Unit I
tities are called Scalars. For example,
VECTOR
ANALYSIS
1 SCALARS AND
VECTORS
In physics, quantities can be classi­
fied into 2 main categories; Scalars
and Vectors. Certain quantities such
as mass, time, temperature, volume,
density etc. can be adequately repre­
sented just by a numerical value, an
amount or a magnitude. Such quan-
if the mass of an object is given as
2 kilograms, the value 2 denotes the
number of times the unit kilogram is
contained in that object.
On the other hand, certain other
quantities such as force, velocity,
displacement, acceleration etc. can­
not be adequately represented by a
numerical value alone. The only way
to meaningfully represent such quan­
tities is by considering them as hav­
ing a direction in addition to the mag­
nitude. Such quantities that have a
magnitude as well as a direction are
called Vectors. For example, if you
walk 2 miles north, then walk 2 miles
 --+
south, you are back at your original
The magnitude of a vector AB is de-
position i.e. your net displacement is
zero. Now instead had we just men­ noted as IABI_

tioned the magnitude and ignored B

the directions, there's no way of tell­
ing your net displacement, you could
have walked 4 miles straight or 4
miles left or whichever way possible. A

Vectors are represented by a directed

y
line segment, basically arrows. The
length of the line segment denotes
the magnitude and the arrowhead in­
dicates the direction of the vector. For
example, a vector from point A to an-
--+
Note that a vector is characterized
other point B is denoted by AB. Point
only by its magnitude and direction,
A is called the tail of the vector and
and not by its position in space. This
point B is called the head of the vector.
implies that as long as the magnitude
and the direction are unaltered, you
are free to move the vector around.
2 SCALAR
MULTIPLICATION
When a vector is multiplied by a posi­
tive scalar quantity (a positive num­
ber), its magnitude gets multiplied by
the scalar quantity and the direction
remains unaltered. Similarly, if a vec­
tor is multiplied by a negative scalar
quantity, its magnitude gets multi­
plied by the scalar quantity and its
direction reverses.
�
The product of a scalar x and vector A
is denoted as xA.
7
3 ADDITION
OF VECTORS
Vector addition isn't as straightfor­
ward as adding scalar quantities. In
case of vectors, the directions have to
be considered as well. Consider the
following example of a solid block
being applied 10 Newtons of force
from 2 directions as shown in the fig­
ure below.
10 N --•-•
10 N
In which direction do you think the
block will move as a result of these
2 forces? Of course, along the dot­
ted line. Now imagine that the block
is being pushed with 20 Newtons of
force from the bottom and 10 New­
tons of force from the side as before.
In which direction do you think the
block will move now? Common sense
says the block will move in a direction
slightly to the left of the dotted line.
What we just did with these 2 ex­
amples is Vector addition. Now let's
consider a general case where 2 vec­
tors are oriented at a certain angle
from one other as shown below.
 Repositioning the vector CD such that
by the vector AD. Now if you try the
the tail (C) of vector CD coincides with
the head (B) of vector AB
same, but by repositioning vector AB
instead of CD, you will get the exact
B
A B same result, proving that vector add­
ition is commutative.
A B
To find the sum or the resultant of
Repositiooing the vector.AB such that

2 vectors, reposition any one of the the tail.(A) of v¢c\'Or·AB coincicles with
the. head (DJ of vector co

vectors such that the tail of the one

B

vector coincides with the head of the

other vector, then the line segment
that completes the triangle denotes We can extend this idea to any no of
the resultant vector. This is called the vectors and this rule is known as the
triangle law of vector addition. In the Polygon law of vector addition.
figure above, we repositioned vector
CD such that the tail (C) of vector CD
coincides with the head (B) of vec­
tor AB. Hence the resultant is given
 cance of the Unit vector is that all vec­
tors in a specific direction are scaled
Repositio.niflg \lectors one
after the other
,.
versions of the unit vector in that dir­
:� /;
A
I ./
;'c
~ ection.
A B
A unit vector in the direction of a vec-

tor A, denoted by A can be obtained as,

4 UNIT VECTOR

Along a particular direction there can S RESOLUTION

be infinite no. of possible vectors with
all of them differing in magnitudes OF VECTOR
from one another. But in any direc­
tion, there can only be a single vector
One big advantage of using vectors is
of unit magnitude, such a vector is
that they can be resolved into any no.
called the Unit vector. The signifi-
of components. In the simplest case, a
vector can be resolved into 2 compo­
y
nent vectors lying in the same plane.

