Electromagnetism

Mehdi Bouguerra Ezzina

2022 - 2023

Contents
1 Introduction 2

2 Electric field 4

3 Gauss’s Law or first Maxwell’s equation 7

4 Electric field and electric potential 9

5 Magnetic force 12

6 Second Maxwell’s equation 15

7 Third Maxwell’s equation 18

8 Amphère’s law 20

9 Fourth Maxwell’s equation 20

10 Electromagnetic waves 21

11 Emission of electromagnetic waves 23

12 Circuits 26

13 Interference and diffraction 29

Summary for static charges 31

Summary for electric and magnetic dipoles 32

Summary for Maxwell’s equations 33

Summary for emission of EM wave 34

Summary for interference and diffraction 35

1
1 Introduction
• Two kinds of electric charges exist: positive (+) and negative (–)
• The electron and a proton are the most common particles carrying a unit of charge. Electrons
are negatively charged and protons are positively charged.
A proton is a composite particle: consists of 2 up and one down quarks with charges of + 23 e
(twice) and − 31 e, but these do not exist unbound.
There also exist neutral particles, e.g. neutron.
• Atoms are electrically neutral and composed of a charged nucleus containing A nucleons (protons
and neutrons) out of which Z protons with a total charge +Ze, and Z electrons.
• No matter what you do, if you transfer charge from one body to another, the total net charge
remains the same. This is known as charge conservation.
• Experiments show that electric charge q always exists only in integer multiples of an elementary
charge e:
q = ne, n ∈ N and e = 1.602 × 10−19 C
• Materials are commonly classified based on the ability of electrons to flow through them. In
conductors they can move relatively freely and insulators keeps the ones deposited at one
location there.
• the Earth is an infinite source of electrons: if there is a pathway between an object and the
Earth, this object is electrically grounded – there is always a supply of electrons to neutralize
a charge.
• Charging by contact:

electrons flow from the copper to neutralize positive charges at the point of contact with the
glass the resulting charge on the copper when the glass is removed is positive.

• Charging by induction:

2
 electrons flow from ground to neutralize the positive charges on the far end of the copper rod
the resulting charge on the copper when the glass is removed is negative.
• A point charge is a hypothetical charge located at a single point in space.
(sr. 1 - ex. 1)
• The charge density is the amount of electric charge per unit length, surface area, or volume:

– line charge density λ: used when only one coordinate is important (wires);

dq = λ dl
Z Z
Q= λ dl = λ dx
L x

– surface charge density σ: used when two coordinates are important (plane, disc);

dq = σ dS
Z Z Z
Q= σ dS = σ dy dx
S x y

– volume charge density ρ: used for charged volumes and when no details are specified (general
laws);
dq = ρ dV
Z Z Z Z
Q= ρ dV = ρ dz dy dx
V x y z

(sr. 1 - ex. 2)
• Coulomb law: The electric force F12 between two resting charges q1 , q2 depend only on their
values and the distance r12 = |r1 − r2 | between them.
q1 q2 q1 q2 r1 − r2
F12 = kE 2 r̂12 = kE 2
r12 |r1 − r2 | |r1 − r2 |

– The magnitude of the force is proportional to the magnitude of both charges: F12 ∝ q1 q2 .
Same-sign charges repel, opposite-sign charges attract.
– The magnitude of the force is inversely proportional to the square of their separation:
F12 ∝ r12 .
12

(sr. 1 - ex. 3 and 4)

• If we have a charge Q that interacts with n charges qi , the total force is just a vector sum of all
Coulomb forces:
F = F1 + F2 + + Fn

This is known as the superposition principle.

(sr. 1 - ex. 5)

3
2 Electric field
• An electric field is a function E(r) that specifies at any point in space a vector E that would
act on a unit charge at this point – the force F = QE(r) is acting on any charge Q in this point.
The same field at the point can be caused by very different charge configurations.
• An electric field associated with a single charge q can simply be deduce from Coulomb’s law:
q r − rq
E(r) = kE 2
|r − rq | |r − rq |

• The superposition principle applies also to E (r):

E(r) = E1 (r) + E2 (r) + ... + En (r)

To determine E (r) one measures F (r) on the probe charge Q. Since the charge does not act on
itself, its field does not contribute to the measured E (r).
• For arbitrary continuous time-independent charge distribution ρ(r) this gives:
X qi · (r − ri ) ρ(r′ ) · (r − r′ ) 3 ′
Z
E (r) = kE → k E d r
i
|r − ri |3 |r − r′ |3
Z
where qi = ρ(ri )∆Vi , ∆Vi → 0, ρ(ri ) d3 r′ = qtotal

(a) discrete (b) continuous

Figure 1: different time-independent charge distribution cases

Example: Electric field calculation

We have an uniformly charged rod, with a total charge q
and charge density λ = Lq . Let’s find the field at point
P at distance y from the rod along its perpendicular bisec-
tor.

There is no field component perpendicular to the shown

plane at point P . Field component in the z direction is
equal to zero too, as all the elementary charges cancel each
other.

So we just need to just sum up dEy components from all the

elementary charges dq = λ dz and get an answer!

