Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

ELECTROMAGNETIC WAVES
& OPTICS.
PREFACE.
The objective of the course is to provide an account of the study of optics taking up
the subject from where the study of electric and magnetic fields finished with the
elucidation of the Maxwell equations and developing these into a wave description of
light. Some basics of the properties of electromagnetic waves will be studied allowing
an exploration of physical optics.
As with any subject, it is necessary to begin by developing a language with which to
discuss the subject matter. In this case it is electromagnetic waves that are the
subject of interest and as for all waves, sound, light, quantum mechanical etc. the
waves have several properties of interest in common. For the purposes of this study
these include at the top of the list the wave amplitude and phase.
The module outline eschews traditional development of the subject by beginning with
physical optics before going on to geometric optics in keeping with its emphasis on
the electromagnetic wave nature of light. The development of the module will
proceed as follows;
i) The module begins by exploring the meaning of the amplitude of the EM
wave, in terms of the electric field (and in terms of the accompanying magnetic field)
and its relationship to the intensity (power per unit area) carried by that wave. This
will enable the development of an understanding of how transmission and reflection
at a dielectric interface may be described and evaluated.
ii) The module will then focus on the phase of the wave and how this is to be
used in circumstances (the most general ones) where the overall optical effect is
produced by a superposition of many waves; specifically in interference and
diffraction effects.
iii) Geometric optics will then, ignoring the wave properties of the light, provide
an understanding of simple optical components and the use of such components in
some common optical instruments.
iv) The course will end by exploring some of the quantum aspects of light and
in particular a simple description of the interaction of light with matter by describing
that interaction in terms of a basic two (quantum) level system.

1
 Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

MAXWELLS EQUATIONS & the WAVE EQUATION

a) Maxwell’s equations in free space.
The starting point of this course is where the first year electromagnetism course
ended, namely with the four Maxwell’s equations and the electromagnetic wave
equation. A review of these equations is therefore a good place to start.
Beginning with the simplest situation of Maxwell’s equations in free space (no
charges or currents ). There are a set of four equations, established by Maxwell, that
relate the magnetic fields and the electric fields. These equations may be written in
differential form as:


E  0 (1.1a)

B  0 (1.2a)

 dB
E   (1.3a)
dt

 dE
  B   0  0 (1.4a)
dt

In free space the equations for E and B are symmetric in form.  0 is the electric

permittivity of free space ( = 8.85  10-12 Fm-1) and  0 is the magnetic permeability

of free space (= 4  10-7 NA-2 )

It is often necessary to use the equations in a form that includes the field sources
i
where  is the charge density in free space and J is the current density, = where i
A
is the current.  and J are source terms for the electric and magnetic fields. The
equations then become;

2
 Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

 
E  (1.1b)
0

B  0 (1.2b)

 dB
E   (1.3b)
dt

 dE
  B  0J  00 (1.4b)
dt

Maxwell’s equations are often written in integral form using Gauss’s Law or Stoke’s
Law with surface or line integrals representing the divergence and curl equations
respectively:

  Q
 .dA   0
E (1.1c)

 
 .dA  0
B (1.2c)

  dB
 .dl   dt
E (1.3c)

   dE 
 .dl  0  iC  dt 
B (1.4c)

where  E and  B are the electric and magnetic flux through the area A enclosed by
the loop (= EA and BA respectively) and Q is the charge enclosed by the surface
implied in 1.1c.

3
 Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

Electromagnetic Wave Equation

Equations 1a – 4a may be used to derive a wave equation in the absence of charge
and current densities and in vacuum which describes all electromagnetic waves
propagating in free space.
One of the two dependent variables (E or B ) may be eliminated by using the vector
identity

             2  (1.5)

where  is any vector field.

For instance, using 1.3a, 1.4a (and 1.1a for   E  0 )

 
  2E 

 
  B  
 E      E    
2



 
 t   B   0  0  2 
 (1.6)
 t   t 

This is recognisably the wave equation of the form

1  2
2  (1.7)
v 2 t 2

Thus the electromagnetic wave equation has been derived for fields obeying
Maxwell’s equations;

2  2E 1  2E
 E  0 0  (1.8)
t 2 c 2 t 2

where
1
c (1.9)
 0 0

is the speed of light in vacuum and given our values for 0 and 0, c = 2.99 108 ms-
1
.

4
 Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

Solutions of the Wave Equation

The fact that this wave equation is obtained from Maxwell’s field equations indicates
that the electric field has wave like solutions with its amplitude varying with space
and time in a related way. The same is true of the magnetic field as a wave equation
for the magnetic field could have been derived using a similar approach to that taken
for the electric field. In fact the two fields, more precisely their time and spatial
variation, sustain one another each acting as a source for the other. The original
sources are the charge and current densities.
There are some familiar solutions to such an equation and perhaps the most
common is the plane wave solution.

a) Plane Wave Solutions.

