Physics II: Electromagnetism

PH 102

Lecture-3

January-May 2025
Review :

What we have learnt so far:

• Line integrals of vector fields along a given path. The results are in general path
dependent.

• Gradient fields are a class of vector fields, whose line integrals are path independent.
The result depends only on the end points. (Fundamental theorem of Gradients).

• Such Gradient fields are called conservative vector fields that satisfy the property :
∇ ⃗ × F ⃗ = 0 and F ⃗ ⋅ .d r ⃗ = 0.
∮

• Surface integrals of vector fields can be calculated from the idea that elementary
surfaces can be treated as vectors . In general d a ⃗ = d x ⃗ × d y⃗ = da n.̂

• In general, line integrals are one-dimensional integrals while the surface integrals
are two-dimensional ones.

IIT Guwahati 2 PH102 (2025)

Surface integrals (Example)
Closed Surface integral

Given a vector A = 2xz x ̂ + 2zx ŷ − yz z,̂ calculate the closed surface integral of it over
the surface de ned by a cube. The cube occupies the space between 0 ≤ x, y, z, ≤ 1.

Solution

IIT Guwahati 3 PH102 (2025)

fi
Surface integrals (Example contd.)

IIT Guwahati 4 PH102 (2025)

Surface integrals (Example contd.)

IIT Guwahati 5 PH102 (2025)

Volume Integrals

Consider a closed surface in space enclosing a volume. Then,

Z Z
~v d⌧ and Td⌧
V V

are examples of volume integrals. Here d⌧ represents elementary volume.

• Here the volume element is obtained from dτ = d a ⃗ ⋅ d x3̂ = dx1dx2dx3 n ̂ ⋅ x3̂

• For the elementary surface area in the xy-plane, the corresponding unit vector n ̂ of area
is in the same direction as z.̂

• In the case of surface area, if d x1̂ and d x2̂ are in the same direction then the elementary
area vanishes i.e. when d x1̂ ∥ d x2̂ . and d a ⃗ is maximum if d x1̂ ⊥ d x2̂ . In the cartesian
coordinate system, x ̂ ⊥ ŷ and hence d a ⃗ = d x ⃗ × d y⃗ = dxdy z.̂

• For the case of elementary volumes, if the unit vector of the elementary area and the
unit vector of height are orthogonal to each other, then the elementary volume vanishes.
In cartesian coordinate system, d a ⃗ ∥ z ̂ and hence dτ = d a ⃗ ⋅ dz z ̂ = da dz is maximum.

IIT Guwahati 6 PH102 (2025)

 ies of the region. In fact, there are two ways we could determine how much is
eing produced: (a) we could count up all the faucets, recording how much each
Volume Integral of a scalar eld
uts out, or (b) we could go around the boundary, measuring the flow at each
oint, and add it all up. You get the same answer either way:
! "
Let’s consider
(faucets withinathe
simple scalar
volume) (flow ϕ(x,
= field y, z) =the2(x
out through + y). Calculate
surface). the volume integral
of this scalar field over a cube with a unit length at the origin, as shown in the figure.
his, in essence, is what the divergence theorem says.
1.3 Integral Calculus

Here,
xample 1.10.the elementary
Check volume
the divergence is given
theorem thedτ
usingby = dx dy dz.
function
z (v)
(ii)
2
v = y x̂ + (2x y + z ) ŷ + (2yz) ẑ 2 1

∫v ∫∫∫
nd a unit
Wecube at the
need ϕ(x, y, z)
origin (Fig. 1.29).
to calculate dτ = 2(x + y) dx dy dz
olution
this case (i)
(iv) (iii)
The integration limits are simple and
∇ · v = 2(x + y), 1 y
are x = [0,1], y = [0,1] and z = [0,1] 1
nd
! !
x
1! 1! 1
2(x + y) dτ = 2 (x + y) d x d y dz, (vi)
V 0 0 0

! ! !
FIGURE 1.29
1 1 1
(x + y) d x = 12 + y, ( 12 + y) dy = 1, 1 dz = 1.
0 0 0
So much for the left side of the divergence theorem.
∫
hus,
∴ separately
integral we must consider ϕ(x, y, x)the
dτ six
= 2faces of the c
! v

∇ · v dτ = 2. ! ! 1 ! 1
IIT Guwahati V (i) 7
v · da = y 2 dy(2025)
PH102 dz = 1 .
fi
Volume Integrals of a scalar eld (example)
Let = 45x2 y and let V denotes a closed region boundedR by the planes 4x +
2y + z = 8, x = 0, y = 0, z = 0. Evaluate the integral V d⌧ .