4
 1 λ dz
dEy = dE cos θ = cos θ
4πϵ0 y + z 2
2

y y
cos θ = =p
r y + z2
2

Z L/2
1 λy dz 1 λL
Ey = =
4πϵ0 (y 2 + z 2 )3/2
p
−L/2 4πϵ0 y 2 + L2 /4
Analysis of the result:
1 q
limit y → ∞: the expression would look like a field of a point charge q: dEy = 4πϵ0 y 2 .
limit L ≫ y: we get a field of an infinitely long line of charge: dEy = 2πϵλ0 y .
• Electric field lines fulfill the followings properties:
– The tangent to the electric field line passing through any point in space gives the direction
of the electric field at that point.
– The electric field lines start on positive charges and end on negative charges.
– The magnitude of the electric field at any point is proportional to the number of field lines
per unit cross-sectional area perpendicular to the lines
(sr. 2 - ex. 2)
• The electric field inside a conductor is zero in the static situation: otherwise the free
electrons would move and arrange so that there is no field and hence no force acting on them.
– Consequence 1: any net charge on a conductor distributes itself on the surface.

– Consequence 2: the electric field is always perpendicular to the surface outside of a

conductor.

In the static case: E| | = 0 and E = E⊥ .

5
• A dipole consists of equal but opposite charges separated by a distance l. The dipole moment
p = Ql and points from the negative to the positive charge.

• Torque acting on a dipole in a uniform electric field:

l l
τ = QE sin θ + QE sin θ = pE sin θ
2 2
τ =p×E

Figure 2: An electric dipole in a uniform electric field

It tries to align a dipole with the electric field:

– If the dipole is parallel to E then the equilibrium is stable.
– If antiparallel, it’s unstable.

• Force acting on a dipole in a non-uniform electric field:

∂E ∂E ∂E
F = QE+ − QE− = Q(E+ − E− ) = Q(lx + ly + lz ) = p · ∇E
∂x ∂y ∂z
It tries to suck the dipole into the stronger electric field.
• On perpendicular bisector at the large distance r ≫ l:
1 p
E=
4πϵ0 r3
The field decreases more rapidly for a dipole than for a single point charge ( r13 vs 1
r 2 ).
(sr. 2 - ex. 5)

6
3 Gauss’s Law or first Maxwell’s equation
• The Coulomb’s force implies action at a distance, i.e. force changes immediately at any
distance once one charge is displaced. But in reality this is not the case, there is a delay that
exist between the moment the charge is displaced before the moment where the other charges
notice it. Therefore Coulomb’s law works in static situation but not when the charges are moving.
We can rewrite our equation in a different form using field E to evade this problem, but to deal
with action at a distance, we need this equation to be local.

• We define flux or electric field through a surface such as:

Z Z
Φ= E(r) · dS = E(r) · dS · cos α
surf ace surf ace

Where α is the angle between E and the normal vector n to dS at each point of the surface (thus
dS is the vector of magnitude dS and direction perpendicular to the area).

kE q
Using Coulomb’s law E = r 3 r,
and we obtain:

dS · cos α
Z Z
Φ = kE q = kE q dΩ
S r2 S

For a charge inside the surface, the full integral is just the full solid angle at which we see this
surface (it will always be equal at 4π).
Flux does not depend on the form of the surface.

• Let’s consider two regions V1 and V2 and a charge Q inside V1 and outside V2 . The electric field
lines go out V1 and they pierces right through V2 (they go in before going out). Therefore the
flux for V1 will be positive and 0 for V2 (positive and negative solid angle cancel each other in
the second case).

7
• For all the practical purposes, flux Φ is:
(
q∈V
Z
4πkE q
E · dS =
S 0 q∈
/V

Using the principle of superposition, we get Gauss’s law in integral form:

Z Z Z
X 1
E · dS = 4πkE q = 4πkE ρ(r) dV ≡ ρ(r) dV
S V ϵ0 V
q∈V

Flux of the electric field over any closed surface is ”equal” to the enclosed charge.
Key to solve the exercise:
R find the surface for which E is a constant (i.e. does not depend on the
coordinates). There by S E · dS = E · S .

• Using the mathematical Gauss’s theorem on the last result we obtain the first Maxwell’s
equation:
ρ
∇ · E = 4πkE ρ ≡
ϵ0
Divergence of the electric field at any point is ”equal” to the charge density at this
point.
Because of multiple solutions, Gauss’s law is insufficient to find the field of a moving charge. We
need other equations.

8
4 Electric field and electric potential
• When a charge Q is moved in an elecctric field, the electric force does work W = Curve F · dl.
R

If the field is created by many charges, the total work is a sum of works done by each charge
separately due to the superposition principle.
• The work done by Q moving near a point charge q is:
(
1
− r1f radialmovement
W = kE Qqi rin
0 circularmovement

The work depends only on the initial and final position (rin and rf ) but not on the
form of the curve!

• The electrostatic force created by a charge (Coulomb’s force) is conservative. We can define
the potential energy U for the electrostatic force.
• The change in potential energy between two points equals the negative of the work done by the
conservative force as an object moves from one to the other:

∆U ≡ Uf − Uin = −W

• The total mechanical work done by many charges:

Qq1 Qq1 Qq2 Qq2
Wtot = kE − kE f + kE in − kE f + ... ≡ Uin − Uf
r1in r1 r2 r2
| {z } | {z }
W1 W2

The work done by the system leads to the change of energy.

