0% found this document useful (0 votes)

14 views

Vector Analysis SuccessClap

Uploaded by

Kallyan Vvp

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

14 views

Vector Analysis SuccessClap

Uploaded by

Kallyan Vvp

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 246

SuccessClap

UPSC MATHEMATICS STUDY MATERIAL

BOOK- 05 Vector Analysis
Updated Version 2.0 (Dated 25-09-2020) (Download from Website)

-----------------------------------------------------

SuccessClap Android App

STUDY, DISCUSS & CLARIFY YOUR
DOUBTS directly with US
Easy to Read and Take Notes
Complete FREE Study Material , Question Bank in our App
Download Android App: Click Here

------------------------------------------------------------
VIDEO COURSES for UPSC MATHEMATICS
Crash Course for UPSC Mathematics 2020
22 Weeks Course for Working Professional
PREVIOUS YEAR VIDEO & TEXT SOLUTIONS 1992-2019
Question Bank Solutions
Full Length Tests, Topic wise Tests
-------------------------------------------------------------------------
For Exam Related Info and Guidance WhatsApp: 9346856874
Table of Contents
E_Vector Analysis 2
01 Curves in Space 104
02 Curvature and Torsion 139
B.Sc Vectors 191
Vector Calculus
9.0 INTRODUCTION
In Science and Engineering we often deal with the analysis of forces and velocities and other
quantities which are vectors. These vectors are not constants but vary with position and time. Hence,
they are functions of one or more variables.
Vector Calculus extends the concepts of differential calculus and integral calculus of real functions
in an interval to vector functions and thus enabling us to analyse problems over curves and surfaces in
three dimension. Vector Calculus finds applications in a wide variety of fields such as fluid flow, heat
flow, solid mechanics, electrostatics etc.
In Vector Calculus we deal mainly with two kinds of functions, scalar point functions and vector
point functions and their fields.

9.1 SCALAR AND VECTOR POINT FUNCTIONS

Definition 9.1 If to each point P( r ) (the point P with position vector r ) of a region R in space there
is a unique scalar or real number denoted by f( r ), then f is called a scalar point function in R.
The region R is called a scalar ﬁeld.

Definition 9.2 If to each point P( r ) of a region R in space there is a unique vector denoted by F( r ),
then F is called a vector point function in R. The region R is called a vector ﬁeld.

Note
1. In applications, the domain of definition of point functions may be points in a region of space,
points on a surface or points on a curve.
2. If we introduce cartesian coordinate system, then r = xi + y j + zk or
r = ( x, y, z ) and instead of F( r ) and f( r ) we can write
F( x, y, z ) = ( F1 ( x, y, z ), F2 ( x, y, z ), F3 ( x, y, z )) or
F( x, y, z ) = F1 ( x, y, z )i + F2 ( x, y, z ) j + F3 ( x, y, z )k
and f( r ) as f( x, y, z )
3. A vector or scalar field that has a geometrical or physical meaning should depend only on
the points P where it is defined but not on the particular choice of the cartesian coordinates.
In otherwords, the scalar and vector fields have the property of invariance under a transformation
of space coordinates.

Examples of scalar field

1. Temperature T within a body is scalar ﬁeld, namely temperature ﬁeld.
2. When an iron bar is heated at one end, the temperature at various points will attain a steady state
and the temperature will depend only on the position.

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 1 5/27/2016 12:03:05 PM

SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
9.2 ■ Engineering Mathematics

3. The pressure of air in earth’s atmosphere is a scalar field called pressure field.
4. f( x, y, z ) = x 3 + y 3 + z 3 − 3 xyz defines a scalar field.
Examples of vector field
1. The velocity of a moving fluid at any instant is a vector point function and defines a vector field.
2. Earth’s magnetic field is a vector field.
3. Gravitational force on a particle in space defines a vector field.
4. F ( x, y, z ) = x 2 i − y 2 j + zk defines a vector field.

Note Vector and scalar functions may also depend on time or on other parameters.

Definition 9.3 Derivative of a Vector Function

f ( t 1 Δt ) 2 f ( t )
A vector function f (t ) is said to be differentiable at a point t, if lim exists.
Δt → 0 Δt
df
Then it is denoted by or f ′ and is called the derivative of the vector function f at t.
dt
Note
1. If f (t ) = f1 (t )i + f 2 (t ) j + f 3 (t )k then f (t ) is differentiable at t if and only if its components f1(t),
d f (t )
f2(t), f3(t) are differentiable at t and = f1′(t )i + f 2′ (t ) j + f 3′(t )k
dt
df d2 f
2. If the derivative of w.r.to t exists, it is denoted by 2 . Similarly, we denote higher derivatives.
dt dt
dc
3. If c is a constant vector, then = 0.
dt
dc
For c = c1 i + c2 j + c3 k and = 0i + 0 j + 0 k = 0.
dt

9.1.1 Geometrical Meaning of Derivative

Let r (t ) be the position vector of a point P with respect to the origin O.
As t varies continuously over a time interval P traces the curve C. Thus, the vector function r (t )
represents a curve C in space.
Q
Let r and r + Δr be the position vectors of neighbouring
points P and Q on the curve C.

Then PQ = OQ − OP
= r + Δr − r
= Δr
Δr
∴ is along the chord PQ.
Δt
P
Δr dr dr
If lim exists, it is denoted by and is in the
Δt → 0 Δt o
dt dt C
directing of the tangent at P to the curve.
Fig. 9.1

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 2 5/27/2016 12:03:13 PM

Vector Calculus ■ 9.3

dr dr
If ≠ 0, then or r ′(t ) is called a tangent vector to the curve C at P.
dt dt
r ′ (t )
The unit tangent vector at P is = = u(t ).
r ′ (t )
Both r ′(t ) and u(t ) are in the direction of increasing t. Hence, their sense depends on the orientation
of the curve C.

9.2 DIFFERENTIATION FORMULAE

If f and g are differentiable vector functions of t and f is a scalar function of t then
d d f dg d d f df
1. ( f ± g) = ± 2. (ff ) = f + f
dt dt dt dt dt dt
d dg d f d dg d f
3. ( f ⋅ g) = f ⋅ + ⋅g 4. ( f × g) = f × + ×g
dt dt dt dt dt dt
d df dg dh
5. ( f ⋅ g × h) = ⋅g×h + f ⋅ ×h + f ⋅g× .
dt dt dt dt

d f d f ds
Note If f is a continuous function of a scalar s and s is a continuous function of t, then = .
dt ds dt
6. Let f (t ) be a vector function. f (t ) changes if its magnitude is changed or its direction is changed
or both magnitude and direction are changed. We shall ﬁnd conditions under which a vector
function will remain constant in magnitude or in direction.
(i) Let f (t ) be a vector of constant length k.
2
Then f ⋅ f = f = k2
Differentiating w.r.to t, we get

df df df df
⋅f +f⋅ = 0 ⇒ 2f ⋅ =0 ⇒ f⋅ =0
dt dt dt dt
df df
∴ = 0 or = is ⊥ to f ⋅
dt dt
(ii) Let f (t ) be a vector function with constant direction and let a be the unit vector in that
direction
Then f (t ) = fa, where f= f
df df da
∴ = a+f .
dt dt dt
da
But a is a constant vector, since its direction is ﬁxed and magnitude is 1. ∴ =0
dt
df df
∴ = a
dt dt
df df df
Now f × = fa × a=f a×a = 0 ( a × a = 0)
dt dt dt
df df
∴ = 0 or is parallel to f .
dt dt

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 3 5/27/2016 12:03:24 PM

9.3 LEVEL SURFACES

Let f be a continuous scalar point function deﬁned in a region R in space. Then the set of all points
satisfying the equation f (x, y, z) = C, where C is a constant, determines a surface which is called a
level surface of f. At every point on a level surface the function f takes the same value C. If C is an
arbitrary constant, the for different values of C, we get different level surfaces of f.
No two level surfaces intersect. For, if f = C1 and f = C2 be two level surfaces of f intersecting at a point P.
Then f(P) = C1 and f(P) = C2 and so f has two values at P which contradicts the uniqueness of value of the
function f. So, f = C1 and f = C2 do not intersect.
Thus, only one level surface of f passes through a given point
For example, if f (x, y, z) represents the temperature of (x, y, z) in a region R of space, then the
level surfaces of equal temperature are called isothermal surfaces.

9.4 GRADIENT OF A SCALAR POINT FUNCTION OR GRADIENT OF A SCALAR FIELD

9.4.1 Vector Differential Operator
The symbolic vector is called Hamiltonian operator or vector differential
operator and is denoted by ∇ (read as del or nabla).

∴ .

It is also known as del operator. This operator can be applied on a scalar point function f (x, y, z)
or a vector point function F( x, y, z ) which are differentiable functions. This gives rise to three ﬁeld
quantities namely gradient of a scalar, divergence of a vector and curl of a vector function.

Definition 9.4 Gradient

If f (x, y, z) is a scalar point function continuously differentiable in a given region R of space, then the
∂f ∂f ∂f
gradient of f is deﬁned by ∇f = i +j +k .
∂x ∂y ∂z
It is abbreviated as grad f. Thus, grad f = ∇f.

Note Since ∇f is a vector, the gradient of a scalar point function is always a vector point function.
Thus, ∇f defines a vector field.
Gradient is of great practical importance because some of the vector fields in applications can be
obtained from scalar fields and scalar fields are easy to handle.

9.4.2 Geometrical Meaning of =f

Let f (x, y, z) be a scalar point function. Let f (x, y, z) = C be a level surface of f. Let P be a point
on this surface with position vector r = xi + yj + zk .
Then the differential d r = dx i + dy j + dz k is tangent to the surface at P.
⎛ ∂f ∂f ∂f ⎞
Now ∇f ⋅ d r = ⎜ i +j +k ⋅ ( dx i + dy j + dz k )
⎝ ∂x ∂y ∂z ⎠⎟

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 4 5/27/2016 12:03:26 PM

Vector Calculus ■ 9.5

∂f ∂f ∂f
= dx + dy + dz = df = 0 [{ f = C ]
∂x ∂y ∂z
∴ ∇f is normal to the surface f (x, y, z) = C at P.
∇f
So, a unit normal to the surface at P is n =
∇f
∇f
There is another unit normal in the opposite direction = − .
∇f

9.4.3 Directional Derivative

The directional derivative of a scalar point function f in a given direction a is the rate of change of f
in that direction. It is given by the component of ∇f in the direction of a
a
∴ the directional derivative = ∇f ⋅ .
a
a ∇f a
Since ∇f ⋅ = cos u, where u is the angle between ∇f and a.
a a
= ∇f cos u
So, the directional derivative at a given point is maximum if cos u is maximum.
i.e., cos u = 1⇒ u = 0.
∴ the maximum directional derivative at a point is in the direction of ∇f and the maximum
directional derivative is ∇f .
Note
1. The directional derivative is minimum when cos u = −1 ⇒ u = p
∴ the minimum directional derivative is − ∇f
2. In fact, the vector ∇f is in the direction in which f increases rapidly.
i.e., outward normal and −∇f points in the direction in which f decreases rapidly.

9.4.4 Equation of Tangent Plane and Normal to the Surface

(i) Equation of tangent plane
Let A be a given point on the surface f(x, y, z) = C.
Let r0 = x0 i + y0 j + z0 k be the position vector of A.
Let P be any point on the tangent plane to the surface at the point A and let r = xi + yj + zk be the
position vector of P.
Then ∇f at A is normal to the surface and r − r0 lies on the tangent plane at A.
∴ the equation of the tangent plane at the point A is ( r − r0 ). ∇f = 0

Note The cartesian equation of the plane at the point A (x0, y0, z0) is
∂f ∂f ∂f
( x − x0 ) + ( y − y0 ) + ( z − z0 ) =0
∂x ∂y ∂z
where the partial derivatives are evaluated at the point (x0, y0, z0).

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 5 5/27/2016 12:03:31 PM

(ii) Equation of the normal at the point A

Let A be a given point on the surface f (x, y, z) = C and let r0 = x0 i + y0 j + z0 k be the position
vector of A.
Let r be the position vector of any point P on the normal at the point A. Then r − r0 is parallel to
the normal at the point A.
∴ the equation of the normal at the point A is ( r − r0 ) × ∇f = 0.
The cartesian equation of the normal at the point A is
x − x0 y − y0 z − z0
= = ,
∂f ∂f ∂f
∂x ∂y ∂z
where the partial derivatives are evaluated at (x0, y0, z0).

9.4.5 Angle between Two Surfaces at a Common Point

We know that the angle between two planes is the angle between their normals.
We deﬁne angle between two surfaces at a point of intersection P is the angle between their tangent
planes at P and hence, the angle between their normals at P.
The angle between two surfaces f(x, y, z) = C1 and g(x, y, z) = C2 at a common point P is the angle
between their normals at the point P.
The normal at P to the surface f(x, y, z) = C1 is ∇f.
The normal at P to the surface g(x, y, z) = C2 is ∇g.
∇f ⋅ ∇g
If u is the angle between the normals at the point P, then cos u =
∇f ∇g
p ∇f ⋅ ∇g
(i) If u = , then the normals are perpendicular and cos u = 0 ⇒ = 0 ⇒ ∇f ⋅ ∇g = 0
2 ∇f ∇g

∴ if two surfaces are orthogonal at the point P then ∇f ⋅ ∇g = 0

p
Conversely, if ∇f ⋅ ∇g = 0, then u = That is they are orthogonal.
2
(ii) If u = 0, the normals at the common point coincide.
∴ the two tangent planes coincide and the surfaces touch at the common point.

9.4.6 Properties of Gradients

If f and g are scalar point functions which are differentiable, then
1. ∇C = 0, where C is constant. 2. ∇(Cf ) = C∇f, where C is a constant.
3. ∇(f ± g) = ∇f ± ∇g 4. ∇(fg) = f ∇g + g∇f
⎛ f ⎞ g ∇f − f ∇g
5. ∇ ⎜ ⎟ = if g ≠ 0
⎝ g⎠ g2

1. ∇C = 0, C is constant.
∂f ∂f ∂f
Proof We know ∇f = i +j +k (1)
∂x ∂y ∂z

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 6 5/27/2016 12:03:35 PM

Vector Calculus ■ 9.7

∂f
= ∑i
∂x
∂C ⎡ ∂C ∂C ∂C ⎤
∴ ∇C = ∑ i =0 ⎢ C is a constant ∂x = 0, ∂y = 0, ∂z = 0 ⎥ ■
∂x ⎣ ⎦
2. ∇Cf = C∇f
∂ ∂f
Proof We have ∇Cf = ∑ i (Cf) = C∑ i = C∇f [ using (1)] ■
∂x ∂x

3. ∇(f ± g) = ∇f ± ∇g
∂
Proof We have ∇( f ± g ) = ∑ i ( f ± g) [ using (1)]
∂x
⎡ ∂f ∂g ⎤ ∂f ∂g
= ∑ ⎢i ±i = ∑i ± ∑i = ∇f ± ∇g
⎣ ∂x ∂x ⎥⎦ ∂x ∂x
∴ ∇( f ± g ) = ∇f ± ∇g ■

4. ∇(fg) = f ∇g + g∇f
∂
Proof We have ∇( fg ) = ∑ i ( fg )
∂x
⎡ ∂g ∂f ⎤
= ∑i ⎢ f +g ⎥
⎣ ∂x ∂x ⎦
⎛ ∂g ⎞ ⎛ ∂f ⎞
= ∑i ⎜ f
⎝ ∂x ⎟⎠ ∑ ⎜⎝ ∂x ⎟⎠
+ i g

∂g ∂f
= f ∑i + g∑ i = f ∇g + g ∇f
∂x ∂x
∴ ∇( fg ) = f ∇g + g ∇f ■

⎛ f ⎞ g ∇f − f ∇g
5. ∇ ⎜ ⎟ =
⎝ g⎠ g2
⎛ f⎞ ∂ ⎛ f⎞
We have ∇⎜ ⎟ = ∑i
∂x ⎝⎜ g ⎠⎟
Proof
⎝ g⎠
⎡ ∂f ∂g ⎤
⎢ g ∂x − f ∂x ⎥
= ∑i ⎢ ⎥
⎣ g2 ⎦
1 ⎡ ∂f ∂g ⎤ g ∇f − f ∇g
= 2 ⎢g∑ i − f ∑i =
g ⎣ ∂x ∂x ⎥⎦ g2

∴ ⎛ f ⎞ g ∇f − f ∇ g ■
∇⎜ ⎟ =
⎝ g⎠ g2

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 7 5/27/2016 12:03:41 PM

WORKED EXAMPLES
EXAMPLE 1
Find grad f for the following functions.
(i) f(x, y, z) 5 3x2y 2 y3z2 at the point (1, 22, 1)
(ii) f(x, y, z) 5 log (x2 1 y2 1 z2) at the point (1, 2, 1).

Solution.
(i) Given f(x, y, z) = 3x2y − y3z2
∂f ∂f ∂f
We know grad f = ∇f = i +j +k
∂x ∂y ∂z
Differentiating f partially w.r. to x, y, z respectively, we get
∂f ∂f ∂f
= 6 xy, = 3x 2 − 3 y 2 z 2 , = −2 y 3 z
∂x ∂y ∂z

∂f
At the point (1, 22, 1), = 6 ⋅1( −2) = −12
∂x
∂f
= 3 ⋅12 − 3 ⋅ ( −2) 2 12 = 3 − 12 = −9
∂y
∂f
= −2 ⋅ ( −2)3 ⋅1 = 16
∂z
∴ at the point (1, 22, 1), ∇f = −12i − 9 j + 16 k .

(ii) Given f(x, y, z) = log (x2 + y2 + z2)

∂f ∂f ∂f
We know grad f = ∇f = i +j +k
∂x ∂y ∂z
Differentiating f partially w.r.to x, y, z respectively, we get,

∂f 1 ∂f 1 ∂f 1
= ⋅ 2x , = ⋅ 2y , = ⋅ 2z
∂x x 2 + y 2 + z 2 ∂y x 2 + y 2 + z 2 ∂z x 2 + y 2 + z 2

∂f 2 ⋅1 2 1
At the point (1, 2, 1), = 2 = =
∂x 1 + 2 + 1
2 2
6 3

∂f 2⋅ 2 4 2
= 2 = =
∂y 1 + 2 + 1
2 2
6 3
∂f 2 ⋅1 2 1
= 2 = =
∂z 1 + 2 + 1
2 2
6 3
1 2 1 1
∴ at the point (1, 2, 1), grad f = i + j + k = [i + 2 j + k ].
3 3 3 3

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 8 5/27/2016 12:03:44 PM

Vector Calculus ■ 9.9

EXAMPLE 2
Find the directional derivative of f (x, y, z) 5 x2yz 1 4xz2 at the point (1, 22, 21) in the direction
of the vector 2 i 2 j 2 2 k .

Solution.
Given f (x, y, z) = x2yz + 4xz2
∂f ∂f ∂f
We know grad f = ∇f = i +j +k
∂x ∂y ∂z
Differentiating f partially w.r.to x, y, z respectively, we get
∂f ∂f ∂f
= 2xyz + 4 z 2 , = x 2z , = x 2 y + 8xz
∂x ∂y ∂z
∂f
At the point (1, 22, 21), = 2 ⋅1( −2)( −1) + 4( −1) 2 = 8
∂x
∂f 2
= 1 ⋅ ( −1) = −1
∂y
∂f 2
= 1 ( −2) + 8 ⋅1( −1) = −2 − 8 = −10
∂z
∴ at the point (1, 22, 21), ∇f = 8i − j − 10 k

Given direction is a = 2i − j − 2k
∴ the directional derivative of f at the point (1, −2, −1) in the direction of a is

a (2i − j − 2k ) 16 + 1 + 20 37
∇f ⋅ = (8i − j − 10k ) ⋅ = =
a 4 +1+ 4 9 3

EXAMPLE 3
r
If r 1 xi 1 yj 1 zk and r 5 r prove that (i) ∇r 5 , (ii) ∇r n 5 nr n22 r ,
r
⎛ 1⎞ r r
(iii) ∇ ⎜ ⎟ 52 3 (iv) ∇(log r ) 5 2 .
⎝r⎠ r r
Solution.
Given r = xi + yj + zk and r= r = x2 + y2 + z2 ⇒ r2 = x 2 + y 2 + z 2 (1)
r
(i) ∇r =
r
∂r ∂r ∂r
We know ∇r = i +j +k
∂x ∂y ∂z
Differentiating (1) partially w.r.to x, we get
∂r ∂r x
2r = 2x ⇒ =
∂x ∂x r

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 9 5/27/2016 12:03:50 PM

∂r y ∂r z
Similarly, = and =
∂y r ∂z r
x y z 1 r
∴ ∇r = i + j + k = [ xi + yj + zk ] =
r r r r r
(ii) ∇r n = nr n − 2 r
∂ n ∂ ∂
We know ∇r n = i (r ) + j (r n ) + k (r n )
∂x ∂y ∂z
⎛ ∂r ⎞ ⎛ ∂r ⎞ ⎛ ∂r ⎞
= i ⎜ nr n −1 ⎟ + j ⎜ nr n −1 ⎟ + k ⎜ nr n −1 ⎟
⎝ ∂x ⎠ ⎝ ∂y ⎠ ⎝ ∂z ⎠
⎡x y z ⎤ n r n −1
= n r n −1 ⎢ i + j + k ⎥ = [ xi + yj + zk ] = nr n − 2 r
⎣r r r ⎦ r

(iii) ∇ ⎜⎛ ⎟⎞ = − 3
1 r
⎝ r⎠ r
⎛ 1⎞ ∂ ⎛ 1⎞ ∂ ⎛ 1⎞ ∂ ⎛ 1⎞
We know, ∇⎜ ⎟ = i ⎜⎝ ⎟⎠ + j ⎜⎝ ⎟⎠ + k ⎜⎝ ⎟⎠
⎝ r⎠ ∂x r ∂y r ∂z r
⎛ 1 ∂r ⎞ ⎛ 1 ∂r ⎞ ⎛ 1 ∂r ⎞
= i ⎜− 2 ⎟ + j ⎜− 2 ⎟ + k ⎜− 2 ⎟
⎝ r ∂x ⎠ ⎝ r ∂y ⎠ ⎝ r ∂z ⎠
1 ⎡x y z ⎤ 1 r
=− i + j + k ⎥ = − 3 ( xi + yj + zk ) = − 3
r 2 ⎢⎣ r r r ⎦ r r

r
(iv) ∇(log r ) =
r2
∂ ∂ ∂
We know, ∇ (log r) = i (log r ) + j (log r ) + k (log r )
∂x ∂y ∂z
⎛ 1 ∂ r ⎞ ⎛ 1 ∂ r ⎞ ⎛ 1 ∂ r ⎞ 1⎡x y z ⎤ r
=i⎜ ⎟ + j⎜ ⎟ +k⎜ ⎟ = ⎢ i + j + k⎥ = 2
⎝ r ∂x ⎠ ⎝ r ∂y ⎠ ⎝ r ∂z ⎠ r ⎣ r r r ⎦ r

EXAMPLE 4
Find the directional derivative of the function 2yz 1 z2 in the direction of the vector i 1 2 j 1 2 k
at the point (1, −1, 3).

Solution.
Given f = 2yz + z2
∂f ∂f ∂f
We know ∇f = i +j +k
∂x ∂y ∂z
Differentiating f partially w.r.to x, y, z respectively, we get

∂f ∂f ∂f
= 0, = 2 z, = 2 y + 2z
∂x ∂y ∂z

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 10 5/27/2016 12:03:55 PM

Vector Calculus ■ 9.11

∂f ∂f ∂f
At the point (1, 21, 3), = 0, = 2(3) = 6 , = 2( −1) + 2 ⋅ 3 = 4
∂x ∂y ∂z
∴ at the point (1, 21, 3), ∇f = 6 j + 4 k

Given direction is a = i + 2 j + 2k
∴ the directional derivative of f at the point (1, −1, 3) in the direction of a is
a (i + 2 j + 2k ) 12 + 8 20
∇f ⋅ = (6 j + 4 k ) ⋅ = =
a 1+ 4 + 4 9 3

EXAMPLE 5
Find the directional derivative of x3 1 y3 1 z3 at the point (1, 21, 2) in the direction of i 1 2 j 1 k .

Solution.
Given f (x, y, z) = x3 + y3 + z3
∂f ∂f ∂f
We know ∇f = i +j +k
∂x ∂y ∂z
Now differentiating f partially w.r.to x, y, z respectively, we get

∂f ∂f ∂f
= 3x 2 , = 3y2 , = 3z 2
∂x ∂y ∂z

∂f ∂f ∂f
At the point (1, −1, 2), = 3 ⋅12 = 3, = 3( −1) 2 = 3, = 3 ⋅ 22 = 12
∂x ∂y ∂z

∴ at the point (1, 21, 2), ∇f = 3i + 3 j + 12k

Given direction is a =i +2j +k

∴ the directional derivative of f at the point (1, −1, 2) in the direction of a is
a (i + 2 j + k ) 3 + 6 + 12 21 6 7 6
∇f ⋅ = (3i + 3 j + 12k ) ⋅ = = = 21 =
a 1+ 4 +1 6 6 6 2
EXAMPLE 6
Find a unit normal vector to the surface x3 1 y3 1 3xyz 5 3 at the point (1, 2, −1).

Solution.
The given surface is x3 + y3 + 3xyz = 3, which is taken as f = C
∴ f = x3 + y3 + 3xyz
We know that ∇f is normal to the surface.
∇f
So, unit normal to the surface is n=
∇f

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 11 5/27/2016 12:04:03 PM

∂f ∂f ∂f
Now ∇f = i +j +k
∂x ∂y ∂z
Differentiating f partially w.r.to x, y, z respectively,
∂f ∂f ∂f
we get, = 3 x 2 + 3 yz , = 3 y 2 + 3 xz , = 3 xy
∂x ∂y ∂z
∂f
At the point (1, 2, 21), = 3 ⋅12 + 3 ⋅ 2( −1) = −3
∂x
∂f ∂f
= 3 ⋅ 22 + 3 ⋅1( −1) = 9 and = 3 ⋅1⋅ 2 = 6
∂y ∂z
∴ at the point (1, 2, 21), ∇f = −3i + 9 j + 6 k
∴ unit normal to the given surface at the point (1, 2, −1) is
−3i + 9 j + 6 k −3i + 9 j + 6 k
n= =
9 + 81 + 36 126
Note If the surface equation is written as x + y + 3xyz − 3 = 0, then we take
3 3

f (x, y, z) = x3 + y3 + 3xyz − 3. Here C = 0.

EXAMPLE 7
Find a unit normal to the surface x2y 1 2xz2 5 8 at the point (1, 0, 2).
Solution.
Given f (x, y, z) = x2y + 2xz2
∂f ∂f ∂f
We know, ∇f = i +j +k
∂x ∂y ∂z
Differentiating f partially w.r.to x, y, z respectively, we get
∂f ∂f ∂f
= 2 xy + 2 z 2 , = x2 , = 4 xz
∂x ∂y ∂z
∂f ∂f 2 ∂f
At the point (1, 0, 2), = 2 ⋅1⋅ 0 + 2 ⋅ 22 = 8, = 1 = 1, = 4 ⋅1⋅ 2 = 8
∂x ∂y ∂z
∴ at the point (1, 0, 2), ∇f = 8i + j + 8k
∴ unit normal vector to the given surface at the point (1, 0, 2) is
∇f 8i + j + 8k 8i + j + 8k
n= = =
∇f 64 + 1 + 64 129
EXAMPLE 8
Find the maximum value of the directional derivative of f 5 x3yz at the point (1, 4, 1).
Solution.
Given f = x3yz
∂f ∂f ∂f
We know, ∇f = i +j +k
∂x ∂y ∂z

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 12 5/27/2016 12:04:07 PM

Vector Calculus ■ 9.13

The directional derivative is maximum in the direction of ∇f and the maximum value = ∇f
Differentiating f partially w.r.to x, y, z respectively, we get

∂f ∂f ∂f
= 3 x 2 yz , = x 3 z, = x3 y
∂x ∂y ∂z

∂f ∂f 3 ∂f 3
At the point (1, 4, 1), = 3 ⋅1⋅ 4 ⋅1 = 12, = 1 ⋅1 = 1 and = 1 ⋅4 = 4
∂x ∂y ∂z

∴ at the point (1, 4, 1), ∇f = 12i + j + 4 k

Maximum value of the directional derivative = ∇f = 12i + j + 4 k = 144 + 1 + 16 = 161

EXAMPLE 9
In what direction from the point (1, 1, 22), is the directional derivative of f 5 x2 2 2y2 1 4z2
maximum? Also ﬁnd the maximum directional derivative.

Solution.
Given f = x2 − 2y2 + 4z2
We know that the directional derivative is maximum in the direction of ∇f. The maximum value
= ∇f
∂f ∂f ∂f
We have ∇f = i +j +k
∂x ∂y ∂z
Differentiating f partially w.r.to x, y, z respectively, we get
∂f ∂f ∂f
= 2 x, = −4 y, = 8z
∂x ∂y ∂z

∂f ∂f ∂f
At the point (1, 1, 22), = 2 ⋅ 1 = 2, = −4 ⋅1 = −4, = 8( −2) = −16
∂x ∂y ∂z

∴ at the point (1, 1, 22), ∇f = 2i − 4 j − 16 k = 2[i − 2 j − 8k ]

∴ the directional derivative is maximum in the direction of 2(i − 2 j − 8k )

Maximum value = ∇f = 2(i − 2 j − 8k ) = 2 1 + 4 + 64 = 2 69

EXAMPLE 10
Find the angle between the surfaces x2 1 y2 1 z2 5 9 and x2 1 y2 2 z 5 3 at the point (2, 21, 2).

