Chapter 2: Wave equations

2.1 Introduction

In this chapter we develop mathematical expressions for wave motion in general but concentrate on
the most useful special case, the harmonic wave. Harmonic wave functions are then adapted to represent
electromagnetic waves, which include light waves. Results from electromagnetism describing the physics
of electromagnetic waves are borrowed to enable a determination of the energy delivered by such waves.

2.2 One-dimensional wave equation

The most general form of a one-dimensional traveling wave, and the differential equation it satisfies,
can be determined in the following way. Consider first a one-dimensional wave pulse of arbitrary (but
time-independent) shape, described by 𝒚′ = 𝒇(𝒙′), fixed to a (moving) coordinate system 𝑶′(𝒙′, 𝒚′) as in
Figure 1a. Consider next that the 𝑶′system, together with the pulse, moves to the right along the x-axis
at uniform speed relative to a fixed coordinate system, 𝑶(𝒙, 𝒚), as in Figure 1b. Here the coordinate 𝒚
could, for example, represent the transverse displacement from equilibrium of a string stretched out along
the x-direction. As it moves, the pulse maintains its shape. Any point on the pulse, such as P, can be
described by either of two coordinates, 𝒙 or 𝒙′, where 𝒙′ = 𝒙 − 𝒗𝒕. The y-coordinate is identical in either
system. From the point of view of the stationary coordinate system, then, the moving pulse has the
mathematical form 𝒚 = 𝒚′ = 𝒇(𝒙′ ) = 𝒇(𝒙 − 𝒗𝒕)

Figure 1 : Translating wave pulses.

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 1

Chapter 2: Wave equations

If the pulse moves to the left, the sign of must be reversed, so that we may write

𝒚 = 𝒇(𝒙 ± 𝒗𝒕) (𝟏)

as the general form of a traveling wave. Notice that we have assumed 𝒙 = 𝒙′ at 𝒕 = 𝟎. The original
shape of the pulse, 𝒚′ = 𝒇(𝒙′), does not vary but is simply translated along the x-direction by the amount
𝒗𝒕 at time 𝒕. The function 𝒇 is any function whatsoever, so that, for example,

all represent traveling waves. Only the first, however, represents the important case of a periodic
wave. We wish to find next the partial differential equation that is satisfied by all such waves, regardless
of the particular function 𝒇. Since 𝒚 is a function of two variables, 𝒙 and 𝑡, we use the chain rule of partial
differentiation and write

′) ′
𝝏𝒙′ 𝝏𝒙′
𝒚 = 𝒇(𝒙 where 𝒙 = 𝒙 ± 𝒗𝒕 so that = 𝟏 and = ±𝒗
𝝏𝒙 𝝏𝒕

Employing the chain rule, the spatial derivative is:

Repeating the procedure to find the second derivative,

Similarly, the temporal derivatives are found:

Combining the results for the two second derivatives, we arrive at the one-dimensional differential wave
equation,

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 2

Chapter 2: Wave equations

Any wave of the form of Eq. (1) must satisfy Eq. (2), regardless of the physical nature of the
wave itself. Thus, to determine whether a given function of 𝒙 and 𝒕 represents a traveling wave, it is
sufficient to show either that it is of the general form of Eq. (1) or that it satisfies Eq. (2).

2.2 Harmonic Waves

Of special importance are harmonic waves that involve the sine or cosine functions,

𝒚 = 𝑨𝒔𝒊𝒏
𝒄𝒐𝒔 [𝒌(𝒙 ± 𝒗𝒕)] (𝟑)

where 𝑨 and 𝒌 are constants that can be varied without changing the harmonic character of the wave.
These are periodic waves, representing smooth patterns that repeat themselves endlessly. Such waves are
generated by undamped oscillators undergoing simple harmonic motion. In addition, the sine and cosine
functions together form a complete set of functions; that is, a linear combination of terms like those in Eq.
(3) can be found to represent any periodic waveform. Such a series of terms is called a Fourier series.
Thus combinations of harmonic waves are capable of representing more complicated waveforms, even a
series of rectangular pulses or square waves.