• The potential energy of a point charge Q near other point charges is:
Qq1 Qq2
U (r) = kE + kE + ... ≡ Q · ϕ(r)
|r − r1 | |r − r2 |

with ϕ the electrostatic potential.

• The electrostatic potential (or just potential) of a point charge q at distance r is:
q q
ϕ(r) = kE ≡
|r| 4πϵ0 |r|

Only differences in potential energy are meaningful, hence only this difference is measurable.
• For arbitrary electrostatic field configuration, it can be obtain as the negative gradient of the
potential:
E = −∇ϕ

It’s not the case for any arbitrary field !

Rr
If we know E we can use: ϕ(r) = − ref E · dr with ref is where ϕ = 0 (often at ±∞)

• Electric field inside a conductor is zero in the static situation (when the charges are at rest).
Any net charge on a conductor distributes itself on the surface.
Static electric fields are always perpendicular to the surface outside of a conductor.
R
As there is no field inside a conductor, for any point in it E dl = 0, and ϕ = const. The whole
conductor is at the same potential and is called an equipotential volume.

9
• For an arbitrary field, the total work over a trajectory depends on it’s form, not on the endpoints
only.
Consider a field configuration in which E(r) is directed along circle with a constant magnitude
depending on the radius r:

For a radial trajectory, AB, displacement is always perpendicular to the field. Work is zero.
Z
WAB = Q E · dl1 = 0
AB

However, for a circular arc BC the work is:

Z
WBC = Q E(r) · rdθ = QE(r) · rθBC
BC

For a full and closed circle the work is non-zero:

Wcircle = Q · 2πr · E(r)

If for some ϕ the electric field is E = −∇ϕ, we would have for a closed trajectory:
Z
W = Q ∇ϕ · dl = Q(ϕB − ϕB ) = 0

This means that E ̸= −∇ϕ.

We call this type of field, ”tornado” field.

• The main reason why we can introduce ϕ for Coulomb field (field from static charges, ”sunny”
fields) is that the electric field will always be radial E || r.
Since there is no ”rotational” component of E, there is no work over a circle.
For an arbitrary closed loop, the parts with E ↑↑ dl (E · dl > 0) and E ↑↓ dl (E · dl < 0) cancel
each other.

• To understand if there is a ”rotational” component for an arbitrary field E we ask ourselves

about its circulation.
• To quantify circulation for arbitrary field E one can use the curl operator.

– For a vector field on a plane:

∂Ey ∂Ex
∇×E≡ −
∂x ∂y

10
 – In 3D, there is three rotation axes, and therefore the curl operator produces a vector:
∂Ez ∂Ey
(∇ × E)x = −
∂y ∂z
∂Ex ∂Ez
(∇ × E)y = −
∂z ∂x
∂Ey ∂Ex
(∇ × E)z = −
∂x ∂y
• For any field E that has no curl:
∇×E=0
and there exist a scalar function f such that:

E = ∇f

the electrostatic potential ϕ = −f .

• Stokes theorem:
Let’s consider an arbitrary closed curve, and split it into many small squares.

Circulation of the squares inside cancel each other, because E · dl for each internal side enters
twice with opposite directions and, on the other hand, the external sides together duplicate the
curve, therefore: Z Z
E · dl = ∇ × E dS
C
with S being the area inside C, dS = dx dy.

11
5 Magnetic force
• A capacitor is a device used to store an electric charge, consisting of one or more pairs of
conductors separated by an insulator.
• A current is a flow of charge, i.e. charged particles.
• In wires the total current I is the amount of charge passing thought the wire’s cross section in a
unit of time:
dQ
I=
dt
In general, current is describe by the current density j, equal to the charge passing through unit
area S || j in a unit of time. If a charged particle with charge density ρ move with velocity v,
current density is:
j=ρ·v
Thus, the current going though a wire with crosse-section S is then:

I =j·S

• The continuity equation (charge conservation law):

∂ρ
+∇·j=0
∂t

• Two long parallel conductors (for example wires) with currents I1 and I2 at distance r attract
(repel) each other if the current flow in the same (opposite) direction with the force per unit
length:
F I1 I2
= 2kM − Ampère′ s force law
L r
kM is a constant similar to kE in Coulomb’s law.

• Force between two small elements of wires is an inverse square law F ∝ 1

r2 , but it depends in a
rather non-trivial way on their directions.
Let’s consider two wires at fixed place r1 and r2 . The maximal force between them is when they
are parallel to each other and perpendicular to the relative vector between them r = r1 − r2 .

12
 The force decreases if we rotate one of the wires to the vector r by angle β. Moreover, it also
decreases if we rotate the second vector in the plane perpendicular to r by angle α.