There are two common ways of writing a so-called plane wave solution to the
equation. The first of these is the cosinusoidal notation for a plane wave travelling
in the z direction;

   z  
E ( z, t )  A0 sin2  t   A0 sinkz  t  (1.10)
  

The important parameters describing the plane wave are;


A0 = amplitude and it should be noted that this quantity is a vector.

 = wavelength which is the spatial period within which the wave repeats.
 = linear frequency (often written f ) which is the reciprocal of the temporal period
1
within which the wave repeats, ie.   where T is the period of the wave.
T
It is possible to compact the plane wave equation by specifying new wave
parameters;
2
k = wavevector , , (in this defining equation the vector character is suppressed as

it is stipulated that the wave is travelling in the z direction).
 = angular frequency, 2,
It is easily seen that in 1.10 if z is increased by m, where m is an integer, the field is
unchanged, similarly if t is increased by mT the field is unchanged. In 1.10 an

5
 Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

arbitrary choice has been made such that the field is zero when z and t are both
zero. This choice may not be possible when predicting the effect of superposition of
two fields if their peaks and troughs do not coincide and in this case a further factor
may be added to 1.10 called the phase shift,  , such that
   z   
E ( z, t )  A0 sin2  t     A0 sinkz  t   (1.10a)
   

Using 1.10 as a trial solution in the wave equation 1.8 leads to

 1 2 
 k 2 A0 sinkz  t    A0 sinkz  t  (1.11)
c2
Giving a relation between c, k and ;

c (1.12)
k


Using a cosine in place of the sine is the trivial matter of a phase difference of .
2

Note the vector character of the amplitude as it represents the electric field, itself a
vector. This is important later when discussing polarisation and is a major distinction
between electromagnetic waves and de-Broglie waves (for example) which are
scalar.
For a plane wave travelling in a general direction, r , 1.10 may be re-written;

  
  
E(r , t )  A0 sin k  r  t  (1.13)

The exponential notation for waves is also very useful and for ease of manipulation
is frequently preferred.

  

 
E(r , t )  A0 exp j k  r  t  (1.14)

6
 Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

where j   1 . The alternative notation i   1 is often used but i may be used for
currents later in the course and the j notation for imaginary numbers will be used in
these notes.
By choosing one of these notations and differentiating wrt position and wrt time gives
E
r
 

 jkA0 exp j k  r  t  (1.15)

and
E
t
  

 jA0 exp j k  r  t  (1.16)

Therefore dividing 1.16 by 1.15 the velocity is found as;

dr  1
c     (1.17)
dt k 00

as found in 1.12.
The phase,  , of a wave is often a useful property and whether using the
cosinusoidal or exponential notation the phase is the argument of the cosine or
exponential, ie.

 
  k  r  t   (1.18)

For a plane wave travelling in the z direction

  kz  t   (1.18b)

Its importance will become apparent later in the course but 1.18 or 1.18b should be
recalled. The phase is the information in the plane wave equation describing the
relative position of one wave wrt another, ie. where the peaks and troughs of one
wave are wrt another of the same wavelength. The phase shift  may be different for
each wave or of course zero.

7
 Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

In Phase 900 out of phase 1800 out of phase.

In the above diagram the phase shift of the three upper waves can be chosen as  =
0. For the waves beneath them, on the left the phase is also  = 0, in the centre  = -
90 and on the right  = 180 .
0 0

A plane wave is so called because, for example , a plane wave travelling in the z
direction has a constant wavefront represented by the xy plane. That is, the
amplitude does not depend on x or y at a particular position z and point in time t. The
xy plane in that case forms a constant phase front and at any particular point (x, y)
across the entire xy plane the waves will all have the same relative phase.
Further, a plane wave extends from - < z < + having no begininning or end! Plane
waves are therefore highly idealised and will not exist as a physical reality. However,
for the monochromatic light beam from a highly collimated laser, the plane wave will
be a superb approximation as such sources are highly directional and the
power/intensity in the beam (which will be seen to be proportional to the square of
the field amplitude) is substantially independent of distance travelled.

