Chapter Two: Optical Waveguides and Fibers

2.1 Introduction
The dramatic reduction of transmission loss in optical fibers coupled with equally important developments
in the area of light sources and detectors has brought about a phenomenal growth of the fiber optic
industry during the past three decades. The birth of optical fiber communication coincided with the
fabrication of low-loss optical fibers and room-temperature operation of semiconductor lasers in 1970.
Ever since, the scientific and technological progress in this field has been so phenomenal that we are
already in the fifth generation of optical fiber communication systems within a brief span of 30 years.
Recent developments in optical amplifiers and wavelength division multiplexing (WDM) are taking us to a
communication system with almost “zero” loss and “infinite” bandwidth. Indeed, optical fiber
communication systems are fulfilling the increased demand on communication links, especially with the
proliferation of the Internet.
2.2 Benefits of Fiber Optics
Fiber-optic cable has many benefits, including:
• Enormous information-carrying capability. The main benefit of fiber-optic cables is their enormous
information-carrying capability. This capacity is dependent upon the bandwidth of the cable.
Bandwidth refers to the range of frequencies that a cable will carry. Electric cables such as coaxial
have a wide bandwidth (up to about 750 MHz), but the bandwidth of fiber-optic cable is much
greater. Data rates in excess of 100 GHz have been achieved, and even higher rates have been
achieved with multiplexing.
• Low loss. Fiber-optic cables have less signal attenuation over a given distance than an equivalent
length of coaxial cable.
• Lightweight. Glass or plastic cables are much lighter than copper cables and offer benefits when low
weight is critical (e.g., aircraft).
• Small size. Practical fiber- optic cables are much smaller in diameter than electrical cables and thus
can be contained in a relatively small space.
• Security. Fiber-optic cables cannot be as easily “tapped” as electrical cables, and they do not radiate
signals that can be picked up for eavesdropping purposes. There is less need for complex and
expensive encryption techniques.
• Interference immunity. Fiber optic cables do not radiate signals, as some electrical cables do, and
cause interference to other cables. They are immune to the picking up of interference from other
sources.
• Greater safety. Fiber-optic cables do not carry electricity. Therefore, there is no shock hazard. They
are also insulators and thus not susceptible to lightning strikes as electrical cables are. They can be
used in corrosive and/or explosive environments without danger of sparks.
There are some disadvantages to fiber-optic cable. High cost is the greatest disadvantage. Otherwise, its
small size and brittleness make it difficult to work with. Special, expensive tools and test instruments are
required.
Page 1 of 14
2.3 Optical Fibre Waveguides
Waveguiding structures are needed at optical frequencies because of intrinsic atmospheric absorption
and the effects of atmospheric disturbance on the propagation of visible and near infra-red radiation.
An optical fibre consists of a core region surrounded by a cladding region. The transmission characteristics
of the optical fibre depend of the size, shape and refractive index profile of the core and cladding.
The core may be of any shape.
• Most fibres have a cylindrical core surrounded by a cylindrical cladding.
• The refractive index of the core is greater than that of the cladding.

Figure 2.1: Illustration of the core and cladding of the fiber optic cable
Outer protective coating layers are applied to the cladding.
The coating material:
• Protects the core and cladding against mechanical damage.
• Provides mechanical strength.
• Provides a seal from water and other external contaminants.
The protective coating does not significantly affect the optical properties of the fibre.
In all fibres the electric fields are concentrated in the core region, and in the cladding region near to the
core/cladding interface.
2.4 Optical Fibre Materials
All electrical conductors absorb optical frequencies of EM radiation. Low-loss dielectric materials must be
used for optical waveguides.
Typical materials:
Silicate glasses: Used extensively for communications applications in the visible and near infra-red. (loss
<1000dB/km from 0.4 to 2.5μm)
Plastic (PMMA) fibres: Plastic fibres based on polymethyl methacrylate are useful in only a limited spectral
range. (loss <1000dB/km from 0.4 to 0.9 μm)
Fluoride glasses: Transparent in the infra-red region, but are toxic and/or hygroscopic. (loss <1000dB/km
from 0.5 to 5 μm)
Page 2 of 14
Chalcogenide glasses: Glasses based on the elements sulphur, selenium and tellurium. Potentially useful
at longer wavelengths(loss <1000dB/km from 1 to 7 μm)
2.5 The Optical Fiber
Figure 2.2a shows an optical fiber, which consists of a (cylindrical) central dielectric core clad by a material
of slightly lower refractive index. The corresponding refractive index distribution (in the transverse
direction) is given by:
𝑛 = 𝑛1 𝑓𝑜𝑟 𝑟 < 𝑎 ∶ 𝑛 = 𝑛2 𝑓𝑜𝑟 𝑟 > 𝑎
Where 𝑛1 and 𝑛2 (< 𝑛1 ) represent respectively the refractive indices of the core and cladding and 𝑎
represents the radius of the core.