Figure 2: Extension of a sine wave in space and time. (a) Section of a sine wave at
a fixed time. (b) Section of a sine wave at a fixed point.

Since 𝒔𝒊𝒏(𝒙) = 𝒄𝒐𝒔(𝒙 − 𝝅/𝟐) the only difference between the sine and cosine functions is a
relative translation of 𝝅/𝟐 radians. It is sufficient in what follows, therefore, to treat only one of these
functions. Accordingly, a section of a sine wave is pictured in Figure 2. In Figure 2a, a section of a wave
with amplitude A is shown at a fixed time, as in a snapshot; in Figure 2b, the time variations of the wave
are pictured at a fixed point x along the wave. In Figure 2a, the repetitive spatial unit of the wave is shown
as the wavelength Because of this periodicity, increasing all x by should reproduce the same wave.

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 3

Chapter 2: Wave equations

Mathematically, the wave is reproduced because the argument of the sine function is advanced by 𝟐𝝅
Symbolically,

Or

It follows that 𝒌𝝀 = 𝟐𝝅, so that the propagation constant k contains information regarding the
wavelength.

𝟐𝝅
𝒌= (𝟒)
𝝀

Alternatively, if the wave is viewed from a fixed position, as in Figure 2b, it is periodic in time with
a repetitive temporal unit called the period T. Increasing all t by T, the waveform is exactly reproduced,
so that

Clearly 𝒌𝒗𝑻 = 𝟐𝝅, and we have an expression that relates the period T to the propagation constant
k and wave velocity 𝒗. The same information is included in the relation

where we have used Eq. (4) together with the reciprocal relation between period 𝑻 and frequency 𝝂

𝟏
𝝂= (𝟔)
𝑻

Related descriptions of wave parameters are often used. The combination 𝝎 = 𝟐𝝅𝝊 is called the
angular frequency, and the reciprocal of the wavelength 𝒌 = 𝟏/𝝀 is called the wave number. Note that
the propagation constant k is related to the spatial period (i.e., the wavelength) of the wave in the same
way that the angular frequency is related to the temporal period T. Therefore, the propagation constant k
is the spatial frequency of the wave. With these relationships it is easy to show the equivalence of the
following common forms for harmonic waves:

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 4

Chapter 2: Wave equations

In any case, the argument of the sine or cosine, which depends on space and time, is called the
phase, 𝝋. For example, in Eq. (7),

𝝋 = 𝒌(𝒙 ± 𝒗𝒕) (𝟏𝟎)

When x and t change together in such a way that 𝝋 is constant, the displacement 𝒚 = 𝑨𝒔𝒊𝒏𝝋 is also
constant. The condition of constant phase evidently describes the motion of a fixed point on the waveform,
which moves with the velocity of the wave. Thus if 𝝋 is constant,

𝒅𝒙
𝒅𝝋 = 𝟎 ⟺ 𝒌(𝒅𝒙 ± 𝒗𝒅𝒕) = 𝟎 ⟹ = ±𝒗
𝒅𝒕

confirming that 𝒗 represents the wave velocity, which is in the negative x-direction when 𝝋 =
𝒌(𝒙 + 𝒗𝒕) and in the positive x-direction when 𝝋 = 𝒌(𝒙 + 𝒗𝒕).

Notice that the waveforms of Eqs. (7) through (9) that use the sine function all represent waves for
which 𝒚 = 𝟎 at position 𝒙 = 𝟎 and time 𝒕 = 𝟎. All of the cosine waveforms in these equations represent
waves for which 𝒚 = 𝑨 at position 𝒙 = 𝟎 and time 𝒕 = 𝟎. As pointed out previously, both situations could
be handled by either the sine or cosine function if an angle of 𝝅/𝟐 is added to the phase. In general, to
accommodate any arbitrary initial displacement, some angle 𝝋𝟎 must be added to the phase. For example,
Eq. (7) with the sine function becomes

Now suppose our initial boundary conditions are such that 𝒚 = 𝒚𝟎 when 𝒙 = 𝟎 and 𝒕 = 𝟎. Then

from which the required initial phase angle 𝝋𝟎 can be calculated as

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 5

Chapter 2: Wave equations

The waveforms in Eqs. (7) to (9) can be generalized further to yield any initial displacement,
therefore, by the addition of an initial phase angle 𝝋𝟎 to the phase. In many cases, the precise phase of the
wave is not of interest. Then 𝝋𝟎 can be set equal to zero for simplicity.