8
 Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

b) Spherical Waves.
Point sources of light act as a source of spherical waves where the constant phase
front is a spherical surface, centred on the point source, that expands in time at the
speed of light, the radius of curvature of the sphere being r  ct after a time t. To
describe such waves mathematically we need to use spherical co-ordinates, r, ,  in
place of cartesian co-ordinates, x, y, z.


r

 y
x

x  r sin cos 
y  r sin sin (1.19)
z  r cos 

In spherical co-ordinates

2 2 2 21   2   1     1 2
     r   sin   2
x 2 y 2 z 2 r 2 r  r  r 2 sin     r sin2   2

9
 Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

In the case of spherical waves the E and B fields have no dependence on  or  as

they are spherically symmetrical and the differentials wrt these variables are then
zero. This is of course why spherical polar co-ordinates are chosen in this particular
case because with this symmetry a simple Laplacian may be obtained.

1   2   2 2 
2  r   (1.20a)
r 2 r  r  r 2 r r

This allows the Laplacian to be written;

1 2
 2E  rE  (1.20b)
r r 2

This may be checked by double differentiating the quantity rE wrt r and showing that
1.20b gives the same operation on E as would 1.20a

This is done because then the wave equation, 1.8, takes on a particularly useful form

1 2 1  2E
rE   2 2 (1.21)
r r 2 c t

Multiplying both sides by r

2 1 2
2
rE   rE  (1.22)
r c 2 t 2

This is the simple one dimensional wave equation where the spatial variable is r and
the field is replaced by the product of field and spatial variable and this means that
the solution is simply
f (r  ct )
rE(r ,t )  f (r  ct )  E (r , t )  (1.23)
r
This represents a wave propagating outwards from a point at a velocity c. There is a
second solution representing a wave travelling inwards towards a point

10
 Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

g (r  ct )
E (r , t )  (1.24)
r
This expression diverges at r = 0 and is therefore unphysical.

The general solution would be a sum of the two

f (r  ct ) g (r  ct )
E(r , t )  C1  C2 (1.25)
r r

A special type for this solution is the Harmonic Spherical Wave solution.

cosk (r  ct )
A
E (r , t )  (1.26)
r

coskr  t 
A
E (r , t )  (1.26a)
r

exp kr  t 
A
E (r , t )  (1.26b)
r
It is noteworthy that in the case of the spherical wave above, in contrast to that
of the plane wave, 1.12, there is a dependence of the amplitude on distance
travelled and as a result the intensity of the light (power per unit area
proportional to the amplitude squared) falls as reciprocal square of the
distance from the point source. This is the well known inverse square law and
comes about as a result of the constant energy from the point source being shared

out over an increasing area, 4r 2 , as the wave travels outward.

Other types of solution are possible but will not interest us further.

11
 Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

Maxwell’s equations in simple media.

Maxwell’s Equations
So far attention has been given to the simple problem of electromagnetic waves in
free space ie vacuum. Most of this course will be concerned with electromagnetic
waves in media. Simple media such as glasses and liquids will now be considered
and they are simple in these terms when they are;

(i) Homogeneous; the same everywhere under translation i.e. the wave
velocity, wavevectors and frequency which must be defined from place to place
remain unchanged and

(ii) Isotropic; the same in all directions, i.e. under rotation. This means that
the field vectors which have directionality by virtue of their vector nature are not
affected by the medium except in the simplest way or put in another way the
polarisation of the wave is unaffected by the medium.

The modification to what has already been done is straightforward. Free space is
homogeneous and isotropic so the new thing here is that the electric and magnetic
fields may interact with the medium and its charges. Simple scalars may be
attributed to the medium in order to account for its electric and magnetic properties.
The simplest way to do this is to note that the action of the electric and magnetic
fields associated with a light wave on a medium is to generate an electric or
magnetic polarisation in that medium. That is, when an electric field, such as that
associated with the electromagnetic wave, is present there may be a separation of
positive and negative charge in the medium leading to a polarisation, P per unit
volume and P, in homogeneous and isotropic media is related to the electric field by

 
P  0 E E (1.27)

where E is the electronic susceptibility

12
 Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

Polarisation per unit volume and electric field are also related to the electric
displacement vector D sometimes used in place of the electric field which is defined
as follows;
   
D   0E   0E  P (1.28)

 is the relative permittivity also known as the dielectric constant of a medium and
thus
  1  E (1.29)

D is a useful way of representing the electric field in the presence of a medium.

In an analogous manner the magnetic field H is often used when dealing with
magnetic media in the place of the magnetic induction, B.
Analagous to the electric field there will be a magnetisation induced in magnetic
materials where the magnetic field of the light wave interacts with the spins of the
atoms of the medium causing them to aligm and thus developing a magnetisation
per unit volume, M.


 
B  0 H  M  (1.30)

and
 
M  M H (1.31)

where M is the magnetic susceptibility.