Figure 2.2(a) A glass fiber consists of a cylindrical central core clad by a material of slightly lower refractive
index. (b) Light rays impinging on the core-cladding interface at an angle greater than the critical angle
are trapped inside the core of the fiber.
We define a parameter ∆ through the following equations.
𝑛1 2 − 𝑛2 2
∆≡
2𝑛2 2
When ∆≪ 1 (as is indeed true for silica fibers when 𝑛1 is very nearly equal to 𝑛2 ) we may write
(𝑛1 + 𝑛2 )(𝑛1 − 𝑛2 ) (𝑛1 − 𝑛2 ) (𝑛1 − 𝑛2 )
∆= ≈ ≈
2𝑛1 2 𝑛1 𝑛2
Meridional Rays and Skew Rays
Meridional rays pass through the central axis of the guide. Meridional rays give rise to modes which can
be designated as TE or TM modes. Two integers are required to specify each mode. Thus, we have TE lm
and TMlm modes.

Page 3 of 14
Figure 2.3: Meridional rays
Skew rays do not pass through the central axis but describe angular helices within the guide. Components
of both B (and hence H) and E can be transverse to the fibre axis. The modes are designated HElm or EHlm
according to whether the magnetic or electric character is more important.

Figure 2.3: Skew Rays

2.6 The Coherent Bundle
A large number of fibers put together form what is known as a bundle. If the fibers are not aligned, i.e.
they are all jumbled up, the bundle is said to be an incoherent bundle. However, if the fibers are aligned
properly, i.e., if the relative positions of the fibers in the input and output ends are the same, the bundle
is said to be a coherent bundle. If a particular fiber in this coherent bundle is illuminated at one of its ends,
there will be a bright spot at the other end of the same fiber. Thus, a coherent bundle will transmit the
image from one end to another (see Figure 2.4). On the other hand, in an incoherent bundle the output
image will be scrambled.
Because of this property, an incoherent bundle can be used as a coder; the transmitted image can be
decoded by using a similar bundle at the output end. In a bundle, since there can be hundreds of thousands
of fibers, decoding without the original bundle configuration would be extremely difficult. Incoherent
bundles are also used in illumination such as in traffic lights or road signs or even for lighting applications
in buildings.

Page 4 of 14
Figure 2.4: A bundle of aligned fibers. A bright (or dark) spot at the input end of the coherent fiber bundle
produces a bright (or dark) spot at the output end. Thus, an image will be transmitted (in the form of bright
and dark spots) through a bundle of aligned fibers, forming, for example, the letter T shown above.
Perhaps the most important application of a coherent bundle is in a fiber optic endoscope where it can
be inserted inside a human body and the interior of the body can be viewed from outside.
The Numerical Aperture (NA)
We return to Figure 2.2b and consider a ray that is incident on the entrance face of the fiber core, making
an angle i with the fiber axis. Let the refracted ray make an angle θ with the same axis. Assuming the
outside medium to have a refractive index n0 (which for most practical cases is unity), we get
sin 𝑖 𝑛1
=
sin 𝜃 𝑛0
Obviously, if this refracted ray is to suffer total internal reflection at the core-cladding interface, the angle
of incidence 𝜑 must satisfy the equation,
𝑛2
sin 𝜑 (= cos 𝜃) >
𝑛1

Since sin 𝜃 = √1 − 𝑐𝑜𝑠 2 𝜃, then

1
𝑛2 2 2
sin 𝜃 < [1 − ( ) ]
𝑛1
And using the above equations
1/2 1/2
𝑛1 𝑛2 2 𝑛1 2 − 𝑛2 2
sin 𝑖 < [1 − ( ) ] =( )
𝑛0 𝑛1 𝑛02
Let im represent the maximum half-angle of the acceptance cone for rays at the input end.