Solution.
The given surfaces are
x2 + y2 + z2 = 9 (1) and x2 + y2 − z = 3 (2)
P(2, −1, 2) is a common point of (1) and (2)
Let f = x2 + y2 + z2 and g = x2 + y2 − z

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 13 5/27/2016 12:04:11 PM

∂f ∂f ∂f
Now, ∇f = i +j +k
∂x ∂y ∂z
Differentiating f partially w.r.to x, y, z respectively we get,
∂f ∂f ∂f
= 2 x, = 2 y, = 2z
∂x ∂y ∂z
∂f ∂f ∂f
At the point (2, 21, 2), = 2 ⋅ 2 = 4, = 2( −1) = −2, = 2( +2) = +4
∂x ∂y ∂z
∴ at the point (2, 21, 2), ∇f = 4 i − 2 j + 4 k
∂g ∂g ∂g
Now ∇g = i +j +k
∂x ∂y ∂z
Differentiating g partially w.r.to x, y, z respectively, we get
∂g ∂g ∂g
= 2 x, = 2 y, = −1
∂x ∂y ∂z
∂g ∂g ∂g
at the point (2, 21, 2), = 2 ⋅ 2 = 4, = 2( −1) = −2, = −1
∂x ∂y ∂z
∴ at the point (2, 21, 2), ∇g = 4 i − 2 j − k
If u is the angle between the surfaces (1) and (2) at (2, −1, 2), then
∇f ⋅ ∇g (4i − 2 j + 4 k ) (4i − 2 j − k ) 16 + 4 − 4 16 8
cos u = = ⋅ = = =
∇f ∇g 16 + 4 + 16 16 + 4 + 1 36 21 6 21 3 21
⎛ 8 ⎞
∴ u = cos −1 ⎜
⎝ 3 21 ⎟⎠

EXAMPLE 11
Show that the surfaces 5x2 2 2yz 2 9x 5 0 and 4x2y 1 z3 2 4 5 0 are orthogonal at the point
(1, 21, 2).

Solution.
The given surfaces are
5x2 − 2yz − 9x = 0 (1) and 4x2y + z3 − 4 = 0 (2)
Let f = 5x2 − 2yz − 9x and g = 4x2y + z3 − 4
To prove (1) and (2) cut orthogonally at the point (1, −1, 2),
i.e., to prove ∇f⋅∇g = 0
∂f ∂f ∂f
Now ∇f = i +j +k
∂x ∂y ∂z
∂f ∂f ∂f
= 10 x − 9, = −2 z and = −2 y
∂x ∂y ∂z
∴ ∇f = (10 x − 9)i − 2 zj − 2 yk

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 14 5/27/2016 12:04:14 PM

Vector Calculus ■ 9.15

∂g ∂g ∂g
and ∇g = i +j +k
∂x ∂y ∂z
∂g ∂g ∂g
= 8 xy, = 4 x 2 and = 3z 2
∂x ∂y ∂z
∴ ∇g = 8 xyi + 4 x 2 j + 3 z 2 k
At the point (1, 21, 2), ∇f = (10 − 9)i − 2 ⋅ 2 j − 2( −1)k = i − 4 j + 2k
and ∇g = 8 ⋅1⋅ ( −1)i + 4 ⋅12 j + 3 ⋅ 22 k = −8i + 4 j + 12k
∴ ∇f ⋅ ∇g = (i − 4 j + 2k ) ⋅ ( −8i + 4 j + 12k ) = −8 − 16 + 24 = 0
Hence, the two surfaces cut orthogonally at the point (1, −1, 2).

EXAMPLE 12
Find a and b if the surfaces ax2 2 byz 5 (a 1 2)x and 4x2y 1 z3 5 4 cut orthogonally at the point
(1, 21, 2).

Solution.
The given surfaces are
ax2 − byz − (a + 2)x = 0 (1) and 4x2y + z3 − 4 = 0 (2)
Let f = ax − byz − (a + 2)x and
2
g = 4x2y + z3 − 4
Given the surfaces (1) and (2) cut orthogonally at the point (1, −1, 2).
∴ ∇f ⋅ ∇g = 0 (3)
∂f ∂f ∂f
Now ∇f = i +j +k
∂x ∂y ∂z
∂f ∂f ∂f
= 2ax − a − 2, = − bz and = −by
∂x ∂y ∂z
∴ ∇f = (2ax − a − 2)i − bzj − byk
∂g ∂g ∂g
and ∇g = i +j +k
∂x ∂y ∂z
∂g ∂g ∂g
= 8 xy, = 4 x 2 and = 3z 2
∂x ∂y ∂z
∴ ∇g = 8 xyi + 4 x 2 j + 3 z 2 k
At the point (1, 21, 2), ∇f = (2a − a − 2)i − b ⋅ 2 j − b( −1)k
⇒ ∇f = ( a − 2)i − 2bj + bk
and ∇g = −8i + 4 j + 12k
∴ ∇f ⋅ ∇g = (( a − 2)i − 2bj + bk ) ⋅ ( −8i + 4 j + 12k )
= −8(a − 2) − 8b + 12b = −8a + 4b + 16
From (3), ∇f⋅∇g = 0 ⇒ −8a + 4b + 16 = 0 ⇒ 2a − b = 4 (4)
Since (1, −1, 2) is a point on the surface f = 0, we get
a + 2b − ( a + 2) = 0 ⇒ 2b = 2 ⇒ b = 1

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 15 5/27/2016 12:04:20 PM

5
∴ (4) ⇒ 2a = 4 + b = 4 + 1 = 5 ⇒ a=
2
5
∴ a= , b =1
2
EXAMPLE 13
Find the angle between the normals to the surface xy = z2 at the points (1, 4, 2) and (−3, −3, 3).

Solution.
The given surface is xy − z2 = 0
∴ f = xy − z2
We know ∇f is normal to the surface at the point (x, y, z)
Let n1, n 2, be the normals to the surface at the points (1, 4, 2) and (−3, −3, 3) respectively.
∴ n1 = ∇f at the point (1, 4, 2)
and n 2 = ∇f at the point ( −3, − 3, 3)
∂f ∂f ∂f
Now ∇f = i +j +k
∂x ∂y ∂z
∂f ∂f ∂f
= y, = x and = −2z
∂x ∂y ∂z
∴ ∇f = yi + xj − 2 zk
At the point (1, 4, 2), ∇f = 4 i + j − 4 k ∴ n1 = 4i + j − 4 k
At the point (23, 23, 3), ∇f = −3i − 3 j − 6 k ∴ n2 = −3i − 3 j − 6 k
If u is the angle between the normals, then
n ⋅n (4i + j − 4 k ) ⋅ ( −3i − 3 j − 6 k )
cos u = 1 2 =
n1 n2 16 + 1 + 16 9 + 9 + 36
−12 − 3 + 24 9 1
= = =
33 54 33 54 22
⎛ 1 ⎞
∴ u = cos −1 ⎜
⎝ 22 ⎟⎠

EXAMPLE 14
Find the directional derivative of the function f 5 xy2 1 yz3 at the point (2, 21, 1) in the direction
of the normal to the surface xlog z 2 y2 1 4 5 0 at the point (21, 2, 1).

Solution.
Given f = xy2 + yz3
∂f ∂f ∂f
∴ ∇f = i +j +k = y 2 i + (2 xy + z 3 ) j + 3 yz 2 k
∂x ∂y ∂z
At the point (2, 21, 1), ∇f = ( −1) 2 i + ( −4 + 1) j + 3( −1)12 k = i − 3 j − 3k
The directional derivative of f in the direction of the normal to the surface xlogz − y2 + 4 = 0 at the
point (−1, 2, 1) is required.

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 16 5/27/2016 12:04:28 PM

Vector Calculus ■ 9.17

Let f = xlogz − y2 + 4
∂f ∂f ∂f x
∴ ∇f = i +j +k = log zi − 2 yj + k
∂x ∂y ∂z z
⎛ −1⎞
At the point (21, 2, 1), ∇f = log1i − 4 j + ⎜ ⎟ k = 0i − 4 j − k = −4 j − k
⎝ 1⎠
∴ a = −4 j − k
a
Required directional derivative is = ∇f ⋅
a
( −4 j − k ) 12 + 3 15
= (i − 3 j − 3k ) ⋅ = =
16 + 1 17 17

EXAMPLE 15
If ∇f 5 2 xyz 3 i 1 x 2 z 3 j 1 3x 2 yz 2 k , then ﬁnd f if f (1, 22, 2) 5 4.

Solution.
Given ∇f = 2 xyz 3 i + x 2 z 3 j + 3 x 2 yz 2 k (1)
∂f ∂f ∂f
But ∇f = i +j +k (2)
∂x ∂y ∂z
Equating the coefﬁcients of i , j , k , from (1) and (2), we get
∂f ∂f ∂f
= 2 xyz 3 (3) = x2 z3 (4) = 3 x 2 yz 2 (5)
∂x ∂y ∂z
Integrating (3) partially w.r.to x, we get
f = x 2 yz 3 + f1 ( y, z ) (6)
Integrating (4) partially w.r.to y, we get,
f = x 2 z 3 y + f 2 ( x, z ) (7)
Integrating (5) partially w.r.to z, we get,
f = x 2 yz 3 + f 3 ( x, y ) (8)
From (6), (7), (8), f is obtained by adding all the terms and an arbitrary constant C, but omitting
f1(y, z), f2(x, z), f3(x, y) and choosing only one of the repeated terms.
Thus, f = x2yz3 + C
Given f (1, −2, 2) = 4
∴ 1 × (−2) × 8 + C = 4 ⇒ C = 4 + 16 = 20
∴ f = x2yz3 + 20

EXAMPLE 16
Find the equation of the tangent plane and the equation of the normal to the surface
x2 2 4y2 1 3z2 1 4 5 0 at the point (3, 2, 1).

Solution.
The given surface is x2 − 4y2 + 3z2 + 4 = 0

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 17 5/27/2016 12:04:36 PM

Let f = x2 − 4y2 + 3z2 + 4

∂f ∂f ∂f
∴ ∇f = i +j +k = 2 xi − 8 yj + 6 zk
∂x ∂y ∂z
At the point (3, 2, 1), ∇f = 6i − 16 j + 6 k
We know that the equation of the tangent plane at the point (x0, y0, z0) is

∂f ∂f ∂f
( x − x0 ) + ( y − y0 ) + ( z − z0 ) =0
∂x ∂y ∂z

∂f ∂f ∂f
Now = 2 x, = −8 y and = 6z
∂x ∂y ∂z

∂f ∂f ∂f
Here (x0, y0, z0) = (3, 2, 1) ∴ = 6, = −16 and =6
∂x ∂y ∂z
∴ the equation of the tangent plane at the point (3, 2, 1) is
( x − 3)6 + ( y − 2)( −16) + ( z − 1)6 = 0
⇒ 3( x − 3) − 8( y − 2) + 3( z − 1) = 0 [dividing by 2]
⇒ 3 x − 8 y + 3 z − 9 + 16 − 3 = 0
⇒ 3x − 8 y + 3z + 4 = 0
The equation of the normal at the point (x0, y0, z0) is
x − x0 y − y0 z − z0
= =
∂f ∂f ∂f
∂x ∂y ∂z
The equation of the normal at the point (3, 2, 1) is
x − 3 y − 2 z −1 x − 3 y − 2 z −1
= = ⇒ = = .
6 −16 6 3 −8 3
EXAMPLE 17
If the directional derivative of
f(x, y, z) 5 a(x 1 y) 1 b(y 1 z) 1 c(z 1 x) has maximum value 12 at the point (1, 2, 1) in the
x 21 y 2 2 z 21
direction parallel to the line 5 5 , ﬁnd the value of a, b, c.
1 2 3
Solution.
Given f = a(x + y) + b(y + z) + c(z + x)

∴ ∂f ∂f ∂f
∇f = i +j +k
∂x ∂y ∂z
⇒ ∇f = ( a + c )i + ( a + b ) j + ( b + c ) k

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 18 5/27/2016 12:04:43 PM

Vector Calculus ■ 9.19

We know that the directional derivative is maximum in the direction of ∇f.

x −1 y − 2 z −1
But given it is maximum in the direction parallel to the line = = .
1 2 3
a+c a+b b+c
∴ = = =K
1 2 3
⇒ a+c = K (1) a + b = 2K (2) b + c = 3K (3)
Adding we get,
a + c + a + b + b + c = K + 2K + 3K

⇒ 2( a + b + c) = 6 K ⇒ a + b + c = 3K (4)

Using (3), (4) ⇒ a + 3K = 3K ⇒ a=0

From (1), 0+c = K ⇒ c=K

From (2), 0 + b = 2K ⇒ b = 2K

Given the maximum value of directional derivative = 12

⇒ ∇f = 12

⇒ ( a + c) 2 + ( a + b) 2 + (b + c) 2 = 12

⇒ ( a + c) 2 + ( a + b) 2 + (b + c) 2 = 144

⇒ K 2 + 4 K 2 + 9 K 2 = 144

144 12
⇒ 14 K 2 = 144 ⇒ K2 = ⇒ K=±
14 14
24 12
∴ a = 0, b = ± ,c=±
14 14
EXAMPLE 18
If u 5 x 1 y 1 z , v 5 x 2 1 y 2 1 z 2 , w 5 xy 1 yz 1 zx , then show that the vectors ∇u, ∇v, ∇w
are coplanar.

Solution.
Given u = x + y + z, v = x2 + y2 + z2, w = xy + yz + zx
∂u ∂u ∂u
Now, ∇u = i +j +k =i + j+k
∂x ∂y ∂z
∂v ∂v ∂v = 2 xi + 2 yj + 2 zk
∇v = i +j +k
∂x ∂y ∂z
∂w ∂w ∂w = ( y + z ) i + ( z + x ) j + ( x + y ) k
∇w = i +j +k
∂x ∂y ∂z

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 19 5/27/2016 12:04:56 PM

We know that three vectors a , b , c are coplanar, if their scalar triple product a ⋅ b × c = 0.
∴ ∇u, ∇v, ∇w are coplanar, if ∇u ⋅ ∇v × ∇w = 0
1 1 1
Now ∇u ⋅ ∇v × ∇w = 2x 2y 2z
y+z z+x x+ y
1 1 1
=2 x y z
y +z z +x x+y
1 1 1
= 2 x +y +z x +y +z x +y +z R2 → R2 + R3
y +z z +x x+y
1 1 1
= 2( x + y + z ) 1 1 1 =0 [since R1 = R2 ]
y +z z +x x+y

∴ the vectors ∇u, ∇v, ∇w are coplanar.

EXERCISE 9.1
1. If f(x, y, z) = 3xz2y − y3z2, ﬁnd ∇f at the point (1, −2, −1).
2. If f = 2xz − y2 ﬁnd grad f at the point (1, 3, 2).
3. Find the directional derivative of f = 3x2 + 2y − 3z at the point (1, 1, 1) in the direction of
2i + 2 j − k .
4. Find the directional derivative of xyz − xy2z2 at the point (1, 2, −1) in the direction of the vector
i − j − 3k .
5. Find the directional derivative of the function f = x2 − y2 + 2z2 at the point P (1, 2, 3) in the
direction of the line PQ where Q = (5, 0, 4).
6. Find the unit normal vector to the surface
(i) x2 + 2y2 + z2 = 7 at the point (1, −1, 2). (ii) x2 + y2 − z2 = 1 at the point (1, 1, 1).
(iii) x + y − z = 1 at the point (1, 1, 1).
2 2
(iv) x2 + y2 = z at the point (1, 2, 5).
7. Find the angle between the surfaces x2 + y + z = 2 and xlogz = y2 − 1 at the point (1, 1, 1).
8. Find the angle between the surfaces 2yz + z2 = 3 and x2 + y2 + z2 = 3 at the point (1, 1, 1).
9. Find the angle between the surfaces xyz = 4 and x2 + y2 + z2 = 9 at the point i + 2 j + 2k .
10. Find the equation of the tangent plane and normal line to the surface xz2 + x2y − z + 1 = 0 at the
point (1, −3, 2).
11. Find the equation of the tangent plane and normal line to the surface 2xz2 − 3xy − 4x = 7 at the
point (1, −1, 2).
12. Find the equation of the tangent plane and normal line to the surface 2z − x2 = 0 at the point P(2, 0, 2).
13. Find f if
(i) ∇f = ( y 2 − 2 xyz 3 )i + (3 + 2 xy − x 2 z 3 ) j + (8 z 3 − 3 x 2 yz 2 )k

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 20 5/27/2016 12:05:02 PM

Vector Calculus ■ 9.21

(ii) ∇f = 2 xyz 3 i + x 2 z 3 j + 3 x 2 yz 2 k if f(1, − 2, 2) = 4

(iii) ∇f = (6 xy + z 3 )i + (3 x 2 − z ) j + (3 xz 2 − y )k
(iv) ∇f = (2 xyz + x )i + x 2 zj + x 2 yk
(v) ∇f = ( y + sin z )i + xj + x cos zk .
14. Find the angle between the normals to the intersecting surfaces xy − z2 − 1 = 0 and y2 − 3z − 1 = 0
at the point (1, 1, 0).
15. Find the angle between the normals to the surface x2 = yz at the points (1, 1, 1) and (2, 4, 1).
16. Find the values of a and b so that the surfaces ax3 − by2z = (a + 3)x2 and 4x2y − z3 = 11 may cut
orthogonally at the point (2, −1, −3).
17. The temperature at any point in space is given by T = xy + yz + zx. Find the direction in which the
temperature changes most rapidly from the point (1, 1, 1) and determine the maximum rate of
change.
18. In what direction is the directional derivative of the function f = x2 − 2y2 + 4z2 from the point
(1, 1, −1) is maximum and what is its value?
19. Find the maximum value of the directional derivative of the function f = 2x2 + 3y2 + 5z2 at the
point (1, 1, −4).
20. Find ∇f at the point (1, 1, 1) if f(x, y, z) = x2y + y2x + z2.
21. Find the directional derivative of f(x, y, z) = x2 − 2y2 + 4z2 at the point (1, 1, −1) in the direction
2i − j − k .
22. Find the directional derivative of the function f = xy + yz + zx in the direction of the vector
2i + 3 j + 6 k at the point (3, 1, 2).
23. Find the directional derivative of f = x2yz + 4xz2+ xyz at (1, 2, 3) in the direction of 2i + j − k .
24. Find the directional derivative of f = x2yz + 4xz2 at the point P(1, −2, −1) in the direction of PQ,
where Q is (3, −3, −3).
25. Find a unit normal to the surface xy3z2 = 4 at the point (−1, −1, 2).
26. In what direction from (3, 1, −2) is the directional derivative of f = x2y2z4 maximum? Find also
the magnitude of this maximum.
27. What is the greatest rate of increase of f = xyz2 at the point (1, 0, 3)?
28. Find the angle between the spheres x2 + y2 + z2 = 29 and
x2 + y2 + z2 + 4x − 6y − 8z − 47 = 0 at the point (4, −3, 2).
29. Find f if ∇f = (6 xy + z 3 )i + (3 x 2 − z ) j + (3 xz 2 − y )k .

ANSWERS TO EXERCISE 9.1

19 29 28
1. −6i − 9 j − 4 k 2. 4i − 6 j + 2k 3. 4. 5.
3 11 21
i − 2 j + 2k 2i + 2 j + k
6. (i) (ii) i + j − k (iii) (iv) 2i + 4 j − 5k
3 3 3 3 5
⎛ 1 ⎞ 3 2
7. cos −1 ⎜ 8. cos −1 9. cos −1
⎝ 30 ⎟⎠ 5 3

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 21 5/27/2016 12:05:16 PM

x −1 y + 3 z − 2
10. 2x − y − 3z + 1 = 0, = =
−2 1 3
x −1 y +1 z − 2 x−2 y z−2
11. 7x − 3y + 8z − 26 = 0, = = 12. 2x − z = 2; = =
7 −3 3 −2 0 1
13. (i) f = xy2 − x2yz3 + 3y + 2z4 + c (ii) f = x2yz3 + 20
x2
(iii) f = 3x2y + xz3 − yz + c (iv) f = x 2 yz + +c (v) f = xy + x sin z + c
2
⎛ 2 ⎞ 13 7 64
14. cos −1 ⎜ 15. cos −1 16. a = − , b =
⎝ 26 ⎟⎠ 3 22 3 9
17. i + j + k , 2 3 18. 2i − 4 j − 8k , 2 21 19. 1652 20. ∇f = 3i + 3 j + 2k
(i + 3 j − k )
25. −
45 86 37
21. 16 22. 23. 24.
7 6 3 11
6
⎛ 19 ⎞
28. u = cos −1 ⎜ 29. f = 3x2y + xz3 − yz + c
⎝ 29 ⎟⎠
26. 96 19 27. 9

9.5 DIVERGENCE OF A VECTOR POINT FUNCTION OR DIVERGENCE OF A VECTOR FIELD

Definition 9.5 If F( x, y, z ) be a vector point function continuously differentiable in a region R of
space, then the divergence of F is deﬁned by
∂F ∂F ∂F
∇⋅F = i ⋅ + j ⋅ + k ⋅
∂x ∂y ∂z
It is abbreviated as div F and thus, div F = ∇⋅ F
∂F ∂F ∂F
If F = F1i + F2 j + F3 k , then ∇⋅F = 1 + 2 + 3
∂x ∂y ∂z
If is a constant vector, then =?F 5 0 and conversely if = ? F 5 0, then F is a constant vector.
F

Note (i) From the deﬁnition it is clear that div F is a scalar point function. So, the divergence of a
vector ﬁeld is a scalar point function. The notation ∇ ⋅ F is not a scalar product in the usual sense, since
∂ ∂ ∂
∇ ⋅ F ≠ F ⋅ ∇. In fact F ⋅ ∇ = F1 + F2 + F3 is a scalar operator.
∂x ∂y ∂z

9.5.1 Physical Interpretation of Divergence

Physical interpretation of divergence applied to a vector field is that it gives approximately the ‘loss’
of the physical quantity at a given point per unit volume per unit time.
(i) If v ( x, y, z ) is the moving fluid at a point (x, y, z), then the ‘loss’ of the fluid per unit volume per
unit time at the point is given by div v . Thus, divergence gives a measure of the outward flux per
unit volume of the flow at (x, y, z).
If there is no ‘loss’ of fluid anywhere, then div v = 0 and the fluid is said to be incompressible.

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 22 5/27/2016 12:05:34 PM

Vector Calculus ■ 9.23

(ii) If v represents an electric flux, div v is the amount of electric flux which diverges per unit volume
in unit time.
(iii) If v represents the heat flux, div v is the rate at which heat is issuing from a point per unit volume.
Definition 9.6 Solenoidal Vector
If div F = 0 everywhere in a region R, then F is called a solenoidal vector point function and R is
called a solenoidal field.

9.6 CURL OF A VECTOR POINT FUNCTION OR CURL OF A VECTOR FIELD

Definition 9.7 If F( x, y, z ) be a vector point function continuously differentiable in a region R, then
the curl of F is deﬁned by

It is abbreviated as curl F
Thus, curl
If F = F1i + F2 j + F3 k , then
curl F = ∇ × F
⎛ ∂ ∂ ∂⎞
= ⎜i +j + k ⎟ × (F1i + F2 j + F3 k )
⎝ ∂x ∂y ∂z ⎠
⎡ ∂F ∂F ⎤ ⎡ ∂F ∂F ⎤ ⎡ ∂F ∂F ⎤
= i ⎢ 3 − 2⎥+ j ⎢ 1 − 3⎥+k ⎢ 2 − 1⎥
⎣ ∂y ∂z ⎦ ⎣ ∂z ∂x ⎦ ⎣ ∂x ∂y ⎦
i j k
∂ ∂ ∂
This is symbolically written as ∇ × F =
∂x ∂y ∂z
F1 F2 F3
If F is a constant vector, then curl F = 0
9.6.1 Physical Meaning of Curl F
If F represents the linear velocity of the point P of a rigid body that rotates about a fixed axis (e.g., top)
with constant angular velocity v, then curl F at P is equal to 2v.
If the body is not rotating, then v = 0 ∴ Curl F = 0
Definition 9.8 Irrotational Vector Field
Let F( x, y, z ) be a vector point function. If curl F = 0 at all points in a region R, then F is said to be an
irrotational vector in R. The vector field R is called an irrotational vector field.

Definition 9.9 Conservative Vector Field

A vector ﬁeld F is said to be conservative if there exists a scalar function f such that F = ∇f

Note
1. In a conservative vector ﬁeld F = ∇f
∴ ∇ × F = ∇ × ∇f = 0 ⇒ F is irrotational.
2. This scalar function f is called the scalar potential of F.
Only irrotational vectors will have scalar potential f.

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 23 6/3/2016 9:11:54 PM

WORKED EXAMPLES
EXAMPLE 1
Prove that ∇ 3 ∇f 5 0, where f is a scalar point function.

Solution.
∂ ∂ ∂ ∂f ∂f ∂f
We have ∇=i + j +k , ∇f = i +j +k
∂x ∂y ∂z ∂x ∂y ∂z
i j k
∂ ∂ ∂
∴ ∇ × ∇f =
∂x ∂y ∂z
∂f ∂f ∂f
∂x ∂y ∂z
⎡ ∂ 2f ∂ 2f ⎤ ⎡ ∂ 2f ∂ 2f ⎤ ⎡ ∂ 2f ∂ 2f ⎤
= i⎢ − ⎥− j⎢ − ⎥ + k ⎢ − ⎥
⎣ ∂y ∂z ∂z ∂y ⎦ ⎣ ∂x ∂z ∂z ∂x ⎦ ⎣ ∂x ∂y ∂y ∂x ⎦
⎡ ∂ 2f ∂ 2f ∂ 2f ∂ 2f ∂ 2f ∂ 2f ⎤
=0 ⎢ Assuming = , = , = ⎥
⎣ ∂y ∂z ∂z ∂y ∂z ∂x ∂x ∂z ∂x ∂y ∂y ∂x ⎦
∴ ∇f is always an irrotational vector.

EXAMPLE 2
Find the divergence and curl of the vector v 5 xyz i 1 3x 2 y j 1 ( xz 2 2 y 2 z ) k at the point
(2, −1, 1).

Solution.
Given v = xyzi + 3x 2 y j + ( xz 2 − y 2 z )k
∂ ∂ ∂
∴ div v = ∇ ⋅ v = ( xyz ) + (3 x 2 y ) + ( xz 2 − y 2 z )
∂x ∂y ∂z
= yz + 3 x 2 + 2 xz − y 2
At the point (2, 21, 1), ∇ ⋅ v = ( −1) ⋅1 + 3 ⋅ 4 + 2 ⋅ 2 ⋅1 − ( −1) 2 = −1 + 12 + 4 − 1 = 14

i j k
∂ ∂ ∂
and Curl v = ∇ × v =
∂x ∂y ∂z
xyz 3 x 2 y xz 2 − y 2 z
⎡∂ ∂ ⎤ ∂ ∂
= i ⎢ (xz 2 − y 2z ) − (3x 2 y ) ⎥ − j ⎡⎢ (xz 2 − y 2z ) − (xyz ) ⎤⎥
⎣ ∂y ∂z ⎦ ⎣ ∂x ∂z ⎦
⎡ ∂ ∂ ⎤
+ k ⎢ (3 x 2 y ) − ( xyz ) ⎥
⎣ ∂ x ∂ y ⎦
= i[0 − 2 yz − 0] − j[ z 2 − 0 − xy ] + k [6 xy − xz ]
= −2 yzi − ( z 2 − xy ) j + (6 xy − xz )k

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 24 5/27/2016 12:05:57 PM

Vector Calculus ■ 9.25

At the point (2, 21, 1),

∇ × v = −2( −1) ⋅1i − (12 − 2( −1)) j + [6 ⋅ 2( −1) − 1⋅ 2]k = 2i − 3 j − 14k

EXAMPLE 3

Show that the vector F 5 (6 xy 1 z 3 )i 1 (3 x 2 2 z ) j 1 (3 xz 2 2 y )k is irrotational.

Solution.
Given F = (6 xy + z 3 )i + (3 x 2 − z ) j + (3 xz 2 − y )k

F is irrotational if curl F = 0

i j k
∂ ∂ ∂
Now curl F = ∇ × F =
∂x ∂y ∂z
6 xy + z 3 (3 x 2 − z ) (3 xz 2 − y )
⎡∂ ∂ ⎤ ⎡∂ ∂ ⎤
= i ⎢ (3 xz 2 − y ) − (3 x 2 − z ) ⎥ − j ⎢ (3 xz 2 − y ) − (6 xy + z 3 ) ⎥
⎣ ∂ y ∂ z ⎦ ⎣ ∂x ∂z ⎦
⎡∂ ∂ ⎤
+ k ⎢ (3 x 2 − z ) − (6 xy + z 3 ) ⎥
⎣ ∂x ∂y ⎦
= i[ −1 + 1] − j[3 z 2 − 3 z 2 ] + k [6 x − 6 x ] = 0.
∴ F is irrotational vector.

EXAMPLE 4
Prove that (i) div r 5 3, (ii) curl r 5 0 where r is the position vector of a point (x, y, z) in space.
Solution.
Given r is the position vector of a point (x, y, z) in space.
∴ r = xi + y j + zk
∂ ∂ ∂
(i) div r = ∇ ⋅ r = ( x) + ( y) + ( z) = 1 + 1 + 1 = 3
∂x ∂y ∂z
i j k
∂ ∂ ∂
(ii) Curl r =
∂x ∂y ∂z
x y z
⎡∂ ∂ ⎤ ⎡∂ ∂ ⎤ ⎡∂ ∂ ⎤
= i ⎢ ( z ) − ( y)⎥ − j ⎢ ( z ) − ( x)⎥ − k ⎢ ( y) − ( x)⎥
⎣ ∂y ∂z ⎦ ⎣ ∂x ∂z ⎦ ⎣ ∂x ∂y ⎦
= i [0 − 0] − j[0 − 0] + k [0 − 0] = 0
∴ r is an irrotational vector.