Example 1:

A traveling wave in a string has a displacement from equilibrium given as a function of distance
along the string x and time t as

Determine the wavelength, frequency, velocity, and initial phase angle. Also find the displacement
at 𝒙 = 𝟏𝟎 𝒄𝒎 and 𝒕 = 𝟎.

Solution

By comparison with Eq. (9), 𝒌 = 𝟑𝝅/𝒎 and 𝝎 = 𝟏𝟎𝝅/𝒔 .Thus,

The initial phase (𝒙 = 𝟎, 𝒕 = 𝟎) is 𝝅/𝟒 The velocity of the wave may be found from 𝒗 = 𝝀𝝂 =
𝟐
( ) 𝟓𝒎𝒔−𝟏 = 𝟑. 𝟑𝟑𝒎𝒔−𝟏 in the positive x-direction (due to the negative sign in the phase). One can also
𝟑
𝟏𝟎𝝅
set the phase 𝜑 = (𝟑𝝅/𝒎)𝒙 − ( ) 𝒕 + 𝝅/𝟒 equal to a constant so that
𝒔

2.4 Harmonic waves as complex functions

Using Euler’s formula, it is possible to express a harmonic wave as the real (or imaginary) part of
the complex function

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 6

Chapter 2: Wave equations

Note that any equation that involves only terms that are linear in and its derivatives will also hold
̃) or 𝒚 = 𝑰𝒎(𝒚
for 𝒚 = 𝑹𝒆(𝒚 ̃). Many mathematical manipulations can be carried out more simply with
exponential functions than with trigonometric functions. As a result, it is common practice to use the
complex waveform Eq. (18) to represent a harmonic wave when doing calculations and then to take the
real or imaginary part of this complex function to recover the physical wave represented by one of the
forms in Eq. (19).

2.5 Plane waves

We wish now to generalize the harmonic wave equation further so that it can represent a waveform
propagating along any direction in space. Since an arbitrary direction involves the three spatial coordinates
𝒙, 𝒚, and 𝒛, we represent the wave “displacement” or disturbance by rather than y; for example,

𝝍 = 𝑨 𝐬𝐢𝐧(𝒌𝒙 − 𝝎𝒕) (𝟐𝟎)

It is important to note that need not represent only physical displacements but could represent any
quantity that varies in space and time such as the difference of air pressure from its equilibrium value (as
in a sound wave) or the strength of an electric or magnetic field (as in a light wave). Equation (20)
represents a traveling wave moving along the + x-direction. At fixed time (for simplicity we take 𝒕 = 𝟎 ),
the wave is described by

𝝍 = 𝑨 𝐬𝐢𝐧(𝒌𝒙) (𝟐𝟏)

When 𝒙 = 𝒄𝒐𝒏𝒔𝒕𝒂𝒏𝒕, 𝝋 = 𝒌𝒙 = 𝒄𝒐𝒏𝒔𝒕𝒂𝒏𝒕. Thus, the surfaces of constant phase are a family of
planes perpendicular to the x-axis. These surfaces of constant phase are often called the wavefronts of the
disturbance. For concreteness, consider a plane sound wave propagating along the x-direction through a
sample of air. The propagation of this sound wave alters the air pressure P as it passes. Let

𝝍 = 𝑷 − 𝑷𝟎 = (𝟏𝟎𝑵/𝒎𝟐 ) 𝐬𝐢𝐧[(𝟐𝝅/𝒎)𝒙 − (𝟔𝟖𝟎𝝅/𝒔)𝒕]

represent the difference in the air pressure from its equilibrium value (𝑷𝟎 ≈ 𝟏𝟎𝟓 𝑵𝒎−𝟐 ). In Figure 5,
𝝍 is plotted as a function of x at a fixed time 𝒕 = 𝟎 and several wavefronts associated with the plane sound
wave are depicted at this same time. Note that at all points on a given plane wavefront 𝝍 have the same
value. In addition, 𝝍 has the same value on all wavefronts separated by a wavelength 𝝀 = 𝟏 𝒎.