Shown below is a vector OA, if we draw

v[Zj
its projections onto any 2 perpendicu­
lar axes in its plane (for convenience
we have picked the x and y axes), we 0
----,::::j·---�-·
U X
� �

get a set of 2 new vectors ou and ov. We can take this one step further and

These 2 new vectors are called the resolve a vector into 3 components
� in 3 dimension. W hat's the advantage
components of vector OA. of doing this? For one, resolving vec­
tors into perpendicular components
allows us to express them in terms of
the Cartesian coordinates, which we
are familiar with.
Consider a vector OB as shown in the OG which lies along the x, y and z axes

figure below. If we resolve this vector respectively.

l:�
into 2 perpendicular vectors, we get 2

component vectors OG and ov. Vector

G ........
.... ...,
... B
OG is along the z- axis, so we'll leave

it as such. But vector ov on the other

hand lies in the x-y plane, so we to
need to resolve it further. Now if we

consider ov as a separate vector and

further resolve it, we get its 2 com-

ponents OE and OF_ This way we have

resolved the vector OB into 3 mutu-

ally perpendicular vectors OE, OF and

Any vector can be resolved into com­
ponents along the coordinate axes in
this manner. By doing this, we have
managed to generalize vectors a lit­
tle bit. What if we somehow incul­
cate the idea of unit vectors from last
section into this whole thing, Can we
generalize vectors even further?
The idea is that the unit vector is
the most basic vector along a direc­
tion and every other vector along that
same direction is just a scaled version
of that unit vector. So if we denote the
" A
unit vectors along x, y and z axes as ax,
" "
aY and az respectively, then any vector
A can be expressed as:
A = (Magnitude in x direction) x ax
+ (Magnitude in y direction) x av
+ (Magnitude in z direction) x az
Math books usually denote the unit
vectors along the 3 axes as i, J and k,
but since these notations have other
meanings in electrical engineering,
" "
the usual practice is to go for ax' aY
"
and az ·
As a consequence of this generaliza­
tion, we can denote a vector simply by
using 3 no's. For example, (5, 2, -4) de-
A A
notes the vector 5 ax + 2 aY - 4 a z· Simi-
A
larly, (0, 1, 2) denotes the vector aY + 2
az · In the later example the x compo-
nent is absent, meaning the vector lies
in the y-z plane.
The magnitude of a vector can be ob­
tained from its components as:
Magnitude of a vector= {Magnitude in x direction)2
+ {Magnitude in y direction)2
+ {Magnitude in z direction)2
This result can be easily obtained by
using the Pythagoras theorem.
6 DOT PRODUCT
Vectors can be multiplied in 2 possible
ways; the Dot product and the Cross
product. The dot product of 2 vec-
tors results in a scalar quantity and
the cross product of 2 vectors results
in another vector. For this reason, the
dot product and the cross product are
also known as the scalar product and
the vector product respectively.
The dot product between 2 vectors is
denoted as A B (read as A dot B). Math­
ematically, the dot product is defined
as:
---+ ---+
A. B = IAIIBI cos 0
IAI = Magnitude of A
-
1B1 = Magnitude of B - -+
cos 0 = cos of the angle between A and B
 -
-B
A
:□
B
......... A

A.B = IAIIBI cos 0

= IAIIBI

A-...., □L B
A

The dot product can be intuitively

_/
B
__ A.B = IAIIBI cos 0
Value between 0 & IAIIBI

thought of as a measure of the simi­

B
larity of two vectors or how well they
f_
l-
A
work together with one another. Con­
sider our "forces on a block" example A.B = IAIIBI cos 90
again.
- -B
=0
B

