Physics II: Electromagnetism

PH 102

Lecture-5

January-May 2025

1
Review
Line element d~r = h1 ê1 du1 + h2 ê2 du2 + h3 ê3 du3

@~
r
@r
ê1 ⌘ r̂ = @~
r
= sin ✓ cos x̂ + sin ✓ sin ŷ + cos ✓ẑẑ
@r
@~
r
Unit vectors:
ê2 ⌘ ✓ˆ = @✓
@~
r
= cos ✓ cos x̂ + cos ✓ sin ŷ sin ✓ẑ
@✓
@~r
ˆ= @
ê3 ⌘ = sin x̂ + cos ŷ
@~r
@

Spherical polar coordinate h1 = hr = 1 , h2 = hθ = r , h3 = hϕ = r sinθ

Cylindrical polar coordinate h1 = hs = 1 , h2 = hϕ = s , h3 = hz = 1

Del operator in curvilinear coordinate system ⃗

∇ = e1̂
1 ∂
+ e2̂
1 ∂
+ e2̂
1 ∂
h1 ∂u1 h2 ∂u2 h3 ∂u3
Symmetry of the problem decides what coordinate system to choose.
IIT Guwahati 2 .
PH102 (2025)
 ∂x ∂y ∂z
Spherical coordinate !
∂vz system
∂v y
" !
∂vx ∂vz
" !
∂v y ∂vx
"
Curl : ∇×v = − x̂ + − ŷ + − ẑ
∂y ∂z ∂z ∂x ∂x ∂y
The expressions for the gradient, divergence, curl and Laplacian in the spherical polar
coordinate system 2 ∂ 2t ∂ 2t ∂ 2t
Laplacian : ∇ t = 2
+ 2+ 2
∂x ∂y ∂z
Use already discussed definitions of gradient, divergence and curl in terms of
(h1, h2 , h3) dl = dr r̂ + r dθ θ̂ + r sin θ dφ φ̂; dτ = r 2 sin θ dr dθ dφ
Spherical.

∂t 1 ∂t 1 ∂t
Gradient : ∇t = r̂ + θ̂ + φ̂
∂r r ∂θ r sin θ ∂φ

1 ∂ 2 1 ∂ 1 ∂vφ
Divergence : ∇ · v = 2 (r vr ) + (sin θ vθ ) +
r ∂r r sin θ ∂θ r sin θ ∂φ
# $
1 ∂ ∂vθ
Curl : ∇×v = (sin θ vφ ) − r̂
r sin θ ∂θ ∂φ
# $ # $
1 1 ∂vr ∂ 1 ∂ ∂vr
+ − (r vφ ) θ̂ + (r vθ ) − φ̂
r sin θ ∂φ ∂r r ∂r ∂θ
! " ! " 2
2 1 ∂ 2 ∂t 1 ∂ ∂t 1 ∂ t
Laplacian : ∇ t= 2 r + 2 sin θ +
r ∂r ∂r r sin θ ∂θ ∂θ r 2 sin2 θ ∂φ 2

Cylindrical. dl = ds ŝ + s dφ φ̂ + dz ẑ; dτ = s ds dφ dz
IIT Guwahati 3 PH102 (2025)
Cylindrical coordinate
1
# system
1 ∂vr ∂
$ #
1 ∂
$
∂vr
+ − (r vφ ) θ̂ + (r vθ ) − φ̂
r sin θ ∂φ ∂r r ∂r ∂θ
The expressions for the gradient, divergence, curl and Laplacian in the cylindrical polar
! " ! " 2
coordinate system 1 ∂ 2 ∂t 1 ∂ ∂t 1 ∂ t
2
Laplacian : ∇ t = 2 r + 2 sin θ +
r ∂r ∂r r sin θ ∂θ ∂θ r 2 sin2 θ ∂φ 2
Use already discussed definitions of gradient, divergence and curl in terms of
(h1, h2 , h3)
Cylindrical. dl = ds ŝ + s dφ φ̂ + dz ẑ; dτ = s ds dφ dz