A motion picture depiction of the sound wave of Figure 5 would show the wavefronts moving in
the positive x-direction at the wave speed 340 m/s. Clearly, plane waves, which have planar wavefronts
that are infinite in extent, are approximations to real waves, which have limited extent in directions that
are transverse to the propagation direction. Treating a wave as a simple plane wave is a useful
approximation if a portion of the wave

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 7

Chapter 2: Wave equations

Figure 5: Air pressure variation induced by a plane sound wave propagating in the x-direction. The plot
shows the difference in air pressure from its equilibrium value as a function of x at and several planar
wavefronts associated with the sound wave are depicted at the same time. Note that the wavefronts
associated with adjacent maxima are separated by one wavelength.

Consider again the wave represented in Eq. (20). Since, for this waveform, the wave disturbance at
⃗ vector in Figure 6a, is the same as for the point x along the
an arbitrary point in space, defined by the 𝒓

x-axis, where 𝒙 = 𝒓𝒄𝒐𝒔𝜽. Eq. (21) may then be written as

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 8

Chapter 2: Wave equations

Figure 6: Generalization of the plane wave to an arbitrary direction. The wave direction is
given by the vector ⃗𝒌 along the x-axis in (a) and an arbitrary direction in (b).

Equation 20 can therefore be generalized if the propagation constant, whose magnitude has already
been determined in Eq. (4), is now considered to be a vector quantity, pointing in the direction of
propagation. Then and the harmonic wave of Eq. (20) becomes

In this form, Eq. (22) can represent plane waves propagating in any arbitrary direction given by as
shown in Figure 6b. In the general case,

where (𝒌𝒙 , 𝒌𝒚 , 𝒌𝒛 ) are the components of the propagation direction and (𝒙, 𝒚, 𝒛) are the components
of the point in space where the displacement 𝝍 is evaluated and 𝒔 is the component of the position vector
along the direction of the propagation of the wave. Note that, in general, 𝒔 represents the distance along a
waveform measured along a direction that is perpendicular to the wavefronts associated with the wave.

A general harmonic wave in three-dimensions can be expressed in complex form as

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 9

Chapter 2: Wave equations

(Recall that the physical waveform is described by the real or imaginary part of the complex form.) The
partial differential equation satisfied by such three-dimensional waves is a generalization of Eq. (2) in the
form

as can easily be verified by computing the second partial derivatives of from Eq. (23). The wave Eq. (24)
is often written more compactly by separating the spatial second derivatives from the wave function 𝝍
by treating them as operators:

𝝏𝟐 𝝏𝟐 𝝏𝟐 𝟏 𝝏𝟐 𝝍
( 𝟐 + 𝟐 + 𝟐) 𝝍 = 𝟐 𝟐
𝝏𝒙 𝝏𝒚 𝝏𝒛 𝒗 𝝏𝒕

𝝏𝟐 𝝏𝟐 𝝏𝟐
The entire operator in parentheses is known as the Laplacian operator, 𝛁 𝟐 = 𝝏𝒙𝟐 + 𝝏𝒚𝟐 + 𝝏𝒛𝟐

and Eq. (24) becomes simply

𝟏 𝝏𝟐 𝝍
𝛁𝟐 𝝍 = (𝟐𝟓)
𝒗𝟐 𝝏𝒕𝟐

2.6 Spherical waves

Harmonic wave disturbances emanating from a point source in a homogeneous medium travel at
equal rates in all directions. As shown in Figure 7, surfaces of constant phase, that is, wavefronts, are
then spherical surfaces

Figure 7: Portions of three spherical wavefronts emanating from a point source

O. The rays indicate that the direction of energy propagation is radially outward
from O.
centered at the source. Such waves, which are of course also solutions to