A 4

�o B
+--

A.B = IAIIBl cos 180

=-IAIIBI
In the first figure, the forces A and B
are in the same direction and hence
they are working together to move
the block. Therefore, the dot product
is maximum in this case. In the sec­
ond figure, the force B is applied at an
angle 8 to the force A, this obviously
is not the best method to move the
block. The force B is not contributing
to the motion of the block as much as
it did in the first case, hence the dot
product is a smaller value, but greater
than zero. In the third figure, the force
B is applied orthogonally to the force
A, so it doesn't contribute to moving
the block (in the direction of A) at all,
hence the dot product here is zero.
In the fourth example, the force B is
applied in opposite direction to force
A, which means force B is not only
not contributing to the motion of the
block, but it's actually negating the
effect of force A. So, the dot product in
this case is negative.
Properties of Dot Product:
• Commutative property: u.v =
v.u
• Distributive property: u.( v +
w) = u.v + u.w
• (u + v).(w + z) = u.w + u.z + v.w
+ v.z
 • k (u.v) = (ku).v = u.(kv), where The second way of multiplying 2 vec­
tors is called the Cross product. As
k is a scalar
mentioned earlier, the cross product
. u.u = lul 2 of two vectors results in another vec­

· ax. ax = ay. ay = az. az = 1 tor.

• ax. ay = ay. az = az. ax = 0 Mathematically, the cross product is

defined as:
---+ �

Example: Ax B= IAI I Bl sin 0 A

-
--
IAI = Magnitude of A

A.B = ( ax + 2ay - 3a 2 ).( 3ax +Say + 7a2 )

IBI = Magnitude of B
- -
sin 0 = sin of the angle between A and B
n = Unit vector perpendicular to both the vectors
= (ax. 3a x + 2a y.Say - 3a 2 • 7a 2 )
= 3 +10-21
= -8 Defining the cross product isn't as
straightforward as with the dot prod­
uct. Given 2 vectors, the cross prod­
7 CROSS PRODUCT uct of the two vectors represents

the area of the parallelogram formed
with these 2 vectors taken as adjacent
sides. And the direction of the cross
product vector is normal to plane of
this parallelogram. The cross prod­
uct can be intuitively thought of as
a measure of the orthogonality of 2
vectors. Closer the angle between the
vectors is to 90 degrees, larger the area
of the parallelogram formed and cor­
respondingly larger the cross product
vector.
--+ -
AxB
--------------------------------�,
,,,
,, ,,,'
, ,,
,,,,'
, ,
�=--===----====;;;..__---->
A
Now there's another small problem,
given 2 vectors A and B, there are
two possible directions that the cross
product vector could point to, up­
ward and downward. This is where
the right hand thumb rule comes in,
which states that "If you curl your fin­
gers of your right hand in such a way
that the index finger points in the dir­
ection of vector A and middle finger
points in the direction of vector B, then
the thumb points in the direction of
AxB". Now if you try the same for BxA, • Distributive property: u x ( v +
the thumb will point in the down­
w) = (uxv) + (uxw)
ward direction i.e. BxA = -AxB.
• k (uxv) = (ku)xv = ux(kv),
-+ -+ where k is a scalar
AxB
• uxu = O

-+

=a
y
-+
B

az = -ay

Example:
Properties of Cross Product:

. uxv = -vxu
A =a x + 2a y - 3az,B = 3a x + Sa y +7az
AX 8 = ( ax + 2a y - 3az) X ( 3ax + Sa y +7az)

ax ay az Vectors can be arranged

as following and the
= 1 2 -3 Determinant gives the
3 5 7 Cross product

= a x (14 +15) - a y {7+9) + az {5-6)

= 29a x - 16a y -az
 it comes to practical application. In
this chapter we'll briefly introduce 2
other coordinate systems: Cylindrical
CURVILINEAR and Spherical. However, for simpli­
city we'll mostly stick with using

COORDINATE cartesian coordinates throughout this

book.