Solution Keep x and y constant and integrate from z = 0 (base of the column
PQ in figure) to z = 8 4x 2y (top of the column PQ)
z
y

n 4x + 2y + z = 8
n

R y
a x P S
2

Next keep x constant and integrate w.r.t y. This amounts to addition of columns
having bases in the xy plane (z = 0) located anywhere from R (where y = 0)
to S (where 4x + 2y = 8 or y = 4 2x), and the integrations from y = 0 to
y = 4 2x.
IIT Guwahati 8 PH102 (2025)
fi
Volume Integrals
Finally add all slabs parallel to yz plane, which amounts to integration from
x = 0 to x = 2. z
Z Z 2 Z 4 2x Z 8 4x 2y
d⌧ = 45x2 ydzdydx
V x=0 y=0 z=0
Z 2 Z 4 2x
2
= 45 x y(8 4x 2y)dydx
x=0 y=0
Z 2
8 2 4x + 2y + z = 8
Here, xmin = ymin = zmin = 0 = 45 x (2 x)3 dx
x=0 3
For 4x = 8 − 2y − z, x is maximum Z 2
when y and z are minimum. = 120 x2 (2 x)3 dx Q
Therefore, xmax = 2. x=0

For a given x, then y has maximum

16
= 120 ⇥ = 128 R
When z is minimum in 2y=8-4x-z. 15 y
Therefore, ymax = 4 − 2x. x P S
For the given x and y, then z has a
maximum of z = 8-4x-2y.

Physically the result can be interpreted as the mass of the region V in which
the density varies according to the formula = 45x2 y.

IIT Guwahati 9 PH102 (2025)

 Problem 1.32 Check the fundamental theorem for gradients, using T = x +
Now we + 2yza3little
4x ymove further,
, the points a =by(0,an0,additional
0), b = (1,small
1, 1), displacement dl2 ;inthe
and the three paths incre-
Fig. 1.28
Important theorems in mental
vectorchange incalculus
T will be (∇T ) · dl . In this manner, proceeding by infinitesimal
(a) (0, 0, 0) → (1, 0, 0) → (1, 1, 0) → (1, 1, 1);
2
steps, we make the journey to point b. At each 2step we compute the gradient of T
(at that(b)
point)
(0, 0,and dot(0,
0) → it into
0, 1) the displacement
→ (0, 1, 1) → (1, dl. . .ythis gives us the change in T .
1, 1);
2
Evidently the total change in T in going from a to b (along the path selected) is
(c) the parabolic path xz = x ; y = x.
2

! b FIGURE 1.30
Fundamental theorem of gradients : (∇T ) · dl
z = T (b) − T (a). z (1.55)
z a

1.3.5
The Fundamental Theorem for Curls
This is the fundamental theorem for gradients; like the “ordinary” fundamental
The fundamental
theorem, it says theorem for curls,(here
that they integral which goesintegral)
a line by the special Stokes’
name of(here
of a derivative the
y y
theorem,
gradient)states thatby the value of the function at the boundaries (a and b).
is given
Geometrical
x (a) ! x Suppose you" wanted
Interpretation: (b) to
x determine the(c)height of
Fundamental theorem for Curls the Eiffel Tower. You could (∇ climb the stairs, using a ruler to measure the rise at
× v) · da = v · dl. (1.57)
each step, and adding them all upFIGURE 1.28left
(that’s the side of Eq. 1.55), or you could
Also known as Stokes’ theorem. place altimeters at the topSand the bottom, and P subtract the two readings (that’s
the right side); you should get the same answer either way (that’s the fundamental
theorem).
1.3.4 As always,
The the integral
Fundamental of a derivative
Theorem (here, the curl) over a region (here, a patch
for Divergences
Incidentally, as we found in Ex. 1.6, line integrals ordinarily depend on the
of surface, S) is equal to the value of the function at the boundary (here, the
path taken from atheorem
The fundamental to b. Butforthe
divergences
right side states
of Eq.that:
1.55 makes no reference to the
perimeter of the patch, P). As in the case of the divergence theorem, the boundary
path—only to the end points. Evidently, gradients have the special property that
term is itself an integral—specifically,
! a closed# line integral.
their
Fundamental theorem for divergences. line integrals are path independent:
Geometrical Interpretation: (∇ Recall that=the vcurl
· v) dτ measures the “twist” of(1.56)
· da. the
" b curl is a whirlpool—if you put a tiny paddle wheel
Also known as Gauss’s theorem vectors v;Corollary
a region 1: of high
V ) · dl is independent
(∇T S of the path taken from a to b.
there, it will rotate. Now,a the integral of the curl over some surface (or, more
or divergence theorem. the flux of2:the #curl
precisely,Corollary ) · dl =that
(∇Tthrough surface)
0, since representsand
the beginning theend
“total amount
points
of swirl,” and we can determine that just
are identical, and as wellTby
hence (b)going around
− T (a) = 0. the edge
# and
finding how much the flow is following the boundary (Fig. 1.31). Indeed, v · dl
is sometimes called the circulation of v.
These theorems are very important and have a wide range of applications
You may have noticed an 2apparent ambiguity in Stokes’ theorem: concerning
in Electrodynamics. Wethe Example
will discuss1.9. Let T = x y , and take point adetail.
to be the origin (0, 0, 0) and b the
boundary line them
integral,onewhich byway oneare in
we supposed to go around (clockwise
point (2, 1, 0). Check the fundamental theorem for gradients.
or counterclockwise)?
10 If we go the “wrong” way, we’ll pick up an overall sign
IIT Guwahati PH102 (2025)
error. The answer is that it doesn’t matter which way you go as long as you are
Fundamental theorem for gradients
• Scalar field (x, y, z).
z
• Start at a and move infinitesimal distance d~r1 along path C
to reach ~ra + d~r1, ~ra is position vector at a C
d~r1(t)
a
(x0 , y0 , z0 )
Recall from first lecture that the scalar function
will change by d = (r ~ ).d~r1 b y
(xn , yn , zn )