Eq. (25), can be represented by the complex waveform

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 10

Chapter 2: Wave equations

Here, 𝒓 is the radial distance from the point source to a given point on the waveform and, as for
plane waves, 𝒌 = 𝟐𝝅/𝝀 and 𝝎 = 𝟐𝝅/𝑻

Note that the constant 𝑨 appearing in Eq. (26) is not the overall amplitude of the wave, which is
instead given by 𝑨/𝒓. The spherical wave, as it propagates further from the source, decreases in amplitude,
in contrast to a plane wave for which the amplitude is constant. If the amplitude at distance 𝒓 from the
point source is 𝑨/𝒓, then the irradiance (𝑾/𝒎𝟐 ) of the wave there is proportional to (𝑨/𝒓)𝟐. That is, Eq.
(26) encodes the familiar inverse square law of propagation for spherical wave disturbances. Clearly, Eq.
(26) is not valid as r approaches zero but rather describes the disturbance at a finite distance from a small
physical source. Over a small enough region (or sufficiently far from the source), the spherical wavefronts
associated with a spherical wave are approximately planar. For this reason, waves emanating from point
sources can be adequately described by plane waveforms when the region of interest is small compared to
the distance from the point source.

2.7 Other harmonic waveforms

2.7.1 Cylindrical Waves

Another useful complex waveform represents a cylindrical wave in which the wavefronts are
outward-moving cylindrical surfaces surrounding a line of symmetry, as shown in Figure 8. Such a wave
takes the form,

Figure 8: Plane waves incident on a slit generate cylindrical waves.

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 11

Chapter 2: Wave equations

𝑨
𝝍= 𝐞𝒊(𝒌𝝆±𝝎𝒕)
√𝝆

Here, 𝝆 represents the perpendicular distance from the line of symmetry to a point on the waveform.

That is, if the z-axis is the line of symmetry, then 𝝆 = √𝒙𝟐 + 𝒚𝟐 . Waves of this form are not exact solutions
to the wave equation given in Eq. (25) and so do not exactly represent physical waves but rather are
approximately valid for large 𝝆. Still, they are useful forms that approximate the wave that emerges from
a slit illuminated by a plane wave.

2.7.2 Gaussian Beams

Another important family of (single-frequency) approximate solutions to the differential wave Eq.
(25) consists of the rather complicated but important Hermite-Gaussians. Hermite-Gaussian waveforms
are, to an excellent approximation, produced by laser systems that use spherical mirrors to form the laser
cavity and are beamlike, in the sense that the beam irradiance is strongly confined in the transverse
direction. We defer a detailed discussion of these beams for now but sketch and indicate some of the most
important features of the simplest Hermite-Gaussian waveform in Figure 9. The parameter 𝝎(𝒛), shown
in Figure 9, is often called the spot size and marks the transverse distance from the axis of the beam to the
point at which the irradiance falls to 𝒆−𝟐 ≈ 𝟎. 𝟏𝟑𝟓 of the maximum irradiance that occurs on the symmetry
axis of the beam. Note that the beam spreads while maintaining nearly spherical wavefronts that change
radius of curvature as the beam propagates.

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 12

Chapter 2: Wave equations

The minimum spot size 𝝎𝟎 occurs at the so-called beam waist, where the wavefronts are planar. The
location and size of the beam waist is determined by the nature of the laser cavity (and subsequent focusing
elements) that form the beam. Note that the half-angle beam divergence is larger for beams with smaller
beam waists. This is an important general feature of the propagation of electromagnetic waves. In regions
close to the symmetry axis, the Gaussian beam can be adequately described by planar waveforms.