F (α, β) = Fmax × sin α × sin β

• Biot-Savart law: Two infinitesimally short wires dl1 and dl2 with currents I1 and I2 going
through them interact with the force:
  
I1 I2 r1 − r2
dF = kM 2 dl1 × dl2 ×
r |r1 − r2 |

Force acting on the first current is directed perpendicularly to both dl1 and the vector dl2 × r.
r = r1 − r2
• If several currents interact, we add the corresponding forces:
   
I1 dl1 × (r0 − r1 ) In dln × (r0 − rn )
dF0 = kM I0 dl0 × + · · · + kM I0 dl0 ×
|r0 − r1 |3 |r0 − rn |3

• We can define the magnetic field B(r) as a force acting on a ”unit current” at a given point:

Ii dli × (r − ri )
dF0 = I0 dl0 × B(r0 ), dBi (r) = kM
|r − ri |3

and the total magnetic field is:

B = B1 + · · · + Bn − superposition principle

The magnetic force is always perpendicular to both the current direction and the
magnetic field.
Use the right hand rule by pointing your hand in the direction of the conventional current I, bend
them in the direction of the magnetic field lines B and your thumb will point in the direction of
the force F.
Similarly you can grasp the wire with your right hand so that your thumb points in the direction
of the conventional current I, then your fingers will encircle the wire in the direction of the
magnetic field B.
• Currents in conductors are basically flows of negatively charged electrons I = ev · nS (e the
electric charge, v the velocity, n the number density and S the cross-section). Therefore, one can
assume that moving charges (electrons) are the source of magnetic field.

13
• A magnetic field B created by a moving charge q1 is:
q1 v1 × r
B = kM
r3
and other charges interact with them via:

F = q2 v2 × B(r)

For continuous current density distribution j this gives:

j(r′ ) × (r − r′ )
Z
B(r) = kM dV
|r − r′ |3

• In conclusion:

– When two charges are at rest, they interact via electric force, according to Coulomb’s law:
q1 q2
F = kE r
r3
– If the charges move, in addition to the electic force, they experience also a magnetic force
(called Lorentz force), according to Ampère’s law with j = qv:

q1 q2 v1 × [v2 × r]
F = kE r + kM q1 q2
r3 r3
If a charged particle moves with velocity v in an environment where both electric and
magnetic fields are present:
F = q(E + v × B)

14
6 Second Maxwell’s equation
• We have seen with Coulomb’s law that will get a problem with an action at a distance. We will
get the same problem if we write down interaction of current at some distance from each other.
• We define the magnetic dipole moment µ of the loop made by the current as:

µ = IA

If a loop has N turns then:

µ = N IA
with A the surface of the loop.

Right hand rule: If the fingers of your right hand are in the direction of the current, then your
thumb indicates the direction of µ.
• In a external magnetic field, a magnetic dipole moment experience a torque τ that rotates µ into
alignment with B:
τ =µ×B

• The potential energy of a magnetic dipole in a magnetic field can be computed from a work done
to change the orientation of a magnetic dipole in a magnetic field:

U = −µ · B

15
• The magnetic force (like all the force that depend on velocity), is in general not conservative and
therefore cannot be represented by a potential energy.
But in the special case in which the torque on a dipole depends on its position relative to the
field, it is possible to define a potential energy for the system consisting of the dipole in the field.
The potential energy is not characteristic of the field but of the dipole in the field ! In general we
cannot define a scalar ”magnetic potential energy” of a point of charge or ”magnetic potential”
of the magnetic field it self.
• The field of a dipole:

• When a electric dipole is cut in half, the positive and the negative charges are isolated in one
piece each.
When a magnetic dipole is cut in half, a new pair of north and south poles appears.

• Second Maxwell’s equation:

∇·B=0
This equation holds for moving charges in any system (and not only for static currents).
Due to the superposition principle, this holds for any field created by many currents.
Using mathematical Gauss’s theorem, we can get back to the integral form:
Z Z
∇ · B dV = B dS = 0
V ∂V

There is no ”magnetic charge”.

16
• Source of field:

• Bio-Savart law satisfies the second Maxwell’s equation, but we can find other solutions that in
addition respect action at the distance.
We can find a new vector A, called vector potential, such that:

B≡∇×A

i.e. such that the equation is satisfied:

∇x ∇y ∇z
∇ · [∇ × A] = ∇x ∇y ∇z ≡ 0
Ax Ay Az

For any well-behaved A (such that ∇x ∇y Ai = ∇y ∇x Ai , · · · ) we immediately recover the desired

relation ∇ · B = 0.

17
7 Third Maxwell’s equation
• Until now, we have derived the two first Maxwell’s equations. Coulomb and Bio-Savart laws
satisfy these and at the same time there exist other solutions that respect action at a distance.
But these equation do not have time derivatives. If we assume they are always correct, how do
we describe fields that change in time ?
• A battery (or any other device creating a potential difference V ) is called a source of electro-
motive force or emf. It makes charges moves, i.e. it induces a current I.
The term ”electromotive force” does not refer to a force measured in Newtons.

Current in a circuit is proportional to the potential difference:

V = IR − Ohm′ s law
where R [ohms] is the resistance of a wire device, V [volts] is the potential difference applied
across the wire or device and I [ampheres] is the current through it.
The potential difference between the terminals of a source, when no current flows to an external
circuit, is called the emf ε of the source (measured in volts).
• Electromagnetic induction:
Consider an electric circuit with no current sources. If we put a magnet near the wire, the circuit
won’t be affected since the current is 0.
However, if we move the magnet, current starts to circulate inside the circuit !