x
• Now, move a little further, by small displacement d~r2 .

• The change in ~ .d~r2 .

will be r

• In this manner, proceeding by infinitesimal steps we reach point b. At each

step computing gradient of (at that point) and dot it into the displacement.

• This gives the total change in as Fundamental theorem of calculus

Rbdf (x)
Recall dx = f (b) f (a)
Z b
a dx
~ ).d~r = (b)
(r (a) Fundamental theorem for gradients
a

Like the ordinary fundamental theorem of calculus, it says that the integral (here line integral) of a
derivative (here the gradient) is given by the value of the function at the boundaries (a and b)
IIT Guwahati 11 PH102 (2025)
Some important corollaries
? Gradients have the special property that their line integrals are path indepen-
Rb
dent: a r ~ .d~r is independent of the path from a to b.
H
? r ~ .d~r = 0, since the beginning and end points are identical and hence
(b) (a) = 0.
Assume path independence and consider the closed Hpath C in Figure 1. Since
~ .d~r = (b)
the starting point a and end point b are same, we get r (a) = 0.
y a=b

path independence =) line integral around closed path=0 C

x
H
Assume r ~ .d~r = 0 for any closed curve. If C1 and C2 are both paths between
a and b, then C1 C2 us is a closed path. So by hypothesis
I Z Z y b
~ .d~r
r = ~ .d~r
r ~ .d~r = 0
r C1
C1 C2 C1 C2
Z Z C2
=) ~ .d~r =
r ~ .d~r
r a
C1 C2
x
line integral around closed path=0 =) path independence
 Fundamental theorem for Curl (Stokes’ Theorem)
Let us find the circulation of a vector field F~ (x, y, z) around a closed curve C.
I Fx (x, y + y, z)
F~ .d~r y+ y

x, y, z)
Fy (x, y, z)
C
y
The fields in the x- direction at bottom and top are

Fy (x +
x
@Fx y
Fx (x, y, z) and Fx (x, y + y, z) = Fx + @y y. Fx (x, y, z)

The fields in the y- direction at left and right are x x+ x

@Fy
Fy (x, y, z) and Fy (x + x, y, z) = Fy + @x x.

Summing around from bottom in anti-clockwise manner

X
C= F~ . ~r = Fx (x, y, z) x + Fy (x + x, y, z) y Fx (x, y + y, z) x Fy (x, y, z) y
X ✓ ◆
~ @Fy @Fx ~ ⇥ F~ ). x yẑ
F . ~r = x y = (r
@x @y
= (r ~ ⇥ F~ ). ~a

This implies that curl can be defined as circulation per unit area...
IIT Guwahati 13 PH102 (2025)
 Fundamental theorem for Curl (Stokes’ Theorem)
Now, if we add these little elementary loops together, the internal line sections
cancel out because the d~r’s are in opposite directions, except on the bounding
line.

This gives the larger bounding contour.

I Z
Stokes’ Theorem: F~ .d~r = ~ ⇥ F~ ).d~a
(r
C S

Corollaries:

R
• (r~ ⇥ F~ ).d~a depends only on the boundary line, not on the particular surface
used.