Figure 9: Gaussian beam propagating in the z-direction. The spot size at the beam waist
(planar wavefront) is defined as 𝜽 = 𝝀/𝝅𝝎𝟎 . The half-angle beam divergence is valid only
in the far field. Note the change in transverse irradiance as the beam propagates to the right.
2.8 Electromagnetic waves

The harmonic waveforms discussed so far can represent any type of wave disturbance that varies in
a sinusoidal manner. Some familiar examples of such disturbances are waves on a string, water waves,
and sound waves. The disturbance 𝝍 may refer to transverse displacements of a string or longitudinal
pressure variations due to a sound wave propagating in a gas—as mentioned earlier. In general, harmonic
waves are produced by sources that oscillate in a periodic fashion. Charged particles oscillating with a
regular frequency emit harmonic electromagnetic waves. For electromagnetic waves (including light), can
stand for either of the varying electric or magnetic fields that together constitute the wave. Figure 10a
depicts a plane electromagnetic wave traveling in some arbitrary direction. From Maxwell’s equations,
which describe such waves, we know that the harmonic variations of the electric and magnetic fields are
always perpendicular to one another and to the direction of propagation given by ⃗𝒌, as suggested by the
orthogonal set of axes in Figure 10a. These variations may be described by the harmonic waveforms

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 13

Chapter 2: Wave equations

⃗
Figure 10: Plane electromagnetic wave described by Eqs. (28) and (29). (a) The electric field 𝑬
⃗ and propagation vector ⃗𝒌 are everywhere mutually perpendicular. (b) Wavefronts
magnetic field ⃗𝑩
for a (linearly polarized) plane electromagnetic wave.

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 14

Chapter 2: Wave equations

where ⃗𝑬 and ⃗𝑩
⃗ represent the electric and magnetic fields, respectively, and ⃗𝑬
⃗ 𝟎 and ⃗𝑩
⃗ 𝟎 are their

amplitudes. Each component of the wave travels with the same propagation vector ⃗𝒌 and frequency 𝝎 and
thus with the same wavelength and speed. Furthermore, electromagnetic theory tells us that the field
amplitudes are related by 𝑬𝟎 = 𝒄𝑩𝟎 where c is the speed of the wave. Figure 10b shows a plane wave
propagating along the positive z-direction with the electric field varying along the x-direction and the
magnetic field varying along the y-direction.

At any specified time and place,

In free space, the velocity 𝒄 is given by

where the constants 𝜺𝟎 and 𝝁𝟎 are, respectively, the permittivity and permeability of vacuum.
Measured values for these constants, 𝜺𝟎 = 𝟖. 𝟖𝟓𝟒𝟐 ∙ 𝟏𝟎−𝟏𝟐 (𝑪𝒔)𝟐 /𝒌𝒈𝒎𝟐and 𝝁𝟎 = 𝟒𝝅 ∙ 𝟏𝟎−𝟕 𝒌𝒈 ∙
𝒎/(𝑨 ∙ 𝒔)𝟐 , provide an indirect method of determining the speed of electromagnetic waves in free space
and yield a value of 𝒄 = 𝟐. 𝟗𝟗𝟖 ∙ 𝟏𝟎𝟖 𝒎𝒔−𝟏 . Recall that light is the term for electromagnetic radiation that
human eyes can “see.” Humans see different wavelengths of light as different colors. Light wavelengths
range from 380 nm (violet) to 770 nm (red).

An electromagnetic wave, of course, represents the transmission of energy. The energy density,
𝒖𝑬 in 𝑱𝒎−𝟑 , associated with the electric field in free space is

and the energy density associated with the magnetic field in free space is

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 15

Chapter 2: Wave equations

These expressions, easily derived for the static electric field of an ideal capacitor and the static
magnetic field of an ideal solenoid, are generally valid. Incorporating Eqs. (30) and (31) into either of the
Eqs. (32) or (33), 𝒖𝑬 and 𝒖𝑩 are shown to be equal. For example, starting with Eq. (33),

The energy of an electromagnetic wave is therefore divided equally between its constituent electric
and magnetic fields. The total energy density is the sum

1 2
𝒖 = 𝒖E + 𝒖B = 𝟐𝒖E = 𝟐𝒖B 𝒐𝒓 𝒖 = 𝜺𝟎 E 2 = 𝐵 (35)
𝜇0

Figure 11: Energy flow of an electromagnetic wave.

In time ∆𝒕 the energy enclosed in the rectangular
volume ∆𝑽 flows across the surface A.