∂t 1 ∂t ∂t
Gradient : ∇t = ŝ + φ̂ + ẑ
∂s s ∂φ ∂z

1 ∂ 1 ∂vφ ∂vz
Divergence : ∇ · v = (svs ) + +
s ∂s s ∂φ ∂z
# $ # $ # $
1 ∂vz ∂vφ ∂vs ∂vz 1 ∂ ∂vs
Curl : ∇×v = − ŝ + − φ̂ + (svφ )− ẑ
s ∂φ ∂z ∂z ∂s s ∂s ∂φ
! " 2 2
2 1 ∂ ∂t 1 ∂ t ∂ t
Laplacian : ∇ t= s + 2 2+ 2
s ∂s ∂s s ∂φ ∂z

IIT Guwahati 4 PH102 (2025)

Examples:
~ n using Cartesian coordinates and the calculation
In Tutorial 1, we evaluated rr
required quite a few steps. Finally, we arrived at the result rr ~ n = nrn 1 r̂.
The result can be arrived at in a single step if we take help of Spherical Polar
Coordinates:
Using the form of the gradient operator in Spherical Polar Coordinate:

~ @ ˆ 1 @ ˆ 1 @
r ⌘ r̂ +✓ +
@r r @✓ r sin ✓ @

~ @ n
rr = r̂ r = r̂ nrn
n 1 ( Since r has no dependence on ✓, .)
@r
In a similar manner, you can show the following:

~ 2 df
r.(r̂f (r)) = f (r) +
r dr
~
r.(r̂r n
) = (n + 2)rn 1
~ ⇥ (r̂f (r))
r = 0
2 df d2 f
r2 f (r) = + 2
r dr dr
r2 r n = n(n + 1)rn 2 .

IIT Guwahati 5 PH102 (2025)

 r̂
Divergence of V~ = 2
�

r
~ is directed radially outward.
At every direction, V �

R
The function has large positive divergence. �

But…
-�

~ V
r. ~ = 1 @
r2 r12 = 1 @
r 2 @r r 2 @r (1) =0
R -�

=) ( ~ V
r. ~ ) d⌧ = 0
V -� -� � � �

Suppose, we integrate over a sphere of radius R, entered at origin: the surface

integral is
H R ⇣R ⌘ ⇣R ⌘
~ .d~a = 1 2 ⇡ 2⇡
V R2 r̂ .(R sin ✓d✓d r̂) = 0
sin ✓d✓ 0
d = 4⇡

R R
But divergence theorem states that ( ~ V
r. ~ )d⌧ = ~ .d~a !
V R
~ ~
R
~ .d~a !
But divergence
V theorem
S states that
V
(r.V )d⌧ = S V
What is happening here? Is divergence
What is theorem
happeningwrong??
here? Is divergence theorem wrong??

IIT Guwahati PH102 (2025)

 r̂
Divergence of V~ = 2
r
The source of the problem is the point r = 0, where the function blows up!
R
~ V~ = 1 @ =) ( ~ V
r. ~ ) d⌧ = 0
r. r 2 @r r2 r12 = 1 @
r 2 @r (1) =0 V

~ V
It is true that r. ~ = 0 everywhere except at the origin. But, right at the origin
the situation is more complicated.

Note that surface integral is independent

R of R; so if divergence theorem is right
(and it is), we should expect (r. ~ V
~ )d⌧ = 4⇡. The entire contribution must
then be coming from the point r = 0.

~ V
r. ~ has the bizarre property that it vanishes everywhere except at one point,
and yet its integral over any volume containing that point is 4⇡ =) “No
Ordinary Function”.

Dirac Delta Function

IIT Guwahati 7 PH102 (2025)

 Dirac Delta Function
3.5

3.0

2.5

2.0

1.5

1.0

0.5

-2 -1 1 2

Paul Dirac

✦ Dirac delta function is a generalised function or distribution introduced by physicist

Paul Dirac.