M09_ENGINEERING_MATHEMATICS-I _CH09_Part A.indd 25 5/27/2016 12:06:10 PM

EXAMPLE 5
Find the value of a if the vector
F 5 (2 x 2 y 1 yz ) i 1 ( xy 2 2 xz 2 ) j 1 ( axyz 2 2 x 2 y 2 ) k is solenoidal.

Solution.
Given F = (2 x 2 y + yz )i + ( xy 2 − xz 2 ) j + ( axyz − 2 x 2 y 2 )k
is solenoidal.
∂ ∂ ∂
∴ ∇⋅F=0 ⇒ ( 2 x 2 y + yz ) + ( xy 2 − xz 2 ) + ( axyz − 2 x 2 y 2 ) = 0
∂x ∂y ∂z
⇒ 4 xy + 2 xy + axy = 0
⇒ 6 xy + axy = 0
⇒ xy(6 + a) = 0 ⇒ (6 + a ) = 0 ⇒ a = −6 [{ x ≠ 0, y ≠ 0]

EXAMPLE 6
Show that F 5 ( y 2 2 z 2 1 3 yz 2 2 x ) i 1 (3 xz 1 2 xy ) j 1 (3 xy 2 2 xz 1 2 z ) k is irrotational
and solenoidal.

Solution.
Given F = ( y 2 − z 2 + 3 yz − 2 x )i + (3xz + 2 xy ) j + (3 xy − 2 xz + 2 z )k .
We have to prove F is irrotational and solenoidal.
i.e., to prove ∇ × F = 0 and ∇ ⋅ F = 0
i j k
∂ ∂ ∂
∇×F =
∂x ∂y ∂z
y 2 − z 2 + 3 yz − 2 x 3xz + 2 xy 3xy − 2 xz + 2 z
= i(3 x − 3 x ) − j[3 y − 2 z − ( −2 z + 3 y )] + k [3z + 2 y − (2 y + 3z )] = 0
∴ F is irrotational.
∂ 2 ∂ ∂
∇⋅F = ( y − z 2 + 3 yz − 2 x ) + (3 xz + 2 xy ) + (3 xy − 2 xz + 2 z )
∂x ∂y ∂z
= −2 + 2 x + ( −2 x + 2) = 0

∴ F is solenoidal.

EXAMPLE 7
If r 5 xi 1 y j 1 z k and r 5 r , prove that r n r is solenoidal if n 5 23 and irrotational for all
values of n.

Solution.
Given r = xi + y j + zk ∴ r= r = x2 + y2 + z2 ⇒ r2 = x2 + y2 + z2 (1)
r n r = r n ( xi + y j + zk ) = r n xi + r n y j + r n zk
∂ n ∂ ∂
∴ div ( r n r ) = ∇ ⋅ ( r n xi + r n y j + r n zk ) = (r x) + (r n y) + (r n z) (2)
∂x ∂y ∂z

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 26 5/19/2016 10:15:07 AM

Vector Calculus ■ 9.27

∂ n ∂r ∂ n ∂r
But ( r x ) = r n + x ⋅ nr n −1 , ( r y ) = r n + y ⋅ nr n −1
∂x ∂x ∂y ∂y
∂ n ∂r
and ( r z ) = r n + z ⋅ nr n −1
∂z ∂z
∂r x ∂r y ∂r z
We have, r2 = x2 + y2 + z2 , = , = , =
∂x r ∂y r ∂z r
∂ n x
∴ ( r x ) = r n + nxr n −1 ⋅ = r n + nx 2 r n − 2
∂x r
∂ n y
( r y ) = r n + nyr n −1 ⋅ = r n + ny 2 r n − 2
∂y r
∂ n z
and ( r z ) = r n + nzr n −1 ⋅ = r n + nz 2 r n − 2
∂z r
Substitute in (2).
∴ div ( r n r ) = r n + nx 2 r n − 2 + r n + ny 2 r n −2 + r n + nz 2 r n − 2
= 3r n + nr n − 2 ( x 2 + y 2 + z 2 ) = 3r n + nr n − 2 ⋅ r 2 = 3r n + nr n = ( n + 3)r n

If n = −3, then div ( r n r ) = 0 ∴ r n r is solenoidal if n = −3

i j k
∂ ∂ ∂
Now ∇ × r n r =
∂x ∂y ∂z
rn x rn y rn z
⎡∂ ∂ ⎤ ⎡∂ ∂ ⎤ ⎡∂ ∂ ⎤
= i ⎢ (r n z ) − (r n y)⎥ − j ⎢ (r n z ) − (r n x)⎥ + k ⎢ (r n y) − (r n x)⎥
⎣ ∂y ∂z ⎦ ⎣ ∂x ∂z ⎦ ⎣ ∂ x ∂ y ⎦

⎛ ∂r ∂r ⎞ ⎛ ∂r ∂r ⎞ ⎛ n −1 ∂r ∂r ⎞
= i ⎜ nzr n −1 − nyr n −1 ⎟ − j ⎜ nzr n −1 − nxr n −1 ⎟ + k ⎜ nyr − nxr n −1 ⎟
⎝ ∂y ∂z ⎠ ⎝ ∂x ∂z ⎠ ⎝ ∂x ∂y ⎠
⎛ y z⎞ ⎛ n −1 x z⎞ ⎛ n −1 x n −1 y ⎞
= i ⎜ nzr n −1 − nyr n −1 ⎟⎠ − j ⎜⎝ nzr ⋅ − nxr
n −1
⎟⎠ + k ⎜⎝ nyr ⋅ r − nxr ⋅ r ⎟⎠
⎝ r r r r
= i( nr n − 2 yz − nr n − 2 yz ) − j ( nr n − 2 xz − nr n − 2 xz ) + k ( nr n − 2 xy − nr n − 2 xy ) = 0

∴ ∇ × ( r n r ) = 0 for all values of n.

Hence, r n r is irrotational for all values of n.

EXAMPLE 8
Prove that F 5 ( y 2 cos x 1 z 3 ) i 1 ( 2 y sin x 2 4 ) j 1 3 xz 2 k is irrotational and ﬁnd its scalar
potential.

Solution.
Given F = ( y 2 cos x + z 3 )i + (2 y sin x − 4) j + 3 xz 2 k

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 27 5/19/2016 10:15:12 AM

i j k
∂ ∂ ∂
Now ∇×F =
∂x ∂y ∂z
y 2 cos x + z 3 2 y sin x − 4 3 xz 2
= i(0 − 0) − j (3 z 2 − 3 z 2 ) + k (2 y cos x − 2 y cos x ) = 0
∴ F is irrotational.
Hence, there exist a scalar function f such that F = ∇f
∂f ∂f ∂f
⇒ ( y 2 cos x + z 3 )i + (2 y sin x − 4) j + 3 xz 2 k = i +j +k
∂x ∂y ∂z

∂f ∂f ∂f
∴ = y 2 cos x + z 3 (1) = 2 y sin x − 4 (2) and = 3 xz 2 (3)
∂x ∂y ∂z
Integrating (1) w.r.to x, f = y 2 sin x + z 3 x + f1 ( y, z ) (4)
Integrating (2) w.r.to y, f = y 2 sin x − 4 y + f 2 ( x, z ) (5)
Integrating (3) w.r.to z, f = xz + f 3 ( x, y )
3
(6)

From (4), (5), (6), f = y 2 sin x + xz 3 − 4 y + c is the scalar potential, where c is an arbitrary constant.

EXAMPLE 9
(i) Find a such that (3 x 2 2 y 1 z ) i 1 (4 x 1 ay 2 z ) j 1 ( x 2 y 1 2 z ) k is solenoidal.
(ii) Find a, b, c if ( x 1 y 1 az ) i 1 (bx 1 2 y 2 z ) j 1 ( 2 x 1 cy 1 2 z ) k is irrotational.

Solution.
(i) Let F = (3 x − 2 y + z )i + (4 x + ay − z ) j + ( x − y + 2 z )k
Given F is solenoidal.
∴ ∇⋅F = 0
∂ ∂ ∂
⇒ (3 x − 2 y + z ) + ( 4 x + ay − z ) + ( x − y + 2 z ) = 0
∂x ∂y ∂z
⇒ 3 + a + 2 = 0 ⇒ a = −5

(ii) Let F = ( x + y + az )i + (bx + 2 y − z ) j + ( − x + cy + 2 z )k

Given F is irrotational.
i j k
∴ ∇×F = 0 ⇒ ∂ ∂ ∂
=0
∂x ∂y ∂z
x + y + az bx + 2 y − z − x + cy + 2 z

⇒ i ⎡ ∂ ( − x + cy + 2 z ) − ∂ (bx + 2 y − z ) ⎤ − j ⎡ ∂ ( − x + cy + 2 z ) − ∂ ( x + y + az ) ⎤
⎢ ∂y ∂z ⎥ ⎢⎣ ∂x ∂z ⎥⎦
⎣ ⎦
⎡∂ ∂ ⎤
+ k ⎢ (bx + 2 y − z ) − ( x + y + az ) ⎥ = 0
⎣ ∂x ∂y ⎦

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 28 5/19/2016 10:15:18 AM

Vector Calculus ■ 9.29

⇒ i (c + 1) − j ( −1 − a) + k (b − 1) = 0
⇒ (c + 1)i + (1 + a) j + (b − 1)k = 0
∴ c + 1 = 0, 1 + a = 0, b − 1 = 0
∴ a = −1, b = 1 and c = −1

EXAMPLE 10
Determine f(r) so that the vector f(r) r is both solenoidal and irrotational.

Solution.
If r is not speciﬁed, it will always represent the position vector of any point (x, y, z).
∴ r = xi + y j + zk and r = r = x2 + y2 + z2 ∴ r2 = x2 + y2 + z2 (1)
∴ f ( r )r = f ( r )( xi + y j + zk ) = f ( r ) xi + f ( r ) y j + f ( r ) zk
Given f(r) r is solenoidal.
∂ ∂ ∂
∴ ∇ ⋅ (f ( r ) r ) = 0 ⇒ ( f (r) x) + ( f (r) y) + ( f (r) z) = 0 (2)
∂x ∂y ∂z
∂ ∂r
But ( f ( r ) x ) = f ( r ) + xf ′( r )
∂x ∂x
∂ ∂r
( f ( r ) y ) = f ( r ) + yf ′( r )
∂y ∂y
∂ ∂r
and ( f ( r ) z ) = f ( r ) + zf ′( r )
∂z ∂z
∂r x ∂r y ∂r z
Differentiating (1) we get, = , = , =
∂x r ∂y r ∂z r
∂ x x2
∴ ( f ( r ) x ) = f ( r ) + xf ′( r ) ⋅ = f ( r ) + f ′( r )
∂x r r
∂ y2
Similarly, ( f (r) y) = f (r) + f ′( r )
∂y r
∂ z2
and ( f (r) z) = f (r) + f ′( r )
∂z r
x2 y2 z2
∴ (2) ⇒ f (r) + f ′( r ) + f ( r ) + f ′( r ) + f ( r ) + f ′( r ) = 0
r r r
f ′( r ) 2
⇒ 3 f (r) + ( x + y2 + z2 ) = 0
r
f ′( r ) 2
⇒ 3 f (r) + ⋅r = 0
r
f ′( r ) 3
⇒ 3 f ( r ) + rf ′( r ) = 0 ⇒ =−
f (r ) r
[here r is real variable.]

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 29 5/19/2016 10:15:24 AM

f ′( r ) 1
Integrating w.r.to ‘r’, we get ∫ dr = −3∫ dr
f (r) r
⇒ log e f ( r ) = −3 log e r + log c
c c
⇒ log e f ( r ) = − log e r 3 + log e c = log e ⇒ f (r) =
r3 r3
where c is the constant of integration.
i j k
∂ ∂ ∂
Now ∇ × (f ( r )r ) =
∂x ∂y ∂z
f ( r )x f ( r ) y f ( r )z
⎡∂ ∂ ⎤ ⎡∂ ∂ ⎤
= i ⎢ ( f (r ) z ) − ( f (r ) y)⎥ − j ⎢ ( f (r) z ) − ( f (r ) x)⎥
⎣ ∂ y ∂ z ⎦ ⎣ ∂ x ∂ z ⎦
⎡∂ ∂ ⎤
+ k ⎢ ( f (r ) y) − ( f (r ) x)⎥
⎣ ∂x ∂y ⎦
⎡ ∂r ∂r ⎤
= ∑ i ⎢ zf ′( r ) ⋅ − y ⋅ f ′( r ) ⋅ ⎥
⎣ ∂ y ∂z ⎦
⎡ y z⎤ ⎡ yz yz ⎤
= ∑ i ⎢ zf ′( r ) ⋅ − y ⋅ f ′( r ) ⋅ ⎥ = ∑ i f ′( r ) ⎢ − ⎥ = 0
⎣ r r⎦ ⎣r r ⎦
c
∴ f ( r )r is irrotational for all f(r) and it is solenoidal for f ( r ) =
, where c is arbitrary constant.
r3
c
Hence, the required function is f ( r ) = 3 , for which f ( r )r is both solenoidal and irrotational.
r

EXERCISE 9.2
1. If F = xy 2 + 2 x 2 yz j − 3 yz 2 k, then find div F and curl F at (1, 1, −1).
2. If F = x 2 y i + y 2 z j + z 2 x k then find curl curl F.
3. Find div F and curl F at (1, 1, 1)
if F = ( x 2 − y 2 + 2 xz )i + ( xz − xy + yz ) j + ( z 2 + x 2 )k.
4. Show that the following vectors are solenoidal.
(i) F = (2 + 3 y )i + ( x − 2 z ) j + xk
(ii) F = ( y 2 − z 2 + 3 yz − 2 x )i + (3 xz + 2 xy ) j + (3 xy − 2 xz + 2 z )k
(iii) F = 3 x 2 yi − 4 xy 2 j + 2 xyzk
5. Find the value of a if F = ay 4 z 2 i + 4 x 3 z 2 j + 5 x 2 y 2 k is solenoidal.
6. If the vector 3 xi + ( x + y ) j − azk is solenoidal, then find a.
7. Show that the following vectors are irrotational.
(i) F = ( y 2 + 2 xz 2 )i + (2 xy − z ) j + (2 x 2 z − y + 2 z )k
(ii) F = (sin y + z )i + ( x cos y − z ) j + ( x − y )k
(iii) F = (4 xy − z 2 )i + 2 x 2 j − 3 xz 2 k

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 30 5/19/2016 10:15:31 AM

Vector Calculus ■ 9.31

8. Find the value of a if F = ( axy − z 2 )i + ( x 2 + 2 yz ) j + ( y 2 − axz )k is irrotational.

9. If F = ( ax 2 + 2 y 2 + 1)i + (4 xy + by 2 z − 3) j + (c − y 3 )k is irrotational, then ﬁnd the values of

a, b, c.
10. Show that F = (2x + 3y + z 2 )i + (3x + 2 y + z ) j + ( y + 2zx )k is irrotational and hence, find its
scalar potential.
11. Prove that F = ( y 2 cos x + z 3 )i + (2 y sin x − 4) j + 3 xz 2 k is irrotational and find its scalar potential.
12. Show that F = (6 xy + z 3 )i + (3 x 2 − z ) j + (3 xz 2 − y )k is irrotational, find its scalar potential.
13. Find the div F and curl F, where F = grad ( x 3 + y 3 + z 3 − 3 xyz ).
1
14. If v = w × r , prove that w = curl v, where w is a constant vector and is the position vector of
the point (x, y, z). 2
15. If r is the position vector of a point (x, y, z) in space and A is a constant vector, prove that A r is
solenoidal.
16. Prove that the vector F = ( x + 3 y )i + ( y − 3 z ) j + ( x − 2 z )k is solenoidal.
17. Show that v = xyz 2 u is solenoidal, where
u = (2 x 2 + 8 xy 2 z )i + (3 x 3 y − 3 xy ) j − (4 y 2 z 2 + 2 x 3 z )k .

ANSWERS TO EXERCISE 9.2

1. 5; − 5i − 6 k 2. 2[ zi + x j + yk ] 3. 6; − 2i + 2k
5. a can be any real number 6. 4 8. 2
9. a = 3, b = −3, c = 2 10. f = x2 + y2 + 3xy + yz + z2x + c
11. f = y2 sinx + xz3 − 4y + c 12. f = 3x2y + xz3 − yz + c
13. div F = b( x + y + z ) Curl F= O

9.7 VECTOR IDENTITIES

We shall list the vector identities into two categories.
(i) ∇ operator applied once to point functions.
(ii) ∇ operator applied twice to point functions.

TYPE 1.
If f and g are scalar point functions we have already proved the following results.
1. ∇c = 0, where c is a constant. 2. ∇(cf) = c∇f, where c is constant.
3. ∇(f ± g) = ∇f ± ∇g 4. ∇(fg) = f∇g + g∇f
g ∇f − f ∇ g
5. ∇ ⎛⎜ ⎞⎟ =
f
⎝ g⎠ f2

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 31 5/19/2016 10:15:38 AM

6. If F and G are vector point functions, then ∇ ⋅ ( F 1 G ) 5 ∇ ⋅ F 1 ∇ ⋅ G .

Proof
⎛ ∂ ∂ ∂⎞
∇ ⋅ (F + G) = ⎜ i + j + k ⎟ ⋅ (F + G)
⎝ ∂x ∂y ∂z ⎠
⎛ ∂F ∂G ⎞ ⎛ ∂F ∂G ⎞ ⎛ ∂F ∂G ⎞
= i ⋅⎜ + + j ⋅⎜ + + k ⋅⎜ +
⎝ ∂x ∂x ⎟⎠ ⎟
⎝ ∂y ∂y ⎠ ⎝ ∂z ∂z ⎟⎠
⎛ ∂F ∂F ∂F ⎞ ⎛ ∂G ∂G ∂G ⎞
= ⎜i ⋅ + j⋅ + k ⋅ ⎟ + ⎜i ⋅ + j⋅ +k⋅
⎝ ∂x ∂y ∂z ⎠ ⎝ ∂x ∂y ∂z ⎠⎟
= ∇ ⋅ F + ∇ ⋅G
Similarly, ∇ ⋅ ( F − G ) = ∇ ⋅ F − ∇ ⋅ G ■

7. If f is a scalar point function and G is a vector point function, then ∇ ? ( fG ) 5 ∇f ? G 1 f ( ∇?G )

Proof Let G = G1 i + G2 j + G3 k, then fG = fG1 i + fG2 j + fG3 k

∂ ∂ ∂
∴ ∇ ⋅ ( fG ) = ( fG1 ) + ( fG2 ) + ( fG3 )
∂x ∂y ∂z
∂G1 ∂f ∂G2 ∂f ∂G3 ∂f
= f + G1 + f + G2 + f + G3
∂x ∂x ∂y ∂y ∂z ∂z
∂f ∂f ∂f ⎛ ∂G ∂G ∂G ⎞
= G1 + G2 + G3 + f ⎜ 1 + 2 + 3 ⎟
∂x ∂y ∂z ⎝ ∂x ∂y ∂z ⎠

∴ ∇ ⋅ ( fG ) = ∇f ⋅ G + f (∇ ⋅ G ) ■

8. If f is a scalar point function and G is a vector point function, then

∇ 3 ( fG ) 5 ∇f 3 G 1 f ( ∇ 3 G )
Proof Let G = G1 i + G2 j + G3 k ∴ fG = fG1 i + fG2 j + fG3 k
i j k
∂ ∂ ∂
Now ∇ × ( fG ) =
∂x ∂y ∂z
fG1 fG2 fG3
⎡∂ ∂ ⎤ ⎡∂ ∂ ⎤ ⎡∂ ∂ ⎤
= i ⎢ ( fG3 ) − ( fG2 ) ⎥ − j ⎢ ( fG3 ) − ( fG1 ) ⎥ + k ⎢ ( fG2 ) − ( fG1 ) ⎥
⎣ ∂y ∂z ⎦ ⎣ ∂x ∂z ⎦ ⎣ ∂x ∂y ⎦
⎡ ∂G3 ∂f ∂G2 ∂f ⎤ ⎡ ∂G3 ∂f ∂G1 ∂f ⎤
= i⎢ f + G3 −f − G2 ⎥ − j ⎢ f + G3 −f − G1 ⎥
⎣ ∂y ∂y ∂z ∂z ⎦ ⎣ ∂x ∂x ∂z ∂z ⎦
⎡ ∂G2 ∂f ∂G1 ∂f ⎤
+k⎢f + G2 −f − G1 ⎥
⎣ ∂x ∂x ∂y ∂y ⎦

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 32 5/19/2016 10:15:46 AM

Vector Calculus ■ 9.33

⎡ ⎛ ∂G ∂G ⎞ ⎛ ∂G ∂G ⎞ ⎛ ∂G ∂G ⎞ ⎤
= f ⎢⎜ 3 − 2 ⎟ i − ⎜ 3 − 1 ⎟ j + ⎜ 2 − 1 ⎟ k ⎥
⎝ ⎠
⎣ ⎝ ∂y ∂z ⎠ ∂x ∂z ⎝ ∂x ∂y ⎠ ⎦
⎛ ∂f ∂f ⎞ ⎛ ∂f ∂f ⎞ ⎛ ∂f ∂f ⎞
+ ⎜ G3 − G2 ⎟ i − ⎜ G3 − G1 ⎟ j + ⎜ G2 − G1 ⎟ k
⎝ ∂y ∂z ⎠ ⎝ ∂x ∂z ⎠ ⎝ ∂x ∂y ⎠
i j k i j k
∂ ∂ ∂ ∂f ∂f ∂f
= f +
∂x ∂y ∂z ∂x ∂y ∂z
G1 G2 G3 G1 G2 G3
∴ ∇ × ( fG ) = f (∇ × G ) + (∇f ) × G. ■

9. If F and G are vector point functions, then

∇( F ⋅ G ) 5 ( F ? ∇)G 1 (G ? ∇) F 1 F 3 ( ∇ 3 G ) 1 G 3 ( ∇ 3 F )

∂f ∂f ∂f ∂f
Proof We know that ∇f = i + j +k = ∑i
∂x ∂y ∂z ∂x
∂
∴ ∇( F ⋅ G ) = ∑ i (F ⋅ G)
∂x
⎡ ∂G ∂F ⎤
= ∑ i ⎢F ⋅ +G⋅ ⎥
⎣ ∂x ∂x ⎦
⎛ ∂G ⎞ ⎛ ∂F ⎞
= ∑⎜F ⋅ ⎟ i + ∑ ⎜⎝ G ⋅ ⎟⎠ i (1)
⎝ ∂x ⎠ ∂x
We know that a × ( b × c ) = ( a ⋅ c )b − ( a ⋅ b )c
∴ ( a ⋅ b )c = ( a ⋅ c )b − a × ( b × c )
⎛ ∂G ⎞ ⎛ ∂G ⎞ ⎛ ∂G ⎞
∴ ⎜⎝ F ⋅ ⎟⎠ i = ( F ⋅ i ) ⎜⎝ ⎟⎠ − F × ⎜⎝ × i⎟
∂x ∂x ∂x ⎠
⎛ ∂G ⎞ ⎛ ∂G ⎞
= ( F ⋅ i) ⎜ + F × ⎜i ×
⎝ ∂x ⎟⎠ ⎝ ⎟
∂x ⎠
⎛ ∂G ⎞ ⎛ ∂⎞ ⎛ ∂G ⎞
∴ ∑ ⎝⎜ F ⋅ ∂x ⎠⎟ i = ⎝⎜ F ⋅ ∑ i ∂x ⎠⎟ G + F × ∑ ⎝⎜ i × ∂x ⎟⎠
= ( F ⋅ ∇)G + F × (∇ × G ) (2)
Interchanging F and G, we get
⎛ ∂F ⎞
∑ ⎜⎝ G ⋅ ∂x ⎟⎠ i = (G ⋅ ∇) F + G × (∇ × F ) (3)

Substituting (2) and (3) in (1) we get

∇( F ⋅ G ) = ( F ⋅ ∇)G + F × (∇ × G ) + (G ⋅ ∇) F + G × (∇ × F )
∴ ∇( F ⋅ G ) = ( F ⋅ ∇)G + (G ⋅ ∇) F + F × (∇ × G ) + G × (∇ × F ) ■

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 33 5/19/2016 10:15:52 AM

10. If F and G are vector point functions then

∇ ? ( F 3 G ) 5 G ? ( ∇ 3 F ) 2 F ?( ∇ 3 G )
i .e., div F 3 G 5 G ? Curl F 2 F ? Curl G .

Proof
∂
∇ ⋅ (F × G) = ∑ i ⋅ (F × G)
∂x
⎛ ∂F ∂G ⎞
= ∑i⋅⎜ ×G + F × ⎟
⎝ ∂x ∂x ⎠
⎛ ∂F ⎞ ⎛ ∂G ⎞
= ∑i⋅⎜ × G⎟ + ∑ i ⋅ ⎜ F × ⎟
⎝ ∂x ⎠ ⎝ ∂x ⎠
In a scalar triple product ⋅ and × can be interchanged.
⎛ ∂F ⎞ ⎛ ∂G ⎞
∴ we get ∇ ⋅ (F × G) = ∑ ⎜ i × ⎟ ⋅G − ∑ ⎜i × ⎟ ⋅F
⎝ ∂x ⎠ ⎝ ∂x ⎠
⇒ ∇ ⋅ ( F × G ) = (∇ × F ) ⋅ G − (∇ × G ) ⋅ F ■
11. If F and G are vector product functions, then
∇ × ( F × G ) = F (∇ ⋅ G ) − G (∇ ⋅ F ) + (G ⋅ ∇) F − ( F ⋅ ∇)G
Proof
∂
∇ × (F × G) = ∑ i × (F × G)
∂x
⎛ ∂F ∂G ⎞
= ∑i ×⎜ ×G + F × ⎟
⎝ ∂x ∂x ⎠
⎛ ∂F ⎞ ⎛ ∂G ⎞
⇒ ∇ × (F × G ) = ∑ i × ⎜ × G⎟ + ∑ i × ⎜ F × ⎟ (1)
⎝ ∂x ⎠ ⎝ ∂x ⎠
We know a × ( b × c ) = ( a ⋅ c )b − ( a ⋅ b )c
⎛ ∂F ⎞ ⎡ ∂F ⎛ ∂F ⎞ ⎤
∴ ∑ i × ⎜⎝ ∂x × G ⎟⎠ = ∑ ⎢⎣(i ⋅ G ) ∂x − ⎜⎝ i ⋅ ∂x ⎟⎠ G ⎥⎦
∂F ⎛ ∂F ⎞
= ∑ (G ⋅ i ) − ∑ ⎜i ⋅ ⎟ G
∂x ⎝ ∂x ⎠ ■
⎛ ∂⎞ ⎛ ∂F ⎞
= G ⋅⎜∑i ⎟ F − ⎜∑i⋅ ⎟ G
⎝ ∂x ⎠ ⎝ ∂x ⎠
⎛ ∂F ⎞
⇒ ∑ i × ⎜⎝ ∂x × G ⎟⎠ = (G ⋅ ∇)F − (∇ ⋅ F )G (2)

⎛ ∂G ⎞ ⎡ ⎛ ∂G ⎞ ∂G ⎤
Similarly, ∑ i × ⎜⎝ F × ∂x ⎟⎠ = ∑ ⎢⎣⎜⎝ i ⋅ ∂x ⎟⎠ F − ∑ (i ⋅ F ) ∂x ⎥⎦

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 34 5/19/2016 10:15:58 AM

Vector Calculus ■ 9.35

⎛ ∂G ⎞ ∂G
= ∑ ⎜i⋅ F − ∑ (i ⋅ F )
⎝ ∂x ⎟⎠ ∂x
⎛ ∂G ⎞ ⎛ ∂⎞
= ⎜∑i⋅ ⎟ F − ∑ ⎜⎝ F ⋅ i ⎟⎠ G
⎝ ∂x ⎠ ∂x
⎛ ∂⎞
= (∇ ⋅ G ) F − F ⋅ ⎜ ∑ i ⎟ G
⎝ ∂x ⎠
⎛ ∂G ⎞
⇒ ∑ i × ⎜⎝ F × ∂x ⎟⎠ = (∇ ⋅ G )F − (F ⋅ ∇)G (3)
Substituting (2) and (3) in (1), we get
∇ × ( F × G ) = (G ⋅ ∇) F − (∇ ⋅ F )G + (∇ ⋅ G ) F − ( F ⋅ ∇)G
∴ Curl F × G = F (div G ) − G (div F ) + (G ⋅ ∇) F − ( F ⋅ ∇)G ■

TYPE II – Identities – ∇ Applied Twice

∂2 f ∂2 f ∂2 f
1. If f is scalar point function, then div grad f 5 ∇ 2 f 5 1 1
∂x 2 ∂y 2 ∂z 2
∂f ∂f ∂f
Proof We know, grad f = ∇f = i +j +k
∂x ∂y ∂z
div (grad f ) = ∇ ⋅ ∇f
⎛ ∂ ∂ ∂ ⎞ ⎛ ∂f ∂f ∂f ⎞
= ⎜i + j + k ⎟ ⋅⎜i + j + k ⎟
⎝ ∂x ∂y ∂z ⎠ ⎝ ∂x ∂y ∂z ⎠
∂ ⎛ ∂f ⎞ ∂ ⎛ ∂f ⎞ ∂ ⎛ ∂f ⎞ ∂ 2 f ∂ 2 f ∂ 2 f
= ⎜ ⎟+ + ⎜ ⎟= + +
∂x ⎝ ∂x ⎠ ∂y ⎝⎜ ∂y ⎠⎟ ∂z ⎝ ∂z ⎠ ∂x 2 ∂y 2 ∂z 2
∂2 f ∂2 f ∂2 f
∴ div (grad f ) = ∇ 2 f = + + ■
∂x 2 ∂y 2 ∂z 2

∂2 ∂2 ∂2
Note ∇ ⋅ ∇ = ∇ 2 = + + is a scalar operator called the Laplacian operator.
∂x 2 ∂y 2 ∂z 2

2. If F is a vector point function, then div curl F 5 0.

Proof Let F = F1 i + F2 j + F3 k, where F1, F2, F3 are scalar functions of x, y, z.

i j k
∂ ∂ ∂
Curl F = ∇ × F =
∂x ∂y ∂z
F1 F2 F3
⎡ ∂F ∂F ⎤ ⎡ ∂F ∂F ⎤ ⎡ ∂F ∂F ⎤
= i⎢ 3 − 2 ⎥− j⎢ 3 − 1⎥+k⎢ 2 − 1⎥
⎣ ∂y ∂z ⎦ ⎣ ∂x ∂z ⎦ ⎣ ∂x ∂y ⎦

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 35 5/19/2016 10:16:02 AM

∴ ∂ ⎛ ∂F3 ∂F2 ⎞ ∂ ⎛ ∂F1 ∂F3 ⎞ ∂ ⎛ ∂F2 ∂F1 ⎞

div Curl F = ∇ ⋅ ∇ × F = ⎜⎝ ∂y − ∂z ⎟⎠ + ∂y ⎜⎝ ∂z − ∂x ⎟⎠ + ∂z ⎜⎝ ∂x − ∂y ⎟⎠
∂x
∂ 2 F3 ∂ 2 F2 ∂ 2 F1 ∂ 2 F3 ∂ 2 F2 ∂ 2 F1
= − + − + −
∂x ∂y ∂x ∂z ∂y ∂z ∂y ∂x ∂z ∂x ∂z ∂y
⎡ ∂ 2 F3 ∂ 2 F3 ∂ 2 F2 ∂ 2 F2 ∂ 2 F1 ∂ 2 F1 ⎤
⇒ div Curl F = 0 ⎢Since = , = , = ⎥ ■
⎣ ∂ x ∂ y ∂y ∂x ∂x ∂z ∂z ∂x ∂y ∂z ∂z ∂y ⎦
3. If F is a vector point function, then
curl (Curl F ) 5 ∇ 3 ( ∇ 3 F ) 5 ∇( ∇ ? F ) 2 ∇ 2 F.