Consider next the rate at which energy is transported by the electromagnetic wave, or its power. In
a time ∆𝒕 the energy transported through a cross section of area A (Figure 11) is the energy associated
with the volume ∆𝑽 of a rectangular volume of length 𝒄∆𝒕. Thus,

or the power transferred per unit area, S, is

We now express the energy density u in terms of 𝑬 and 𝑩, as follows, making use of Eqs. (31) and (35):

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 16

Chapter 2: Wave equations

Inserting this result into Eq. (37),

The power per unit area, S, when assigned the direction of propagation, is called the Poynting vector.
Since this direction is the same as that of the cross product of the orthogonal vectors, ⃗𝑬 and ⃗𝑩
⃗ we can
write, finally,

Note that since this relation involves the product of two waveforms, it does not hold for waveforms
written in complex form. Because of the rapid variation of the electric and magnetic fields, whose
frequencies are 𝟏𝟎𝟏𝟒 to 𝟏𝟎𝟏𝟓 Hz in the visible spectrum, the magnitude of the Poynting vector in Eq. (39)
is also a rapidly varying function of time. In most cases, a time average of the power delivered per unit
area is all that is required. This quantity is called the irradiance, 𝑬𝑒

where the angle brackets denote a time average and we have expressed the fields as sine functions
of the phase. The average of the functions 𝐬𝐢𝐧𝟐 𝜽 or 𝐜𝐨𝐬𝟐 𝜽 over a period is easily shown to be 1/2 so that

The alternative forms of Eq. (42) are expressed for the case of free space. They apply also to a
medium of refractive index n if 𝜀0 is replaced by 𝑛2 𝜀0 and c is replaced by the velocity c/n. Notice that
these changes leave the first of the alternative forms invariant.

Example 2

A laser beam of radius 1 mm carries a power of 6 kW. Determine its average irradiance and the
amplitude of its 𝑬 and 𝑩 fields.

Solution

The average irradiance

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 17

Chapter 2: Wave equations

2.9 Light polarization

As we have noted, the fields associated with electromagnetic waves are vector quantities such that,
at every point in the wave, the electric field, the magnetic field, and the direction of energy propagation
⃗ ×𝑩
are mutually perpendicular, with the direction of energy propagation being the direction of 𝑬 ⃗⃗ . In order
to completely specify the electromagnetic wave, it is sufficient to specify the electric field since the
⃗ is known. The direction of the electric field
magnetic field and Poynting vector can be determined once 𝑬
is known as the polarization of the wave. For example, consider an electric field propagating in the positive
z-direction and polarized in the x-direction,

According to Maxwell’s equations, the magnetic field associated with this electric field would be

and Eq. (40) would give the Poynting vector as

The polarization of an electromagnetic wave determines the direction of the force that the
electromagnetic wave exerts on charged particles in the path of the wave through application of the Lorentz
⃗
force law. This law states that the electromagnetic force on a particle of charge 𝑸 moving with velocity 𝑽
in an electromagnetic field is

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 18

Chapter 2: Wave equations

Unless the speed of the charged particle is a significant fraction of the speed of light, the magnitude
of the electric force on the particle will be much larger than that of the magnetic force. The electric force
on the charged particle is along the direction of the polarization of the wave and so must be perpendicular
to the direction of propagation of the wave. Many optical applications depend critically on the nature and
manipulation of the polarization of electromagnetic waves. In summary, electromagnetic waves are
produced by oscillating (in general, simply accelerating) charge distributions and carry energy (and
momentum) as they travel. These waves exert forces on charged particles in the wave path.

Linear and Elliptical Polarizations

We conclude this section with but a brief discussion of the basic nature of the polarization of
harmonic electromagnetic fields. The electric field of Eq. (43) is said to be linearly polarized along the x-
direction since the direction of the electric field is always along the x-direction. As illustrated in Figure
12a, light may be linearly polarized along any line that is perpendicular to the direction of wave
propagation. In the figure, the electric field ⃗𝑬 made up of equal parts along the x- and y-directions,
oscillates along a line making an angle of 45° with the x-axis. The electric field vector in the 𝒛 = 𝟎 plane
is shown every eighth of a period over one period, 𝑻 = 𝟐𝝅/𝝎.