✦ Dirac Delta function is a function that is zero everywhere except one point and at
that point it can be thought of as either undefined or as having an infinite value.

IIT Guwahati 8 .
PH102 (2025)
Dirac Delta Function

A real function on R is called Dirac Delta Function if

⇢
0 if x 6= 0;
(x) =
1 if x = 0.

and
Z 1
(x)dx = 1.
1
“In nitely high, in nitesimally
narrow spike with area 1”

This of course is a heuristic definition.

Not well de ned at x=0, nor even continuous at x = 0.

In a traditional sense, it is not a function and can be called a generalised function

or a distribution.

IIT Guwahati 9 PH102 (2025)

fi
fi
fi
Then, how to “see” them?

The best way to look at a delta function is as a limit of a sequence of functions.

We give a few such examples:

? We can have a sequence of function as 2 R2 (x)

8
< 0 if x  21n ;
R1 (x)
Rn (x) = 2n 1 if 21n < x < 1
2n ;
1
:
0 if x > 21n .
1
2
1
4
1
4
1
2
x

For a fixed n, it represents a rectangle of height n and width between 21n to

1
2n . As n ! 1, width decreases but height increases in such a proportion that
the area always remains 1. So, as n ! 1, Rn ! .

IIT Guwahati 10 PH102 (2025)

on is defined
(t)by
= the
0
f (x) = t 6
=
following
2πe 0
property
e
(1)
1
f (y) dy dk for “nice” functions for
f (•) x >
Some representations −∞
1 t =Step 0 and
−∞
of Delta
Dirac delta function
0 where
t 6= 0 the central 2 is that
a (usually inconsequentia
( 1 Functions
can by reorganization be read as an assertion
(t) = 18.031 (1)
Z 1 the
t = 0stipulation that
Integral representation
t2 :
Haynes Miller−ik(y−x)
! +∞ and Jeremy Orlo↵ ⎧
1
dt (t) = 1 2π −∞ e dk = δ(y − x)
(2)
Z t
t1 2 ⎨ −1 f
The
1 dt The
(t) =unit
t history
o otherwise).
1 step function
of the δ-function
It is “infinitely
1 can in this
peaked” ε(x)(2)
at tsense
= 0bewith2θ(x)
≡traced
the −to1 the
back = early 0 a
+1 f
1820’s. Fourier, of course, was concerned with the Fourier
theory of integral
heat
⎩
representation
conduction,
ro otherwise).
ou can view1.1Itthis
isDefinition
“infinitely
functionpeaked” at t =of0 Gaussian
as a limit with the
Limiting
You but
case
can view this ϵfunction
→the
by (1828 0):δ-function hadof
as a limit intruded for a second time into a physical theory;
Gaussian
George Green
be odd
noticed
(i.e.,
that the
that
solution
vanish
offunction,
the Poisson
at the origin).
u(t): equation ∇ ϕ(x) = ρ(x),
As D
2
Let’s start
considered
with
1
the definition of the
ε(•)
unit step

(t) = lim p1 to describe

t2e
2
t /2the electrostatic
2 ( potential generated by a given charge
(3)
(t) = lim p
distribution
!0!0 Heaviside)
2⇡2⇡
e
ρ(x),
/2 2
can be have
obtained
u(t) = by0 remarked,
for t < 0 Limit
superposition(3) of
of the
the step
Gaussian
potentials function
generated (wit
function
by a population of point charges; i.e., 1 that
for tthe
> 0 general problem can be reduced
to the
acquainted
special problem
as a student of electrical engineering)
We do not define u(t) at t = 0. Rather, at t = 0Limit of Lorentz
we think of it as function
in transition
between10 close
1 ✏1. ✏ relationship ∇ ϕ(x; y) supplied
= δ(x − y) by the calculus:
2
and
(t) = lim . (4)
(t) It=is✏!0
lim t2
⇡ the+unit
called ✏2 .
step2function (4)
because it takes a unit step at t = 0. $charge
It is sometimes x
where ✏!0now ⇡ the
t2 + ✏
δ-function is being used to describe a “unit point
thecalled
roperty of positioned
delta the function
Heaviside function.
is the The graph of relation
following u(t) is simple.
at the point y.” Thus came into beingθ(x the − “theory
a) =of Green’sδ(y −
operty Zoffunctions,”
the delta functionimportant
which—with θ(x)
is the following
input relation
u(t) by Kirchhoff (physical optics, in the
1 −∞
dtf (t) and
Z 1880’s) (t) Heaviside
= f (0) (transmission lines, in the (5)
1890’s)—became, as it remains,
Heaviside step function θ(x)
one of the principal consumers of applied distribution theory. ⇕
dtf (t) (t) = f (0) xt (5)
x