IIT Guwahati 14 PH102 (2025)

Example
Verify Stokes’ theorem when S is the rectangle with vertices at (0, 0, 0), (1, 1, 0), (0, 0, 1),
and (1, 1, 1), and F~ = yz x̂ + xz ŷ + xyẑ.
z
Direct Method:
(0,0,1
H H (iii)
Line integral C F~ .d~r = C yzdx + xzdy + xydz
(1,1,1
over path (i) + (ii) + (iii) + (iv): (iv)

R
F~ .d~r = 0, since z = dz = 0 on (i). (0,0,0
(i) y
(ii)
R R1
~
F .d~r = 0 1.1dz = 1, since x = 1, y = 1, dx = 0 = dy. (i)
(ii)
R R R0 x (1,1,0
F~ .d~r = ydx + xdy = 1 xdx + xdx = 1, since
(iii) (iii)
y = x, z = 1, dz = 0.
R
(iv)
F~ .d~r = 0, since x = 0, y = 0, on (iv).
I Z Z Z Z !
F~ .d~r = + + + F~ .d~r = 0
C (i) (ii) (iii) (iv)
By Stokes’ Theorem :

x̂ ŷ ẑ Z
~ ⇥ F~ =
r @
@x
@
@y
@
@z = x̂(x x) + ŷ(y y) + ẑ(z z) = 0 ) ~ ⇥ F~ ).d~a = 0
(r
S
yz xz xy
 Fundamental theorem for Divergence (Gauss’s Theorem)

Consider a vector field F~ (x, y, z).

z F~ (x, y, z)
Say, flow of water...

We want to find out the

flux of F~ (x, y, z). z

To do that, take an infinitesimal aa =

d~ x xz ŷy
volume element like a cube as shown. y

..and look at the opposite x

faces (coloured in figure) first.
x y

The left most face (orange) is at a fixed value of y and the incoming flux into
this surface is F~ (x, y, z). ~a = F~ (x, y, z).n̂ a = (Fx x̂+Fy ŷ +Fz ẑ).( x z ŷ) =
Fy (x, y, z) x z.

The right most face (blue) is at a fixed value of y + y and the flux going out
of this surface is F~ (x, y + y, z). ~a = F~ (x, y + y, z).n̂ a = (Fx x̂ + Fy ŷ +
Fz ẑ).( x z ŷ) = Fy (x, y + y, z) x z.

IIT Guwahati 16 PH102 (2025)

 Fundamental theorem for Divergence (Gauss’s Theorem)
z F~ (x, y, z)
So, the outward flux from the blue face is

F~ (x, y + y, z). ~a = Fy (x, y + y, z) x z z

✓ ◆ a= x y ŷ
@Fy d~aa =
= x xz ŷy
= Fy + y x z
@y yy
x
Net flux out of the box through these two
y
opposite (orange and blue) faces xx
✓ ◆
@Fy @Fy @Fy
Fy x z + Fy + y x z= y x z= V⌧
@y @y @y

Net flux going out through the sides parallel to yz plane: ( @F

@x ) x y z
x

Net flux going out through the sides parallel to xy plane: ( @F

@z ) z y x
z

✓ ◆
@Fx @Fy @Fz
Summing over all the faces: F~ .n̂ a = + + ~ F~ ⌧
x y z = r.
all surfaces @x @y @z
Flux over a closed surface can bePwritten as aP sum over the surfaces of elemental
volumes that make the volume: F~ .n̂ a = (r. ~ F~ ) ⌧
Z I
In the limit x, y, z ! 0 ~ F~ )d⌧ =
(r. F~ .n̂da Gauss’s Divergence Theorem
V S
Fundamental theorem for Divergence (Gauss’s Theorem)
What does it mean?
We want to find the total outward flux of the vector field F~ (~r) across a surface
H
S that bounds a volume V: S F~ .d~a d~a

d~a
d~a
d~a is • normal to the local surface element

• points everywhere out of the volume

d~a

Gauss’s theorem tells us that we can calculate the flux of vector field across a
surface S., by considering the total flux generated inside the volume V.
Z I
Gauss’s Theorem: ~ F~ )d⌧ =
(r. F~ .n̂da
V S

Volume integrals are easier than the surface integrals: computational efficiency!
IIT Guwahati 18 PH102 (2025)
How to “see” this?
If we sum over the volume elements, this results in a sum over the surface
elements!

Note: if two elementary surfaces touch,

their d~a vectors are in opposite directions!