Therefore in a medium
  
B  0 1  M H  0H (1.32)

Where  is the relative magnetic permeability of the medium and

  1  M  (1.33)
thus

 B
H (1.34)
0
The above analysis means the each medium has a value for the electric permittivity
and for the magnetic permeability that differs from the values given above for
vacuum. The permittivity which in free space was  0 is now  0 and the permeability

13
 Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

in free space  0 becomes 0 within a medium. Non magnetic media have   1

and so the permeability remains unchanged, that is, there is no induced

magnetisation per unit volume and  M  0 .

The action of electric and magnetic fields on media may be

summarised as follows;
Electric Field Magnetic Field
The effect on the medium is ;
Polarisation per unit volume Magnetisation per unit volume
   
P  0 E E M  M H

We may introduce new field descriptions within the medium

Electric Displacement Magnetic Field Induction
  
D   0E  P


 
B  0 H  M 

Using the above

   
D   0E 1   E  B   0H 1  M 

And a new material property is defined using the above

Permittivity (Dielectric constant) Permeability
   
D   0E B  0H

  1 E    1 M 
Thus, the Maxwell equations and everything that flows from them in the
preceding section may be modified by replacing 0 with 0 and by replacing 0
with 0 for example when dealing with the wave equation in the next section.
In summary Maxwell’s equations, 1.1b – 1.4b become:
 
E  (1.1c)
 0

B  0 (1.2c)

 dB
E   (1.3c)
dt

  dE
  B  0 J  0  0 (1.4c)
dt

14
 Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

The newly defined D and H fields may be used to write the Maxwell equations in an
even more compact form recalling their definition.

   B
D   0E and H
0


D   (1.1d)

B  0 (1.2d)

 dB
E   (1.3d)
dt

   dD 
  H   J  
 (1.4d)
 dt 

In the case of inhomogenous or anisotropic media such as crystals the situation

regarding  (or equivalently  or n ) is more complex and it is necessary to modify the
dielectric constant (or equivalently the refractive index) to account for new
phenomena arising from inhomogeneity or anisotropy.

(i) This will lead to the use of a tensor to describe the dielectric
constant/refractive index when discussing birefringence and related phenomena.

(ii) Indeed, where the medium is absorbing or conducting it is the case that the
dielectric constant will be complex,     j , and the imaginary part related to the

loss or conductivity,  , by    0  .

These complexities will be dealt with when they arise. For now the simple isotropic,
homogenous medium approximation will be sufficient. Generally, apart from
magnetic materials it is the case that   1. In particular, glasses, which are of great
interest in optics, are generally non magnetic.

15
 Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

Electromagnetic Wave Equation in Simple Media

To obtain the wave equation in a simple medium it is sufficient to replace  0 with

0 and  0 with  0 . The wave equation in a simple medium now becomes

 
  2
E 1  2
E
 2E   0 0  (1.8a)
t 2 v 2 t 2

with a wave velocity in the medium of

1 c c
v   (1.9a)
 0  0  n

Here n is the refractive index and in non-magnetic media where   1

n     (1.35)

NB.  (the relative permittivity) will vary with frequency but we ignore this for now
and note that the relative permittivity and dielectric constant of a medium are one
and the same.
c 
It is worth noting here that v   in a medium of refractive index, n, and if an
n k
electromagnetic wave passes from free space into this medium, that is from a region
 c 
where v  c  to a region (the medium) where v   this can either involve 
k n k

changing to or k changing to nk. It is the latter that changes with k becoming nk or
n

 becoming . The frequency of the wave does not alter. Generally in these notes
n
the wavevector and wavelength in free space is denoted as k0 and 0 respectively
and in a medium of refractive index, n, there is a new wavevector and wavelength
given by

k  nk 0 and   0
n

16
 Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

The Relationship Between E and H in the Electromagnetic Field.

The plane wave solution may be analysed further by assuming that in free space (or
in a simple medium ) the direction of the field vectors do not change as the wave
propagates and also by assuming that B is described by the same plane wave
(wavevector, angular frequency) as is E. Taking the E field to be pointing in the x
direction, ie Ex is the only component that exists and with the plane wave
propagating in the z direction (as usual).
Using equation 1.3a
To form the vector product   we may use the mnemonic

   
y z x y

Ey Ez Ex Ey

dB
E  
dt

 dE z dE y   dE x dE z   dE y dE x  dB dB dB
  ,   ,      x , y , z
 dy dz   dz dx   dx dy  dt dt dt

There are only two components of the B field, y and z, as the above does not allow
an x component.

dE x dBy
 jkEx     jBy (1.36)
dz dt

Recalling that for a plane wave propagating in the z direction the amplitude does not
vary with x or y the above simplifies to

dE x dB
 0   z  jBz (1.37)
dy dt

17
 Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

The B field is at right angles to the E field, pointing in the y direction and both E and
B are at right angles to the direction of propagation.
Further it can be seen that
Ex  1
 c  (138)
By k 00

This is a very important relation between E and B.