∫
. This is easy toI have
see. First of all, (t) vanishes everywhere
0 to =
sketched this history 6
in order to maked clear that what I propose
goes −
θ(x froma) δ(x − a)
efore, it does δ(y)
We
not
to do
will dy
use =
matter
in these
u(t)θ(x)
as an idealized model of a natural system that
what
pages values
stands theapart,
function dx
fin(t)
1 very
quickly. In reality it will make quite
a smooth both such
transition, spirit
as theand by intent, from the
following.
This is easy
−∞ to
trend of see.
say First
recent of =all,f (0)(t)
f developments, and vanishes
Then f everywhere
is fashioned from much ruder fabric. My
0. You
fore,
can then
objective
The
notis matter
to it
promote
central
(t) (t)
a point
1 is, in this light, equivalent to
1(t).
of2view—a
(0)
computational technique—that came(2025)
de IIT it
the does
integral
Guwahati because does what
not values
depend on11 the
t, and function
you f (t) PH102
Dirac Delta Function: Properties
? For a continuous function f (x),
Z 1
f (x) (x)dx = f (0)
1

This means that for a continuous function f (x), the product f (x) (x) is zero
everywhere except at x = 0. It follows: f (x) (x) = f (0) (x).

⇢ Z 1
? Translation: 0 if x 6= a
(x a) = with (x a)dx = 1
1 if x = a 1
Z 1
Therefore the first property tells us f (x) (x a)dx = f (a)
1
Z 1
f (x) (x a)dx = f (a)
1
? Although itself is not a legitimate function, integrals over are perfectly
acceptable. In fact two expressions
R1 involving delta
R 1 functions (say, D1 (x) and
D2 (x)) are called equal if 1 f (x)D1 (x)dx = 1 f (x)D2 (x)dx, for all f (x).

IIT Guwahati 12 PH102 (2025)

Dirac Delta Function: Properties
1
Scaling : (kx) = (x), where k is any constant. Note : Delta function has meaning
|k| only under integral sign

Proof: Chose an arbitrary test function f (x) and consider the integral:
Z 1
f (x) (kx)dx
1

Let y ⌘ kx, so that x = y/k and dx = dy/k. If k > 0, the integration limits are
unchanged but if k < 0, the xZ=11 implies y = 1,Zand 1 vice versa. Restoring
dy
the proper order of the limits: f (x) (kx)dx = ± f (y/k) (y)
1 1 k
Z 1 Z 1
dy 1 1
f (x) (kx)dx = ± f (y/k) (y) = ± f (0) = f (0)
1 1 k k |k|
1 1
= ± f (0) = f (0)
|k| the same purpose as (1/|k|) (x):
Therefore, under the integral sign,k (kx) serves
Z 1 Z 1 
1
f (x) (kx) = f (x) (x) .
1 1 |k|
Infact, this property tells us ( x) = (x).
IIT Guwahati 13 PH102 (2025)
Dirac Delta Function: in three dimensions

3
Generalize in 3-D: (~r) = (x) (y) (z)

3-D Dirac Delta is zero everywhere except at origin (0,0,0), with its volume integral being 1
Z Z 1 Z 1 Z 1
3
(~r)d⌧ = (x) (y) (z)dxdydz = 1
all space 1 1 1

R1
Generalizing 1
f (x) (x a)dx = f (z) in 3-D:

Z
f (~r) 3 (~r r~0 )d⌧ = f (r~0 )
all space

IIT Guwahati 14 PH102 (2025)

Dirac Delta Function: in three dimensions
Let us get back to the divergence paradox :

~
Recall that r. r̂
= 0, if ~r 6= 0.
r2

The one and only point where divergence is non-zero is origin.