Proof Let F = F1 i + F2 j + F3 k, where F1, F2, F3 are scalar functions.

i j k
∂ ∂ ∂
Then Curl F =
∂x ∂y ∂z
F1 F2 F3
⎛ ∂F ∂F ⎞ ⎛ ∂F ∂F ⎞ ⎛ ∂F ∂F ⎞
= i⎜ 3 − 2 ⎟ − j⎜ 3 − 1⎟ + k ⎜ 2 − 1⎟
⎝ ∂y ∂z ⎠ ⎝ ∂x ∂z ⎠ ⎝ ∂x ∂y ⎠
i j k
∂ ∂ ∂
∴ Curl curl F =
∂x ∂y ∂z
∂F3 ∂F2 ∂F1 ∂F3 ∂F2 ∂F1
− − −
∂y ∂z ∂z ∂x ∂x ∂y
⎡ ∂ ⎛ ∂F ∂F ⎞ ∂ ⎛ ∂F ∂F ⎞ ⎤
= ∑i ⎢ ⎜ 2 − 1⎟ − ⎜ 1 − 3 ⎟⎥
⎣ ∂y ⎝ ∂x ∂y ⎠ ∂z ⎝ ∂z ∂x ⎠ ⎦
⎡ ∂ F2 ∂ F1 ∂ F1 ∂ F3 ⎤
2 2 2 2
= ∑i ⎢ − 2 − 2 + ⎥
⎣ ∂y ∂x ∂y ∂z ∂z ∂x ⎦
⎪⎧ ∂ F2 ∂ F3 ⎛ ∂ F1 ∂ F1 ⎞ ⎪⎫
2 2 2 2
= ∑i⎨ + −⎜ 2 + 2 ⎟⎬
⎩⎪ ∂y∂x ∂z∂x ⎝ ∂y ∂z ⎠ ⎭⎪
⎧⎪ ∂ 2 F ∂ 2 F2 ∂ 2 F3 ⎛ ∂ 2 F1 ∂ 2 F1 ∂ 2 F1 ⎞ ⎫⎪
= ∑ i ⎨ 21 + + − + 2 + 2 ⎟⎬
⎪⎩ ∂x ∂x∂y ∂x∂z ⎜⎝ ∂x 2 ∂y ∂z ⎠ ⎪⎭

⎪⎧ ∂ ⎛ ∂F ∂F ∂F ⎞ ⎛ ∂ F ∂ F ∂ F ⎞ ⎪⎫
2 2 2
= ∑ i ⎨ ⎜ 1 + 2 + 3 ⎟ − ⎜ 21 + 21 + 21 ⎟ ⎬
⎩⎪ ∂x ⎝ ∂x ∂y ∂z ⎠ ⎝ ∂x ∂y ∂z ⎠ ⎭⎪
⎪⎧ ∂ ⎛ ∂2 ∂2 ∂ 2 ⎞ ⎪⎫
= ∑ i ⎨ (∇ ⋅ F ) − ⎜ 2 + 2 + 2 ⎟ F1 ⎬
⎪⎩ ∂x ⎝ ∂x ∂y ∂z ⎠ ⎪⎭

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 36 5/19/2016 10:16:06 AM

Vector Calculus ■ 9.37

= ∑i {
∂
∂x
(∇ ⋅ F ) − ∇ 2 F1 }
∂⎞
⎛
= ⎜ ∑ i ⎟ (∇ ⋅ F ) − ∇2
⎝ ∂x ⎠
(∑ iF )1

∴ ∇ × (∇ × F ) = ∇(∇ ⋅ F ) − ∇ 2 F ■

WORKED EXAMPLES
EXAMPLE 1
⎛ 1⎞ n
Prove that ∇ ⎜ n ⎟ 52 n12 r .
⎝r ⎠ r
Solution.
We have r = xi + y j + zk and r2 = x2 + y2 + z2

∂r x ∂r y ∂r z
∴ = , = , =
∂x r ∂y r ∂z r
⎛ 1⎞ ∂ ⎛ 1⎞ ∂ ⎛ 1⎞ ∂ ⎛ 1⎞
∴ ∇⎜ n ⎟ = i ⎜ n ⎟ + j ⎜ n ⎟ + k ⎜ n ⎟
⎝r ⎠ ⎝
∂x r ⎠ ∂y r⎝ ⎠ ∂z ⎝ r ⎠
⎛ − n ∂r ⎞ ⎛ − n ∂r ⎞ ⎛ − n ∂r ⎞
= i ⎜ n +1 ⎟ + j ⎜ n +1 ⎟ + k ⎜ n +1 ⎟
⎝ r ∂x ⎠ ⎝ r ∂y ⎠ ⎝ r ∂z ⎠
n ⎡ x y z⎤ n n
=− ⎢⎣i r + j r + k r ⎥⎦ = − r n + 2 ( xi + y j + zk ) = − r n + 2 r
r n +1
⎛ 1⎞ n
∴ ∇ ⎜ n ⎟ = − n+ 2 r
⎝r ⎠ r

⎛ 1⎞ n
Note We have ∇ ⎜ n ⎟ = − n+ 2 r
⎝r ⎠ r
If n = 1, 2, 3, 4, …
⎛ 1⎞ 1 ⎛ 1⎞ 2 ⎛ 1⎞ 3 ⎛ 1⎞ 4
Then ∇ ⎜ ⎟ = − 3 r, ∇⎜ 2 ⎟ = − 4 r , ∇⎜ 3 ⎟ = − 5 r, ∇ ⎜ 4 ⎟ = − 6 r and so on..
⎝ r⎠ r ⎝r ⎠ r ⎝r ⎠ r ⎝r ⎠ r

EXAMPLE 2
Prove that ∇ 2 ( r n ) 5 n( n 11) r n22 .

Solution.
We have r = xi + y j + zk and r2 = x2 + y2 + z2

∂r x ∂r y ∂r z
∴ = , = , =
∂x r ∂y r ∂z r

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 37 5/19/2016 10:16:10 AM

∂ n ∂r
∴ ∇( r n ) = ∑ i ( r ) = ∑ i nr n −1
∂x ∂x
n−2
= ∑ i nr n −1 x
= nr n − 2 ∑ xi = nr n − 2 ( xi + y j + zk ) = nr r if n ≥ 3 (1)
r
∇ 2 ( r n ) = ∇ ⋅ (∇r n ) = ∇ ⋅ ( nr n − 2 r )
= n[∇r n − 2 ⋅ r + r n − 2 (∇ ⋅ r )]
= n[( n − 2)r n − 4 r ⋅ r + r n − 2 3] [ using (1)]

= n[( n − 2)r n − 4 r 2 + 3r n − 2 ] = nr n − 2 [n − 2 + 3] = n ( n + 1)r n − 2

Note We have ∇( r n ) = nr n − 2 r
If n = 1, 2, 3, 4, …
1
∇( r ) = r ∇( r 2 ) = 2 ⋅ r 2 − 2 r = 2r , ∇( r 3 ) = 3rr , ∇( r 4 ) = 4 r 2 r …∇( r n − 2 ) = ( n − 2)r n − 4 r , etc.
r ,

EXAMPLE 3
⎛ ⎛ 1 ⎞⎞ 3
Prove that ∇ ? ⎜ r ∇ ⎜ 3 ⎟ ⎟ 5 4 .
⎝ ⎝ r ⎠⎠ r
Solution.
⎛ 1⎞ 3 ⎛ 1 ⎞ −4
We have ∇ ⎜ 3 ⎟ = − 5 r , ∇ ⎜⎝ r 4 ⎟⎠ = r 6 r and ∇ ⋅ r = 3
⎝r ⎠ r
⎛ ⎛ 1 ⎞⎞ ⎛ −3 ⎞ ⎛ −3 ⎞
∴ ∇ ⋅ ⎜ r∇ ⎜ 3 ⎟ ⎟ = ∇ ⋅ ⎜ r 5 r ⎟ = ∇ ⋅ ⎜ 4 r ⎟
⎝ ⎝ r ⎠⎠ ⎝ r ⎠ ⎝r ⎠
⎡ ⎛ 1⎞ 1 ⎤
= −3 ⎢∇ ⎜ 4 ⎟ ⋅ r + 4 ∇ ⋅ r ⎥
⎣ ⎝ r ⎠ r ⎦
⎡ 4 3 ⎤ ⎡ 4 3⎤ ⎡ −4 3 ⎤ 3
= −3 ⎢ − 6 ( r ⋅ r ) + 4 ⎥ = −3 ⎢ − 6 r2 + 4 ⎥ = −3 ⎢ 4 + 4 ⎥ = 4
⎣ r r ⎦ ⎣ r r ⎦ ⎣r r ⎦ r

EXAMPLE 4
If f and c satisfy Laplace equation, prove that the vector (f ∇ c − c ∇ f) is solenoidal.

Solution.
Given f and c satisfy Laplace equation.

∂ 2f ∂ 2f ∂ 2f ∂2c ∂2c ∂2c

∴ + + =0 (1) and + + =0 (2)
∂x 2 ∂y 2 ∂z 2 ∂x 2 ∂y 2 ∂z 2
To prove (f ∇ c − c ∇ f) is solenoidal, we have to prove div (f ∇ c − c ∇ f) = 0
Now div (f ∇ c − c ∇ f) = ∇⋅(f ∇ c − c ∇ f)

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 38 5/19/2016 10:16:15 AM

Vector Calculus ■ 9.39

= ∇ ⋅ (f∇c) − ∇ ⋅ (c∇f)
= ∇f ⋅ ∇c + f(∇ ⋅ ∇c) − [∇c ⋅ ∇f + c(∇ ⋅ ∇f)]
= f∇ 2 c − c∇ 2f [{ ∇f ⋅ ∇c = ∇c ⋅ ∇f]
=0 [from (1) and ( 2)]

∴ (f ∇ c − c ∇ f) is solenoidal.

EXAMPLE 5
d 2f 2 df
Show that ∇2 f ( r ) 5 2
1 .
dr r dr
Solution.
r
We have, ∇f ( r ) = f ′ ( r )
r
r ⎛ f ′ ( r) ⎞ f ′( r )
∴ ∇ 2 f ( r ) = ∇ ⋅ ∇f ( r ) = ∇ ⋅ f ′ ( r ) = ⎜ ∇ ⎟⎠ ⋅ r + (∇ ⋅ r )
r ⎝ r r
⎛ f ′( r ) ⎞ 3f ′( r )
= ⎜∇ ⎟ ⋅r + [{ ∇ ⋅ r = 3]
⎝ r ⎠ r
⎛ r ∇f ′ ( r ) − f ′ ( r ) ∇r ⎞ 3 f ′( r ) ⎡ ⎛ f ⎞ g ∇f − f ∇g ⎤
=⎜
⎝ ⎟⎠ ⋅ r + ⎢{ ∇ ⎜⎝ g ⎟⎠ = g2 ⎥
r2 r ⎣ ⎦
⎛ r r⎞ ⎡ r
⎜⎝ rf ′′( r ) − f ′( r ) ⎟⎠ ⋅ r 3 f ′( r ) ⎢⎣{ ∇f ′( r ) = f ′′( r ) r
= r r +
r2 r r⎤
and ∇r = ⎥
[rf ′′( r ) − f ′( r )]r ⋅ r 3 f ′( r ) r⎦
= +
r3 r
[rf ′′( r ) − f ′( r )] 2 3 f ′( r )
= r + [{ r ⋅ r = r 2 ]
r3 r
rf ′′( r ) − f ′( r ) 3f ′( r ) 2f ′( r ) d 2 f 2 df
= + = f ′′( r ) + = 2 +
r r r dr r dr

9.8 INTEGRATION OF VECTOR FUNCTIONS

d
Let f (t ) and F (t ) be two vector functions of a scalar variable t such that F (t ) = f (t ). Then F (t ) is
dt
called an indeﬁnite integral of f (t ) with respect to t and is written as ∫ f (t )dt = F (t ) + c , where c is an
arbitrary constant vector independent of t and is called the constant of integration.
The deﬁnite integral of f (t ) between the limits t = t1 and t = t2 is given by
t2

∫ f (t ) dt = ⎡⎣ F (t ) ⎤⎦
t2
t1
= F (t 2 ) − F (t1 ).
t1

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 39 5/19/2016 10:16:24 AM

As in the case of differentiation of vectors, in order to integrate a vector function, we integrate its
components.
If f (t ) = f1 (t ) i + f 2 (t ) j + f 3 (t ) k, then

∫ f (t ) dt = i∫ f (t ) dt + j ∫ f
1 2 (t ) dt + k ∫ f 3 (t ) dt

9.8.1 Line Integral

An integral evaluated over a curve C is called a line integral. We call C as the path of integration. We
assume every path of integration of a line integral to be piecewise smooth consisting of ﬁnitely many
smooth curves.

Definition 9.10 A line integral of a vector point function F ( r ) over a curve C, where r is the position
vector of any point on C, is defined by ∫ F ⋅ dr
C
If F = F1 i + F2 j + F3 k and r = x i + y j + zk , then
dr = dx i + dy j + dz k and ∫ F ⋅ dr = C∫ F dx + F
C
1 2 dy + F3 dz

Here F1, F2, F3 are functions of x, y, z, where x, y, z depend on a parameter t ∈ [a, b], since r (t ) is the
equation of the curve C.
⎛ dx dz ⎞
b
dy
Then we can write ∫ F ⋅ dr = ∫ ⎜ F1 + F2 + F3 ⎟ dt .
C a
⎝ dt dt dt ⎠
If the path of integration C is a closed curve, we write ∫ instead of ∫
C
.
C
Note
dr
1. Since is a tangent vector to the curve C the line integral ∫ F ⋅ dr is also called the tangential
dt C
line integral of F over C and line integral is a scalar.
2. Two other types of line integrals are also considered. ∫ F × dr and ∫ f dr are vectors.
C C

WORKED EXAMPLES
EXAMPLE 1

If F 5 3 xy i 2 y 2 j , evaluate ∫ F ? dr where C is the arc of the parabola y 5 2x

C
2
from (0, 0) to
(1, 2).

Solution.
Given F = 3xy i − y 2 j
r = x i + y j , where r is the position vector of any point (x, y) on y = 2x2.
∴ dr = dx i + dy j
and F ⋅ dr = (3 xy i − y 2 j ) ⋅ ( dxi + dy j ) = 3xydx − y2dy

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 40 5/19/2016 10:16:31 AM

Vector Calculus ■ 9.41

∴ ∫ F ⋅ dr = C∫ (3xy dx − y dy)
2

Equation of C is y = 2x2 ∴ dy = 4x dx.

1
A (1, 2)
∴ ∫ F ⋅ dr = ∫ (3x ⋅ 2 x dx − 4 x ⋅ 4 x dx )
2 4

C 0
1

= ∫ (6 x 3 − 16 x 5 ) dx
0
O
1
⎡ x x ⎤
4
3 8 9 − 16 6
7 (0, 0) x=1
= ⎢6 − 16 ⎥ = − = =−
⎣ 4 6 ⎦0 2 3 6 6
Fig. 9.2
EXAMPLE 2
If F 5 (3 x 2 1 6 y ) i 2 14 yz j 1 20 xz 2 k , evaluate ∫ F ? dr from (0, 0, 0) to (1, 1, 1) along the curve
C given by x 5 t, y 5 t2, z 5 t3. C

Solution.
Given F = (3 x 2 + 6 y )i − 14 yz j + 20 xz 2 k
and r = xi + y j + zk ∴ dr = dx i + dy j + dz k
and F ⋅ dr = ⎣⎡(3 x 2 + 6 y )i − 14 yz j + 20 xz 2 k ⎦⎤ ⋅ ⎣⎡ dxi + dy j + dzk ⎦⎤
= (3 x 2 + 6 y )dx − 14 yzdy + 20 xz 2 dz
∴ ∫ F ⋅ dr = C∫ (3x + 6 y ) dx − 14 yz dy + 20 xz 2 dz
2

C
Given x = t, y = t2, z = t3 is the curve.
∴ dx = dt, dy = 2t dt, dz = 3t2 dt
When x = 0, t = 0 and x = 1, t = 1. Limits for t are t = 0, t = 1
1
∴ ∫ F ⋅ dr = ∫ (3 ⋅ t + 6 ⋅ t )dt − 14 ⋅ t ⋅ 2t dt + 20t ⋅ 3t dt
2 2 5 7 2

C 0
1
⎡ t3 t7 t 10 ⎤
= ∫ (9t − 28t + 60t ) dt = ⎢9 − 28 + 60 ⎥ = 3 − 4 + 6 = 5.
1
2 6 9
0 ⎣ 3 7 10 ⎦ 0

EXAMPLE 3

∫ (y dx 2 x 2 dy ) around the triangle whose vertices are (1, 0), (0, 1),
2
Evaluate the line integral
C
(−1, 0) in the positive sense.

Solution.
Given the path C consists of the sides of the Δ ABC, where A(−1, 0), B(1, 0) and C(0, 1).
Equation of AB is y = 0
y − 0 x −1
Equation of BC is = ⇒ y = −x + 1
0 −1 1− 0

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 41 5/19/2016 10:16:36 AM

y −1 x − 0 y
Equation of CA is = ⇒ y = x +1
1− 0 0 +1 C (0, 1)

∫ F ⋅ dr = AB∫ ( y dx − x dy)
2 2

+ ∫ ( y dx − x dy)
2 2

+ ∫ ( y dx − x dy)
2 2
O
CA (−1, 0)A B (1, 0)

Fig. 9.3
On AB, y = 0, ∴ dy = 0 and x varies from −1 to 1
1

∴ ∫
AB
( y 2dx − x 2dy ) = ∫ 0 dx = 0
−1

On BC, y = −x + 1 ∴ dy = −dx and From B to C, x varies from 1 to 0.

0 0
∴
BC
∫ ( y 2 dx − x 2 dy ) = ∫ ( − x + 1) 2 dx − x 2 ( − dx ) = ∫ ( x 2 − 2x + 1 + x 2 )dx
1 1
0

= ∫ ( 2x 2 − 2x + 1)dx
1
0
⎡ x3 x2 ⎤ ⎛2 ⎞ 2
= ⎢ 2 − 2 + x ⎥ = 0 − ⎜ − 1 + 1⎟ = −
⎣ 3 2 ⎦1 ⎝ 3 ⎠ 3

On CA, y = x + 1 ∴ dy = dx and From C to A, x varies from 0 to −1

−1 −1

∴ ∫ ( y 2 dx − x 2 dy ) = ∫ ( x + 1) 2 dx − x 2 dx = ∫ ( x + 2x + 1 − x )dx
2 2

CA 0 0
−1

= ∫ ( 2x + 1)dx = [ x 2 + x ]0 = 1 − 1 − 0 = 0
−1

⎛ 2⎞ 2
∴ ∫ F ⋅ dr = 0 + ⎜⎝ − 3 ⎟⎠ + 0 = − 3
C

EXAMPLE 4

If F 5 (3 x 2 1 6 y ) i 2 14 yz j 1 20 xz 2 k , evaluate ∫ F ? dr , where C is the straight line joining

C
(0, 0, 0) to (1, 1, 1).

Solution.
Given F = (3 x 2 + 6 y )i − 14 yz j + 20 xz 2 k
r = x i + y j + zk ∴ dr = dx i + dy j + dzk

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 42 5/19/2016 10:16:40 AM

Vector Calculus ■ 9.43

∴ F ⋅ dr = ⎣⎡(3 x 2 + 6 y )i − 14 yz j + 20 xz 2 k ⎦⎤ ⋅ ⎣⎡ dxi + dy j + dzk ⎦⎤

= (3 x 2 + 6 y )dx − 14 yz dy + 20 xz 2 dz

∴ ∫ F ⋅ dr = C∫ (3x + 6 y )dx − 14 yz dy + 20 xz 2 dz
2

Equation of the line joining (0, 0, 0) to (1, 1, 1) is

x−0 y−0 z−0
= = ⇒ x = y = z = t , say
1− 0 1− 0 1− 0
∴ dx = dt, dy = dt, dz = dt
At the point (0, 0, 0), t = 0 and at the point (1, 1, 1), t = 1
1

∴ ∫ F ⋅ dr = ∫ (3t + 6t )dt − 14t dt + 20t dt

2 2 3

C 0

= ∫ (3t 2 + 6t − 14t 2 + 20t 3 )dt

0
1
1
⎡ t4 t3 t2 ⎤ 11 13
= ∫ ( 20t − 11t + 6t )dt = ⎢ 20 − 11 + 6 ⎥ = 5 − + 3 =
3 2

0 ⎣ 4 3 2 ⎦0 3 3

Definition 9.11 Work Done by a Force B

If F ( x, y, z ) is a force acting on a particle which is moved along arc AB then ∫ F ⋅ dr gives the total
work done by the force F in displacing the particle from A to B. A

Conservative force field

A line integral ∫ F ⋅ dr is independent of path in domain D if and only if F = ∇f for some scalar
C
function f defined in D. Such a force field is called a conservative field.
In the conservative field the total work done by F from A to B is

∫ F ⋅ dr = C∫ ∇f ⋅ dr
C
B

⎛ ∂f ∂f ∂f ⎞
= ∫⎜ dx + dy + dz
C ⎝ ∂x ∂y ∂z ⎠⎟
B
= ∫ df = ∫ df
C A

∴ ∫ F ⋅ dr = [f] = f( B) − f( A)
B
A
A
C Fig. 9.4

So, in a conservative ﬁeld the work done depends on the value of f at the end points A and B of the
path, but not on the path.

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 43 5/19/2016 10:16:44 AM

Note
1. f is scalar potential.
2. If F is conservative, then F = ∇f ⇒ ∇ × F = ∇ × ∇ f = 0
∴ F is irrotational.
3. If C is a simple closed curve and F is conservative, then ∫ F ⋅ dr = 0 .
C

WORKED EXAMPLES
EXAMPLE 5
Show that F 5 ( e x z 2 2 xy ) i 2 ( x 2 2 1) j 1 ( e x 1 z ) k is a conservative ﬁeld. Hence, evaluate
∫ F ? dr where the end points of C are (0, 1, −1) and (2, 3, 0).
C

Solution.
To prove that F is conservative, we have to prove ∇ × F = 0
i j k
∂ ∂ ∂
Now ∇× F =
∂x ∂y ∂z
e x z − 2 xy 1 − x 2 ex + z
= i[0] − j (e x − e x ) + k ( −2 x + 2 x ) = 0
Hence, F is conservative. ∴ F = ∇f
∂f ∂f ∂f
⇒ (e x z − 2 xy )i + (1 − x 2 ) j + (e x + z )k = i +j +k
∂x ∂y ∂z
∂f ∂f ∂f
∴ = e x z − 2 xy (1) = 1 − x2 (2) = ex + z (3)
∂x ∂y ∂z
Integrating (1) w. r. to x, f = zex − x2y + f1(y, z)
Integrating (2) w. r. to y, f = (1 − x2)y + f2(x, z)
z2
Integrating (3) w. r. to z, f = ex z + + f3 ( x, y )
2
z2
∴ f = ze x − x 2 y + y + +C
2

∫ F ⋅ dr = [ f ]
(2, 3, 0)
∴ (0, 1, −1)
C

z2
= [ze x − x 2 y + y + + c ](( 02,,13,,−01))
2
⎡ ⎛ 1 ⎞⎤ 1 19
= ⎢0 − 22 ⋅ 3 + 3 + C − ⎜ −1 − 0 + 1 + + C ⎟ ⎥ = −12 + 3 − = − .
⎣ ⎝ 2 ⎠⎦ 2 2

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 44 5/19/2016 10:16:51 AM

Vector Calculus ■ 9.45

EXAMPLE 6
If F 5 (4 xy 2 3 x 2 z 2 ) i 1 2 x 2 j 2 2 x 3 z k , then check whether the integral ∫ F ? d r is independent
of the path C. C

Solution.
Given F = (4 xy − 3x 2 z 2 ) i + 2x 2 j − 2x 3 zk

i j k
∂ ∂ ∂
Now ∇× F =
∂x ∂y ∂z
4 xy − 3 x 2 z 2 2x2 −2 x 3 z

⎧∂ ∂ ⎫
= i ⎨ ( −2 x 3 z ) − (2 x 2 )⎬ − j
⎩ ∂y ∂z ⎭
∂
∂x { ∂
}
( −2 x 3 z ) − (4 xy − 3 x 2 z 2 )
∂z
⎧∂ ∂ ⎫
+ k ⎨ (2 x 2 ) − (4 xy − 3 x 2 z 2 )⎬
⎩ ∂ x ∂ y ⎭
= i{0 − 0} − j{−6 x 2 z + 6 x 2 z} + k{4 x − 4 x} = 0

∴ F is conservative.
Hence, ∫ F ⋅ dr is independent of the path C.
C

EXAMPLE 7
Show that F 5 (2 xy 1 z 3 ) i 1 x 2 j 1 3xz 2 k is a conservative ﬁeld. Find the scalar potential and
work done in moving an object in this ﬁeld from (1, −2, 1) to (3, 1, 4).

Solution.
Given F = (2 xy + z 3 )i + x 2 j + 3 xz 2 k

i j k
∂ ∂ ∂
Now ∇× F =
∂x ∂y ∂z
2 xy + z 3 x2 3 xz 2
⎡∂ ∂ ⎤ ⎡∂ ∂ ⎤
= i ⎢ (3 xz 2 ) − ( x 2 ) ⎥ − j ⎢ (3 xz 2 ) − (2 xy + z 3 ) ⎥
⎣ ∂y ∂z ⎦ ⎣ ∂x ∂z ⎦
⎡∂ ∂ ⎤
+ k ⎢ ( x 2 ) − (2 xy + z 3 ) ⎥
⎣ ∂ x ∂ y ⎦
= i[0 − 0] − j[3 z 2 − 3 z 2 ] + k [2 x − 2 x ] = 0

∴ F is conservative.
So, there exists a scalar function f such that F = ∇f.

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 45 5/19/2016 10:16:56 AM

∂f ∂f ∂f
⇒ (2 xy + z 3 )i + x 2 j + 3 xz 2 k = i +j +k
∂x ∂y ∂z

∂f ∂f ∂f
∴ = 2 xy + z 3 (1) = x2 (2) = 3 xz 2 (3)
∂x ∂y ∂z
Integrating (1) partially w.r.to x, f = x2y + z3x + f1(y, z)
Integrating (2) partially w.r.to y, f = x2y + f2(x, z)
Integrating (3) partially w.r.to z, f = xz3 + f3(x, y)
∴ f = x2y + xz3 + C
Since F is conservative, work done by the force F from (1, −2, 1) to (3, 1, 4) is equal to

[f]((13,, −1,24, )1) = ⎡⎣ x 2 y + xz 3 + C ⎤⎦ (1, − 2, 1)

( 3, 1, 4 )

= 32 ⋅1 + 3 ⋅ 43 + C − ⎡⎣(12 ( −2) + 1⋅13 ) + C ⎤⎦ = 9 + 192 + 1 = 202 units.

EXERCISE 9.3
1. Prove that if F = f∇c, then F ⋅ (∇ × F ) = 0.
2. Prove that Curl (f grad f) = 0.
3. Show that ∇ ⋅ (f∇c − c∇f) = f∇ 2 c − c∇ 2f.