The magnetic induction, B is sometimes dropped and H  B is used instead.

0
In this case
Ex 0
  0  377 (1.39)
Hy 0

where 0 is the impedance of free space.

The impedance of a simple medium is then defined by analogy

0
 (1.40)
 0

Noting that for most media which are non-magnetic  = 1, 1.40 becomes

 
 0  0 (1.41)
 n
and in a medium the relationship between E and H is

E = H (1.42)

Power Flow in the EM Wave & the Poynting Vector

To confer a physical reality upon the electromagnetic fields represented by these
solutions to the electromagnetic wave equation it is necessary to recognise that the
electric and magnetic fields have an energy associated with them.
In earlier electromagnetism courses where the energy in the electric field between
two parallel plates separated by a distance d and with a potential difference V was
found as;

18
 Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

1
U QV (1.43)
2
This is seen by noting that the charging of the capacitor involved the movement of
the charge Q through a potential difference V where Q is now the charge stored on
the capacitor and is given by,

 A
Q  CV  0 V (1.44)
d
Where a parallel plate capacitor is assumed with vacuum between two plates of area
A separated by a distance d

1 0 A 2
U V (1.45)
2 d

V
This energy is stored in a volume V  Ad by an electric field E  therefore the
d
energy density associated with the electric field, u E , is

U 1
uE    0E 2 (1.46)
Ad 2

The same energy density may be assumed to be associated with any electric field in
free space.
Likewise for the magnetic field, we compute the energy associated with a current
carrying toroid

1 B2
uB  (1.47)
2 0

The electric and magnetic field energies exchange between one another as the
electromagnetic field propagates and their maximum values are therefore equal

u E max  u B max (1.48)

and the electromagnetic field energy density, u, is simply the sum of the two

19
 Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

1  B 2  B2
u  uE  uB  0E 2   0E 2  (1.49)
2   0  0

It is required to know the amount of energy per unit time (power) crossing unit area,
Wm-2 in order to find the intensity of a light wave. To obtain this value consider a
plane wave travelling in a direction perpendicular to an area A. In a time t a
paralleliped of length ct and of volume ctA containing an energy U  uctA
passes through the area A. Therefore the energy per unit time per unit area or power
P
per unit area, flowing across this area is;
A

P U  E2 0 2 E 2
S   uc  c 0 E 2  0  E  (1.50)
A tA 00 0 0

Where the previously discovered relation between the velocity of light in free space,

1 0
c and the already defined  0  377 , electromagnetic impedance
00 0

of free space was used. Or using the magnetic field

B2 0 1 2 1 2 E2
S  uc  c  B  0 B  (1.51)
0 0  2  2 0
0 0

1
S EB (1.52)
0
Making the reasonable assumption that the power is propagating in the same
direction as the wave this may be written in vector notation
 1  
S E B (1.53)
0

And S or S is called the Poynting vector that shows the power per unit area or
intensity in the electromagnetic field as it flows in a direction perpendicular to both E
and B as defined by their cross product.
NB. Right hand rule: Index finger E, middle finger B and thumb S.

20
 Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

Finally it is necessary to note that any instrument (including our eyes) that
measure/sense light intensity actually respond to the time averaged electric field
squared and in this case the relation between light intensity (as measured) and the
electric field is strictly given by

E2 E 02
I  (1.54)
0 20

in free space where the triangular brackets indicate a time average which gives the 2
in the denominator for a sin or cosine time variation,
0
and in a medium of refractive index n or impedance  
 0

by

E 02 nE02
I  (1.55)
2 20

21
Electromagnetic Waves & Optics: Lecture Notes © Kevin Donovan

Highlights of Part 1

The Electromagnetic Wave Equation

 2E 1  2E
 2E   0  0 
t 2 c 2 t 2

The Velocity of Light

1
c  2.9979  108 ms1
00

The Phase Velocity in a Medium

1 c c
v  
 0  0  n

The Refractive Index n    

Ex 0
The Impedance of Free Space   0  377
Hy 0

Ex 0
The Impedance of a Medium  
Hy  0

Ex 0 
For a Non Magnetic Medium   0
Hy  0 n

Ex
Relation between E and B in free space c
By

Ex c
Relation between E and B in a medium v 
By n

E 02
Relation between E and Intensity in free space I
20

E 02 nE02
Relation between E and Intensity in a medium I 
2 20

22