sin 𝑖𝑚 = (𝑛1 2 − 𝑛2 2 )1/2 = 𝑛1 √2∆

and we have assumed n0 = 1; i.e., the outside medium is assumed to be air. Thus, if a cone of light is
incident on one end of the fiber, it will be guided through it provided the half-angle of the cone is less than
im . This half-angle is a measure of the light-gathering power of the fiber.
We define the numerical aperture (NA) of the fiber by the following equation:

𝑁𝐴 = sin 𝑖𝑚 = √𝑛1 2 − 𝑛2 2 = 𝑛1 √2∆

Example
For a typical step-index (multimode) fiber with n1 ≈ 1.45 and Δ ≈ 0.01, we get

𝑁𝐴 = sin 𝑖𝑚 = √𝑛1 2 − 𝑛2 2 = 𝑛1 √2∆= 0.205 𝑠𝑜 𝑡ℎ𝑎𝑡 𝑖𝑚 = 120

Thus, all light entering the fiber must be within a cone of half-angle 12°.

Page 5 of 14
In a short length of an optical fiber, if all rays between i = 0 and im are launched, the light coming out of
the fiber will also appear as a cone of half-angle im emanating from the fiber end.
2.7 Types of Fiber-Optic Cables
There are two basic ways of classifying fiber-optic cables: The first method is by the index of refraction,
which varies across the cross section of the cable. The second method of classification is by mode, which
refers to the various paths the light rays can take in passing through the fiber. Usually these two methods
of classification are combined to define the types of cable.
Step Index Cable. Step index refers to the fact that there is a sharply defined step in the index of refraction
where the fiber core and the cladding interface. It means that the core has one constant index of refraction
N1 and the cladding has another constant index of refraction N2. When the two come together, there is a
distinct step (see Fig. 2.5). If you were to plot a curve showing the index of refraction as it varies vertically
from left to right across the cross section of the cable, you would see a sharp increase in the index of
refraction as the core is encountered and then a sharp decline in the index of refraction as the cladding is
encountered.
Graded Index Cable. Here, the index of refraction of the core is not constant. Instead, it varies smoothly
and continuously over the diameter of the core (see Fig. 2.6). As you get closer to the center of the core,
the index of refraction gradually increases, reaching a peak at the center and then declining as the other
outer edge of the core is reached. The index of refraction of the cladding is constant.
Cable Mode. Mode refers to the number of paths for the light rays in the cable. There are two
classifications: single mode and multimode. In single mode, light follows a single path through the core; in
multimode, the light takes many paths.
Each type of fiber-optic cable uses one of these methods of rating the index or mode. In practice, there
are three commonly used types of fiber-optic cable: multimode step index, single-mode step index, and
multimode graded index.
Multimode Step Index Cable. The multimode step index fiber cable is probably the most common and
widely used type. It is also the easiest to make and therefore the least expensive. It is widely used for short
to medium distances at relatively low pulse frequencies.
The main advantage of a multimode stepped index fiber is its large size. Typical core diameters are in the
50- to 1000-μm range. Such large-diameter cores are excellent at gathering light and transmitting it
efficiently. This means that an inexpensive light source such as an LED can be used to produce the light
pulses. The light takes many hundreds or even thousands of paths through the core before exiting.
Because of the different lengths of these paths, some of the light rays take longer to reach the other end
of the cable than do others. The problem with this is that it stretches the light pulses. This is called
dispersion.

Page 6 of 14
Figure 2.5 A step index cable cross-section. Figure 2.6 Graded index cable cross section