SYSTEMS
1 CYLINDRICAL
So far in this book, while defining COORDINATE SYSTEM
vectors and other stuff, we used car­
tesian or rectangular coordinate sys­
In Cylindrical coordinat e syst em, a
tem. While cartesian coordinate sys­
point is space is specified in terms of,
tem is the simplest and the most
popular coordinate system, these are • the distance of the point from the
not the most convenient to use when z-axis (P)
• the angle a half plane containing z
the point makes with the x-axis
in the anticlockwise direction
(8). This angle is called the
azimuth.
• the distance of the point from the
xy plane (z) (same as in cartesian 0
z
t--.._.---1► Y
coordinates)

In any coordinate system a point can

be defined by the intersection of 3
orthogonal surfaces. In case of carte­
sian coordinates, these surfaces were
3 planes. For cylindrical coordinate
system these surfaces are a cylinder,
a plane and a half plane as shown to 00
, the azimuth angle can be varied
below.
from O to 2 n and the distance from
the xy plane can be varied from - 00 to
00
p = constant

z = constant Cylindrical to Cartesian co­

ordinates:

To cover the entire space, the radius

of the cylinder P can be varied from 0
 z z
Unit vectors

ap= cos0 ax + sin0 av

p (p, 0 J z) ae = -sin0 ax + cos0 av
A A

az = az

X
2 SPHERICAL
Cylindrical to Cartesian
COORDINATE SYSTEM
x = p cos0 p = -vx2 + y2
y = p sin0 4 a. 0 = tan- 1 (y/x)
z=z z=z In Spherical coordinate system, a
point is space is specified in terms of,
Vectors in Cylindrical coord­ • the distance of the point from
inates: the origin (r)
• the angle the line joining the z
point to the origin makes with
the z- axis (8).
• the angle a half plane contain­
P(r, 0,¢)
ing the point makes with the
x- axis in the anticlockwise
direction (<P) (same as the
azimuth angle in cylindrical
coordinates)

In spherical coordinate system, a

point is defined by the intersection of
a sphere, a cone and a half plane as
shown below.
 z 0 to 2 rr radians.
0 = constant

Spherical to Cartesian coord­

inates:
r = constant
z

P(r,0,cp)

To cover the entire space, the radius of

the sphere (r) can be varied from Oto X
00
, the slant angle of the cone (0) can Spherical to Cartesian
be varied from Oto II radians, and the x = r sin0 cos</> r = ✓x2 + y2 + z2
azimuth angle (Q>) can be varied from y = r sin0 sin</> 4 1111 0 = tan- 1( ✓x2 + yz /z)
z = r cos0 </> = tan- 1 (y/x)
Vectors in Spherical coordin­
ates:

z Unit vectors
a r = Sin0 COS</> a X + Sin0 Sin¢ aV + COS0 a l
a0 = COS0 COS</> a x + COS0 Sin¢ aV - Sin0 a l
a¢= -sin¢ a x + cos¢ av
Position vector

X
 sociates a scalar value or a magnitude
VECTOR to every point in space.

CALCULUS Example: Distribution of tempera­

ture in a room. If Temperature in a
room is given by scalar field T = xy 2 z 3 ,
1 SCALARAND then at point (1,1,1) the temperature
is 1 unit and at another point (2,3,1),
VECTOR FIELDS
the temperature is 18 units and so on

What is a field? A field is a physical

quantity that can be specified every­
where in space as a function of posi­
tion (x, y and z coordinates). And
there are basically 2 types of fields:
Scalar and vector fields.
A scalar field (or a scalar function) as-
 Example: Velocity of flow at different
points in a fluid. If velocity of flow is
A A A

given by V = xy ax + xy 2 z ay - z 3 az,
then at point (1,1,1) the velocity is de-
A A A

noted by the vector ax + aY - az· At an-

A

other point (1,2,3), the velocity is 2 ax

A A

+ 12 ay_27 az and so on.