In general, electric fields may be elliptically polarized in the sense that, over time, the electric field
vector traces out an ellipse as the wave propagates. The special case of an electromagnetic wave that is
circularly polarized is illustrated in Figure 12b. In this graph, the component of the electric field

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 19

Chapter 2: Wave equations

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 20

Chapter 2: Wave equations

Figure 12: Electric field polarization. Evolution of the electric field vector over one period at a fixed plane
is shown for a wave with (a) linear polarization and (b) circular polarization.

along the y-direction is always 𝝅/𝟐 out of phase with the x-component of the electric field, leading
to circular polarization. Note that both circular and linear polarizations represent limiting cases of elliptical
polarizations. An investigation of the general case of arbitrary elliptical polarization is left as an exercise
(see problem 24).

Unpolarized Light

Often the individual atoms in a source, at a given instant, emit light with differing random
polarizations. The light coming from such a source is then a superposition of electromagnetic fields with

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 21

Chapter 2: Wave equations

differing and randomly distributed polarizations. Such light is said to be randomly polarized or,
commonly, unpolarized. If a certain electromagnetic field consists of the superposition of fields with many
different polarizations, of which one or more predominates, we say the field is partially polarized.
Polarized light can be produced by passing unpolarized light through one of a variety of optical systems
that transmit only a particular polarization of light.

2.10 Doppler Effect

The familiar Doppler Effect for sound waves has its counterpart in light waves, but with an important
difference. Recall that when dealing with sound waves, the apparent frequency of a source increases or
decreases depending on the motion of both source and observer along the line joining them. The frequency
shift due to a moving source is based physically on a change in transmitted wavelength. The frequency
shift due to a moving observer is based physically on the change in speed of the sound waves relative to
the observer. The two effects are physically distinct and described by different equations. They are also
essentially different from the case of light waves. The difference between the Doppler Effect in sound and
light waves is more than the difference in wave speeds. Whereas sound waves propagate through a material
medium, light waves propagate in vacuum. As soon as the medium of propagation is removed, there is no
longer a physical basis for the distinction between moving observer and moving source. There is one
relative motion between them that determines the frequency shift in the Doppler Effect for light. The
derivation of the Doppler Effect for light requires the theory of special relativity and so is not carried out
here. The result2 is expressed by

Where 𝝀′ is the Doppler-shifted wavelength and is the relative velocity between source and observer.
The sign of 𝒗 is positive when they are approaching one another and negative when they are separating
from one another.
When 𝒗 ≪ 𝒄 Eq. (44) is approximated by

The Doppler Effect is especially important when used to determine the speed of astronomical sources
emitting electromagnetic radiation. The redshift is the shift in wavelength of such radiation toward longer
wavelengths, due to a relative speed of the source away from us.

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 22

Chapter 2: Wave equations

Example 3

Light from a distant galaxy shows the characteristic lines of the oxygen spectrum, except that the
wavelengths are shifted from their values as measured using laboratory sources. In particular, the line
expected at 513 nm shows up at 525 nm. What is the speed of the galaxy relative to the earth?

Solution

Since the apparent 𝝀 is larger (the frequency less), the galaxy is moving away from the earth with a
speed of approximately 7020 km/s. Another situation in which the Doppler Effect is of pivotal importance
is the Doppler broadening of the spectral lines associated with the light emitted by the fast-moving atoms
of a gas. Since such atoms have a range of velocities relative to a laboratory detector, a range of detected
frequencies, corresponding to the range of atomic velocities, will result even if the atoms emit nearly
single-frequency electromagnetic radiation. Finally, we mention Doppler weather radar, in which a source
emits an electromagnetic radio wave towards moving raindrops and other particulates. In turn, the
particulates reflect the radio wave back towards the source. The difference in wavelength of the emitted
waves and that of those detected after reflection is directly related to the velocity of the particulates relative
to the source of the radar.

PHYS4117 « Advanced Optics » Level III by Dr Alexandre MANDO KONGNE 23