Therefore the d~a vectors cancel whenever

there are two surfaces in touch

Thus the sum over surface elements gives the overall bounding surface!
Z I
~ F~ d⌧ =
r. F~ .d~a
V Surface of V

Seems reasonable! Because, the “boundary” of a line are its endpoints, and
boundary of a volume is a closed surface!

IIT Guwahati 19 PH102 (2025)

Example
Verify the divergence theorem when F~ = xx̂ + y ŷ + z ẑ and S is the surface
composed of the upper half of the sphere of radius a and centred at the origin,
together with the circular disc in xy-plane centred at the origin and of radius a.
✓ ◆
@F x @F y @F z z
~ F~ =
r. + + =1+1+1=3 S1
@x @y @z n̂1
Z
2 3
) ~ ~
r.F d⌧ = 3.(vol. of hemisphere) = 3 ⇡a = 2⇡a3
V 3
To check the result, we need to calculate the surface
(0,0,0)
integral of F~ over the closed surfaces S1 and S2 . y
x ẑ S2
xx̂ + y ŷ + z ẑ xx̂ + y ŷ + z ẑ
Normal vector on S1 (hemisphere) : n̂1 = p 2 =
x + y2 + z2 a S2 n
0o
Normal vector on S2 (disc at the base): n̂2 = ẑ. z =
s e
a u
e c
Surface integral for the flux through S1 + S2 : 0b
=
I Z 2 2 2 Z Z
~ x +y +z
F .d~a = da + ( z)da = ada
S1 +S2 S1 a S2 S1

So, the value of the surface integral is a(area of S1 ) = a(2⇡a2 ) = 2⇡a3 .

IIT Guwahati 20 PH102 (2025)
Summary

Surface encloses volume Curve encloses surface Points enclose curve

Z Z I Z Z
F~ .d~a = ~ F~ d⌧
r. F~ .d~r = ~ ⇥ F~ ).d~a
(r ~ .d~r = (b)
r (a)
S V C S C

Remember: ✓ ◆
~ = @ @ @
In Cartesian Coordinates, with r x̂ + ŷ + ẑ
@x @y @z
x̂ ŷ ẑ
~ ~ @Fx @Fy @Fz ~ ⇥ F~ = @ @ @ ~ @ @ @
r.F = + + r @x @y @z r = x̂ + ŷ + ẑ
@x @y @z @x @y @z
Fx Fy Fz
IIT Guwahati 21 PH102 (2025)
 v · da = dx y dy = 4.
! ! & '
Take home exercises 1
0 3
2
0 1
2 1 1 3
The total flux is = z dz (1 − y) y dy = (9) = .
!
Question 1 : Calculate 2 0 integral
the volume 0 of the scalar field2ϕ(x, y, 12
z) = x y 28z 3 over the
prism as shownsurface v · da
in the = 16 + 0 + 12 − 12 + 4 = 20.
figure.
z
E 3
D
C
(c) Volume
Hint : UseIntegrals. A volume integral is an expression of the form
the constraint
!
(x+y)=1 along AB or CD
T dτ, (1.51)
V O 1
1 volumeBelement.yIn Carte-
where T is a scalar function and dτ is an infinitesimal A
sian coordinates,
34 Chapter
x 1 Vector Analysis
dτ = d x d y dz. (1.52)
Occasionally,
For example, if we
T iscome across
the density of the volume(which
a substance integrals
FIGURE mightof1.24
vector
vary from point to
functions
z
point), thenover a certain
the volume volume.
integral wouldThey aretotal
give the defined
mass. as
Occasionally we shall 2
encounter volume integrals of vector functions:
! ! ! ! !
v dτ = (vx x̂ + v y ŷ + vz ẑ)dτ = x̂ vx dτ + ŷ v y dτ + ẑ vz dτ ;
2 2
Problem 1.29 Calculate the line integral of the function (1.53) v = x x̂ + 2yz ŷ + y ẑ
2
from the (x̂,
origin
ŷ, andtoẑ)the point (1,1,1) by three different routes:
because the unit vectors
⃗ are
2 constants, they
Question 2: Given F (x, y, z) = y x ̂ + (2xy + z ) ŷ + 2yz z,̂2come outside the integral.
2
y
calculate the(a)
volume 0) → (1,
(0, 0,integral of it 0) →the(1,
0,over 1, 0)at→
cube the(1, 1, 1).
origin with x
sides of length 2 (as shown below)
(b) (0, 0, 0) → (0, 0, 1) → (0, 1, 1) → (1, 1, 1).
IIT Guwahati 22 FIGURE 1.30(2025)
PH102
Thank You