-0.382 +0.459
3 //Ill I I�\\\\\"�
///II I I!\,\\\.""
Note that it is not necessary for a sca­ 2 ////I I I
/.///II I
l \ \ \'\�""
I\\\ ,'X""-"­
..,,,,,-,-,, / / I I I \ , ,.....,_....,._......._
lar field to have non zero values at
every point in space. ....._....._.._ -- .... "

......_,, '- ' I / / //,,,,,

\ I I I
'-'-'-'\\\II//////
2
Similarly, a vector field (or a vector - , "- '- , \ \ \ t I I I / / / /
'-''-\\ \ \ t t II/// I
function) is a field in which a vec­ -3 ,, \ \ \ \ \ t f I I I I//
-3 -2 -1 0 1 2 3

tor can be assigned to every point in

space.
 of a function f with respect to x, y & z
af af af
2 PARTIAL -- -
are denoted as ax, ay & az respectively.

DERIVATIVE & THE By considering the other variables as

V OPERATOR constant, we are essentially reducing

For single variable functions, the de­
rivative measures the change in the
function value with respect to the
the multi variable function to a sin­
gle variable function. And therefore,
the partial derivative measures the
change in the function value with re­
spect to any one variable.

I
independent variable. The partial de­
rivative is the equivalent of the ordin­ )

ary derivative for multi variable func­

·.· ,,
" ,:;:J�t...
f(!<;y f{x,y

tions. It is basically the derivative of )

. .

the function with respect to any one y

l�nsta'"
: � X
y = constant ·i(
of its variables with the other vari­
ables held constant. Partial derivative
Using partial derivatives, we can 3 GRADIENT
define a new operator called the del
or the nabla operator, denoted by the
symbol V. In Cartesian coordinates, A multi variable function has mul­
the del operator is defined as: tiple derivatives at every point, one
t"7 a a a with respect to every variable it de­
= ax aX + ay ay+ az az
A A A

V pends on. For example, a function f(x,

V isn't really a vector, it is rather
a differential operator. When applied
to a single variable function, it is the
same as the standard derivative. But
for a multi-variable function, it can be
applied in 3 different forms (gradient,
divergence, curl) and each of these
forms has a different significance.
y) has 2 derivatives iJf/ iJx and iJf/
iJy at every point (shown in the fig­
ure below). Each of these derivatives
correspond to the rate of change of
the function with respect to a particu­
lar variable. The resultant or the sum
of these derivatives denotes the max­
imum rate of change of the function
when all the variables are considered
all at one. That is the gradient.
In other words, the Gradient of a
Vf(x,y, z ) = X :� + Y :� + i :�
multi variable function is a vector
that points in the direction of greatest Example: f = 2x + yz
increase (steepest slope) of the func­
tion at a point. It is denoted by symbol
V f. The gradient is analogous to the
slope for single variable functions.
f(x, y) So at a point say P(5,-3, 9), the gradi­
ent or the max slope is denoted by the
,.. ,.. ,..

vector 2 ax + g ay_ 3 az·

In another way, the gradient can be
thought of as an operator that con­
y
X verts a scalar field (scalar function)
into a vector field (vector function).

The gradient can be calculated as:

 Scalar field f Gradient Vector field l7f
form of the gradient. On a vector field
the del operator can be applied in 2
ways, first of which is the Divergence.

Divergence is nothing but the dot

product between the del operator and
a vector field.
Properties of Gradient:
• V( u + v) = Vu + Vv
• V(uv) = (Vu)v + (Vv)u
• k Vu= V(ku) ,where k is a scalar quantity
Intuitively, the Divergence represents
the outward flow of a vector field
3 DIVERGENCE from an infinitesimal volume at a
given point in a vector field. In other
We saw how the del operator can be words, divergence is a measure of the
applied on a scalar function in the extent to which a point (which is es-
sentially a tiny volume) behaves as a
source of the vector field.
To understand the concept of diver­
gence better, imagine a vector field as
a fluid flow as shown below.
------ '\ t'\ '
ume. For example, at point A, all the
field lines are pointed away from the
volume, which means point A is act­
ing as source of the flux, therefore the
divergence at that point is a positive
value. At point B, some of the field
,--.,,,-,,;/' I' /' /' /