2.8 Modes
The V parameter for a step index circular waveguide is defined as
2𝜋𝑎
𝑉= (𝑛1 2 − 𝑛2 2 )1/2
𝜆0
where a is the radius of the core.
The condition for single mode propagation in this case is V < 2.405
This condition arises from the fact that the electric field amplitudes in the core must be written in terms
of Bessel functions and the value 2.405 represents the first zero of the J0(x) Bessel function.
[This is equivalent to the condition V<π/2 for planar waveguides which represents the first zero of the
cosine function.]
When V>>1 the number of modes N that can propagate is given by
𝑉2
𝑁≈
2
Multimode and Single Mode
Multimode fibres have a core diameter such that the number of modes that can propagate is very large.
Single mode fibres have a core diameter sufficiently small to ensure that only one mode can propagate.
The number of modes that can propagate in multimode fibres is so large that they effectively form a
continuum of ray angles within the guide.
Example
Consider a fibre with n1= 1.48 and n2 = 1.46 and a core radius of 125 μm operating in the infra-red at a
wavelength of 1.2 μm.
V=159
Since V>>1, the number of modes N that can propagate is given by
𝑉2
𝑁≈ = 12597
2
Page 7 of 14
2.9 Intermodal Dispersion
Consider the propagation of rays corresponding to TE and TM modes.
These meridional rays will follow a zig-zag path along the fibre core.

Figure 2.7: Intermodal dispersion

The ray will propagate with a velocity of magnitude (c/n1) along the ray path. The component of this
velocity along the fibre axis is (c/n1)cos(90°-θ) = (c/n1)sinθ .
The largest component of velocity will be when θ = 900 and corresponds to the lowest order, or
fundamental mode.
The smallest component of velocity will be when θ =θC and corresponds to the highest order mode.
For a length L of fibre the ray with the highest velocity component will take a time Δτmin where
𝐿𝑛1 𝐿𝑛1
∆𝜏𝑚𝑖𝑛 = = to travel along the fibre.
𝑐 sin 900 𝑐

Similarly the ray with the lowest velocity component will take a time ∆𝜏𝑚𝑎𝑥 where
𝐿𝑛1 𝐿𝑛1 2
∆𝜏𝑚𝑎𝑥 = =
𝑐 sin 𝜃𝑐 𝑐𝑛2
Hence the highest order mode lags behind the lowest order mode by an amount ∆𝜏 𝑆𝐼 where
𝐿𝑛1
∆𝜏 𝑆𝐼 = ∆𝜏𝑚𝑎𝑥 − ∆𝜏𝑚𝑖𝑛 = (𝑛 − 𝑛2 )
𝑐𝑛2 1
and the superscript ‘SI’ refers to a step index fibre. It can be shown that
𝐿𝑛1 𝐿(𝑁𝐴)2
∆𝜏 𝑆𝐼 ≈ ∆=
𝑐 2𝑐𝑛1
2.10 Information Capacity
If the input pulses are too closely spaced it is not possible to separate the broadened pulses at the output.
The maximum information that can be transmitted is limited by the pulse broadening of the system.
Let T be the time separation between pulses.
The pulse repetition rate is 1/T.
For simplicity, assume that the input pulse width is narrow compared to T. Let the output pulse width be
Δτ. If T<2Δτ then the two neighbouring pulses will merge and will be indistinguishable.

Page 8 of 14
The maximum pulse repetition rate is given by 1/(2Δτ).
Example
A step index fibre with n1= 1.48 and n2 = 1.46.
∆𝜏 𝑆𝐼 𝑛1
= (𝑛 − 𝑛2 ) = 67𝑛𝑠/𝑘𝑚
𝐿 𝑐𝑛2 1
Bandwidth-Length Product
For this fibre with a length of 1 km the minimum pulse separation is
𝑇 ≈ 2∆𝜏 𝑆𝐼 = 135𝑛𝑠
Giving a maximum pulse repetition rate of
1
≈ 7.4 𝑀𝐻𝑧
2∆𝜏 𝑆𝐼
If the fibre lenth is increased to 2 km then ∆𝜏 𝑆𝐼 is doubled, T is doubled and the maximum pulse repetition
rate reduces to 3.7 MHz.
This step index fibre has a bandwidth-length product of 7.4 MHz-km.
Profile Dispersion
The quantity Δ is defined by
𝑛1 2 − 𝑛2 2
∆=
2𝑛1 2
Both refractive indices n1 and n2 are functions of wavelength.
But the change in n1 with wavelength may not be the same as the change in n2 with wavelength.
Hence Δ will depend on wavelength. The variation of Δ with wavelength gives rise to profile dispersion.
In practice many fibres are manufactured such that the core and cladding are made from the same basic
material (e.g.silica). In such cases the dependence of Δ on wavelength is very small and can usually be
neglected.
2.1 Waveguide Dispersion
Waveguide dispersion arises because the wave propagation velocity itself depends on wavelength
regardless of any refractive index changes in the waveguide materials.
If the wavelength increases the effective velocity of the mode along the guide will decrease and vice versa.
Hence, we will have waveguide dispersion.
2.12 Waveguide Dispersion in Single Mode Fibre
For a step index single mode fibre waveguide dispersion is negative and becomes increasingly negative
with wavelength.