But do we know the value of the divergence at origin?

Assume that it is k 3 (~r)

Z ✓ ◆ I
~ r̂ r̂
Divergence theorem =) r. 2 d⌧ = 2
.d~a
V r S r
4⇡
Z
=) k 3
(~r)d⌧ = 4⇡ =) k = 4⇡
V

⇣ r̂ ⌘
~
r. = 4⇡ 3
(~r)
r 2

IIT Guwahati 15 PH102 (2025)

Few examples:
R3
1. Evaluate 0
x3 (x 2)dx.

The delta function picks out the value of x3 at the point x = 2, so the integral
is 23 = 8. Note however, if the upper limit had been 1 (instead of being 3),
the answer would be 0, because the spike would then be outside the domain of
integration.
R6 2
2. Evaluate 2
(3x 2x 1) (x 3)dx.
R1 2
Recall that 1
f (x) (x a)dx = f (a). Here f (x) = (3x 2x 1), a = 3 and it
R6
lies between the limits of the integration. Therefore 2 (3x2 2x 1) (x 3)dx =
f (3) = 20.
R2
3. Evaluate 2
(2x + 3) (3x)dx.
Z 2 Z 2 ✓ ◆
3 t dt
Change variable x = t/3. Then (2x + 3) (3x)dx = 2 +3 (t) = 1
2 2
3
3 3
Alternatively, you can use (3x) = (x)/3 and proceed accordingly.
R
4. Evaluate J = (r 2
+ 2) ~ r̂2 d⌧ . Here V is a sphere of radius R centred at origin.
r.
V r
R 2
J = V (r + 2)4⇡ 3 (~r)d⌧ = 4⇡(0 + 2) = 8⇡
IIT Guwahati 16 PH102 (2025)
Electrostatics

17
Fundamental forces in nature

• Gravitational
• Electromagnetic
• Weak
• Strong

We will study the nature of electromagnetic forces in this course

https://wordpress.com/
IIT Guwahati 18 PH102 (2025)
Concept of Electrostatic
The primary goal of this course is to understand set of four equations known as the
Maxwell’s equations
~ E~ ⇢
r. =
✏0
~ ⇥E
~ @B
r =
@t
~ B
r. ~ = 0
@ ~
E
~ ⇥B
r ~ = µ0 J~ + µ0 ✏0
@t
‣ Situations described by these equations can be extremely complicated and to start with
we will simplify life by assuming that nothing depends on time - “static case”
~ E~ ⇢
r. =
✏0 ~ B
r. ~ = 0
~ ⇥E
r ~ ~ ⇥B ~
= 0 r = µ0 J~
Electrostatics Magnetostatics

‣ Electricityand magnetism are distinct phenomena so long as charges and currents are static.
Independence of E and B does not appear until there are charges or currents. Only when there
are suf ciently rapid changes in the charges and currents with time, will E and B depend on each
other!
IIT Guwahati 19 PH102 (2025)
fi
Some important points to note before we start

✓ Each particle in the Universe carries with it a number of properties. They

determine how the particle interacts with each of the four forces. For the force of
gravity, this property is mass. For the force of electromagnetism, the property
is called electric charge.

✓ For the purposes of this course, we can think of electric charge as a real number, q
∈ R. Importantly, charge can be positive or negative. It can also be zero, in which
case the particle is unaffected by the force of electromagnetism.

✓ The SI unit of charge is the Coulomb, denoted by C. At a fundamental level,

Nature provides us with a better unit of charge. This follows from the fact that
charge is quantised: the charge of any particle is an integer multiple of the charge
carried by the electron: e = 1.60217657×10−19 C. i.e. q=n e.