- - -- - ----­ ----
- --- - ' 1' '\ t _.,. /' I I' /1 / --­ lines are flowing into the volume and
-- --.,..__�---- ' iA .r / 1 I' /" ------
---- ---- ---- ----.. - i 1 ,/' --- - __,....._. -
""----- ---.... some are flowing out of the volume,
--- --- ---
...__ � --
..:..__... -.-.B-� -�-..... but because there are more outward
--
ITT-
flowing field lines, the net outward
-- ---
---- ----
� \ '-... ----... ----...:
flow is positive, therefore divergence
t \, "'-! '-... "---....
is positive at point B as well (but it has
less magnitude compared to point A).
Now if you consider a small spherical At point C, there are equal no. of field
volume, the difference between the lines flowing into the volume as there
outward flow and the inward flow i.e. are field lines flowing out of the vol­
the net outward flow gives the diver­ ume. Hence the divergence at point C
gence of the flow in the small vol- 1s zero.
The above example is only for better r7-v =(a"a + a ,..
+
a" "'
)( a + yz
"'
+ 3xz
"'
a 2)
V • ax x ay a V az a 2 • X x aV
understanding of the concept, in real­ acx) acyz) ac3xz)
=-+--+--
ax ay az
ity the divergence has nothing to do
with the no. of field lines entering or = 1 +z +3x
exiting the volume, it has more to do The divergence operator converts a
with the magnitude & direction of the vector field into a scalar field.
field lines. Vector field V Divergence Scalar field V. V

Positive _Q,ivergence

\l/
17. V > O

.,_..0---+
Zero Qivergence
17. V = O

,,.!/
--.O---+
Negative_2ivergence

\!/
17. V < O

..
/i\ /!\ /f\
--.Q.__

" "
Properties of Divergence:
Example: V = x ax+ yz ay + 3xz az • V.(A + B) = V.A + V.B
➔ - ➔ -

➔ ➔ ➔
• V.{fA)= JV.A+ A.VJ ,where/is a scalarfunction
 below. Now if you consider a small
spherical ball that is free is to rotate
4CURL
in any direction, it will rotate differ­
ently depending on its location in the
The Curl is the cross product between fluid. The field vectors acting on the
the del operator and a vector field. sphere determines both direction and
the speed at which it rotates. Magni­
tude of the curl vector denotes the
speed of rotation and the direction of
ax ay az
= a a a the curl denotes the axis of rotation of
ax oy oz
Vx Vv V2 the sphere.
= (a V2 _ a VY) a _ (a V 2_ a Vx) a + (a VY_ a V x ) a
OY OZ X OX OZ y OX oy Z

The curl of a vector field describes

--------------------------------------------------------------------------------
the rotational tendency of a vector
--------------------------------------------------------------------------------
field at a point in 3 d space. Consider
a fluid flow as shown in the figure
 the field, if any. If you think about
the motion of the earth around the
sun, the curl is analogous to the rota­
tion of the earth and not its revolution
around the sun.
,.
Example: V = x ax+ yz ay + 3xz az
a � a � a � � �
17
V x V = ( ax ax + a ay + az a2 ) X (X a x + yz ay + 3 XZ a� 2)
y
a x av az
a
- a
- a
ox oy oz
X yz 3xz

It is important to note that curl refers

to the microscopic rotation of the ball
= -yaX - 3z ay
at a point in the vector field (i.e. as if The divergence operator converts a
translational motion of the ball from vector field into another vector field.
the point is restricted) and not the
macroscopic circulation of the ball in
Properties of Curl: Line integral or the path integral is
➔ -+ ➔ -+ the integral of a function evaluated
• V X (A+ B) = (V X A)+ (V X B)
along a curve. For single variable
➔ ➔ ➔
• V X (fA) = f{V X A)+ VJ X A ,where/is a scalar function
functions, the line integral can only
be evaluated over a straight-line path

S MORE PROPERTIES (for obvious reasons) as shown.

height of each
f(x) B rectangle= f(x)
f t(x) dx
• v'.(v' X A ) = 0 (i.e. Divergence of Curl =O) ,,._.._
A \

l
• v' X (v'f) = 0 (i.e. Curl of Gradient= 0) .. It,,.