4. Prove that ∇ × (f∇c) = ∇f × ∇c .

5. Prove that ∇ × [ f ( r )r ] = 0.
∫ F ⋅ dr, where F = (2 xy + z )i + x j + 3 xz 2 k, along the straight line joining the points
3 2
6. Evaluate
C
(1, −2, 1) and (3, 2, 4).
7. Find ∫ F ⋅ dr where F = (2 y + 3)i + xz j + ( yz − x)k along the line joining the points (0, 0, 0) to
C
(2, 1, 1).
8. Find the work done in moving a particle in the force field F = 3 x 2 i + (2 xz − y ) j − zk from t = 0 to
t = 1 along the curve x = 2t2, y = t, z = 4t3.
9. Show that F = (2 xy + z 3 )i + x 2 j + 3 xz 2 k is conservative. Find its scalar potential and find the
work done in moving a particle from (1, −2, 1) to (3, 1, 2).
10. Find the work done by the force F = − xyi + y 2 j + zk in moving a particle over a circular path
x 2 + y 2 = 4, z = 0 from (2, 0, 0) to (0, 2, 0).
11. Find the work done when a force F = ( x 2 − y 2 + x )i − (2 xy + y ) j moves a particle in the xy plane
from (0, 0) to (1, 1) along the curve y2 = x. If the path is y = x, whether the work done is different
or same. If it is same, state the reason.
12. Find the total work done in moving a particle in a force field given by F = 3 xyi − 5 z j + 10 xk
along the curve x = t2 + 1, y = 2t2, z = t3 from t = 1 to t = 2.

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 46 5/19/2016 10:17:03 AM

Vector Calculus ■ 9.47

13. For the vector function F = 2 xyi + ( x 2 + 2 yz ) j + ( y 2 + 1)k determine

C
∫ F ⋅ dr around the unit
circle with centre at the origin in the xy plane.
14. Evaluate ∫ F ⋅ dr, where F = ( x − 3 y)i + ( x − 2 y) j and C is the closed curve in the xy plane.
C
x = 2 cos t, y = 2 sin t and t = 0 to t = 2p.

15. Prove that ∇ 2 ⎛⎜ ⎞⎟ = 0.

1
⎝ r⎠
16. Prove that ∇ × (∇r n ) = 0.
17. If F = 5 xyi + 2 y j , then evaluate ∫ F ⋅ dr, where C is the part of the curve y = x2 between x = 1 and
C
x = 2.

18. Show that the vector ﬁeld F, where

F = ( y + y 2 + z 2 )i + ( x + z + 2 xy ) j + ( y + 2 xz )k, is conservative and ﬁnd its scalar potential.

ANSWERS TO EXERCISE 9.3

13
6. 211 [Hint: F is conservative] 7. 5 8. 9. 34
6
16 −2 −2
10. 11. , , F is conservative. 12. 303
3 3 3
135
13. 0 14. 24p 17.
4
18. f = xy + xy2 + yz + xz2 + c.

9.9 GREEN’S THEOREM IN A PLANE

Green’s theorem gives a relation between a double integral over a region R in the xy plane and the line
integral over a closed curve C enclosing the region R. It helps to evaluate line integral easily.

Statement of Green’s theorem

If P(x, y) and Q(x, y) are continuous functions with continuous partial derivatives in a region R in the
xy plane and on its boundary C which is a simple closed curve then
⎛ ∂Q ∂P ⎞
∫ (P dx 1Q dy ) 5R∫∫ ⎜⎝ ∂x 2 ∂ y ⎠⎟ dxdy
C

where C is described in the anticlockwise sense (which is the positive sense).

Green’s theorem in a plane

Proof Let R be the region in the xy-plane bounded by the simple closed curve C traced in the
anticlockwise sense, which is the positive sense. We assume any line parallel to the axes meet the
curve in not more than two points. The curve C consists of two arcs APB and BQA as in ﬁgure.
Let y = f1(x) and y = f2(x) be the equations of these arcs.

M09_SIVARAMAKRIS_9789332526419_SE_C09_Part-A.indd 47 5/19/2016 10:17:08 AM

Clearly, f1(x) ≤ f2(x) in [a, b]

b ⎡ f2 ( x )
y = f2(x)
∂P ∂P ⎤
Now, ∫∫R ∂y dx dy = ∫a ⎢⎢ f ∫( x ) ∂y dy ⎥⎥ dx Q
⎣1 ⎦
b f2 ( x ) A R B
= ∫ [ P ( x, y ) ] dx
a f1 ( x ) P
b C
y = f1(x)
= ∫ ⎡⎣ P (( x, f 2 ( x ) ) − P (( x, f1 ( x ) )⎤⎦ dx
a
b b

= ∫ P (( x, f 2 ( x ) ) dx − ∫ P (( x, f1 ( x ) ) dx O x=a x=b
a a
Fig. 9.5
b

However, ∫ P ( x, f
a
2 ( x ) ) dx is numerically equal to the line integral
∫ P ( x, y ) dx taken along the
AQB

curve AQB.

But the positive sense is BQA (anticlockwise)

b
∴
∫ P (x , f
a
2 ( x ) ) dx = − ∫
BQA
P( x , y ) dx

Similarly, ∫ P ( x, f ( x)) dx = ∫
a
1
APQ
P ( x, y ) dx

∴ ∂P
∫∫ ∂y dy = − ∫
R APB
P ( x, y )dx − ∫
BQA
P ( x, y )dx

⎧⎪ ⎫⎪
= − ⎨ ∫ P( x , y )dx + ∫ P( x , y )dx ⎬ = − ∫ P( x , y ) dx
⎪⎩ A PB BQA ⎪⎭ C

∂P
⇒ ∫ P( x, y)dx = −∫∫ ∂y dx dy
C R
(1)

Now, we regard the curve C as constituted of the arcs QAP and PBQ.
Let their equations be x = f1(y) and x = f2(y)

Then f1(y) ≤ f2(y) in [c, d]

∂Q
y=d
⎡ x = f2 ( y ) ∂Q ⎤
∫∫R ∂x dx dy = ∫ ⎢ x =f∫ ( y ) ∂x dx ⎥⎥ dy
y=c ⎢
⎣ 1 ⎦

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 48 5/19/2016 10:20:10 AM

Vector Calculus ■ 9.49

d Q
= ∫ [ Q( x, y ) ]x = f2 ( y ) dy y=d
x =f ( y )

c
1
x = φ1(y) x = φ2(y)
A R B
d
y=c
= ∫ [Q(f2 ( y ), y ) − f (f1 ( y ), y ) ] dy P
c C

d d

= ∫ Q(f2 ( y ), y ) dy − ∫ Q (f2 ( y ), y ) dy
c c
O

d
Fig. 9.6
But, ∫ Q (f2 ( y ), y ) dy is the line integral ∫ Q( x, y ) dy
c PBQ
d

and ∫ Q (f ( y), y) dy
c
2 is the line integral ∫ Q( x, y ) dy
PAQ

However, the positive sense of arc is QAP.

d
∴ ∫ Q(f ( y), y) dy = − ∫
c
2
QAP
Q( x, y ) dy

∂Q
∴ ∫∫ ∂x dx dy = ∫
R PBQ
Q( x, y )dy + ∫
QAP
Q( x, y )dy = ∫ Q( x, y )dy
C

∂Q
∴ ∫C Q( x, y)dy = ∫∫R ∂x dx dy (2)

Adding the equations (1) and (2), we get

∂P ∂Q
∫ P ( x, y)dx + ∫ Q( x, y)dy = −∫∫ ∂y dx dy + ∫∫ ∂x dx dy
C C R R

⇒ ⎡ ∂Q ∂P ⎤
∫ P dx + Q dy = ∫∫ ⎢⎣ ∂x − ∂y ⎥⎦ dx dy
C R

Note We have proved the theorem by taking a simple closed region. The theorem is also valid in a
region which can be divided into regions enclosed by simple closed curves.

1
Corollary Area of the region R bounded by C is = ∫∫ dxdy = 2 ∫ (xdy − ydx )
R C

∂P ∂Q
Proof In Green’s theorem, take P = −y and Q = x. ∴ = −1 and =1
∂y ∂x
Then
∫ ( − ydx + xdy ) = R∫∫ (1 + 1)dxdy = 2R∫∫ dxdy
C

∴ 1 ■
2 C∫
( xdy − ydx ) = ∫∫ dxdy
R

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 49 5/19/2016 10:20:16 AM

9.9.1 Vector Form of Green’s Theorem

Let F = Pi + Q j and r = xi + y j

∴ d r = dx i + dy j and F .d r = P dx + Q dx

i j k
∂ ∂ ∂ ⎡ ∂Q ⎤ ⎡ ∂P ⎤ ⎡ ∂Q ∂P ⎤
Now, ∇× F = = i ⎢0 − − j ⎢0 − +k ⎢ − ⎥
∂x ∂y ∂z ⎣ ∂z ⎥⎦ ⎣
⎥
∂z ⎦ ⎣ ∂x ∂y ⎦
P Q 0
⎛ ∂Q ∂P ⎞ ⎛ ∂Q ∂P ⎞ ⎡ ∂Q ∂P ⎤
= i (0) − j (0) + k ⎜ − = −
⎝ ∂x ∂y ⎠⎟ ⎜⎝ ∂x ∂y ⎟⎠
k ⎢{ ∂z = 0; ∂z = 0 ⎥
⎣ ⎦

∴ ∂Q ∂P
∇× F ⋅k = −
∂x ∂y

∴ Green’s theorem becomes ∫ F ⋅ dr = ∫∫ ∇ × F ⋅ k d R,

C R
where d R = dx dy

WORKED EXAMPLES
EXAMPLE 1

∫ [( x 2 y 2 )dx 1 2 xydy ], where C is the closed curve of the

2
Using Green’s theorem evaluate
C
region bounded by y2 5 x and x2 5 y.

Solution.
⎛ ∂Q ∂P ⎞
Green’s theorem is ∫ ( Pdx + Qdy) = R∫∫ ⎜⎝ ∂x − ∂y ⎟⎠ dxdy.
C
y
y = x2

The given line integral is ∫ [( x 2 − y 2 )dx + 2 xydy ]

A y2 = x
C
Q
Here P = x 2 − y 2 and Q = 2 xy
P
∂P ∂Q o x=1 x
∴ = −2 y and = 2y
∂y ∂x

∂Q ∂P
∴ − = 2y + 2y = 4y
∂x ∂y
Fig. 9.7

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 50 5/19/2016 10:20:23 AM

Vector Calculus ■ 9.51

∴ ∫ (x − y 2 ) dx + 2 xydy = ∫∫ 4 ydxdy
2

C R
1 x

=∫ ∫ 4 ydydx
0 x2
x 1
1
⎡ y2 ⎤ 1
⎡ x 2 x5 ⎤ ⎡1 1⎤ 3
= 4∫ ⎢ ⎥ dx = 2∫ ( x − x 4 )dx = 2 ⎢ − ⎥ = 2 ⎢ − ⎥ =
0
⎣ 2 ⎦ x2 0 ⎣ 2 5 ⎦ 0 ⎣ 2 5⎦ 5

EXAMPLE 2
⎛p ⎞
Evaluate ∫ [(sin x 2 y )dx 2 cos xdy ],
C
where C is the triangle with vertices ( 0 , 0 ), ⎜ , 0 ⎟
⎝2 ⎠
⎛p ⎞
and ⎜ , 1⎟ .
⎝2 ⎠

Solution.
Green’s theorem is ∫ ( Pdx + Qdy ) = ⎛ ∂Q ∂P ⎞
C
∫∫ ⎜ − ∂y ⎟⎠ dxdy
R ⎝ ∂x

Given line integral is ∫ [(sin x − y )dx − cos xdy ]

C
Here P = sin x − y and Q = −cos x
∂P ∂Q
∴ = −1 and = sin x
∂y ∂x
∂Q ∂P
∴ − = sin x + 1
∂x ∂y

∫ [(sin x − y )dx − cos xdy ] = R∫∫ (sin x + 1)dxdy

Equation of OB is y − 0 = x − 0 ⇒ y =
2x
0 −1 p p
0− y y=
2x
2 y=1 π
p ⎛π ⎞
B ⎜ , 1⎟
Equation of AB is x= ⎝2 ⎠
2
py p π
x=
In this region R, x varies from to
2 2 P Q
and y varies from 0 to 1.
⎡
1 p/2
⎤
∴ ∫ [(sin x − y )dx − cos xdy ] = ∫ ⎢ ∫ (sin x + 1)dx ⎥ dy o
π
x
(0, 0)
C 0 ⎢
⎣ py / 2 ⎥⎦
1

= ∫ [ − cos x + x ]py / 2 dy
p/2
Fig. 9.8
0

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 51 5/19/2016 10:20:27 AM

⎡⎛ p p⎞ ⎛ py py ⎞ ⎤
1

= ∫ ⎢⎜ − cos + ⎟ − ⎜ − cos + ⎟ dy
0⎣⎝ 2 2⎠ ⎝ 2 2 ⎠ ⎥⎦
⎛p py py ⎞
1

= ∫ ⎜ + cos − ⎟ dy
⎝2 2 2 ⎠
0
1
⎡ py ⎤
⎢p sin
2 p y2 ⎥ p 2 p p p 2 p 2 p
= ⎢ y+ − ⎥ = + sin − = + − = +
⎢ 2 p 2 2 ⎥ 2 p 2 4 2 p 4 p 4
⎣ 2 ⎦0

EXAMPLE 3

∫e
2x
Evaluate by Green’s theorem (sin ydx 1 cos ydy ), C being the rectangle with vertices
C
⎛ p⎞ ⎛ p⎞
(0, 0), (p, 0), ⎜ p, ⎟ and ⎜ 0 , ⎟ .
⎝ 2⎠ ⎝ 2⎠

Solution.
Green’s theorem is ∫ ( Pdx + Qdy ) = ⎛ ∂Q ∂P ⎞
C
∫∫ ⎜ − ∂y ⎟⎠ dxdy
R ⎝ ∂x

The given line integral is ∫ e − x (sin ydx + cos ydy )

Here P = e−x sin y and Q = e−x cos y

∂P ∂Q
∴ = e − x cos y and = −e − x cos y
∂y ∂x

∂Q ∂P
∴ − = −e − x cos y − e − x cos y = −2e − x cos y
∂x ∂y
y
∴ e − x (sin ydx + cos ydy ) =
∫ ∫∫ −2e
−x
cos ydxdy
⎛ π⎞ C ⎛ π⎞
C R ⎜⎝0, ⎟⎠ B ⎜π, ⎟
⎝ 2⎠
2
p
2 p
x=0 x=π
= −2 ∫ ∫ e − x cos y dxdy
0 0 A
⎡ p2 ⎤ p O (π, 0) x
⎢ ⎥⎡ ⎤ (0, 0)
= −2 ⎢ ∫ cos y dy ⎥ ⎢ ∫ e − x dx ⎥
⎣0 ⎦ ⎣0 ⎦ Fig. 9.9
p
p/2 ⎡ e ⎤
−x
⎛ p ⎞ −p
= −2 [sin y ]0 ⎢ −p
⎥ = 2 ⎜⎝ sin ⎟⎠ (e − e ) = 2(e − 1)
0

⎣ −1 ⎦ 0 2

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 52 5/19/2016 10:20:31 AM

Vector Calculus ■ 9.53

EXAMPLE 4
Find the area bounded between the curves y2 5 4x and x2 5 4y using Green’s theorem.

Solution.
We know, by Green’s theorem the area bounded by a simple closed curve C is
1
2 C∫
( xdy − ydx )

Here C consists of the curves C1 and C2.

1 1
∴ area = ⎡ ∫ xdy − ydx + ∫ xdy − ydx ⎤ = [ I1 + I 2 ]
2 ⎢⎣C1 C2 ⎥
⎦ 2
On C1: x2 = 4y
1
∴ 2xdx = 4dy ⇒ dy = xdx y
2
and x varies from 0 to 4. x 2 = 4y
y 2 = 4x

∴ I1 = ∫ xdy − ydx
C1
y=4 P
(4, 4)
C2
4 2
1 x
= ∫ x⋅ xdx − dx x=4
0
2 4 C1
4
⎛ x2 x2 ⎞
4 4 2
x 1 ⎡ x3 ⎤ o
= ∫ ⎜ − ⎟ dx = ∫ dx = ⎢ ⎥ x
⎝ 2 4⎠ 4 4 ⎣ 3 ⎦0 (0, 0)
0 0

64 16
= =
4⋅3 3
1
On C2: y2 = 4x ∴ 2ydy = 4dx ⇒ dx = ydy
2 Fig. 9.10
and y varies from 4 to 0.

∴ I2 = ∫ xdy − ydx
C2

0
y2 1
=∫ dy − y ⋅ y dy
4
4 2
4
⎛ y2 y2 ⎞
0 0
y2 1
4
1 ⎡ y3 ⎤ 16
= ∫ ⎜ − ⎟ d y = ∫ − dy = ∫ y 2 dy = ⎢ ⎥ =
⎝ 4 2 ⎠ 4 40 4 ⎣ 3 ⎦0 3
4 4

1 ⎡16 16 ⎤ 16
∴ area = + =
2 ⎢⎣ 3 3 ⎥⎦ 3

EXAMPLE 5

∫ (3x 2 8 y 2 )dx 1 ( 4 y 2 6 xy )dy , where C is the

2
Verify Green’s theorem in the plane for
C
boundary of the region bounded by x 5 0, y 5 0, x 1 y 5 1.

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 53 5/19/2016 10:20:34 AM

Solution.
Green’s theorem is ∫ ( Pdx + Qdy ) = ⎛ ∂Q ∂P ⎞
C
∫∫ ⎜ − ∂y ⎟⎠ dxdy
R ⎝ ∂x

The given integral is ∫ (3 x 2 − 8 y 2 )dx + ( 4 y − 6 xy )dy y

C
(0, 1)
Here P = 3x2 − 8y2 and Q = 4y − 6xy B
x+y=1
∂P ∂Q
∴ = −16 y and = −6 y
∂y ∂x
∂Q ∂P
∴ − = −6 y + 16 y = 10 y
∂x ∂y (1, 0)
o x
1 1− x
⎛ ∂Q ∂P ⎞ A
∴ ∫∫ ⎜
R ⎝ ∂x
− ⎟ dxdy = ∫
∂y ⎠ 0
∫
0
10 y dy dx
Fig. 9.11
1− x 1
⎡y ⎤
1 2 1
⎡ (1 − x ) ⎤3
−5 5
= 10∫ ⎢ ⎥ dx = 5∫ (1 − x ) 2 dx = 5 ⎢ ⎥ = [0 − 1] =
0
⎣ 2 ⎦0 0 ⎣ −3 ⎦ 0 3 3

⇒ ⎛ ∂Q ∂P ⎞ 5
∫∫ ⎜ − ∂y ⎟⎠ dxdy = 3
R ⎝ ∂x
(1)

We shall now compute the line integral ∫ Pdx + Qdy

∫ Pdx + Qdy = C∫ (3x − 8 y 2 )dx + ( 4 y − 6 xy )dy

2
Now
C

= ∫ (3 x − 8 y 2 )dx + ( 4 y − 6 xy )dy + ∫ (3 x − 8 y 2 )dx + ( 4 y − 6 xy )dy

2 2

OA AB

+ ∫ (3 x − 8 y 2 )dx + ( 4 y − 6 xy )dy = I1 + I 2 + I 3
2

On OA: y = 0 ∴ dy = 0 and x varies from 0 to 1.

1
1
⎡ x3 ⎤
∴ I1 = ∫ 3 x dx = 3 ⎢ ⎥ = 1
2

0 ⎣ 3 ⎦0

On AB: x + y = 1 ⇒ y=1−x ∴ dy = −dx and x varies 1 to 0.

∴ I 2 = ∫ (3 x 2 − 8(1 − x ) 2 )dx + [4(1 − x ) − 6 x(1 − x )]( − dx )

= ∫ [3 x 2 − 8(1 − x ) 2 − 4(1 − x ) + 6( x − x 2 )]dx

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 54 5/19/2016 10:20:38 AM

Vector Calculus ■ 9.55

0
⎡ 8(1 − x )3 (1 − x ) 2 ⎛ x 2 x3 ⎞ ⎤
= ⎢ x3 − −4 + 6⎜ − ⎟⎥
⎣ −3 −2 ⎝ 2 3 ⎠ ⎦1
⎡ 8 ⎧ ⎛ 1 1⎞ ⎫⎤ 8 8
= ⎢0 + + 2 + 0 − ⎨1 + 6 ⎜ − ⎟ ⎬ ⎥ = + 2 − 1 − 1 =
⎣ 3 ⎩ ⎝ 2 3 ⎠ ⎭⎦ 3 3
On BO: x = 0 ∴ dx = 0 and y varies from 1 to 0
0

I 3 = ∫ 4 ydy = 2 ⎡⎣ y 2 ⎤⎦1 = −2
0
∴
1

8 5
∴ ∫ Pdx + Qdy = C∫ (3x − 8 y 2 )dx + ( 4 y − 6 xy )dy = 1 + − 2 =
2
(2)
3 3

(1) and (2) give the same value. Hence, Green’s theorem is veriﬁed.

EXAMPLE 6

∫ ( xy 1 y ) dx 1 x 2 dy , where C is the boundary, of the area between

2
Verify Green’s theorem for
C
y 5 x2 and y 5 x.

Solution.
y
Green’s theorem is
⎛ ∂Q ∂P ⎞ y = x2
∫ ( Pdx + Qdy) =
C
∫∫ ⎜
R ⎝ ∂x
− ⎟ dxdy
∂y ⎠ y=x

The given line integral is

∫ ( xy + y ) dx + x 2 dy
2
A (1,1)
C
Q
Here P = xy + y2 and Q = x2 x=1
∂P ∂Q
∴ = x + 2y and = 2x P
∂y ∂x
O (0,0) (1,0) x
∂Q ∂P
∴ − = 2x − x − 2 y = x − 2 y
∂x ∂y
x Fig. 9.12
⎛ ∂Q ∂P ⎞ ⎡ 2 y2 ⎤
1 x 1
∴ ∫∫ ⎜ − ⎟ dxdy = ∫ ∫ ( x − 2 y ) dy dx = ∫ ⎢ xy − ⎥ dx
R ⎝ ∂x ∂y ⎠ 0 x2 0
⎣ 2 ⎦ x2
1
1 1
⎡ x5 x 4 ⎤ 1 1 1
= ∫ [ x 2 − x 2 − ( x 3 − x 4 )] dx = ∫ [( x 4 − x 3 )] dx = ⎢ − ⎥ = − = −
0 0 ⎣ 5 4 ⎦ 0 5 4 20
⎛ ∂Q ∂P ⎞ 1
∫∫ ⎜ − ∂y ⎟⎠ dxdy = − 20
R ⎝ ∂x
(1)

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 55 5/19/2016 10:20:41 AM

We shall now compute the line integral ∫ Pdx + Qdy

∫ Pdx + Qdy = C∫ ( xy + y ) dx + x 2 dy
2
Now
C

= ∫ ( xy + y ) dx + x 2 dy + ∫ ( xy + y ) dx + x 2 dy = I1 + I 2
2 2

C 1 C2

On C1: y = x2, ∴ dy = 2x dx and x varies from 0 to 1.

I1 = ∫ ( x ⋅ x 2 + x 4 ) dx + x 2 ⋅ 2 x dx
0

= ∫ ( x 3 + x 4 + 2 x 3 ) dx
0
1
1
⎡ x 4 x5 ⎤ 3 1 19
= ∫ (3 x 3 + x 4 ) dx = ⎢3 + ⎥ = + =
0 ⎣ 4 5 ⎦ 0 4 5 200
On C2: y = x, ∴ dy = dx and x varies from 1 to 0.
0
0
⎡ x3 ⎤
1

∴ I 2 = ∫ ( x ⋅ x + x ) dx + x dx = ∫ 3 x dx = 3 ⎢ ⎥ = −1
2 2 2

1 0 ⎣ 3 ⎦1

19 1
∴ ∫ Pdx + Qdy = 20 − 1 = − 20
C
(2)

(1) and (2) give the same value.

Hence, Green’s theorem is veriﬁed.

9.10 SURFACE INTEGRALS

Suppose a surface is bounded by a simple closed curve C, then we can regard the surface as having
two sides separated by C. One of which is arbitrarily chosen as the positive side and the other is the
negative side. If the surface is a closed surface, then the outerside is taken as the positive side and the
inner side is the negative side. A unit normal at any point of the positive side of the surface is denoted
by n and is called the outward drawn normal and its direction is considered positive.
Any integral which is evaluated over a surface is called a
surface integral.

Definition 9.12 Surface Integral F

n
Let S be a surface of finite area which is smooth or piece-
P
wise smooth (e.g. a sphere is a smooth surface and a cube is a
ds
piecewise smooth surface). Let F ( x, y, z ) be a vector point
function defined at each point of S. Let P be any point on the
S
surface and let n be the outward unit normal at P. Then the
surface integral of F over S is defined as ∫∫ F ⋅ n dS
S Fig. 9.13

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 56 5/19/2016 10:20:46 AM

Vector Calculus ■ 9.57

If we associate a vector dS (called vector area) with the differential of surface area dS such that
dS = dS and direction of dS is n, then
dS = n dS

∴ ∫∫ F ⋅ n dS can also be written as S∫∫ F ⋅ dS

Note
1. In physical application the integral ∫∫ F ⋅ dS is called the normal ﬂux of F through the surface S,
S
because this integral is a measure of the volume emerging from S per unit time.

9.10.1 Evaluation of Surface Integral

To evaluate a surface integral over a surface it is usually expressed as a double integral over the
orthogonal projection of S on one of the coordinate planes. This is possible if any line perpendicular
to the coordinate plane chosen meets the surface S in not more than one point.
Let R be the orthogonal projection of S on the xy plane.
Then the element surface dS is projected to an element area dx dy in the xy plane as in ﬁg.
∴ dx dy = dS cos u, where u is the angle between the planes of dS and xy−plane.
Let n be the unit normal to dS and k is the unit normal to the xy−plane.
Since angle between the planes is equal to the angle between the normals,
u is the angle between the normals n and k.
n⋅k
∴ cos u =
k
n k n
z
= n ⋅ k [Since n = 1, k = 1]
ds
We take the acute angle between the normals and P

So, we take n ⋅ k
dx dy
∴ dx dy = dS n ⋅ k ⇒ dS =
n⋅k
o
dx dy
Hence, ∫∫ F ⋅ n dS = ∫∫ F ⋅ n y
S R n⋅k
Similarly, taking the projection on the yz and zx
dx
planes, we get
dy dz dy
∫∫
S
F ⋅ n dS = ∫∫ F ⋅ n
R n⋅i
x

Fig. 9.14
dz dx
and ∫∫
S
F ⋅ n dS = ∫∫ F ⋅ n
R n⋅ j
Corollary
dx dy dy dz dz dx
The surface area ∫∫ dS = R∫∫
S n⋅k
= ∫∫
R n⋅i
= ∫∫
R n⋅ j
1 2

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 57 5/19/2016 10:20:59 AM

9.11 VOLUME INTEGRAL

Any integral which is evaluated over a volume bounded by a surface is called a volume integral.
If V is the volume bounded by a surface S, then

∫∫∫
V
f( x, y, z )dV and ∫∫∫ F dV are called volume integrals.
V
If we divide V into rectangular blocks by drawing planes parallel to the coordinate planes, then
dV = dx dy dz.

∴ ∫∫∫
V
fdV = ∫∫∫ f( x, y, z ) dxdydz
V

If F = F1 i + F2 j + F3 k

then ∫∫∫
V
FdV = i ∫∫∫ F dxdydz + j ∫∫∫ F dxdydz + k ∫∫∫ F dxdydz
V
1
V
2
V
3

WORKED EXAMPLES
EXAMPLE 1
Evaluate ∫∫ F ⋅ n dS if F 54 yi 118 z j 2 xk and S is the surface of the plane 3x 1 2y 1 6z 5 6
S
contained in the ﬁrst octant.

Solution.
Given F = 4 yi + 18 z j − xk and the surface 3x + 2y + 6z = 6. z
C
Let f = 3x + 2y + 6z

Let R be the projection of S in the xy plane. n

S
∴ R is the Δ AOB
∴ dx dy
∫∫ F ⋅ n dS = R∫∫ F ⋅ n o B
y
S n⋅k R
3x + 3y = 6
where n is unit normal to S and k is the unit normal to xy−plane. A
x

Normal to the surface is ∇f = i ∂f + j ∂f + k ∂f = 3i + 2 j + 6 k

Fig. 9.15
∂x ∂y ∂z

∇f 3i + 2 j + 6 k 1
∴ unit normal is n = = = (3i + 2 j + 6 k )
∇f 9 + 4 + 36 7
1
∴ F ⋅ n = (4 yi + 18 z j − xk ) ⋅ (3i + 2 j + 6k )
7
1 6
= (12 y + 36 z − 6 x ) = ( 2 y + 6 z − x )
7 7

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 58 5/19/2016 10:21:10 AM

Vector Calculus ■ 9.59

1 6
n ⋅ k = (3i + 2 j + 6 k ) ⋅ k =
7 7
6 dx dy
∴ ∫∫ F ⋅ n dS = R∫∫ 7 (2 y + 6 z − x) 6
= ∫∫ (2 y + 6 z − x)dx dy
R
S
7
We have 3x + 2y + 6z = 6 y
B
⇒ 6z = 6 − 3x − 2y
(0, 3)
∴ 2y + 6z − x = 2y + 6 − 3x − 2y − x = 6 − 4x
∴
S
∫∫ F ⋅ n dS = R∫∫ (6 − 4 x) dx dy 3x + 2y = 6

The plane 3x + 2y + 6z = 6 meets the xy−plane z = 0 in

line AB.
∴ the equation of AB is 3x + 2y = 6
O A (2, 0) x
∴ the point A is (2, 0) and the point B is (0, 3) R

6 − 3x Fig. 9.16
Now 3x + 2y = 6 ⇒ y=
2

6 − 3x
∴ In R, x varies from 0 to 2 and y varies from 0 to
2
6 −3 x 6 −3x
2 2 2 2
∴ (6 − 4 x ) dy dx = 2∫
∫∫ F ⋅ n dS = ∫ ∫ ∫ (3 − 2 x ) dy dx
S 0 0 0 0
6 −3x
2 2
= 2 ∫ [ (3 − 2 x ) y ] dx
0 0

(6 − 3 x )
2

= 2∫ (3 − 2 x ) dx
0
2
2

= 3∫ (3 − 2 x )( 2 − x ) dx
0

= 3∫ (6 − 7 x + 2 x 2 ) dx
0
2
⎡ 7x2 x3 ⎤
= 3 ⎢6 x − +2 ⎥
⎣ 2 3 ⎦0
⎡ 4 8⎤ ⎡ 16 ⎤
= 3 ⎢6 × 2 − 7 × + 2 × ⎥ = 3 ⎢12 − 14 + ⎥ = −6 + 16 = 10
⎣ 2 3⎦ ⎣ 3⎦

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 59 5/19/2016 10:21:14 AM

EXAMPLE 2

Evaluate ∫∫ F ⋅ n dS
S
if F 5 yz i 1 zx j 1 xyk and S is part of the surface x2 1 y2 1 z2 5 1,

which lies in the ﬁrst octant.