IIT Guwahati 20 PH102 (2025)

Electrostatics: Coulomb’s law
Suppose we have a point charge q is at rest. Then what is the force on a test charge ~ r
Coulomb’s law
Q
1 qQ
F~ = r̂ r~
4⇡✏0 r 2 (based on experiments)
q

✏0 = 8.85 ⇥ 10 12 C2
is called permittivity of free space ~r
0
~r
N m2

~r = ~r
0
~r
✦ Between two charges at rest, force is directly proportional to the (separation vector)
product of the charges
✦ Force is inversely proportional to the square of the distance
between.
✦ The force is along the straight line from one charge to another

What happens when there are many point charges?

IIT Guwahati 21 PH102 (2025)

Electrostatics: Principle of Superposition
If we have electric charges, say q1, q2, q3, . . . as the source charges as shown below

Then the force exerted by these charges on another charge, Q can be calculated by the
principle of superposition.

where, F1: force on Q due to q1

F2: force on Q due to q2 and so on

The force on any charge is the vector sum of the Coulomb forces from each of the other
charges present.

IIT Guwahati 22 PH102 (2025)

Electric eld
Q
Let us write the force on Q due to q as follows:
q
r~
1 qQ
F~ = r̂ 1 q
r̂ ~ r)Q
4⇡✏0 r
= Q ⌘ E(~
2
4⇡✏0 r 2
~r
0
~r
~ r) is called the electric field of the source charge q.
• E(~

• The field is a function of the position ~r, since the

separation vector ~r depend on the location of the charge Q (field point).

• While it takes two charges to feel a force, it takes only one charge to produce
~ r) = q 2 r̂ at point ~r.
a field. A charge at the origin produces the field E(~ 4⇡✏0 r

• The field due to charge q is non-zero everywhere, not just where there is
another charge to feel the field.

This leads us to an important conclusion that electric force is a long range force like
gravitational force.
IIT Guwahati 23 PH102 (2025)
fi
Electric eld
• If there are many charges,invoke the superposition principle: the field at some
~r due to many charges will be the (vector) sum of the fields due to each one.
Q
1
✓
q1 Q q21 Q
◆ ~ri
F~ = F~1 + F~2 + · · · =
4⇡✏0 r1
2
r̂ 1+
r2
2
r̂ 2 + ···
q1
qi
q2 ~r
✓ ◆ 0
Q q1 q21
=
4⇡✏0 r1
2
r̂ 1 + 2 r̂ 2 + · · ·
r2
~ri

n
X
1 qi
=Q
4⇡✏0 r2
r̂ i ~ r)
= QE(~
i=1 i

~ r), which is a vector quantity, depends on the location of the

• Note that E(~
field point and is determined by the configurations of the source charges qi .

• To measure a field is easier: put a known test charge q at ~r, equate the force
~ If q = 1C, the force and E
it experiences to q E. ~ are numerically equal but
dimensionally di↵erent. That is why “field is the force on a unit charge”.

IIT Guwahati 24 PH102 (2025)

fi
 0 if x 6= 0;
(x) =
Summary 1 if x = 0.

A real function andDirac Delta Function if

on R is called
⇢ Z 1
0 if x 6= 0; (x)dx = 1.
(x) =
1 if x = 0. 1

and
Z Z 1 Z 1 
Properties 1 Z 1 1
f (x) (x a)dx = f (a) f (x) (kx) = f (x) (x) .
1 (x)dx =11. 1 |k|
1

1 qQ
Coulomb’s law F~ = r̂ ✏0 = 8.85 ⇥ 10 12 C2
is called per
4⇡✏0 r 2 N m2

Principle of Superposition
Total force on charge q1 due to presence of q2 and q3 is given by

F~ = F~12 + F~13
Force on q1 due to q3
Force on q1 due to q2

If there are many charges, the field at some position r ⃗ due to many charges will be the
vector sum of fields due to each one.

IIT Guwahati 25 PH102 (2025)

Thank You