser system is the ratio of the output light power to input pump power. It
therefore depends on how effectively the pump power is converted into producing a
population inversion.

Problems
4.1 The part of the electromagnetic spectrum that is of interest in the laser field starts from
the submillimeter wave region and goes down in wavelength to the x-ray region. This covers
the following regions in succession: (1) far infrared; (2) near infrared; (3) visible; (4)
ultraviolet (uv); (5) vacuum ultraviolet (vuv); (6) soft x-ray; (7) x-ray. Draw a chart (to the
scale) indicating all the regions in the units of (i) Angstroms, (ii) Hertz, (iii) electron volts
(ev), and (iv) meters.

4.2 When in thermal equilibrium (at T=300 K), the ratio of the level populations N 2/N1 for
some particular pair of levels is given by 1/e. Calculate the frequency ν for this transition. In
what region of the e.m. spectrum does this frequency fall?

4.3 The value of signal gain coefficient of a certain laser amplifier is 0.29/m. What’s physical
meanings of it?

4.4 Calculate the small signal gain co-efficient required to sustain laser oscillations in a
medium which is 0.15 meter long, given that the effective loss coefficient is 0.15 m -1, the
reflectivity of rear mirror is 100% and that of output mirror is 80%.

4.5 For a two level system, suppose N2 = 10N1, where N1 and N2 denote the number of atoms

4.28.
 Lasers: Introductory Concepts

(per unit volume) in level 1 and level 2 respectively. The transition between these two levels
corresponds to the frequency of 5x104 Hz. Calculate the ratio of spontaneous emission and
stimulated emission. Also calculate the temperature required to produce this population
inversion.

4.6 In a two levels system, the levels are separated by an energy E 2 - E1. Find the ratio of
population inversions at room temperature if transition between levels occur at (i) λ = 0.55
µm (ii) λ = 3 µm

4.7 When two quantum energy levels, E1 and E2 of an atom are separated by an energy gap
∆E = E2 - E1, and a large number of such atoms are in thermal equilibrium at temperature T,
then the relative number of atoms N 1 and N2 in the two energy levels are given by the
Boltzmann ratio N2/N1 = e-∆E/kT. Evaluate this ratio for the following cases:

(a) transition occurs at the frequency of 3100 MHz, and the temperature is 300 K. What is
the fractional population difference (N2-N1)/ N1?

(b) Consider the same situation, except that ν = 9000 MHz and the temperature is 4 K.
What is (N2-N1)/ N1?

(c) Calculate the Boltzmann ratio N2/ N1 for λ = 5500 Ao, and T = 300 K.

(d) What temperature required to make (N2 equal to 8 % of N1 in part (c) ?

4.8 The irradiance of light becomes half after passing through a laser media. Calculate the
small gain coefficient, explain for obtaining negative answer in part.

4.9 If the irradiance of light becomes double after passing once through a laser amplifier 0.5
m long., assuming no losses in the media:

(a) Calculate the small signal gain co-efficient.

(b) Calculate the small signal gain co-efficient, if the length of the media is doubled.

(c) By doubling the length of the media, is the small signal gain co-efficient doubled or not?
Give comprehensive comments in either case.

4.29
Unit-4

Books for further reading

J. Wilson and J.F.B. Hawkes, Optoelectronics: An Introduction, (Prentice-Hall International,

London, 1983)

O. Svelto, Principles of Lasers, (Plenum Press, New York, 1989)

P. W. Milonni and J. H. Eberly, Lasers, (John Wiley & Sons, New York, 1991).

A. E. Seigman, Lasers and Masers, (1971).

D. C. O’Shea, W. R. Callen, and W. T. Rhodes, Introduction to Lasers and Their

Applications, (Addison-wesley, London, 1978)

4.30.