➔ -➔ -➔ ➔ ➔ -➔
• v'.(A x B) = B .(v' x A ) - A .(v' x B )
I
I
I
I
I
I
I
I
➔ ➔ ➔ I

• v' X (v' X A) = v'(v'. A) - v'2A

I

B dx

The line integral is evaluated by con­

6 LINE INTEGRAL sidering a bunch of small rectangular
strips of width dx and suitable height
(which will be equal to the value of
 f(x, y)
the function at the point i.e. f(x)), so
as to fill up the area. The area of an
,�
height of each
rectangle = f(x,y)

individual strip is therefore f(x) dx A' B'

(width x height) and hence the area

A'!'-:B'
of our interest is simply the inte­ :
:
:
: LJ,W,,1,,1,.,
B
�.------. y
gral (summation) of these small areas
I
I
I
A �I
B
; __,/
along the path. A VJ.di
dy

For multivariable functions, the line
integral can be evaluated along any 2-
dimensional path, not just a straight
line.
X
The line integral for multivariable
functions is evaluated in the same
manner as with single variable func­
tions, except that in this case, the
line element (width of the rectangular
strip) is no longer dx or dy, but a func­
tion of dx and dy.
In general, the line integral of a func­ To summarize, the line integral of a
tion f along a curve C can be evaluated scalar function is nothing but the area
as: of the fence created by a curve path
and its projection on the function.
ff
C
dl
Now let's move on to line integration
f(x,y) of a vector field. The line integral of a
vector field can be interpreted as the

J B
f(x,y)dy
amount of work that a force field does
on an object as it moves along a curve.
J
A
B
f(x,y)dx
A
When you try to move a block along
A': a curve C in space at constant speed
through a force field, a force always
I
I
I
I
I
I

acts on the block, making it easier or

I
I

------'----
i..,----;-'
' -...-�-----.....
B
I
harder to move the block depending
J
'7
: ,
'/ B
f(x,y)dl on the directions of the forces at a
A
point. If the force acts opposite to the
X
direction of the path, then you have to
do work to keep the block moving. On
the other hand, if the direction of the
force is in the direction of the path,
then your job becomes easier because
you're being aided by an outside force.
vector field and the differential curve
element integrated along the entire
curve gives the total work done.
F ➔ Vector field
dr ➔ Differential path element

z Line integral= f F. ch-

'/
--- --- B
7 SURFACE INTEGRAL
\ f

For scalar functions, the surface in­

\ A
Work done at a
point= F. cfi'. tegral is nothing but the volume en­
X closed by a surface and its projection
As mentioned in the previous chapter, on the function. The simplest case is
the dot product is a measure of how when the area is planar or 2-dimen­
well 2 vectors work with each other, sional as shown.
hence the dot product between the
 f(x, y) the surface into a bunch of curves or
lines, allowing us to use the line inte­
gral.

...
Area= J fdx

. I
.. y
I I

\J___J
Surface integral
�
= Volume enclosed

X
To find the surface integral in such a
case, the surface is sliced using planes
of small thickness either along the x­
axis or the y-axis (along the x-axis in
the figure). That way we can convert
The sliced area is given by the line in­
tegral,
ff dx
Here di = dx, since the lines are
straight and parallel to the x-axis. To
obtain the volume, all we need to do
is to combine these areas, which can then the volume corresponding to
be done by integrating them with re­ each differential area is the volume of
spect to dy. Therefore, the surface in­ the French fry shaped element (paral­
tegral is given by: lelopiped) above it. The volume of this