The current orientation is such that the magnetic field of the current compensates the change of
magnetic field of the magnet that passes through the loop. - Lenz’s law

18
• The current cannot be induced by a magnetic force because the charges inside the wire are at
rest. To produce current, we need an electric force. In other word, change in the magnetic field
must create an electric field inside the wires, which then act on the electrons in the wire.
• Faraday’s law:
R
The emf ε = E · dl around
R a closed path is equal to the negative of the time rate of change of
the magnetic flux Φ ≡ B · dS enclosed by the path:

dΦ
ε=−
dt

• Faraday’s law of induction:

The change of the magnetic flux through a surface induces the circulation of an electric field
along its boundary (i.e. in the circuit):
Z Z
d
E dl = − B dS
dt Σ

Let’s consider a surface Σ with boundary ∂Σ in R3 , and a vector field A = (P, R, Q). By
Stokes’s theorem: Z Z
[∇ × A] · dS = A · dS
Σ ∂Σ
It says that the circulation of A along ∂Σ is equal to the flux of the curl of A through Σ.
• Third Maxwell’s equation:
So that Faraday’s law can hold for any surface, we need:
∂B
∇×E=−
∂t
The electric field is no longer conservative in presence of changing magnetic field !
The work over a closed loop is not equal to 0.

19
8 Amphère’s law
• To find the last Maxwell’s equation, we need the magnetic conterpart of Gauss’s law - Ampère’s
circuital law:
Consider aRclosed curve ∂Σ that encloses a surface Σ. The circulation of magnetic fiels along the
boundary ∂Σ B · dl is proportional to the current through Σ:
Z Z
B · dl = 4πkM j · dS ≡ µ0 Iencl
∂Σ Σ

with I the current through the surface.

This is equivalent to Bio-Savart’s law.

• Application tip’s:
– Ampère’s law, like Gauss’s law, is always a valid statement (in absence of varying fields).
But as a calculation tool it is limited mainly to systems with a high degree of symmetry.
The first step in applying Ampère’s law is to identify useful symmetry.
– Choose an integration path reflects the symmetry. Search for path where B has con-
stant magnitude along the entire path or along segments of the path. Make sure your
integration path passes through the point where you wish to evaluate the magnetic field.
– Use symmetry to determine the direction of B along the integration path. With a smart
choise of path, B will be either parallel or perpendicular to the path.
– Determine the enclosed current Iencl . Be careful with signs. Let the fingers of your right
hand curl along the direction of B so that your thumb shows the direction of positive current.
If you have a solid constructor and your integration path does not enclose the full current,
you can use the current density (current per unit area) multiplied by the enclosed area.

9 Fourth Maxwell’s equation

• The displacement current is the rate of change of the electric induction (- displacement field):
1 ∂ΦE
Idis = j · S ≡
4πkE dt
the displacement current density is define such as :
1 ∂E
jdis ≡
4πkE dt

• Fourth Maxwell’s equation:

 
1 ∂E kM ∂E
∇ × B = 4πkM j+ ≡ 4πkM j +
4πkE dt kE dt

The equation implies that even in the absence of matter currents j, variable electric fiels will
cause magnetic field. The creation of magnetic field without currents becomes possible.

20
10 Electromagnetic waves
• Remember that if there are charges far from us, we cannot know directly what happens to them.
We can only measure fields here and now.
Thus if a charge moves, the field that she creates will change from Eold to Enew . But for some
time we would not know this and still would be measuring Eold . So how does the field change
propagate to us ?
Moreover, Faraday’s law of induction and displacement current alone suggest that the field exists
even if there are no charges (ρ = 0) or current (j = 0): variable E-field creates B-field → variable
B-field creates E-field → · · ·
This leads to the main prediction of Maxwell’s theory: Electromagnetic field can live and
propagate in vacuum!
• We define the fundamental velocity:
r
kE 1
c= =√
kM ε0 µ0

• In vaccum, E and B satisfy the wave equation:

 2 
∂
−∆ f =0
c2 ∂t2
where f is some function.
• Any well-behaved function can be represented as a sum of harmonic solutions:
Z ∞
f (r) = Ak cos(k · r + ϕk ) dk
−∞

Therefore, to study the general solution, we can consider only these.

k is called the ”wave vector” and point in the direction of propagation.
• Characteristics of waves:
In a harmonic solution f = A cos(kx ± ωt + ϕ0 )

– A is the amplitude of the wave.

2π
– The solution cos(kx − ϕ) is a periodic funtion with period λ = k − wavelength.
2π
– For the time dependance, the period is T = ω .
– For oscillators with period T, the quantity ω is the angular frequency.
– ω is related to usual frequency ν as ω = 2πν.
2πc
– For EM waves: ω = k · c ≡ λ (and λ = cT ≡ νc )

• An electromagnetic wave is a ”frozen” field configuration E

B that moves as a whole in space
with velocity c.
• Properties of electromagnetic waves:

– An electromagnetic wave is a coupled wave of E and B changing in time: E produces B,

which in return produces E and so on · · ·
– Electromagnetic waves are transverse, i.e. E ⊥ B ⊥ k.
k, E, B is a right handed system: k or v is directed along [E × B].

