Physics 332, Electromagnetic Theory II Lecture 3

1 Magnetic flux
Before we talk about the mathematical description for the induction, we need
to learn the concept of the magnetic flux. In Faraday’s experiments, we know
that a changing magnetic field is crucial, but the magnetic field is a vector
which is hard to handle. Moreover, as shown in the last lecture, some energy
transfer (or emf) is required to generate a current, and energy is a scalar not
a vector. Therefore, the question is: can we find a scalar quantity associated
with the magnetic field? Once we find it, it will be easier to relate it to emf.
This quantity is the magnetic flux.
The magnetic flux is the total amount of the magnetic field passing
through a particular surface S, which can be defined as:
Z
ΦB = ~ · d~a,
B (1)
S

where A is any surface in the space. It is easy to check that the magnetic
flux through a closed surface is always zero.
I Z
~
B · d~a = ~
∇ · BdV = 0, (2)
S V

where V is the volume enclosed by the closed surface S. As a reminder,

we use the right-hand rule to determine the normal vector of d~a, i.e., if the
fingers of the right hand refers to the positive direction around the boundary
of d~a, the thumb indicates the direction of d~a.

2 Dimensional analysis
Before arriving at the final expression, we can try to ’guess’ what equation
we should expect from the dimensional analysis. The basic idea is that for
any physical equation, every term needs to have the same dimension (unit)
so that they can be put together in an equation. We can use the dimensional
analysis to find the possible relation between magnetic flux (ΦB ) and emf ε.
Let us define [X] representing the unit of the quantity X. From the
Lorentz force, we have
F = q(E ~ + ~v × B).
~ (3)
Therefore we immediately know the E has the same unit of v times B, and
we can express it as:
[E] = [v][B]. (4)

by Wei-Cheng Lee Page 1

Physics 332, Electromagnetic Theory II Lecture 3

From the definition of ε, we have

[ε] = [E][L] = [v][B][L], (5)

where [L] refers to the unit of length. Since we know

[L]
[v] = , (6)
[t]

where [t] is the unit of time, we can substitute it into Eq. 5 to obtain

[B][L2 ] [ΦB ]
[ε] = = . (7)
[t] [t]

Therefore we find that dΦB /dt has the same unit of ε, and dΦB /dt also
describes the change of the magnetic field. As a result, a reasonable guess
will be
dΦB
ε=a , (8)
dt
where a is a dimensionless value. We will see that this guess is pretty good,
but special care is required for the sign of a, which can be taken care of by
a special rule called Lenz’s law.

3 Lenz’s law
Here is the statement of Lenz’s law (from Griffiths’s book): If a current
flows, it will be in such a direction that the magnetic field it produces tends
to counteract the change in flux that induced the emf. Following from Lenz’s
law, the constant a in Eq. 8 should be negative.
Lenz’s law is in fact a generalization of Newton’s third law. In order to
change the magnetic field, there must be a force doing this job (e.g. one has
to push the magnet toward the loop of wire). Such a force acts on the loop of
wire via the change of the magnetic flux, and the the loop needs to produce
opposing reaction force in response. Therefore, the current produced by the
loop of wire must produce a magnetic field which counteracts the change in
magnetic flux, and this is just what Lenz’s law states.
Lenz’s law is very useful for the prediction of the direction of the current,
but it does not give us the amount of the current. Next we will show that
the constant a in Eq. 8 is in fact just -1.

by Wei-Cheng Lee Page 2

Physics 332, Electromagnetic Theory II Lecture 3

Figure 1:

4 Derivation of flux rule and Faraday’s law

Let’s consider the textbook example of a moving loop in a static magnetic
field. When the whole loop is inside the magnetic field, the magnetic flux
does not change at all even as the loop is still moving. In this case no current
will be generated. However, as some part of the loop is outside the magnetic
field, a current will appear. Of course, if whole loop is outside the magnetic
field, no current will appear.
Let’s consider the case having induced current shown in Fig. 1. The
magnetic flux at a fixed time is just B times the total area of the loop under
the magnetic field. Therefore we have:

ΦB (t) = Bhs(t). (9)

Therefore
dΦB (t) ds(t)
= Bh = −Bhv, (10)
dt dt
where v = − ds(t)
dt
because if the loop is moving to the right, s(t) keeps
decreasing as the time increases. On the other hand, we can use the definition
of ε I
ε = f~ · d~l, (11)

where f~ is the total force per unit charge. In this case, the only force is the
magnetic force
f~ = f~mag = (~v + I/q)
~ ~
× B, (12)

by Wei-Cheng Lee Page 3

Physics 332, Electromagnetic Theory II Lecture 3

~ is the velocity associated with the current. However, one can easily
where I/q
check that the second part of Eq. 12 does not contribute to the emf because
its direction is always perpendicular to d~l. Therefore
I
dΦB (t)
ε = f~ · d~l = vBh + 0 + 0 + 0 = vBh = − , (13)
dt
and we obtain the flux rule
dΦB (t)
ε=− . (14)
dt
This result seems to suggest that the magnetic field does the work to produce
the emf, but it does not. Magnetic field never does work for sure. In this
case, the one who pulls the loop does the work. Therefore in this case, no
new physics emerges. It is just again an application of Lorentz force and
transfer of energy from mechanical work to electricity.
The following discussion is very crucial. We have just proved the flux
rule for the case of static magnet and a loop moving to the right, and we
find the emf is driven by the Lorentz force. But what if we hold the loop
still but move the magnet to the left? These two cases must give the same
result due to the concept of special relativity, but the physical interpretations
are totally different. In the first case, it is the Lorentz force that drives the
current, but in the second case, there is no Lorentz force because the loop
does not move. Then how can the current be driven? Since the magnetic
field can not move a charge at rest, the only possibility is an electric field
generated by the changing magnetic field! (the magnetic field is changing
now because we are moving it.) Therefore, the emf now should be written in
terms of E~ as: I
ε= E ~ · d~l = − dΦB (t) . (15)
dt
Applying Stokes’ theorem, we obtain
I
~ · d~l = − dΦB (t)
E
dt
Z Z ~
(∇ × E)~ · d~a = − ∂ B · d~a
∂t
~
∇×E ~ = − ∂B . (16)
∂t
The last line is Faraday’s law. Note that the partial derivative instead of total
derivative with respect to time t is introduced to emphasize that B ~ is the only
thing changing with time. And this Faraday’s law is the first correction to

by Wei-Cheng Lee Page 4

Physics 332, Electromagnetic Theory II Lecture 3

Maxwell’s equations with the time dependence. The physical meaning is that
a changing magnetic field can generate a non-conservative electric field. For
the first case in which the magnetic field does not change, no non-conservative
electric field is generated. The Lorentz force is the source to produce the emf.
In the case where Faraday law applies, the changing magnetic field creates
a non-conservative electric field that produces the emf. Even though they
both produce the same emf, the physical interpretations are totally different.
In summary, the universal flux rule tells us that a change in magnetic
flux always produces an emf regardless the way the magnetic flux changes.
That is:
Z
d ~ · d~a]
ε = − [ B
dt A
= εF + εL . (17)

where εL is the emf produced by any magnetic flux changed unrelated to a

changing field for which no non-conservative electric field is generated. εF is
the emf produced according to Faraday’s law for which a non-conservative
electric field is generated by a changing magnetic field.
Z ~ Z
∂B ~ · d~a
εF = − ( ) · d~a = (∇ × E) (18)
A ∂t A

by Wei-Cheng Lee Page 5