J f fdx dy
element is fdS and hence the volume
corresponding to the entire surface is
simply the integral of this differential
This equation is called the double in­ volume.
tegral.
JitdS
But not all surfaces are planar, some s
surfaces maybe spherical or conical or
anything else, in such a case we would
require a more generalized method
to find the surface integral than the
double integral. The idea here is to
......
divide the whole surface into small
differential surfaces each of area dS,
The surface integral of vector field
can be interpreted as the amount
of field flowing (think of it as fluid
flow) through a surface per unit time.
When the vector field is normal to the sur­
face, the flow will be maximum and when
the surface is parallel to the surface, there
will be no flow. So, to calculate the total Hence the surface integral of a vector
field flowing through the surface, all we field is given by:
need to do is to add up the component of
F ➔ Vector field
the vector field that are perpendicular to els ➔ Surface normal vector
the surface.
Surface integral= ff F. dS
the vector field that are perpendicular to
8 DIVERGENCE
the surface.
THEOREM

According to the Divergence Theorem

"The outward flux of a vector field
through a closed surface is equal to the
volume integral of the divergence of the
vector field over the region enclosed by
that closed surface 11 •

Hence the surface integral of a vector Mathematically, the Divergence the­

field is given by: orem can be written as:
F ➔ Vector field
dS ➔ Surface normal vector f ft. dS = f (v. ft) dv
S V

Surface integral= J Js F. dS It is also known as Gauss's theorem or

Ostrogradsky's theorem.
Although the statement and the or negative or zero. In our figure, we
mathematical formula may look com­ have shown the divergence at 3 points
plicated, the intuition behind this A, B and C. Now if we integrate (sum
theorem is pretty straightforward. up) the divergence at all such points
throughout the volume, what we get
Consider any 3 dimensional object
is the total divergence or the total
placed a vector field. We have gone
source like nature of the object as a
with a potato shaped object as shown
whole, which is essentially how much
in the figure. Now let's consider the
field originates from the object. This
right hand side of the equation first.
is exactly the same quantity we obtain
We have already learnt that the di­
by taking the surface integral of the
vergence of a vector field is measure
field across the full surface.
of the outward going or the source
like behavior at a point in the field.
As this object under our consider­
ation is placed a vector field, every
point (small volume) inside the object
will have a divergence, either positive

J (17.F) dv
The Divergence theorem in layman's
terms can be stated as "The total out­
ward field flowing out from the surface
of an object is the equal to the sum of
the fields flowing out from every single
point inside the volume of the object".
What this theorem (and the one in
the next section) really does is find a
correlation between the smaller phe­
nomenon that occurs on inside of an
object to the larger phenomenon on
the outer periphery. But this is not
the reason we engineers use the diver­
gence theorem, for us the divergence
theorem provides a tool for convert-
--------------------------------------------------------------------------------
ing surface integrals which are often
difficult to compute into an easier vol-
--------------------------------------------------------------------------------
ume integral. This is especially use-
ful when we deal with some familiar
shape or if the divergence results in a
simple function.
An important thing to be noted is that
the divergence theorem is only applic­
able to closed surfaces.

9 STOKE'S THEOREM
The Stoke's theorem states that "the
line integral of a vector field round
that a closed path is equal to the sur­
face integral of the curl of field over
any surface bounded by that closed
path".
Mathematically, the Stoke's theorem
can be written as:
fft.ciJ = f(v x ft) .dS
C S
The idea behind the Stoke's theorem
is quite similar to the divergence
theorem. Consider a planar surface
placed in a vector field, there will be a
rotational effect at every point on the
surface which is given by the curl of
the field. Now if we integrate (sum up)
the curl at all points throughout the
surface, what we get is the total rota­
tional effect along the periphery due
to the vector field. This is the same
as the line integral of the vector field
along the outer boundary.
The reason why this is true can be
better understood with the help of the
figure below. If you drop a ball near
to the boundary close to point A, the
ball will rotate in clockwise direction
due to the curl at the point, given it's the vector field is just acting along the
acting in isolation. But if you consider periphery, not throughout the sur­
the curl at all points on the surface, face.
a ball dropped near point A can no
longer keep spinning at A, the curl at
point B will inevitably push the ball
from point A to point B and then it'll
move from point B to point C and
so on. In effect the ball dropped at f(v x ft). cis
the boundary will only keep moving
along the boundary and not the inner
surface. The curls on the surface can­
cel each other out, making it seem like