21
 The electric field E and the magnetic field B in a harmonic wave are perpendicular to the
direction of propagation.
– The amplitudes of the fields are equal: E = cB.
– Each wave (with fixed k) has only two degrees of freedom. Two independent component
of E.

• Apart from wavelength, EM waves are parametrized by their polarization.

An electric field has two independent degrees of freedom Ax , Ay (let k = kez ). They can have
different phases (because they are independent). Depending on them, we can observe different
pattern of propagation.

– If the phases of the two Ax and Ay waves are equal,qthe total E oscillates in a fixed plane
defined by Emax = (Ax , Ay , 0) with amplitude A = A2x + A2y . This is line polarization.

– If the amplitude are equal Ax = Ay and the phase difference is ϕx − ϕy = ± π2 , we would

see that E rotates in the xy plane clockwise (right handed polarization, if + π2 ) or counter-
clockwise (left handed, if − π2 ). This is circular polarization.

– In general we observe elliptic polarization, with axis lengths and directions defined by
Ex and Ey magnitudes and phases.

• Electromagnetic wave can interact with charged particles and, therefore, transfer energy and
momentum. This implies that electromagnetic waves carry energy and momentum by themselves.
Z 2
E + c2 B2
EEM f ield = dV − energy of EM f ield
8πkE
E×B
P= (Poynting vector) − energy f lux of EM f ield
4πkM outward of V

The poynting vector Ris also the momentum density of the EM wave. If you absorb the wave, you
transfer momentum P dV to your system.

22
11 Emission of electromagnetic waves
• When electromagnetic waves are emitted ?
If two particle are at rest, the electric field is static thus there is no EM waves. We need time
varying fields, i.e. moving charges.
But an uniformly moving charge has its own inertial frame where she is at rest. Although the
field change in time, the charge and the field move together as a whole system with velocity v.
There is no action at a distance because everything moves together. Thus we still don’t have
emission of EM waves.
To emit EM waves, the charges has to move with non zero acceleration.
• Moving charges Q ↔ an electric current I:

dQ d(ρ · S · l) dl
I= = = ρS = ρSv
dt dt dt
where ρ is the charge density, S is the wire cross section and v the charge velocity.
accelerating charges Q ↔ an alternating electric current I:

dI d(ρSv) dv
= = ρS = ρSa ̸= 0
dt dt dt
where a is the charge acceleration.
• An inductor is an electrical component, which stores energy in a form of magnetic field.

– The current I creates magnetic field, localized inside the inductor:

N
Bl = µ0 N I ⇒ B = µ0 I = µ0 In
l
where N is the number of wire turns and n the linear density of wire turns.
– The total magnetic flux though the solenoid is then:

ΦB = N · B · S = (nl) · (µ0 In) · S = µ0 Sln2 I = µ0 V n2 I

where S is the cross section of the solenoid and V is its volume.

– The total magnetic flux is then proportional to the current: ΦB = LI. If we change the
current, the magnetic flux changes, creating a non zero potential difference between the
ends of the inductor according to Faraday’s law:
dΦ dI
UL = − ⇒ UL = −L
dt dt
The coefficient L is called (self )-inductance and is measured in henry = ohm · second =
second
volt · ampere .

23
• A capacitor is a device that can store electric charge, and normally consists of two conducting
objects (usually plates or sheets) placed near each other but not touching. The capacitor store
energy in form of electric field.

– Using the Gauss’s law:

Z
Q Q
= E · dS = E · S ⇒ E =
ϵ0 surf ace ϵ0 S

– The potential difference between the plates which are at a distance l in the capacitor is:
l
U =E·l = Q
ϵ0 S
Q
– The potential difference is proportional to the charge: U = C. The coefficient C is called
capacitance and is measured in f arad = coulomb
volt .

• Let’s consider an LC circuit, consisting of an inductor L and a capacitor C connected together.

– If the potential in the capacitor is UC and in the inductor UL , then:

Q dI
0 = UC + UL = −L
C dt

Using the fact that I is the current of charge that goes out of the capacitor, I = − dQ
dt , we
get:
d2 Q Q
L 2 + =0
dt C

24
 – This is the equation of a harmonic oscillator.

• Direct current produces stationary EM fields, hence there is no EM waves. they can be emitted
and received by electric circuits with alternating current.
– for alternating current I and U change periodically: ∝ cos(ωt + ϕ)

– I is proportional to the velocity ve of electrons and U is proportional to acceleration ae of

electrons.

– Electrons inside the wire perform harmonic oscillation:

xe (t) = w0 cos(ωt + ϕ0 )
ve = −ω sin(ωt + ϕ0 )
ae = −ω 2 x0 cos(ωt + ϕ0 )
The field produced by such motion is also periodic in time:
2π √
E(r, t) ∝ cos(ωt + ϕ(r)) − periodic with period T = = 2π LC
ω
ω 1
√
– LC circuit produces EM waves with frequency ν = 2π = √
2π LC
and wavelength λ = 2 LC.