Page 9 of 14
Figure 2.9: Waveguide Dispersion in Single Mode Fibre
The wavelength for minimum total dispersion occurs at a higher wavelength than that corresponding to
the minimum material dispersion.
Through a suitable choice of a and Δ it is possible to make the value of λmin lie anywhere between 1.3 μm
and 1.6 μm. Such fibres are referred to as dispersion shifted fibres.
Depressed Cladding Fibres
More complicated structures such as the depressed cladding fibre can produce a dispersion curve which
exhibits zero dispersion at two wavelengths. Such fibres are referred to as dispersion flattened fibres.

Figure 2.10: Depressed Cladding Fibres

Polarization Dispersion
Single mode fibres with nominal circular symmetry support two modes with orthogonal polarizations.
𝑥 𝑦
These are known as the 𝐻𝐸11 and 𝐻𝐸11 modes where the principal axes x and y are determined by the
symmetry of the fibre cross-section.
These modes generally have different phase velocities. Hence the two modes exhibit different delays
∆𝜏𝑥 𝑎𝑛𝑑 ∆𝜏𝑦 .
A pulse launched into the fibre will suffer dispersion due to polarization mode dispersion (PMD).
Typical values:
Less than 1ps/km in conventional single mode fibres.
Greater than 1ns/km in high birefringence polarization maintaining fibres

Page 10 of 14
2.13 Losses in Optical Fibres
The intensity of light propagating along an optical fibre attenuates due to a variety of mechanisms that
can be classified as intrinsic or extrinsic losses. Intrinsic losses are inherent in the fibre itself. Extrinsic
losses usually result from distortion of the fibre from the ideal straight line configuration.
The absorption losses over a length L of fibre can be expressed as
𝐼 = 𝐼0 𝑒𝑥𝑝(−𝛼𝐿)
where I0 is the initial irradiance, I is the irradiance after a distance L, and α is the absorption coefficient
(which may be wavelength dependent).
For optical fibres this is more usually expressed in the form
10 log(𝑃𝑖 /𝑃𝑓 )
𝑎𝑡𝑡𝑒𝑛𝑢𝑎𝑡𝑖𝑜𝑛, 𝛼 =
𝐿
where Pi is the optical power launched into the fibre and Pf is the power remaining after a length L of fibre.
Fibre attenuation is usually expressed as dB/km.
Absorption Losses in Silica Fibres
Optical fibres suffer intrinsic losses due to absorption. Absorption loss is due to interaction of the light
with electronic states or molecular vibration modes on the core material itself or residual impurities in the
core.
The resulting energy loss is converted into heat. For silica fibres strong absorption takes place in the
ultraviolet due to electronic transitions, and the fundamental absorption edge occurs at 165 nm. Relatively
little absorption takes place in the visible spectrum.
The presence of Si-O, Si-OH and OH multiphonon vibrations produces strong absorption bands in the
infrared at wavelengths of 0.72, 0.95, 1.13, 1.22, 1.38 and 1.73 μm.
Residual transition metal impurities such as the transition metal ions (Fe, Cu, Co, Ni, Mn, Cr, V) contribute
to the absorption. Dopant materials such as P2O5, B2O3, GeO3 used to modify the physical and chemical
properties of the glass contribute to the absorption.
Scattering Losses in Silica Fibres
Silica glasses are amorphous materials resulting in density and compositional fluctuations. These density
variations and compositional changes lead to microscopic variations in the refractive index which in turn
cause scattering.
These scattering centres are dimensionally smaller than the optical wavelength and lead to Rayleigh
scattering.
Waveguide Imperfections
Waveguide imperfections, such as changes in the radius eccentricity, ellipticity and core index profile as
well as impurities or air bubbles in the core and cladding also cause scattering.