Solution. z

Given F = yz i + zx j + xyk and the surface is x2 + y2 + z2 = 1

Let f = x2 + y2 + z2
∂f ∂f ∂f
The normal to the surface is ∇f = i +j +k
∂x ∂y ∂z o
y
= 2 xi + 2 y j + 2 zk R
x2 + y2 = 1
∴ unit normal is ∇f 2 xi + 2 y j + 2 zk
n= = x
∇f 4x2 + 4 y2 + 4z2 Fig. 9.17
2( xi + y j + zk ) [{ x2 + y2 + z2 = 1]
= = xi + y j + zk
2 x +y +z 2 2 2

∴ F ⋅ n = ( yzi + zx j + xyk ) ⋅ ( xi + y j + zk )
= xyz + xyz + xyz = 3 xyz

The projection of the surface of the sphere in the ﬁrst octant into the xy plane is R, which is the
quadrant of the circle x2 + y2 = 1, z = 0, x ≥ 0, y ≥ 0 and k is the unit normal to R.
dx dy
∫∫ F ⋅ n dS = R∫∫ F ⋅ n
y
∴
S n⋅k
dx dy x2 + y 2 = 1
= ∫∫ 3xyz
R n⋅k
But n ⋅ k = ( xi + y j + zk ) ⋅ k = z
∴ 1
∫∫ F ⋅ n dS = R∫∫ 3xyz z
dx dy
O
S (1, 0) x
R
= ∫∫ 3xy dx dy
R
1 1− x 2
Fig. 9.18
=∫ ∫ 3xy dx dy
0 0

1− x 2
⎡y 2 ⎤
1
3
1

= 3∫ ⎢ ⎥
2 ∫0
xdx = x (1 − x 2 )dx
0
⎣ 2 ⎦0
1
3 ⎡x2 x4 ⎤ 3 ⎛ 1 1⎞ 3 1 3
= ⎢ − ⎥ = ⎜⎝ − ⎟⎠ = ⋅ =
2⎣ 2 4 ⎦0 2 2 4 2 4 8

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 60 5/19/2016 10:21:23 AM

Vector Calculus ■ 9.61

EXAMPLE 3

Evaluate ∫∫ F ⋅ n dS, where F 5 4 xz i 2 y 2 j 1 yzk and S is the surface of the cube bounded by the
planes x 5 0, x 5 1, y 5 0, y 5 1, z 5 0, z 5 1.

Solution.
Given F = 4 xz i − y 2 j + yzk
S is the surface of the cube, which is piecewise smooth surface consisting of six smooth surfaces.

∫∫ F ⋅ n dS = ABEF
S
∫∫ F ⋅ n dS + OCDG
∫∫ F ⋅ n dS
z
+ ∫∫
BCDE
F ⋅ n dS + ∫∫
OAFG
F ⋅ n dS
G D

+ ∫∫
OABC
F ⋅ n dS + ∫∫
DEFG
F ⋅ n dS F
E
k
j
On the face ABEF: x = 1, n=i o
C y
∴ F ⋅ n = (4 xzi − y 2 j + yzk ) ⋅ i = 4 xz = 4 z
i
A B
and dy dz dy dz
dS = = = dy dz x
n ⋅i i ⋅i
Fig. 9.19
1
⎡ 2⎤
1 1
∴ ∫∫ F ⋅ n dS = ∫ ∫ 4 z dz dy = 4 ⋅ [ y ] ⎢ z ⎥ = 4 ⋅1⋅ 1 = 2
1

ABEF 0 0
0
⎣ 2 ⎦0 2

On the face OCDG: x = 0, n = −i

F ⋅ n = (4 xzi − y j + yzk ) ⋅ ( −i ) = −4 xz = 0
2

∴ ∫∫
OCDG
F ⋅ n dS = 0

On the face BCDE: y = 1, n= j

dx dz dx dz
dS = = = dx dz
∴ n⋅ j j⋅ j
and F ⋅ n = (4 xzi − y 2 j + yzk ) ⋅ j = − y 2 = −1
1 1
∴ ∫∫ F ⋅ n dS = ∫ ∫ ( −1)dx dz = − [ x ]0 [ z ]0 = −1
1 1

BCDE 0 0

On the face OAFG: y = 0, n=−j

∴ F ⋅ n = (4 xzi − y 2 j + yzk ) ⋅ ( − j ) = y 2 = 0
∴
OAFG
∫∫ F ⋅ n dS = 0

On the face DEFG: z = 1, n=k

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 61 5/19/2016 10:21:32 AM

∴ F ⋅ n = (4 xzi − y 2 j + yzk ) ⋅ k = yz = y
dx dy dx dy
and dS = = = dx dy
n⋅k k ⋅k
1
1 1
⎡ y2 ⎤ 1 1
∴ ∫∫ F ⋅ n dS = ∫ ∫ y dx dy = [ x ]0 ⎢ ⎥ = 1× =
⎣ 2 ⎦0
1

2 2
DEFG 0 0

On the face OABC: z = 0, n = −k

F ⋅ n = (4 xz i − y 2 j + yzk ) ⋅ ( − k ) = − yz = 0
∴ ∫∫
OABC
F ⋅ n dS = 0

∴ 1 3
∫∫
S
F ⋅ n dS = 2 + ( −1) + = .
2 2

9.12 GAUSS DIVERGENCE THEOREM

The divergence theorem enables us to convert a surface integral of a vector function on a closed
surface into volume integral.

Statement of Gauss divergence theorem

Let V be the volume bounded by a closed surface S. If a vector function F is continuous and has
continuous partial derivatives inside and on S, then the surface integral of F over S is equal to the
volume integral of divergence of F taken throughout V.

i.e.,
S
∫∫ F ? d S 5 ∫∫∫
V
=? FdV

If n is the outward normal to the surface dS = n dS

∴
∫∫ F ⋅ n dS = ∫∫∫
S V
∇ ⋅FdV n2
Q S2 z = f2(x, y)
C′
Proof Let F = F1i + F2 j + F3 k P z = f1(x, y)
S1

∴ F ⋅ n = F1 (i ⋅ n ) + F2 ( j ⋅ n ) + F3 ( k ⋅ n ) n1

O
and F ⋅ n dS = F1 (i ⋅ n )dS + F2 ( j ⋅ n )dS + F3 ( k ⋅ n )dS y

= F1 dy dz + F2 dz dx + F3 dx dy R
C
∂F1 ∂F2 ∂ F3
But ∇⋅F = + + x
∂x ∂y ∂z
Fig. 9.20

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 62 5/19/2016 10:21:40 AM

Vector Calculus ■ 9.63

Hence, Gauss theorem in Cartesian form is

⎛ ∂F ∂F ∂F ⎞
∫∫ ( F dy dz + F
S
1 2 dz dx + F3 dx dy ) ≡ ∫∫∫ ⎜ 1 + 2 + 3 ⎟ dx dy dz
V
⎝ ∂x ∂y ∂z ⎠
We shall assume that S is a closed surface such that any line drawn parallel to coordinate axes cuts S
in almost two points. The lines drawn parallel to Z-axis touching the surface S determine the curve C ′
on it and intersect the xy-plane along the curve C. Now, the curve C ′ divides the surface S into two
parts S1 and S2.
S1 and S2 are called the lower and upper surfaces.
Let z = f1(x, y) and z = f2(x, y) be the equations of S1 and S2, respectively.
The projection of S on the xy-plane is the region R bounded by C.
∂F
Now consider the triple integral ∫∫∫ 3 dx dy dz over the volume V enclosed by S.
V
∂z

∂F3 ⎡ z = f2 ( x , y ) ∂F3 ⎤
∫∫∫ ∂z
dx dy dz = ∫∫R ⎢⎢ z = f ∫( x , y ) ∂z ⎥⎥ dx dy
V ⎣ 1 ⎦

= ∫∫ [ F3 ( x, y, z ) ]
z = f2 ( x , y )
dx dy
z = f1 ( x , y )
R

= ∫∫ ⎡⎣ F3 ( x, y, f 2 ( x, y ) ) − F3 ( x, y, f1 ( x, y ) )⎤⎦ dx dy
R

∂F3
⇒
∫∫∫ dx dy dz = ∫∫ F3 ( x, y, f 2 ( x, y ) ) dx dy − ∫∫ F3 ( x, y, f1 ( x, y ) ) dx dy (1)
V
∂z R R

Let a line parallel to the z-axis meet S1 at the point P and S2 at the point Q. Let dS1 and dS2 be element
surface at P and Q, respectively and their projections in the xy-plane be dx dy.
Let n1 be the outward unit normal at P to S1 and n 2 be the outward unit normal at Q to S2.
Let the angle between n 2 and kbe g 2 and g 2 is acute, since k is unit vector in the direction of
the positive z-axis.
Then dx dy = cosg 2 dS2 = k⋅n 2 d S2

Let the angle between n1and kbe g1and it is obtuse. [{ k is upward and n1 is downward]
∴ dx dy = −cosg1 dS1 = − k⋅n1 d S1

Hence, ∫∫ F ( x , y f
R
3 2 ( x , y ) ) dx dy = ∫∫ F3 k ⋅ n 2 dS 2
S2

and ∫∫ F ( x, y f ( x, y)) dx dy = − ∫∫ F k ⋅ n
R
3 1
S1
3 1 dS1

Substituting in (1), we get

∂F3
∫∫∫
V
∂z
dx dy dz = ∫∫ F3 k ⋅ n2 dS2 + ∫∫ F3 k ⋅ n1 dS1
S2 S1

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 63 5/19/2016 10:21:51 AM

⇒ ∂F3
∫∫∫ V
∂z
dx dy dz = ∫∫ F3 k ⋅ n dS
S
(2)

Similarly, projecting S on the yz- and zx-planes, we get

∂F2
∫∫∫ ∂y
V
dx dy dz = ∫∫ F2 j ⋅ n dS
S
(3)

∂F1
and ∫∫∫
V
∂x
dx dy dz = ∫∫ F1 i ⋅ n dS
S
(4)

Adding equations (2), (3) and (4), we get

⎛ ∂F1 ∂F2 ∂F3 ⎞
∫∫∫ ⎜⎝ ∂x
V
+
∂y
+
∂z ⎠⎟
dx dy dz = ∫∫ ( F1i + F2 j + F3 k ) ⋅ n dS
S

⇒ ∫∫∫ ∇ ⋅ F dV = ∫∫ F ⋅ n dS
V S

9.12.1 Results Derived from Gauss Divergence Theorem

The following results are immediate consequence of Gauss divergence theorem:
(1) ∫∫S f n dS = ∫∫∫
V
∇ f dV (2) ∫∫S F × n dS = −∫∫∫
V
∇ × F dV

where f is the scalar point function deﬁned in the region V enclosed by the closed surface S.

Solution.
(1) ∫∫ f n dS = ∫∫∫ ∇ f dV .
S V

Gauss divergence theorem is

∫∫∫ ∇ ⋅ F dV = ∫∫ F ⋅ n dS
V S
(1)

Let F = f a, where a is an arbitrary constant vector.

∴ (1) becomes ∫∫∫ (∇ ⋅ f

V
)
a dS = ∫∫ f a ⋅ n dS
S
(2)

Now, ∇ ⋅ f a = ∇f ⋅ a + f(∇ ⋅ a) = ∇f ⋅ a [{ ∇ ⋅ a = 0]

∴ ∫∫∫ (∇ ⋅ f a) dV = ∫∫∫ (∇f ⋅ a) dV

V V
(3)

and ∫∫ F ⋅ n dS = ∫∫ f a ⋅ n dS = ∫∫ f n dS ⋅ a
S S S
(4)

∴ Using (3) and (4) in (2), we get

∫∫∫ ∇f ⋅ a dV = ∫∫ f n dS ⋅ a
V S

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 64 5/19/2016 10:21:59 AM

Vector Calculus ■ 9.65

⇒ a ⋅ ∫∫∫ ∇f dV = a ⋅ ∫∫ f n dS
V S

⇒ ∫∫∫ ∇f dV = ∫∫ f
V
n dS [{ a is arbitrary ]
s

2. ∫∫S F × n dS = −∫∫∫
V
∇ × F dV

Gauss divergence theorem is ∫∫∫ ∇ ⋅ F dV = ∫∫ F ⋅ n dS (1)

V S
Let F = a × F , where a is an arbitrary constant vector.
∴ ∇ ⋅ F = ∇ ⋅ (a × F ) = F ⋅ (∇ × a) − a ⋅ (∇ × F ) = −a ⋅ (∇ × F ) [{ ∇ × a = 0]
and F ⋅ n = a × F ⋅ n = a ⋅ (F × n)
∴ (1) becomes − ∫∫∫ a ⋅ (∇ × F ) dV = ∫∫ (a ⋅ F × n ) dS
V S

⇒ −a ⋅ ∫∫∫ ∇ × F dV = a ⋅ ∫∫ F × n dS
V S

⇒ − ∫∫∫ ∇ × F dV = ∫∫ F × n dS
V S

⇒
∫∫ F × n dS = −∫∫∫ ∇ × F dV .
S V

If S is closed surface, then prove that Eq.

(1) ∫∫ dS 5 ∫∫∫ ∇? n dV
S V
(2) ∫∫ dS 5 0 (3) ∫∫ r 3 n dS 5 0
S S
r
∫∫∫ (∇ 3 n )dV 5 0 (5) ∫∫ 3 ? n dS 5 0 ∫∫ r n dS 5 4∫∫∫ r r dV
4 4
(4) (6)
V S r S V

(7) ∫∫ f ( r ) r 3 n dS 5 0
S
(8) ∫∫ ( = r
2
? n) dS 5 6V (9) ∫∫ ( = 3 r ) ? n dS 5 0
S S

Solution.
(i) To prove ∫∫ dS 5 ∫∫∫ = ? n dV .
S V

Gauss divergence theorem is ∫∫ F ⋅ n dS = ∫∫∫ ∇ ⋅ F dV (1)

S V

Let F = n ∴ ∇⋅F = ∇⋅n and F⋅n = n⋅n = 1

∴ (1) becomes ∫∫ dS = ∫∫∫ ∇ ⋅ n dV .

S V

(2) To prove ∫∫ dS 5 0.
S

We have ∫∫∫ ∇f dV = ∫∫ f n dS (1)

V S

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 65 5/19/2016 10:22:13 AM

Let f = 1.
⎛ ∂ ∂ ∂⎞
∴ ∇f = ⎜ i + j + k ⎟ (1) = 0 ∴
⎝ ∂x ∂y ∂z ⎠ ∫∫∫ ∇f dV = 0
V

∴ ∫∫ S
n dS = 0 ⇒
∫∫ dS = 0 [using (1)]
S

(3) To prove ∫∫ r 3 n dS 5 0.
S

We have ∫∫ F × n dS ≡ −∫∫∫ ∇ × F dV
S V

Let F = r and r = xi + yj + zk

i j k
∴ ∂ ∂ ∂ = i (0 − 0) + j(0 − 0) + k(0 − 0) = 0
∇× F =
∂x ∂y ∂z
x y z
∴ ∫∫∫ ∇× F dv = 0
V
∴ ∫∫ F × n dS = 0 ⇒ ∫∫ r × n dS = 0 [using (1)]
S S

(4) To prove ∫∫ ∇ 3 n dV 5 0.
V

We have ∫∫ F × n dS = −∫∫∫ ∇ × F dV
S V
(1)

Let F=n ∴ F×n = n×n = 0

∴ ∫∫ F × n dS = 0 ∴ ∫∫∫ ∇ × F dV =0 ⇒ ∫∫∫ ∇ × n dV
V
=0 [using (1)]
S V

r
(5) To prove ∫∫ ? n dS 5 0.
S r3
Gauss divergence theorem is ∫∫ F . n dS = ∫∫∫ Δ ⋅ F dS (1)
S V

r ⎛ r⎞ 1 ⎛ 1⎞
Let F = ∴ ∇ ⋅ F = ∇ ⋅ ⎜ 3 ⎟ = 3 (∇ ⋅ r ) + ∇ ⎜ 3 ⎟ ⋅ r
r3 ⎝r ⎠ r ⎝r ⎠
∂ ∂ ∂
∴ ∇⋅r = ( x) + ( y) + ( z) = 1 + 1 + 1 = 3
∂x ∂y ∂z
∂r ∂r x
Now r 2 = x 2 + y 2 + z 2 ⇒ 2r = 2x ⇒ =
∂x ∂x r
∂r y ∂r z
Similarly, = and =
∂y r ∂z r

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 66 5/19/2016 10:22:26 AM

Vector Calculus ■ 9.67

⎛ 1⎞ ∂ 3 ∂ ∂
∴ ∇ ⎜ 3 ⎟ = ∇( r −3 ) = i ( r ) + j (r 3 ) + k (r 3 )
⎝r ⎠ ∂x ∂y ∂z
∂r ∂r ∂r
= i ( −3) r 4 + j ( −3) r 4 + k ( −3) r 4
∂x ∂y ∂z
3 ⎡x y z ⎤ 3 3
=−
4 ⎢
i + j + k ⎥ = − 5 [ xi + yj + zk ] = − 4 r
r ⎣r r r ⎦ r r
⎛r⎞ 3 3 3 3
∴ ∇ ⋅ F = ∇ ⋅ ⎜ 3 ⎟ = 3 − 5 (r ⋅ r ) = 3 − 3 = 0
⎝r ⎠ r r r r

∴ ∫∫∫ ∇ ⋅ F dV = 0
V
⇒ ∫∫ F ⋅ n dS = 0
S
[using (1)]

r ⎛ r⎞
∴ ∫∫ r
S
3
⋅ n dS = ∫∫∫ ⎜ ∇ ⋅ 3 ⎟ dS = 0.
V ⎝ r ⎠

∫∫ r n dS 5 4∫∫∫ r 2 r dV .
4
(6) To prove
S V

We have ∫∫∫ ∇f dV = ∫∫ f n dS
V S

∂ 4 ∂ ∂
Let f = r 4 ∴ ∇f = i (r ) + j (r 4 ) + k (r 4 )
∂x ∂y ∂z
x y z
= 4 r 3 ⋅ i + 4 r 3 ⋅ j + 4 r 3 ⋅ k = 4 r 2 [ xi + yj + zk ] = 4 r 2 r
r r r

∴ ∫∫∫ 4r r dV = ∫∫ r 4 n dS ⇒ 4∫∫∫ r 2 r dV = ∫∫ r 4 n dS
2
(1) becomes,
V S V S

(7) To prove: ∫∫ f ( r ) r 3 ndS 5 0.

We have ∫∫ F × n dS = −∫∫∫ ∇ × F dV
S V
(1)

Let F = f(r)r, r = xi + yj + zk
i j k
∂ ∂ ∂ = i ⎡ ∂ f (r) z − ∂ f (r) y ⎤
∴ ∇× F = ∑ ⎢ ∂y ∂z ⎥
∂x ∂y ∂z ⎣ ⎦
f ( x) x f ( x) y f ( x) z
⎡∂ ∂ ⎤ ⎡ ∂z ∂r ⎤ ⎡ ∂y ∂r ⎤
Now ⎢ f ( r ) z − f ( r ) y ⎥ = ⎢ f ( r ) + zf 9( r ) ⎥ − ⎢ f ( r ) + y f ′( r ) ⎥
⎣ ∂y ∂z ⎦ ⎣ ∂y ∂y ⎦ ⎣ ∂z ∂z ⎦
y z f ′( r )
= 0 + z f ′( r ) − 0 − yf ′( r ) ⋅ = [ yz − yz ] = 0
r r r

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 67 5/19/2016 10:22:38 AM

⎡ ∂ ∂ ⎤
∴ i ⎢ ( f ( r ) z ) − ( f ( r ) y )⎥ = 0
⎣ ∂y ∂z ⎦
⎡∂ ∂ ⎤ ⎡∂ ∂ ⎤
Similarly, j ⎢ (f ( r )z ) − (f ( r ) y )⎥ = 0 and k ⎢ ( f ( r ) y ) − ( f ( r ) x )⎥ = 0
⎣ ∂ x ∂ z ⎦ ⎣ ∂x ∂z ⎦

∴ ∇× F = 0 ∴ ∫∫∫ ∇ × F dV = 0
V

∴ (1) becomes ∫∫ F × n dS = 0 ⇒ ∫∫ f (r ) r × n dS = 0
S
S

(8) To prove ∫∫ ( = r 2 ⋅ n) dS 5 6V .
S

Gauss divergence theorem is ∫∫ F ⋅ n dS = ∫∫∫ ∇ ⋅ F dV

S V
(1)

Let F = ∇r 2 and r = xi + yj + zk ⇒ r2 = x2 + y2 + z2
∴ ∇ ⋅ F = ∇ ⋅ ∇r 2
∂ 2 ∂ ∂
Now, ∇r 2 = i ( x + y2 + z2 ) + j ( x2 + y2 + z2 ) + k ( x2 + y2 + z2 )
∂x ∂y ∂z
= 2 xi + 2 yj + 2 zk = 2[ xi + yj + zk ] = 2r
∴ ∇ ⋅ ∇r = ∇ ⋅ 2r = 2∇ ⋅ r
2

But ⎛ ∂ ∂ ∂⎞
∇⋅r = ⎜i +j + k ⎟ ⋅ ( xi + yj + zk ) = 1 + 1 + 1 = 3
⎝ ∂x ∂y ∂z ⎠

∴ ∇ ⋅ ∇r 2 = 2 ⋅ 3 = 6 ∴ ∫∫∫ ∇ ⋅ F dV = ∫∫∫ 6 dV
V V
= 6V

∴ (1) becomes ⇒
∫∫ F ⋅ n dS = 6V
S
∫∫ ∇r
2
⋅ n dS = 6V
S

WORKED EXAMPLES
EXAMPLE 1
Let V be the region bounded by a closed surface S. Let f and g be scalar point functions that
together with their derivatives in any directions are uniformly continuous within the region V.
Then
∫∫∫ ( f ∇ g 2 g ∇ f ) dV 5 ∫∫ ( f ∇ g 2 g ∇f ) ? n dS.
2 2

V S
Solution.
Gauss divergence theorem is
∫∫∫ ∇ ⋅ F dV = ∫∫ F ⋅ n dS
V S

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 68 5/19/2016 10:22:50 AM

Vector Calculus ■ 9.69

Put F = f ∇g ∴ ∇ ⋅ F = ∇ ⋅ ( f ∇g ) = f ( ∇ ⋅ ∇ g ) + ∇ f ⋅ ∇ g = f ∇ 2 g + ∇ f ⋅ ∇ g
and F ⋅ n = ( f ∇g ) ⋅ n
∴ by divergence theorem becomes
∫∫∫ (f ∇ g + ∇f ⋅ ∇g ) dV = ∫∫ (f ∇g ⋅ n ) dS
2
(1)
V S

Interchanging f and g, we get

∫∫∫ (g ∇ f + ∇g ⋅ ∇f ) dV = ∫∫ ( g ∇f ⋅ n ) dS
2
(2)
V S

(1) − (2) ⇒ ∫∫∫ (f ∇ g − g ∇ f ) dV = ∫∫ (f ∇g − g ∇f ) ⋅ n dS

2 2
(3)
V S

Note This result is known as Green’s theorem.

Equation (1) is called Green’s ﬁrst identity and equation (3) is called Green’s second identity.

EXAMPLE 2
1 r
Prove that ∫∫∫ dV 5 ∫∫ 2 ? n dS.
V r2 S r
Solution.
Gauss divergence theorem is
∫∫∫ ∇ ⋅ F dV = ∫∫ F ⋅ n dS
V S
(1)

r
Put = r −2 r. Then ∇ ⋅ F = ∇ ⋅ ( r −2 r ) = (∇ ⋅ r )r −2 + ∇r −2 ⋅ r
F=
r2
∂ ∂ ∂
If r = xi + yj + zk , then ∇ ⋅ r = (x ) + (y ) + (z ) = 1 + 1 + 1 = 3
∂x ∂y ∂z
and r2 = x2 + y2 + z2
∂r ∂r x , ∂r y ∂r z
∴ 2r = 2x ⇒ = = and =
∂x ∂x r ∂y r ∂z r
∂ −2 ∂ ∂
∇ r −2 = i ( r ) + j ( r −2 ) + k ( r −2 )
∂x ∂y ∂z
∂r ∂r ∂r
= i ( −2)r −3 + j ( −2)r −3 + k ( −2)r −3
∂x ∂y ∂z
x y z −2 2r
= −2r −3 i − 2r −3 j − 2r −3 k = 4 ( xi + yj + zk ) = − 4
r r r r r

∴ ⎛ −2 ⎞ 3 2 3 2 1
∇ ⋅ F = 3r −2 + ⎜ 4 r ⋅ r ⎟ = 2 − 4 × r 2 = 2 − 2 = 2
⎝r ⎠ r r r r r

∴ (1) becomes 1 r
∫∫∫ r
V
2
dV = ∫∫
S r2
⋅ n dS

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 69 5/19/2016 10:23:00 AM

EXAMPLE 3
Using divergence theorem, evaluate ∫∫ F ? n dS, where F 5 4 x z i 2 y 2 j 1 y z k and S is the surface
S
of the cube bounded by the planes x 5 0, x 5 2, y 5 0, y 5 2, z 5 0, z 5 2.

Solution.
Gauss divergence theorem is ∫∫ F ⋅ n dS = ∫∫∫ ∇ ⋅ F dV
S V

Given F = 4 xzi − y j + yzk

∴ ∂ ∂ ∂
∇⋅F = (4 xz ) + (−y 2 ) + ( yz ) = 4 z − 2 y + y = 4z − y
∂x ∂y ∂z
2 2 2
∴
∫∫ F ⋅ n dS = ∫ ∫ ∫ (4z − y ) dx dy dz
0 0 0

2 2

= ∫ ∫ ( 4 z − y ) [ x ]0 dy dz
2

0 0

2 2

= ∫ ∫ ( 4 z − y )2 dy dz
0 0

2
2
⎡ y2⎤
= 2∫ ⎢ 4 zy − ⎥ dz
0 ⎣
2 ⎦0
⎛ 4⎞
2 2

= 2∫ ⎜ 4 z ⋅ 2 − ⎟ dz = 2 ⋅ ∫ (8z − 2) dz
0
⎝ 2⎠ 0
2
⎡ 8z 2 ⎤ ⎡ 4 ⎤
= 2⎢ − 2z ⎥ = 2 ⎢8 ⋅ − 2 ⋅ 2⎥ = 2[16 − 4] = 2 × 12 = 24.
⎣ 2 ⎦0 ⎣ 2 ⎦

EXAMPLE 4

∫∫ F ⋅ n dS where F 5 x i 1 y j 1 z 3 k and S is the

3 3
Using Gauss divergence theorem, evaluate
S
sphere x2 1 y2 1 z2 5 a2.

Solution.
Gauss divergence theorem is ∫∫ F ⋅ n dS = ∫∫∫ ∇ ⋅FdV
S V

Given F=x i +y j+z k

3 3 3

∴ ∂ 3 ∂ ∂
( x ) + ( y 3 ) + ( z 3 ) = 3x + 3 y + 3 z = 3( x + y + z )
2 2 2 2 2 2
∇⋅ F =
∂x ∂y ∂z
∴
∫∫ F ⋅ n dS = ∫∫∫ + y 2 + z 2 ) dx dy dz
2
3( x
S V

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 70 5/19/2016 10:23:06 AM

Vector Calculus ■ 9.71

We shall evaluate this triple integral by using spherical polar coordinates.

x = r sin u cos f, y = r sin u sin f, z = r cos u
∂( x , y , z )
dx dy dz = dr d u df = r sin u dr d u df
2
then
∂( r , u, f)
and x2 + y2 + z2 = r2
Here r varies from 0 to a, u varies from 0 to p and f varies from 0 to 2p.
2p p a
∴
∫∫ F ⋅ n dS = ∫ ∫ ∫ 3r
z
4
sin udrd udf
S 0 0 0

2p p a P
= 3 ∫ df∫ sin ud u∫ r 4 dr
0 0 0
θ
a r
⎡ r5 ⎤
= 3 [ f ]0 [ − cos u]0
2p p
⎢ ⎥ o
⎣ 5 ⎦0
y
a5 φ
= 3 ⋅ 2p(−− cos p + cos 0) ⋅
5
a 5
12p 5 x M
= 6p ⋅ 2 ⋅ = a
5 5 Fig. 9.21

Note We have div F = 3( x 2 + y 2 + z 2 ). Since the equation of the surface is x 2 + y 2 + z 2 = a 2 , we

cannot take div F = 3a 2 because F is deﬁned in the volume inside and on S.
But x 2 + y 2 + z 2 = a 2 is true only for points on S.

EXAMPLE 5
Verify Gauss divergence theorem for F 5 4 xzi 2 y 2 j 1 yzk over the cube bounded by x 5 0,
x 5 1, y 5 0, y 5 1, z 5 0, z 5 1.