25
12 Circuits
• For now we looked at an isolated LC system − not connected to any external force and not
losing energy to the outside world. But we are interested in LC system that emits EM waves
and, therefore, loses energy. So, for both detection and emission of EM waves we need to account
for external force and energy losses.
• We can determine the resistance R, measured in Ohms (Ω), of a wire from the Ohm’s law
V = I · R:
V El l
R= = =ρ
I I S
with ρ the resistivity of the material, l the length of the wire and S his cross section.
This is the simplest model. There exist non linear resistors, whose relation is not linear: I =
V
Rnonlinear (V ) , or superconductors which have R = 0.

• There are two Kirchhoff rule for electrical circuits that immediately follow from the laws of
conservation of charge and energy.
– Junction rule: The sum of incoming currents flowing into a junction is equal to the sum
of outgoing currents.
– Loop rule: the directed sum of the potential differences (voltages) around any closed loop
of a circuit is zero.
• Resistors in series and in parallel:

• Inductances in series and in parallel:

• Capacitors in series and in parallel:

26
• If we have n connected elements, the formulas for total X = R, L, or C change to :
XT = X1 + X2 + · · · + Xn
1
XT = 1 1 1
X1 + X2 + ··· + Xn

• Let’s consider an RLC circuit:

Its an LC circuit with an added resistance R. Its equation is similar to the one of a pendulum
with a friction (damped oscillator):

From analogy with the pendulum, we expect that oscillation cannot last forever and eventually
decay, having lost all its energy. If the energy losses are sufficiently low (i.e. γ < ω) the solution
is: q
−γt
Q(t) = Q0 · e|{z} · cos ω02 − γ 2 t
decay
| {z }
oscillations

The amplitude decreases by e−γT after each period, and the actual angular frequency becomes
smaller: q
ω = ω02 − γ 2 < ω0

• To maintain the oscillations and compensate the energy losses we need to add an energy source.
Let’s add a source V (t) to the circuit (it may be a generator or an EM wave that we want to
detect and which creates potential difference):

It’s equation is analogous to a pendulum in which we push the bob with some force F (t):
Q dQ d2 Q
V (t) = + R + L 2
|{z} C
|{z} dt}
| {z dt }
| {z
external force
potential term energy losses kinetic term

27
 Since we want to create an alternating current, let’s consider the harmonic external force V (t) =
V0 cos ωt:
V0 R 1
Q̈ + 2γ Q̇ + ω02 Q = cos ωt, γ= , ω0 = √
L 2L LC

V0 V0
L L
Q(t) = Q0 cos(ωt+∆ϕ) ≡ p 2 cos(ωt+∆ϕ) = q cos(ωt+∆ϕ)
2 2 2
(ω0 − ω ) + 4γ ω 2 1 R2 2
( LC − ω 2 )2 + L2 ω

– The oscillations caused by an harmonic force have the same angular frequency ω as
the force. And the same period T = 2πω .
– Driven oscillations have some phase shift ∆ϕ = − arctan L(ωωR
2 −ω 2 ) from the force oscillation.
0

– the also have amplitude linearly proportional to the external force Q0 ∝ A(ω,Vω00 , γ) , with the
coefficient A being some function of ω, ω0 and γ. The amplitude depends on the relation
between ω and ω0 .
This means that oscillation with tuned frequency ω ≈ ω0 experience resonance, i.e. their
amplitude becomes very large.
Depending on the properties of our oscillator, we can amplify the amplitude for some
oscillation and reduce that of the unwanted ones thanks to filters.

• Let’s consider an AC circuit:

Its an LC circuit with an added energy source AC: V (t) = V0 cos ωt.
Computing the sum of voltages on the inductor and the capacitor, we get:
dI Q
L + = V0 cos ωt
dt C
Taking the derivative w.r.t time, we get:

d2 I I
L + = −ωV0 sin ωt
dt2 C
The solution of this equation is I(t) = I0 sin ωt, with the amplitude:

V0 ω VL0 1
I0 = 1 = , with ω0 = √
ωL − ωC ω 2 − ω02 LC

28
13 Interference and diffraction
• A stationary wave can be seen as the combination of two waves moving in opposite direction but
sharing the same properties.
sin(ωt − kx) + sin(ωt + kx) = sin(ωt) × 2 cos(kx)
| {z } | {z }
time oscillations amplitude(x)

If kx = nπ, n = 0, ±1, · · · , the amplitude becomes twice larger − constructive interference.

If kx = n + 12 π, the two waves cancel each other − destructive interference.

In reality, waves cannot have infinite length, their energy would be infinite, so out function
sin(ωt − kx) is a simplication.
• Let’s discuss the energy of EM field. Since it’s a separate entity, the total energy of a system of
charges is equal to:
total energy = kinetic energy of charges + energy of the EM f ield
The energy of the EM field is the measured amount of work which can be done by the field on
the charged particles. Or equivalently, the work done over charges to create the field.
• Let’s study two examples:

29
• The energy density of an EM field is:

E2 B2
uEM = +
8πkE 8πkM

Electric field then can perform work over charges and give the energy back.

∆Welectic = qE(rf − rin )

if the charge q moves from rin to rf .

Although magnetic field cannot perform work over charges, it shoudn’t be considered alone. If
we change the magnetic field in time, we create an electric field according to the third Maxwell’s
law, which can perform work.
• The EM waves that we produce are not ”monochromatic wave” in reality (i.e. a single wave with
some ω). Instead we have wave packages of finite size which can be represented as interference
of many waves.
If we localize our wave package inside a length L, we will have a lot of waves with frequencies ω
which differ from the main frequency ω0 by:
2πc
∆ω ≃
L
These waves interfere and produce a non zero field, only inside the length L !