Page 11 of 14
Figure 2.11: Waveguide imperfections
Scattering loss due to waveguide imperfections is more or less independent of the wavelength.
Bending Losses
In multimode fibres the principal bending loss mechanism is through mode coupling.
For a mode propagating near to cut-off, a bend in the fibre can couple energy into a non-guided mode, i.e
a mode for which θ < θC. The light is refracted out of the core into the cladding and energy is lost to the
cladding. Since higher order modes are closer to cut-off than lower order modes, it is the higher order
modes that are most likely to be affected.
For a fibre bent into a circular arc, the loss is strongly dependent on the radius of curvature and can often
be expressed as
𝑅
𝛼𝐵 = 𝐶𝑒𝑥𝑝 (− )
𝑅𝐶
Where
αB is the absorption coefficient due to bending,
R is the bend radius
RC is a constant which depends upon the fibre.
Butt Joint - Fresnel Losses
May be permanent or temporary. Two fibre ends mechanically held in close proximity to one another.
Losses will occur due to Fresnel reflection at the fibre/air interfaces.
Assuming normal incidence the fraction of light reflected RF at each interface is given by
𝑛1 − 𝑛0 2
𝑅𝐹 = ( )
𝑛1 + 𝑛0
The energy transmitted at each interface will be
𝑇𝐹 = 1 − 𝑅𝐹
Since each joint has two interfaces we will have
𝑇 = (1 − 𝑅𝐹 )2 for the resulting transmission.

Page 12 of 14
Typical losses are of the order of a few tenths of a decibel.
Butt Joint - Longitudinal Misalignment

Figure 2.12: Butt Joint- Longitudinal Misalignment

Assume that the radiation spreads out uniformly from one fibre into a cone of half angle αmax. The fraction
FS of light transmitted by the first fibre and captured by the second fibre is given by
𝑑/2
𝐹𝑆 = { }
𝑑/2 + 𝑆𝑡𝑎𝑛𝛼𝑚𝑎𝑥
Ignoring Fresnel reflection losses
The insertion loss will then be given by
𝐿𝑜𝑠𝑠𝑙𝑜𝑛𝑔 = −10 log(𝐹𝑆 )𝑑𝐵
Butt Joint - Lateral Misalignment

Figure 2.13: Butt Joint - Lateral Misalignment

Assume that the radiation coupled from one fibre into the other is proportional to the area of overlap at
the ends of the fibre.
The fraction FL of light transmitted by the first fibre and captured by the second fibre is given by
1
2 𝑎 𝑎 𝑎 2 2
𝐹𝐿 = {cos −1 ( ) − [1 − ( ) ] }
𝜋 𝑑 𝑑 𝑑

ignoring Fresnel reflection losses.

The insertion loss will then be given by
𝐿𝑜𝑠𝑠𝑙𝑎𝑡 = −10 log(𝐹𝐿 )𝑑𝐵
Butt Joint -Angular Misalignment

Figure 2.14: Butt Joint -Angular Misalignment

Page 13 of 14
The effects of angular misalignment are not so easy to calculate but a reasonably accurate expression for
the angular coupling efficiency FA is
𝛽 𝛽
𝐹𝐴 = 1 − 1 = 1−
𝜋(𝑁𝐴)
𝜋𝑛1 (2Δ)2
where β is the angular misalignment in radians. The insertion loss will then be given by
𝐿𝑜𝑠𝑠𝑎𝑛𝑔 = −10 log(𝐹𝐴 )𝑑𝐵
The expression predicts that smaller the value of Δ the greater will be the effect of angular misalignment.
This is as would be expected from the relationship between Δ and the numerical aperture of the fibre.
Alignment Losses
The effects of longitudinal and lateral misalignment on fibre connection losses in multimode fibres can be
expressed in terms of the normalised parameters (S/d) and (D/d) respectively, where d is the core
diameter.

Figure 2.16: Alignment losses

The loss when there is no misalignment is due, in each case, to Fresnel reflection losses.
Joining Losses
Fusion Splice
Fibre ends are held in close proximity, heated up to about 2000°C and pushed together.
Ends fuse to form a single length of fibre and are allowed to cool. Any slight misalignment of fibres tends
to be corrected by surface tension of glass in liquid state. Joint is mechanically weak and must be
supported e.g with a metal or glass tube glued over joint.
Suitable for both multimode and single mode fibres. Transmission loss can be less than 0.1 dB.

Page 14 of 14