Solution.
Gauss divergence theorem is
S
∫∫ F ⋅ n dS = ∫∫∫
V
∇ ⋅FdV

Given F = 4 xzi − y 2 j + yzk

∴ ∂ ∂ ∂
∇⋅ F = (4 xz ) + ( − y 2 ) + ( yz ) = 4 z − 2 y + y = 4 z − y
∂x ∂y ∂z
1 1 1
∴ ∫∫∫ ∇ ⋅FdV = ∫ ∫ ∫ (4 z − y) dx dy dz
V 0 0 0
[{ dV = dx dy dz ]

1 1 1 1

= ∫ ∫ ( 4 z − y ) [ x ]0 dy dz = ∫ ∫ [4 z − y ] dy dz
1

0 0 0 0

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 71 5/19/2016 11:59:51 AM

1 1
1
⎡ y2 ⎤ 1
⎡ 1⎤ ⎡ z2 1 ⎤ 1 3
= ∫ ⎢ 4 zy − ⎥ dz = ∫ ⎢ 4z − ⎥ dz = ⎢ 4 − z ⎥ = 2 − =
0
⎣ 2 ⎦ 0 0
⎣ 2 ⎦ ⎣ 2 2 ⎦ 0 2 2
3
⇒ ∫∫∫
V
∇ ⋅ F dv =
2
(1)

z
We shall now evaluate ∫∫ F ⋅ n dS
S D
G
Here the surface S consists of the six faces of the cube.
F
E
∴
∫∫ F ⋅ n dS = ∫∫ F ⋅ n dS + ∫∫ F ⋅ n dS
S S1 S2
O
+ ∫∫ F ⋅ n dS + ∫∫ F ⋅ n dS C y
S3 S4 B
A
+ ∫∫ F ⋅ n dS + ∫∫ F ⋅ n dS x
S5 S6
Fig. 9.22
We shall simplify the computation and put it in the form of a table.

Face Equation Outward normal n F⋅n dS

S1 = ABEF x=1 i 4xz = 4z dy dz
S2 = OCDG x=0 −i −4xz = 0 dy dz
S3 = BCDE y=1 j −y = −1
2
dx dz
S4 = OAFG y=0 −j y2 = 0 dx dz
S5 = DEFG z=1 k yz = y dx dy
S6 = OABC z=0 −k −yz = 0 dx dy
1
1 ⎡z ⎤
1 1 2
1
∴
∫∫ F ⋅ n dS = ∫ ∫ 4 z dy dz = 4 [ y ]0 ⎢ ⎥
⎣ 2 ⎦0
= 4 ⋅1⋅ = 2
2
S1 0 0

∫∫ F ⋅ n dS = S∫∫ 0 dy dz = 0
S2 2

1 1

∫∫ F ⋅ n dS = ∫ ∫ −1 dx dz = − [ x ] [ z ] = −1
1 1
0 0
S 3 0 0

∫∫ F ⋅ n dS = S∫∫ 0 dx dz = 0
S4 4

1
1 ⎡y ⎤
1 1 2
1
∫∫ F ⋅ n dS = ∫ ∫ y dx dy = [ x ]0 ⎢
⎣ 2 ⎥⎦ = 2
S5 0 0 0

and ∫∫ F ⋅ n dS = S∫∫ 0 dx dy = 0
S6 6

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 72 5/19/2016 11:59:58 AM

Vector Calculus ■ 9.73

∴ 1 3
∫∫
S
F ⋅ n dS = 2 + 0 + ( −1) + 0 + + 0 =
2 2
(2)

From (1) and (2), ∫∫ F ⋅ n dS = ∫∫∫

S V
∇ ⋅FdV

Hence, Gauss’s divergence theorem is veriﬁed.

EXAMPLE 6
Verify divergence theorem for F 5 x 2 i 1 zj 1 yzk over the cube formed by the planes x 5 61,
y 5 61, z 5 61.

Solution.
Gauss divergence theorem is ∫∫ F ⋅ n dS = ∫∫∫ ∇ ⋅FdV
S V
Given F = x 2i + zj + yzk
∴ ∂ ∂ ∂
∇ ⋅ F = ( x 2 ) + ( z ) + ( yz ) = 2 x + 0 + y = 2 x + y
∂x ∂y ∂z
1 1 1
∴ ∫∫∫ ∇ ⋅FdV =
V
∫ ∫ ∫ (2 x + y) dx dy dz
−1 −1 −1

1 1 1 1

∫ ∫ ⎡⎣ x + yx ⎤⎦ −1 dy dz = ∫ ∫ [1 + y − (1 − y ) ]dy dz = ∫ ∫
1 1 1
= 2
2 y dy dz = 0
−1 −1
−1 −1 −1 −1

⎡a ⎤
⎢ ∫ f ( x )dx = 0 if f ( x ) is odd function, Here y is odd function ⎥
⎣ −a ⎦
⇒ ∫∫∫
V
∇ ⋅FdV = 0 (1)

∴ shall now compute ∫∫ F ⋅ n dS

We H E
S
S is the surface consisting of the six faces of the cube. G
F
∫∫ F ⋅ n dS = ∫∫ F ⋅ n dS + ∫∫ F ⋅ n dS
S S1 S2
z

y
o
+ ∫∫ F ⋅ n dS + ∫∫ F ⋅ n dS x
D
S3 S4 A

+ ∫∫ F ⋅ n dS + ∫∫ F ⋅ n dS B C
S5 S6
Fig. 9.23
We shall simplify the computations and put it in the form of a table.
Faces Equation Outward normal n F?n dS
S1 = BCFG x=1 i x2 = 1 dy dz
S2 = ADEH x = −1 −i −x2 = −1 dy dz
S3 = CDEF y=1 j z dz dx
S4 = ABGH y = −1 −j −z dz dx
S5 = EFGH z=1 k yz = y dx dy
S6 = ABCD z = −1 −k −yz = y dx dy

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 73 5/19/2016 12:00:06 PM

1 1
∴
∫∫ F ⋅ n dS = ∫ ∫ dy dz = [ y ] [ z ] = (1 + 1) (1 + 1) = 4
1 1
−1 −1
S1 −1 −1
1 1

∫∫ F ⋅ n dS = ∫ ∫ −1dy dz = − [ y ] [ z ] = −[1 + 1] [1 + 1] = −4
1 1
−1 −1
S2 −1 −1
1 1

∫∫ F ⋅ n dS =
S3
∫ ∫ zdz dx = 0
−1 −1
[{ z is odd function ]

1 1 1 1

∫∫ F ⋅ n dS =
S4
∫ ∫ − zdz dx = − ∫ ∫ zdz dx = 0
−1 −1 −1 −1
1 1

∫∫
S
F ⋅ n dS = ∫ ∫ y dx dy = 0
−1 −1
[{ y is odd function ]
5
1 1

∫∫ F ⋅ n dS =
S6
∫ ∫ y dx dy = 0
−1 −1

∴
∫∫ F ⋅ n dS = 4 − 4 + 0 + 0 + 0 + 0 = 0
S
(2)

From (1) and (2), ∫∫ F ⋅ n dS = ∫∫∫ ∇ ⋅FdV

S V

Hence, Gauss’s divergence theorem is veriﬁed.

EXAMPLE 7
Verify divergence theorem for the function F 5 4 xi 2 2 y 2 j 1 z 2 k taken over the surface of the
region, bounded by the cylinder x2 1 y2 5 4 and z 5 0, z 5 3.

Solution.
Gauss divergence theorem is
S
∫∫ F ⋅ n dS = ∫∫∫
V
div FdV

Given F = 4 xi − 2 y 2 j + z 2 k ∴ ∂ ∂ ∂
∇⋅ F = (4 x ) + ( −2 y 2 ) + ( z 2 )
∂x ∂y ∂z
y
= 4 − 4 y + 2z

and z varies from 0 to 3,

Also given x 2 + y 2 = 4

⇒ y2 = 4 − x2 ⇒ y = ± 4 − x2
x
and y=0 ⇒ x =4 2
⇒ x = ±2

2 4− x2 3
∴
∫∫∫ ∇ ⋅FdV =
V
∫ ∫ ∫ (4 − 4 y + 2 z ) dz dy dx
−2 − 4 − x 2 z = 0
Fig. 9.24

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 74 5/19/2016 12:00:16 PM

Vector Calculus ■ 9.75

4 − x2 3
2
⎡ z2 ⎤
= ∫ ∫
−2 − 4 − x 2
⎢
⎣
( 4 − 4 y ) z + 2 ⎥ dy dx
2 ⎦0
2 4 − x2

= ∫ ∫
−2 − 4 − x 2
[( 4 − 4 y ) ⋅ 3 + 9] dy dx

2 4 − x2

= ∫ ∫
−2 − 4 − x 2
[ 21 − 12 y ] dy dx

4 − x2
⎡
2
y2 ⎤
= ∫ ⎢ 21 y − 12 ⎥ dx
−2
⎣ 2 ⎦− 4 − x2

( )
2

= ∫ [ 21 4 − x 2 + 4 − x 2 − 6( 4 − x 2 − ( 4 − x 2 )] dx
−2

= ∫ 42
−2
4 − x 2 dx

2 ⎡{ 4 − x 2 is even function ⎤
= 84∫ 4 − x dx 2 ⎣ ⎦
0

2
⎡x 4 x⎤ p
= 84 ⎢ 4 − x 2 + sin −1 ⎥ = 84 [0 + 2 sin −1 1 − 0] = 84 ⋅ 2 = 84p
⎣2 2 2 ⎦0 2

∫∫∫
V
∇ ⋅FdV = 84 p (1)

We shall now compute the surface integral

S
∫∫ F ? n dS.
S consists of the bottom surface S1, top surface S2 and the curved surface S3 of the cylinder.
z
On S1: Equation is z = 0, n = − k S2

∴ F ⋅ n = −z2 = 0 ⇒
∫∫
S
F ⋅ n dS = 0
1

On S2: Equation is z = 3, n = k S3

∴ F ⋅ n = z 2 = 9, dS =
dx dy
=
dx dy
= dx dy
n⋅k k ⋅k O
y
∴
∫∫ F ⋅ n dS = ∫∫ 9 dx dy = 9∫∫ dx dy
S2 S2 S2
S1
x
= 9 (area of the circle S2) = 9 p 22 = 36p.
Fig. 9.25

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 75 5/19/2016 12:00:21 PM

On S3: Equation of the cylinder is x2 + y2 = 4

Let f = x2 + y2
∴ ∂ ∂ ∂f
∇f = i f+ j f+ k = i 2 x + 2 yj + 0 k = 2( xi + yj )
∂x ∂y ∂z
∇f 2( xi + yj ) 2( xi + yj ) 1
∴ the normal n = = = = ( xi + yj )
∇f 2 x 2 + y 2 2 4 2
1
∴ F ⋅ n = (4 xi − 2 y 2 j + z 2 k ) ⋅ ( xi + yj ) = 2 x 2 − y 3
2
Since S3 is the surface of a cylinder x + y2 = 4, we use cylindrical polar coordinates to evaluate
2

∫∫ F ⋅ n dS
S3 z
∴ x = 2 cos u, y = 2 sin u, z = z ∴ dS = 2 du dz
u varies from 0 to 2p and z varies from 0 to 3
3 2p
∴
∫∫ F ⋅ n dS = ∫ ∫ (2 ⋅ 4 cos u − 8sin 3 u) 2d u dz 2d θ
2
ds dz
S3 0 0
S3
3 2p

= 16∫ ∫ (cos
2
u − sin 3 u) d u dz
0 0
O
y
3 2p
⎡1 + cos 2u 1 ⎤
= 16∫ ∫ ⎢⎣ − (3 sin u − sin 3u) ⎥ d u dz 2d θ
0 0
2 4 ⎦ x
2p
⎡1 ⎛ cos 3u ⎞ ⎤
3
sin 2u ⎞ 1 ⎛ Fig. 9.26
= 16∫ ⎢ ⎜ u + ⎟ − ⎜ −3 cos u + ⎟ dz
0 ⎣2
⎝ 2 ⎠ 4⎝ 3 ⎠ ⎥⎦ 0

⎧1 ⎡ sin 4p ⎤ 1⎡ cos 6p ⎛ coss 0 ⎞ ⎤ ⎫

= 16∫ ⎨ ⎢ 2p + − 0 ⎥ − ⎢ −3 cos 2p + − ⎜ −3 cos 0 + ⎟ ⎬ dz

0⎩ ⎣
2 2 ⎦ 4⎣ 3 ⎝ 3 ⎠ ⎥⎦ ⎭
⎛ 3 1 3 1⎞
3

= 16∫ ⎜ p + − − + ⎟ dz
⎝ 4 12 4 12 ⎠
0
3

= 16p∫ dz = 16p [ z ]0 = 16p × 3 = 48p

∫∫ F ⋅ n dS = 36p + 48p = 84p

S
(2)

From (1) and (2), ∫∫ F ⋅ n dS = ∫∫∫ ∇ ⋅FdV

S V
Hence, Gauss’s divergence theorem is veriﬁed.
EXAMPLE 8
Verify Gauss divergence theorem for F 5 a( x 1 y )i 1 a( y 2 x ) j 1 z 2 k over the region bounded
by the upper hemisphere x2 1 y2 1 z2 5 a2 and the plane z 5 0.

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 76 5/19/2016 12:00:26 PM

Vector Calculus ■ 9.77

Solution.
Gauss divergence theorem is
∫∫ F ⋅ n dS = ∫∫∫
S V
∇ ⋅FdV

Given F = a( x + y )i + a( y − x ) j + z 2 k
∂ ∂ ∂ 2
∴ ∇⋅F = (a( x + y )) + (a( y − x )) + ( z ) = a + a + 2 z = 2( a + z )
∂x ∂y ∂z
∴ ∫∫∫
V
∇ ⋅FdV = ∫∫∫ 2(a + z ) dV
V

= 2a∫∫∫ dV + 2∫∫∫ z dV
V V
a a − x2
2 a2 − x 2 − y 2

= 2aV + 2 ∫ ∫ ∫ z dz dy dx
− a − a2 − x 2 0

a2 − x 2 a2 − x 2 − y 2
a
⎡ z2 ⎤
= 2aV + 2 ∫ ∫ ⎢ ⎥ dy dx
− a − a2 − x 2
⎣ 2 ⎦0
a2 − x 2
2p 3 ⎡ 2 3⎤
a

= 2a a +∫ ∫ ( a 2 − x 2 − y 2 ) dy dx ⎢⎣{ V = 3 pa ⎥⎦
3 −a − a2 − x 2

a2 − x 2
4pa 4
a

= +∫2 ∫ ( a 2 − x 2 − y 2 ) dy dx [{ a2 − x 2 − y 2 is even in y ]
3 −a 0

a2 − x 2
4pa 4
a
⎡ y3 ⎤
= + 2 ∫ ⎢( a 2 − x 2 ) y − ⎥ dx
−a ⎣
3 3 ⎦0

4pa 4
a
⎡ ( a 2 − x 2 )3 / 2 ⎤
= + 2 ∫ ⎢( a 2 − x 2 ) a 2 − x 2 − ⎥ dx
−a ⎣ ⎦
3 3
4pa 4
a
⎡ ( a 2 − x 2 )3 / 2 ⎤
= + 2 ∫ ⎢( a 2 − x 2 )3 / 2 − ⎥ dx
−a ⎣ ⎦
3 3
4pa 4
a
2
= + 2 ⋅ ∫ ( a 2 − x 2 )3 / 2 dx
3 3 −a

4pa 4 4
a
4pa 4 8
= + × 2∫ ( a 2 − x 2 )3 / 2 dx = + I [{ (a2 − x 2 )3/ 2 is even]
3 3 0 3 3
a
where I = ∫ ( a 2 − x 2 )3 / 2 dx
0

Put x = a sin u ∴ dx = a cos udu

p
When x = 0, sin u = 0 ⇒ u = 0 and when x = a, sin u = 1 ⇒ u =
2

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 77 5/19/2016 12:00:31 PM

p
2
∴ I = ∫ ( a 2 − a 2 sin 2 u)3 / 2 a cos ud u
0
p
2
= ∫ a3 cos3 u ⋅ a cos ud u
0
p/2
4 −1 4 − 3 p 3 1 p 3pa 4
= a4 ∫
0
cos 4 u d u = a 4 ⋅ ⋅
4 4−2 2
⋅ = a4 ⋅ ⋅ =
4 2 2 16
4pa 4 8 3pa 4 (8 + 3) 4 11 4
∫∫∫ ∇ ⋅F =
V 3
+ ⋅
3 16
=
6
pa = pa
6
(1)

Now we shall compute the double integral ∫∫ F ⋅ n dS

S z
S consists of S1 and S2
∴
∫∫ F ⋅ n dS = ∫∫ F ⋅ n dS + ∫∫ F ⋅ n dS
S2
1 2
S S 1 S 2

On S1: z = 0, n = − k
∴ F ⋅ n = ( a( x + y ) i + a ( y − x ) j + z 2 k ) ⋅ ( − k ) = − z 2 = 0
∴
∫∫
O
F ⋅ n dS = 0 y
S 1
S1
On S2: x2 + y2 + z2 = a2
x
Let f = x2 + y2 + z2
∴ Fig. 9.27
∇f = 2 x i + 2 y j + 2 z k
= 2( x i + y j + z k )
∇f 2( xi + yj + zk ) xi + yj + zk xi + yj + zk z
∴ n= = = and n ⋅ k = ⋅k =
∇f 2 x + y + z 2 2 2 a a a
xi + yj + zk
F ⋅ n = [ a( x + y ) i + a ( y − x ) j + z 2 k ] ⋅
a
z3 z3
= ( x + y) x + ( y − x) y + = x 2 + y 2 +
a a
dx dy
∫∫S F ⋅ n dS = R∫∫ F ⋅ n n ⋅ k , where R is the projection of S2 on the xy−plane.
2

∴ ⎛ z 3 ⎞ dx dy
∫∫S F ⋅ n dS = R∫∫ ⎜⎝ x + y2 + ⎟
2

2
a⎠ z
a
⎛ a( x 2 + y 2 ) ⎞
= ∫∫ ⎜⎝
R z
+ z 2 ⎟ dx dy
⎠

⎛ a( x 2 + y 2 ) ⎞
=
R
∫∫ ⎜⎝
z
+ [a 2 − x 2 − y 2 ]⎟ dx dy
⎠

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 78 5/19/2016 12:00:37 PM

Vector Calculus ■ 9.79

Changing to polar coordinate, we have

x = r cos u, y = r sin u, r2 = x2 + y2 and dx dy = r dr du

a 2p
⎧ ar 2 2 ⎫
∴ ∫∫ F ⋅ n dS = ∫ ∫ ⎨ 2 2 + ( a − r )⎬ r dr d u
2

S2 0 0 ⎩ a −r ⎭
a 2p
⎧ − a( a 2 − r 2 ) + a3 ⎫
=∫ ∫⎨ + ( a 2 − r 2 )⎬ r dr d u
0 0 ⎩ a −r
2 2
⎭
a 2p
⎧ a3 ⎫
=∫ ∫ ⎨ − a a 2
− r 2
+ + ( a 2 − r 2 )⎬ r dr d u
0 0 ⎩ a −r
2 2
⎭
2p a
⎧ a 3
⎫
= ∫ d u∫ ⎨ − a a 2 − r 2 + + ( a 2 − r 2 )⎬ r dr
0 0⎩ a −r
2 2
⎭
a

= [u]0 ∫ ( −a
2p
a2 − r 2 )r + a3 (a2 − r 2 ) −1/ 2 r + (a2 − r 2 ) r ] dr
0

⎧⎪ a a a3
a a
⎫⎪
= 2p ⎨∫ + (a2 − r 2 )( −2r )dr − ∫ (a2 − r 2 ) −1/ 2 ( −2r )dr + ∫ (a2 r − r 3 )dr ⎬
⎩⎪ 0 2 2 0 0 ⎪⎭
⎧ a ⎡ (a2 − r 2 )3/ 2 ⎤ a a3 ⎡ (a2 − r 2 )1/ 2 ⎤ a ⎡ r 2 r 4 ⎤ a ⎫
= 2p ⎪ ⎢ ⎥ − ⎢ ⎥ + ⎢a
2
− ⎥ ⎪
⎨2 3 2 1 ⎣ 2 4 ⎦0 ⎬
⎪ ⎣ ⎢ ⎥ ⎢ ⎥ ⎪
⎩ 2 ⎦0 ⎣ 2 ⎦0 ⎭
⎡a a4 a4 ⎤
= 2p ⎢ (0 − a3 ) − a3 (0 − a) + − ⎥
⎣3 2 4⎦
⎡ a 4
a ⎤
4
11a4 11pa4
= 2p ⎢ − + a4 + ⎥ = 2p × =
⎣ 3 4⎦ 12 6

∴ 11pa 4 11pa 4
∫∫ F ⋅ ndS = 0 +
S 6
=
6
(2)

From (1) and (2),

S
∫∫ F ⋅ n dS = ∫∫∫
V
∇ ⋅FdV

Hence, Gauss’s divergence theorem is veriﬁed.

EXAMPLE 9

∫∫ x dy dz 1 x 2 y dz dx 1 x 2 z dx dy over the surface z 5 0, z 5 h, x2 1 y2 5 a2.

3
Evaluate
S
Solution.
We know Gauss divergence theorem in cartesian form is

⎛ ∂F ∂F ∂F ⎞
∫∫ F dy dz + F
S
1 2 dz dx + F3 dx dy = ∫∫∫ ⎜ 1 + 2 + 3 ⎟ dx dy dz
V ⎝ ∂x ∂y ∂z ⎠

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 79 5/19/2016 12:00:40 PM

∫∫ x
z
Given surface integral is 3
dy dz + x 2 y dz dx + x 2 z dx dy z=h
S

Here F1 = x 3 , F2 = x 2 y, F3 = x 2 z
∂F1 ∂F2 ∂F3
∴ = 3x 2 , = x2 , = x2
∂x ∂y ∂z
∂F1 ∂F2 ∂F3
∴ + + = 3x 2 + x 2 + x 2 = 5 x 2 O
∂x ∂y ∂z y

∫∫ F dy dz + F
S
1 2 dz dx + F3 dx dy = ∫∫∫ 5 x 2 dx dy dz
V x
z=0

h a a2 − y 2
Fig. 9.28
=5∫ ∫ ∫
2
x dx dy dz
z = 0 y = − a x = − a2 − y 2

h ⎡ a a2 − y 2 ⎤
= 5∫ ∫ ⎢ 2 ∫ x 2 dx ⎥ dy dz [{ x 2 is even ]
⎢
0 y=−a ⎣ 0
⎥⎦
a2 − y 2
⎡ x3 ⎤
h a

= 10 ∫ ∫ ⎢ ⎥ dydz
z=0 y=−a ⎣
3 ⎦0
h a
10
3 z∫= 0 y =∫− a
= ( a 2 − y 2 )3 / 2 dydz

10
h
⎡ a 2 ⎤
= ∫ ⎢ 2∫ ( a − y ) dy ⎥
2 3/ 2
dz
3 0 ⎣ 0 ⎦
a a
20 20 20 h
=
3
[ z ]0h ∫ (a2 − y 2 )3/ 2 dy = 3 h∫ ( a2 − y 2 )3/ 2 dy = 3 × I
0 0

where I = ∫ ( a 2 − y 2 )3 / 2 dy
0

Put y = a sin u ∴dy = a cos u du

p
When y = 0, sin u = 0 ⇒ u = 0 and when y = a, sin u = 1 ⇒ u=
2
p/2 p/2

∴I= ∫ ( a 2 − a 2 sin 2 u)3/ 2 a cos ud u = a 4 ∫ cos u cos ud u

0 0
p/2
3 1 p 3pa 4
= a4 ∫0
cos 4 ud u = a 4 ⋅ ⋅ ⋅ =
4 2 2 16
20 3pa 4 5 4
∴ ∫∫
S
F1 dy dz + F2 dz dx + F3 dx dy =
3
h×
16
= pa h
4

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 80 5/19/2016 12:00:44 PM

Vector Calculus ■ 9.81

9.13 STOKE’S THEOREM

Stoke’s theorem gives a relation between line integral and surface integral.

Theorem 9.1 If S is an open surface bounded by a simple closed curve C and if F is continuous having
continuous partial derivatives in S and on C, then ∫ F? dr 5 S∫∫ curl F? n dS,
C
where C is traversed in the positive direction.

Proof Let F = F1i + F2 j + F3 k and r be the position vector of any point P on S.

∴ r = xi + yj + zk ⇒ dr = dxi + dyj + dzk
∴ ( )
F ⋅ dr = F1i + F2 j + F3 k ⋅ ( dxi + dyj + dzk ) = F1dx + F2 dy + F3 dz
∴ ∫ F ⋅ d r = ∫ (F dx + F dy + F dz )
1 2 3
z
n
C C C
Let z = f(x, y) be the equation of the surface S enclosed
by the curve C.
C
Any line parallel to Z-axis intersects the surface in at
most one point. The positive direction of the normal n
is that it makes an acute angle with the positive Z-axis
O
(or k ). y
The projection of S on the xy-plane is a region R
enclosed by C′. R
C′
Now,
∫ F dx = ∫ F ( x, y, z )dx
C
1
C
1
x
Fig. 9.29
= ∫ F1 ((x , y , f (x , y )) dx =
C′
∫ P(x , y )dx
C′

where P ( x, y ) = F1 ( x, y f ( x, y ) )

By Green’s theorem,
∂P [{ Q = 0 here]
∫ P(x , y )dx = ∫∫ − ∂y dx dy
C′ R

But P ( x, y ) = F1 ( x, y f ( x, y ) )

∂P ∂F1 ∂F1 ∂f
∴ = + ⋅ [{P(x, y) = F1(x, y, z) and z = f(x, y)] (1)
∂y ∂y ∂z ∂y
⎛ ∂F1 ∂F1 ∂f ⎞
∴ ∫ P(x , y )dx = −∫∫ ⎜⎝ ∂y
C′ R
+ ⋅
∂z ∂y ⎟⎠
dx dy (2)

Now ∫∫ ∇ × F ⋅ n dS = ∫∫ ∇ × ( F i + F
S S
1 2 j + F3 k ) ⋅ n dS

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 81 5/19/2016 12:01:00 PM

Consider ∫∫ ( = × F1 i ) ⋅ n dS
S

i j k
∂ ∂ ∂ ⎛ ∂F1 ⎞ ⎛ ∂F1 ⎞ ∂F1 ∂F1
But ∇ × F1i =
∂x ∂y ∂z = i (0) − j ⎜⎝ 0 − ∂z ⎟⎠ + k ⎝⎜ 0 − ∂y ⎠⎟ = ∂z j − ∂y k
F1 0 0

∴ ∂F1 ∂F
(∇ × F1i ) ⋅ n = j ⋅n − 1 k ⋅n (3)
∂z ∂y
We have r = xi + yj + zk = xi + yj + f ( x, y )k [since z = f(x, y)]

∴ ∂r ∂f
= j+ k
∂y ∂y

But ∂ r is a tangent vector to S at P, and hence, ∂ r is ⊥ to n. ∴ ∂r

⋅n = 0
∂y ∂y ∂y
∂f ∂f
Substituting in (4), we get j ⋅ n + k ⋅n = 0 ⇒ j ⋅n = − k ⋅n
∂y ∂y
∂F1 ⎛ ∂f ⎞ ∂F ⎛ ∂F ∂f ∂F1 ⎞
∴ (3) ⇒ ∇ × F1i ⋅ n = − k ⋅ n⎟ − 1 k ⋅ n = − ⎜ 1 + k ⋅n
∂z ⎝⎜ ∂y ⎠ ∂y ⎝ ∂z ∂y ∂y ⎠⎟

∴ ⎛ ∂F1 ∂f ∂F1 ⎞
∫∫ (∇ × F i ) ⋅ n dS = −∫∫ ⎜⎝ ∂z
S
1
S
+
∂y ∂y ⎠⎟
(k ⋅ n ) dS

⇒ ⎛ ∂F1 ∂f ∂F1 ⎞
∫∫ (∇ × F i ) ⋅ n dS = −∫∫ ⎜⎝ ∂z
S
1
R
+
∂y ∂y ⎠⎟
dx dy (5)

From (2) and (5), we get

∫ F dx = ∫∫ ∇ × F i ⋅ n dS
C′
1
S
1

Similarly, ∫ F dy = ∫∫ (∇ × F
C′
2
S
2 j ) ⋅ n dS (6)

and ∫ F dz = ∫∫ (∇ × F k ) ⋅ n dS
C′
3
S
3 (7)

Adding (5), (6), and (7), we get

∫ F dx + F dy + F dz = ∫∫ ∇ × ( F i + F
C′
1 2 3
S
1 2 )
j + F3 k ⋅ n dS

⇒ ∫ F ⋅ d r = ∫∫ ∇ × F ⋅n dS
C S

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 82 5/19/2016 12:01:10 PM

Vector Calculus ■ 9.83

Note
If S is the region R in the xy−plane, bounded by the simple closed curve C, then n = k is the outward
unit normal.
∴ Stoke’s theorem in the plane is ∫ F ⋅ dr = R∫∫ Curl F ⋅ k dR,
C
which is Green’s theorem.