• Interference can happen between the separate parts of a single wave, we call this phenomenon
diffraction.
It follows from the Huygens–Fresnel principle: every point on a wavefront is itself the source of
spherical wavelets, and the secondary wavelets expanding from different points mutually interfere.

• When a plane wave passes through a small hole, the resulting wave propagates in all directions.
These spherical waves are isotropic and have amplitude that falls as 1r with r being the distance
to the source.

30
Summary for static charges
Electrostatic fields are electric fields that do not change with time. Such fields are present when systems
of charged matter are stationary, or when electric currents are unchanging. In that case, Coulomb’s
law fully describes the field.

• For interaction of two charges:

– Coulomb’s force:
q1 q2 q1 q2 r1 − r2
F12 = k 2 r̂12 = k |r − r |2 |r − r |
r12 1 2 1 2

measured in N (Newtons).
– Potential energy of two charges:
q1 q2
U = kE
|r1 − r2 |

measured in J (Joules).
– Force from the potential energy:
F = −∇U

• For one source charge:

– Coulomb’s electric field:
q r − rq
E(r) = k
|r − rq |2 |r − rq |
measured in N/C or V /m.
– Potential of a charge:
q
ϕ(r) = kE
|r|
measured in V (1 volt = 1 J/C)
– Field from the potential:
E = −∇ϕ

If we have more than one source charge, we add all up (or integrate) following the superposition
principle !

31
Summary for electric and magnetic dipoles
• Electric dipoles:

– Dipole moment: p = q · l
– Torque in E-field: τ = p × E
– Potential energy: U = −p · E
– Equilibrium: If the dipole is parallel to E then the equilibrium is stable, if antiparallel - it’s
unstable.
– Cutting in two: If we cut the dipole in two pieces, we get two opposite charges - sources of
electric field.
• Magnetic dipoles:

– Dipole moment: µ = I · A
– Torque in B-field: τ = µ × B
– Potential energy: U = −µ · B
– Equilibrium: If the dipole is parallel to B then the equilibrium is stable, if antiparallel - it’s
unstable.
– Cutting in two: If we cut the dipole in two pieces, we get two dipole - no magnetic monopole
(or souce) has been found.

32
Summary for Maxwell’s equations
• To summarize, the Maxwell’s equations are :

– ∇ · E = 4πkE ρ (Gauss′ s law)

– ∇·B=0 (no magnetic charges)
– ∇·E= − ∂B
∂t (F araday ′ s law)
kM ∂E
– ∇ · B = 4πkM j + kE dt (Ampère′ s law, extend by M axwell)
Or in integral form (which we used up till now):

– ΦE ≡ S EdS = 4πkE q ≡ ϵq0

R
R
– ΦB ≡ S BdS = 0
– ε ≡ E dl = − dΦ
R B
dt
– B dl = µ0 I + µ0 ϵ0 dΦ
R E
dt

These equations allow us to find electric and magnetic forces given a distribution of charges and
currents.
• The Maxwell’s equation are local, i.e. they do not assume action at a distance, and can therefore
be postulated for arbitrary moving charges in any reference frame.
• Faraday’s induction law and displacement current already suggest that the field does exist: it
has non-trivial dynamics even if there are no charges at all.
• If there are charges far from us, we cannot know directly what happens to them. We can only
measure fields here and now.
• In vacuum, variable electic field can create magnetic field (4th law), which in turn creates electric
field (3rd law), and so on (electromagnetic waves).

33
Summary for emission of EM waves
• Accelerating charges (or varying in time current) are necessary to produce EM waves.

• A simple way to produce an alternating current is using an LC circuit where :

– Inductance L:
∗ is a coefficient of proportionality between magnetic flux and current ΦB = LI.
∗ measured in [H] = henry.
∗ opposes changes in the current according to Faraday’s law:
dI
U = −L
dt
– Capacitance C:
∗ is a coefficient of proportionality between accumulated charge and potential difference
Q = CU .
∗ measured in [F ] = f arad.
∗ leads to the potential difference:
Q
U=
C
• An LC circuit can contain an oscillating current with frequency ω = √1
LC
and produce EM waves
with the same frequency.

34
Summary for interference and diffraction
• Real waves have the form of wave packages, consisting of various waves with different ω(k). If
you localize a wave in the box with size L, you spread the frequencies in the range ∆ω ≃ 2πc
L
(∆k ≃ 2πL ).

• Wave nature of EM fields lead to such phenomena as interference and diffraction.

• Interference: two waves may amplify each other if they oscillate together, or instead cancel each
other, if they are out of phase (have phase difference π).
• Diffraction follows from the Huygens-Fresnel principle: each point of the wavefront is a source
of spherical waves. The total field is an interference of all such secondary waves.
• Diffraction means that geometrical optics is no longer valid. Diffraction may happen when a
light ray passes through a slit with the width comparable to the wavelength.

• Spherical waves are isotropic and have amplitude that falls as 1/r with r being the distance to
the source.

35