Cartesian form of Stoke’s theorem

If F = F1i + F2 j + F3k , then
i j k
∂ ∂ ∂ ⎛ ∂F3 ∂F2 ⎞ ⎛ ∂F3 ∂F1 ⎞ ⎛ ∂F2 ∂F1 ⎞
Curl F =
∂x ∂y ∂z = i ⎜⎝ ∂y − ∂z ⎟⎠ − j ⎜⎝ ∂x − ∂z ⎟⎠ + k ⎝⎜ ∂x − ∂y ⎠⎟
F1 F2 F3
and F ⋅ dr = F1 dx + F2 dy + F3 dz
∴ the cartesian form of Stoke’s theorem is
C
∫ (F dx + F dy + F dz )
1 2 3

⎡⎛ ∂F3 ∂F2 ⎞ ⎛ ∂F ∂F ⎞ ⎛ ∂F ∂F ⎞ ⎤
= ∫∫ ⎢⎜
S ⎣ ⎝ ∂y
− ⎟
∂z ⎠
dydz + ⎜ 1 − 3 ⎟ dzdx + ⎜ 2 − 1 ⎟ dxdy ⎥
⎝ ∂z ∂x ⎠ ⎝ ∂x ∂y ⎠ ⎦
Note
If F = Pi + Qj and r = xi + yj , then d r = dx i + dy j and F ⋅ dr = P dx + Q dy

i j k
∂ ∂ ∂ ⎛ ∂Q ∂P ⎞ ⎛ ∂Q ∂P ⎞
Curl F = = i (0 − 0) − j (0 − 0) + k ⎜ − = − k
∂x ∂y ∂z ⎝ ∂x ∂y ⎠⎟ ⎝⎜ ∂x ∂y ⎠⎟
P Q 0
∴ ∂Q ∂P
Curl F ⋅ k = −
∂x ∂y
⎛ ∂Q ∂P ⎞
∴ Stokes theorem in the plane is ∫ ( P dx + Q dy ) = ∫∫ ⎜ − dxdy
C R ⎝ ∂x ∂y ⎠⎟
which is Green’s thorem.

WORKED EXAMPLES
EXAMPLE 1
Prove that r ⋅ dr 5 0, where C is the simple closed curve.
C
Solution.
Let r be the position vector of any point P(x, y, z) on C. ∴ r = xi + yj + zk .

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 83 5/19/2016 12:01:18 PM

Stokes theorem is ∫ F ⋅ d r = S∫∫ curl F ⋅ n dS

C
Here F = r .
i j k
∂ ∂ ∂
∴ Curl F = Curl r = = i (0 − 0) − j (0 − 0) + k (0 − 0) + (0 − 0) = 0
∂x ∂y ∂z
x y z
∴ ∫ r ⋅ dr = 0
C

EXAMPLE 2
If A is solenoidal, then prove that ∫∫ ∇2 A ? ndS
ds 52 ∫ Curl A ? dr .
S C

Solution.
Given A is solenoidal. ∴ ∇⋅A = 0
We know ∇ × (∇ × A ) = ∇(∇ ⋅ A ) − ∇ 2 A = −∇ 2 A

Stoke’s theorem is ∫∫ ∇ × F ⋅ n dS = ∫ F ⋅ d r
S C

Putting F = ∇ × A, we get ∇ × F = −∇ 2 A

∴ ∫∫ −∇ A ⋅ n dS = ∫ ∇ × A ⋅ d r
2

S C

⇒
∫∫ ∇ A ⋅ n dS = − ∫ Curl A ⋅ d r
2

S C

EXAMPLE 3
Prove that ∫ f d r 52∫∫ ∇f3 n dS .
C S

Solution.
Stoke’s theorem is ∫ F ⋅ d r = ∫∫ Curl F ⋅ n dS = ∫∫ ∇ × F ⋅ n dS
C S S

Put F = f a, where a an arbitrary constant vector.

∴
∫ (f a) ⋅ d r = ∫∫ ∇ × f a ⋅ n dS
C S

We know curl f a = ∇ × fa = ∇f × a + f∇ × a = ∇f × a [{ ∇ × a = 0]
∴ ∫ (f a) ⋅ d r = ∫∫ (∇f × a) ⋅ n dS
C S

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 84 5/19/2016 12:01:28 PM

Vector Calculus ■ 9.85

⇒
∫ f a ⋅ d r = −∫∫ (a × ∇f) ⋅ n dS
C S

⇒ a ⋅ ⎛ ∫ f dr ⎞ = − ∫∫ a ⋅ ( ∇f × n ) dS [Interchanging dot and cross]

⎝C ⎠
S

⇒
a ⋅ ⎜ ∫ f d r ⎟ = −a ⋅ ∫∫ ∇f × n dS = a ⋅ ⎜ − ∫∫ ∇f × n dS ⎟
⎝C ⎠ S ⎝ S ⎠

∴ ∫ fd r
C
= − ∫∫ ∇f × n dS
S
[{ a is arbitrary]

EXAMPLE 4
If S is the surface of the sphere x2 1 y2 1 z2 5 a2, then show that ∫∫ Curl F ⋅ n dS 5 0.
S
Solution.
Suppose the sphere is cut by a plane into two parts S1 and S2 and let C be the curve binding these two
parts.
n

Then ∫∫ Curl F ⋅ n dS = ∫∫
S S
Curl F ⋅ n dS + ∫∫ Curl F ⋅ n dS
S
1 2 S1

By Stoke’s theorem, ∫∫ Curl F ⋅ n dS = ∫ F ⋅ dr

S1 C

and ∫∫
S
Curl F ⋅ n dS = − ∫ F ⋅ dr , because for S
C
2 C S2
2

the positive sense of the curve C is the opposite direction of C in S1 n

Fig. 9.30
∴
∫∫ Curl F ⋅ n dS = C∫ F ⋅ dr − C∫ F ⋅ dr = 0
S

EXAMPLE 5

∫ ( xydx 1 xy
2
Evaluate dy ) by Stoke’s theorem, where C is the square in the xy−plane with
C
vertices (1, 0), (21, 0), (0, 1), (0, 21).

Solution.
Stoke’s theorem is ∫ F ⋅ dr = S∫∫ curl F ⋅ n dS
C

∫ ( xy dx + xy dy) and r = xi + yj ∴
2
Given dr = dxi + dyj .
C

Here F ⋅ dr = xy dx + xy 2 dy ∴ F = xy i + xy 2 j

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 85 5/19/2016 12:01:37 PM

i j k
∴ ∂ ∂ ∂
Curl F = ∇ × F = = i ( 0 − 0) − j ( 0 − 0) + k ( y 2 − x )
∂x ∂y ∂z
xy xy 2 0
⇒ Curl F = ( y 2 − x )k
Also given C is the square in the xy plane with vertices (1, 0), (−1, 0), (0, 1), (0, −1).

∴ n = k and dS = dx dy y

∴ Curl F ⋅ n = ( y 2 − x )k ⋅ k = y 2 − x B (0, 1)
y=x+1 y = −(x − 1)
∴
∫∫ Curl F ⋅ n dS = ∫∫ ( y − x) dx dy
2

S R (−1, 0)
(1, 0)
where R is the region inside the square. x′ x
C O A

∫ xy dx + xy dy = R∫∫ ( y − x ) dx dy
2 2
That is y=x−1
C y = −(x + 1) D (0, −1)
We shall now evaluate this double integral.
y′
Fig. 9.31
Equation of AB in intercept form is
x y
+ =1 ⇒ x + y = 1 ⇒ y = −x + 1 ⇒ y = −(x − 1)
1 1
x y
Equation of BC is + =1 ⇒ y−x=1 ⇒ y=x+1
−1 1
x y
Equation of CD is + =1 ⇒ x + y = −1 ⇒ y = −(x + 1)
−1 −1
x y
Equation of AD is + =1 ⇒ y − x = −1 ⇒ y=x−1
1 −1
0 x +1 1 − ( x −1)

∴ ∫ ( xy dx + xy dy) = ∫
2
∫ ( y 2 − x ) dydx + ∫ ∫ ( y 2 − x ) dydx
C −1 − ( x +1) 0 x −1

x +1 − ( x −1)
⎡ y3
0
⎤ 1
⎡ y3 ⎤
= ∫ ⎢ − xy ⎥ dx + ∫ ⎢ − xy ⎥ dx
−1 ⎣ ⎦ − ( x +1) 0 ⎣ ⎦ x −1
3 3

∫ 3 {⎡⎣ ( x + 1) }
0
1
= 3
− ( −( x + 1))3 ⎤⎦ − x [ x + 1 − ( −( x + 1)) ] dx
−1

⎧1
1

+ ∫ ⎨ ⎡⎣( −( x − 1))3 − ( x − 1)3 ⎤⎦ − x[ − ( x − 1) − ( x − 1)]} dx

0⎩
3

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 86 5/19/2016 12:01:42 PM

Vector Calculus ■ 9.87

{ }
0
1
= ∫
−1
3
[( x + 1)3 + ( x + 1)3 ] − x[( x + 1) + ( x + 1)] dx

∫{ }
1
1
+ − [( x − 1)3 + ( x − 1)3 ] + x[ x − 1 + x − 1] dx
0
3

⎡2 ⎤ ⎡ 2 ⎤
0 1

= ∫−1 ⎢⎣ 3 ( x + 1) − 2 x( x + 1)⎥⎦ dx + ∫0 ⎢⎣ − 3 ( x − 1) + 2 x( x − 1)⎥⎦ dx

3 3

0 1
⎡ 2 ( x + 1) 4 ⎛ x3 x 2 ⎞ ⎤ ⎡ 2 ( x − 1) 4 ⎛ x3 x 2 ⎞ ⎤
=⎢ − 2 ⎜ + ⎟ ⎥ + ⎢− + 2⎜ − ⎟ ⎥
⎣3 4 ⎝ 3 2 ⎠ ⎦ −1 ⎣ 3 4 ⎝ 3 2 ⎠ ⎦0
2 ⎛ 1⎞ ⎧ ⎡1 ( −1) 2 ⎤ ⎫ 2 ⎡ ⎛ 1 ⎞ ⎤ ⎡1 1 ⎤
= ⎜⎝ ⎟⎠ − 2 ⎨0 − ⎢ ( −1) +
3
⎥ ⎬ − ⎢0 − ⎜⎝ ⎟⎠ ⎥ + 2 ⎢ − ⎥
3 4 ⎩ ⎣3 2 ⎦⎭ 3 ⎣ 4 ⎦ ⎣3 2⎦
1 2 1 2 2 1
= − +1+ + −1 = =
6 3 6 3 6 3

EXAMPLE 6

Evaluate ∫ [( x 1 y )dx 1 ( 2 x 2 z )dy 1 ( y 1 z )dz ]

C
where C is the boundary of the triangle

with the vertices (2, 0, 0), (0, 3, 0) and (0, 0, 6), using Stoke’s theorem.

Solution. z
Stoke’s theorem is C (0, 0, 6)

∫ F ⋅ dr = S∫∫ Curl F ⋅ n dS ,
C
where S is the surface of the triangle ABC bounded
by the curve C, consisting of the sides of the triangle o 3 B
in the ﬁgure. 2 (0, 3, 0) y
Given F ⋅ d r = ( x + y )dx + (2x − z )dy + ( y + z )dz A
(2, 0, 0) x y
Here F = ( x + y )i + (2 x − z ) j + ( y + z )k x + = 1
2 3
i j k Fig. 9.32
∂ ∂ ∂
∴ Curl F =
∂x ∂y ∂z
x + y 2x − z y+z
⎡∂ ∂ ⎤ ⎡∂ ∂ ⎤ ⎡∂ ∂ ⎤
= i ⎢ ( y + z ) − (2 x − z ) ⎥ − j ⎢ ( y + z ) − ( x + y ) ⎥ + k ⎢ (2 x − z ) − ( x + y ) ⎥
⎣ ∂y ∂z ⎦ ⎣ ∂x ∂z ⎦ ⎣ ∂x ∂y ⎦
= i[1 − ( −1)] − j[0 − 0] + k (2 − 1)] = 2i + k

x y z
Equation of the plane ABC is + + =1 [intercept form]
2 3 6

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 87 5/19/2016 12:01:47 PM

x y z ∂f 1 ∂f 1 ∂f 1
∴ f= + + , = , = , =
2 3 6 ∂x 2 ∂y 3 ∂z 6
∂f ∂f ∂f = 1 i + 1 j + 1 k = 1 (3i + 2 j + k )
∴ ∇f = i +j +k
∂x ∂y ∂z 2 3 6 6
1
(3i + 2 j + k )
∴ ∇f 1
n= = 6 = (3i + 2 j + k )
∇f 1 14
9 + 4 +1
6

∴ 1 1 7
Curl F ⋅ n = (2i + k ) ⋅ (3i + 2 j + k ) = (6 + 1) =
14 14 14
∴ 7 7 dx dy
∫∫ Curl F ⋅ n dS = S∫∫
S 14
dS =
14
∫∫
R n⋅k
where R is the orthogonal projection of S on the xy−plane.
1 1
But n⋅k = (3i + 2 j + k ) ⋅ k =
14 14

∴ 7 dxdy
∫∫ Curl F ⋅ n dS =
S
∫∫
14 R 1
14
1
= 7 ∫∫ dx dy = 7 × Area of Δ OA B = 7 ⋅ ⋅ 2 ⋅ 3 = 21
R 2
∴ ∫ [(x + y )dx + (2x − z )dy + ( y + z )dz ] = 21.
C

EXAMPLE 7
Using Stoke’s theorem, evaluate ∫ F ⋅ dr , where F 5 y 2 i 1 x 2 j 2 ( x 1 z ) k and C is the boundary
C

of the triangle with vertices at (0, 0, 0), (1, 0, 0), (1, 1, 0).

Solution.
Given F = y 2 i + x 2 j − ( x + z )k
Stoke’s theorem is

∫ F ⋅ dr = ∫∫ curl F ⋅ n dS
C S

i j k
∂ ∂ ∂
Now curl F = ∇ × F =
∂x ∂y ∂z
y2 x2 −x − z

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 88 5/19/2016 12:01:53 PM

Vector Calculus ■ 9.89

⎡∂ ∂ ⎤ ⎡∂ ∂ ⎤
= ⎢ ( − x − z ) − ( x 2 )⎥ i − ⎢ ( − x − z ) − ( y 2 )⎥ j
⎣ ∂ y ∂ z ⎦ ⎣ ∂ x ∂ z ⎦
⎡∂ ∂ ⎤
+ ⎢ ( x 2 ) − ( y 2 )⎥ k
⎣ ∂ x ∂ y ⎦
= (0) i − [ −1] j + [2 x − 2 y ] k = j + 2 ( x − y ) k .

Given C is the boundary of the triangle formed by the points (0, 0, 0), (1, 0, 0) and (1, 1, 0) which lie
in the xy−plane. ∴ n=k
B(1, 1)
∴ curl F ⋅ n = 2 ( x − y ) y=x
∴ ∫ F ⋅ dr = ∫∫ 2 ( x − y) dx dy
C S

Equation of OB is y = x
O A(1,0)
1 x
Fig. 9.33
∴ ∫ F ⋅ dr = 2∫ ∫ ( x − y) dy dx
C 0 0

x
⎡ y2⎤
1

= 2∫ ⎢ xy − ⎥ dx
0 ⎣
2 ⎦0
1
1
⎡ x2 ⎤ 1
x2 ⎡x3 ⎤ 1
= 2∫ ⎢ x 2 − − 0 ⎥ dx = 2∫ dx = ⎢ ⎥ = .
0 ⎣ ⎦ ⎣ 3 ⎦0 3
2 0
2

EXAMPLE 8
Verify Stoke’s theorem for F 5 ( y 2 z 1 2) i 1 ( yz 1 4) j 2 xzk , where S is the surface of the
cube x 5 0, x 5 2, y 5 0, y 5 2, z 5 0 and z 5 2 above the xy−plane.

Solution. z
Given F = ( y − z + 2)i + ( yz + 4) j − xzk .
G D
Stoke’s theorem is
F
E
∫
C
F ⋅ dr = ∫∫ curl F ⋅ n dS
S
O
i j k C y
A
∂ ∂ ∂ B
Now curl F = x
∂x ∂y ∂z
Fig. 9.34
y−z+2 yz + 4 − xz

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 89 5/19/2016 12:01:58 PM

⎡∂ ∂ ⎤ ⎡∂ ∂ ⎤
= i ⎢ ( − xz ) − ( yz + 4) ⎥ − j ⎢ ( − xz ) − ( y − z + 2) ⎥
⎣ ∂ y ∂ z ⎦ ⎣ ∂ x ∂ z ⎦
⎡∂ ∂ ⎤
+ k ⎢ ( yz + 4) − ( y − z + 2) ⎥
⎣ ∂ x ∂ y ⎦
= i [(0 − y )] − j [ − z − ( −1)] + k (0 − 1) = − yi + ( z − 1) j − k

We shall compute
S
∫∫ Curl F ⋅n dS .
Given S is the open surface consisting of 5 faces of the cube except the face OABC.

∫∫ Curl F ⋅n dS = ∫∫
S S
Curl F ⋅n dS + ∫∫ Curl F ⋅n dS + ∫∫ Curl F ⋅n dS
S S
1 2 3

+ ∫∫ Curl F ⋅n dS + ∫∫ Curl F ⋅n dS
S4 S5

Face Equation Outward normal n F?n dS

S1 = ABEF x=2 i −y dy dz
S2 = OCDG x=0 −i y dy dz

S3 = BCDE y=2 j z−1 dx dz

S4 = OAFG y=0 −j −(z − 1) dx dz
S5 = DEFG z=2 k −1 dx dy

2
⎡ −y 2 ⎤
2 2 2 2
∴
∫∫ Curl F ⋅n dS = ∫ ∫ − y dy dz = ∫ dz ⋅ ∫ ( − y ) dy = [ z ]0 ⎢ ⎥ = 2( −2) = −4
2

S1 0 0 0 0 ⎣ 2 ⎦0
2
2 ⎡y ⎤
2 2 2 2 2

∫∫ Curl F ⋅n dS = ∫0 ∫0 y dy dz = ∫0 ∫0
dz y dy = [ z ]0 ⎢ ⎥ = 2⋅⋅ 2 = 4
⎣ 2 ⎦0
S2

2 ⎡ ( z − 1) ⎤
2 2 2 2 2

∫∫ Curl F ⋅n dS = ∫ ∫ ( z − 1) dz dx = ∫ dx ⋅ ∫ ( z − 1) dz = [ x ]0 ⋅ ⎢
S3 0 0 0 0
⎣ 2 ⎥⎦ 0

= 2⋅
1
2
{ }
( 2 − 1) 2 − ( −1) 2 = 1 − 1 = 0

2 2

∫∫
S
Curl F ⋅n dS = ∫ ∫ −( z − 1) dz dx = 0
0 0
[as above]
4

2 2

∫∫ Curl F ⋅n dS = ∫ ∫ −1 dx dy = − [ x ]0 [ y ]0 = −4
2 2
and
S5 0 0

∴ ∫∫S Curl F ⋅n dS = −4 + 4 + 0 + 0 − 4 = −4 (1)

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 90 5/19/2016 12:02:05 PM

Vector Calculus ■ 9.91

We shall now compute the line integral over the simple closed curve C bounding the surface consisting
of the edges OA, AB, BC and CO in z = 0 plane

∴
C
∫ F ⋅ dr = OA∫ F ⋅ dr + AB∫ F ⋅ dr + BC∫ F ⋅ dr + CO∫ F ⋅ dr
Now F ⋅ dr = ⎡⎣( y − z + 2)i + ( yz + 4) j − xzk ⎤⎦ ⋅ ⎡⎣ dxi + dy j + dzk ⎤⎦
= ( y − z + 2)dx + ( yz + 4)dy − xzdz

⇒ F ⋅ dr = ( y + 2)dx + 4 dy [{ z = 0]

On OA: y = 0 ∴ dy = 0 and F ⋅ dr = 2dx and x varies from 0 to 2

2
∴ ∫ F ⋅ dr = ∫ 2dx = 2 [ x ]0 = 4
2

OA 0

On AB: x = 2 ∴ dx = 0 and F ⋅ dr = 4 dy and y varies from 0 to 2

∫ F ⋅ d r = ∫ 4dy = 4 [ y ]0 = 8
2
∴
AB 0

On BC: y = 2 ∴ dy = 0 and F ⋅ dr = 4 dx and x varies from 2 to 0

0
∴ ∫ F ⋅ dr = ∫ 4 dx = 4 [ x ]2 = 4( −2) = −8
0

BC 2

On CO: x = 0 ∴ dx = 0, F ⋅ dr = 4 dy and y varies from 2 to 0

∴ ∫ F ⋅ dr = ∫ 4dy = 4 [ y ]2 = −8
0

CO 2

∴ ∫ F ⋅ dr = 4 + 8 − 8 − 8 = −4
C
(2)

From (1) and (2), ∫ F ⋅ dr = S∫∫ Curl F ⋅n dS

Hence, Stoke’s theorem is veriﬁed.

EXAMPLE 9
Verify Stoke’s theorem for F 5 ( x 2 2 y 2 ) i 1 2 xy j in the rectangular region in the xy plane
bounded by the lines x 5 0, x 5 a, y 5 0, y 5 b.

Solution.
Given F = ( x 2 − y 2 )i + 2 xy j

Stoke’s theorem is
∫ F ⋅ dr = S∫∫ Curl F ⋅ n dS
C

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 91 5/19/2016 12:19:34 PM

i j k y
∂ ∂ ∂ y=b
Curl F = B
∂x ∂y ∂z C

x2 − y2 2 xy 0 x=a
x=0
= i( 0 − 0) − j ( 0 − 0) + k ( 2 y + 2 y ) = 4 y k
A
Since the surface is a rectangle in the xy−plane, normal n = k O y=0 (a, 0) x

∴ Curl F ⋅ n = 4 yk ⋅ k = 4 y Fig. 9.35

a b

∫∫S Curl F ⋅n dS = ∫ ∫ 4 y dx dy
0 0
b
a b
⎡ y2 ⎤
⇒ ∫∫S Curl F ⋅n dS = ∫0 ∫0
dx 4 y dy = [ x ]a
0
4 ⎢ ⎥ = 2ab
⎣ 2 ⎦0
2
(1)

We shall now compute the line integral.

∴ ∫ F ⋅ dr = OA∫ F ⋅ dr + AB∫ F ⋅ dr + BC∫ F ⋅ dr + CO∫ F ⋅ dr

C
Now F ⋅ dr = ( x 2 − y 2 )dx + 2 xydy
On OA: y = 0 ∴ dy = 0 and F ⋅ dr = x 2 dx and x varies from 0 to a
a
a
⎡ x3 ⎤ a3
∴ ∫ ⋅ = ∫0 = =
2
F dr x dx ⎢ ⎥
OA ⎣ 3 ⎦0 3

On AB: x = a ∴ dx = 0 and F ⋅ dr = 2aydy and y varies from 0 to b

b
b
⎡ y2 ⎤
∴ ∫AB ⋅ = ∫0 2aydy = 2a ⎢⎣ 2 ⎥⎦ 0 = ab
2
F dr

On BC: y = b ∴ dy = 0 and F ⋅ dr = ( x 2 − b 2 )dx and x varies from a to 0

0
0
⎡ x3 2 ⎤ ⎛ a3 2 ⎞ a3
∴ ∫ ⋅ = ∫a − = − = − − = −
2 2
⎜ ⎟
2
F dr ( x b ) dx ⎢ b x ⎥ 0
⎝ 3
b a
⎠
ab
BC ⎣3 ⎦a 3
On CO: x = 0 ∴ dx = 0 and F ⋅ dr = 0
∴ ∫
CO
F ⋅ dr = 0

a3 a3
∴ ∫ F ⋅ dr =
C 3
+ ab 2 + ab 2 −
3
= 2ab 2 (2)

From (1) and (2), ∫ F ⋅ dr = S∫∫ Curl F ⋅n dS

C
Hence, Stoke’s theorem is veriﬁed.

Note Stoke’s theorem in the plane is Green’s theorem. This is indeed Green’s theorem veriﬁcation.

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 92 5/19/2016 12:19:39 PM

Vector Calculus ■ 9.93

EXAMPLE 10
Verify Stoke’s theorem for the vector ﬁeld F 5 (2 x 2 y ) i 2 yz 2 j 2 y 2 z k over the upper half
surface x2 1 y2 1 z2 5 1, bounded by its projections on the xy−plane.

Solution. z
Stoke’s theorem is

∫ F ⋅ dr = ∫∫S curl F ⋅ n dS
C
S

Given F = (2 x − y )i − yz 2 j − y 2 zk
O y
i j k C
∂ ∂ ∂
∴ Curl F =
∂x ∂y ∂z x

2 x − y − yz 2 − y 2 z Fig. 9.36
⎡∂ ∂ ⎤
= i ⎢ ( − y 2 z ) − ( − yz 2 ) ⎥
⎣ ∂ y ∂ z ⎦
⎡∂ ∂ ⎤ ⎡∂ ∂ ⎤
− j ⎢ ( − y 2 z ) − (2 x − y ) ⎥ + k ⎢ ( − yz 2 ) − (2 x − y ) ⎥
⎣ ∂x ∂z ⎦ ⎣ ∂ x ∂ y ⎦
= i[ − 2 yz + 2 yz ] − j[0 − 0] + k [0 − ( −1)] = k
∴ F ⋅n = k ⋅n
The surface is the upper hemisphere x2 + y2 + z2 = 1
dxdy
∫∫ Curl F ⋅n dS = S∫∫ k ⋅ n dS = R∫∫ k ⋅ n
S k ⋅n
,

where R is the projection of S on the xy−plane.

∴ R is the circle x2 + y2 = 1 in the xy−plane.
∴ ∫∫ Curl F ⋅n dS = R∫∫ dx dy
S
⇒ ∫∫ Curl F ⋅n dS = area of the circle = p ⋅ 1
S
2
=p (1)

Now C is the circle x2 + y2 = 1 in the z = 0 plane.

Parametric equations are x = cos u, y = sin u, 0 ≤ u ≤ 2p
∴ ∫ F ⋅ dr = C∫ [(2 x − y)dx − yz dy − y zdz ] = ∫ (2x − y )dx [{ z = 0]
2 2

C C
Now x = cos u ⇒ dx = −sin udu
2p
∴ ∫ F ⋅ dr = ∫ (2 cos u − sin u)( − sin u) du
C 0
2p

= ∫ ( −2 sin u cos u + sin u) d u

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 93 5/19/2016 12:19:42 PM

= ∫ ( −2 sin u cos u + sin u) d u

0
2p
⎡ 1 − cos 2u ⎤
= ∫ ⎢⎣ − sin 2u +
0
2 ⎥⎦ d u
2p
⎡ cos 2u 1 ⎛ sin 2u ⎞ ⎤
=⎢ + ⎜u − ⎟
⎣ 2 2 ⎝ 2 ⎠ ⎥⎦ 0

⇒ ∫ F ⋅ dr = 12 ⎡⎢⎣(cos 4p − cos 0) + 2p − sin24p − 0⎤⎥⎦ = 12 [1 − 1 + 2p] = p

C
(2)

From (1) and (2), ∫ F ⋅ dr = S∫∫ Curl F ⋅n dS

C
Hence, Stoke’s theorem is veriﬁed.

EXAMPLE 11
Verify Stoke’s theorem for F 5 ( x 2 1 y 2 ) i 2 2 xy j taken around the rectangle bounded by the
lines x 5 6a, y 5 0 and y 5 b.

Solution.
C y=b B
Stoke’s theorem is
∫ F ⋅ dr = S∫∫ Curl F ⋅ n dS
C x = −a x=a

Given F = ( x 2 + y 2 )i − 2 xy j O x
D A
(−a, 0) (a, 0)
i j k
Fig. 9.37
∂ ∂ ∂
∴ Curl F =
∂x ∂y ∂z
x2 + y2 −2 xy 0
= i[0 − 0] − j (0 − 0) + k ( −2 y − 2 y ) = −4 y k
Since S is the rectangular surface, n = k

∫∫ Curl F ⋅n dS = S∫∫ −4 y k ⋅ k dx dy
S
b
b a
⎡ y2 ⎤
= −4∫ ∫ y dx dy = −4 ⎢ ⎥ [ x ]− a = −2b 2 ⋅ 2a = −4 ab 2
a

0 −a ⎣ 2 ⎦0

∫∫ Curl F ⋅n dS = − 4ab
S
2 (1)

We shall now compute the line integral

C
∫ F ⋅ dr .
Now F ⋅ dr = [( x 2 + y 2 )i − 2 xy j ] ⋅ [dxi + dy j ] = ( x + y )dx − 2 xy dy
2 2

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 94 5/19/2016 12:19:46 PM

Vector Calculus ■ 9.95

∴ ∫ F ⋅ dr = AB∫ F ⋅ dr + BC∫ F ⋅ dr + CD∫ F ⋅ dr + DA∫ F ⋅ dr

On AB: x=a ∴ dx = 0 and F ⋅ dr = −2ay dy and y varies from 0 to b

b
b
⎡ y2 ⎤
∴ ∫ ⋅ = ∫0 − = − ⎢ ⎥ = − ab
2
F dr ( 2 a ) ydy 2 a
AB ⎣ 2 ⎦0

On BC: y=b ∴ dy = 0 and F ⋅ dr = ( x 2 + b 2 )dx and x varies from a to −a

−a −a
⎡ x3 2 ⎤
∴ ∫ F ⋅ dr = ∫a ( x + b ) dx = ⎢⎣ 3 + b x ⎥⎦ a
2 2

BC
1 −2 3
= ( − a 3 − a 3 ) + b 2 ( − a − a) = a − 2ab 2
3 3
On CD: x = −a ∴ dx = 0 and F ⋅ dr = 2aydy and y varies from b to 0
0
0
⎡ y2 ⎤
∴ ∫CD ⋅ = ∫b = ⎢ ⎥ = a(0 − b ) = − ab
2 2
F dr 2 ay dy 2 a
⎣ 2 ⎦b

On DA: y=0 ∴ dy = 0 and F ⋅ dr = x 2 dx and x varies from −a to a

a
a a
⎡ x3 ⎤ 2 3
∴ ∫ F ⋅ dr = ∫− a = ∫0 = ⎢ ⎥ = a
2 2
x dx 2 x dx 2
DA ⎣ 3 ⎦0 3
2 2
∴ ∫ F ⋅ dr = − ab − a3 − 2ab 2 − ab 2 + a3 = −4 ab 2
2

C 3 3
From (1) and (2), ∫ F ⋅ dr = S∫∫ Curl F ⋅n dS
C
Hence, Stoke’s theorem is veriﬁed.

EXAMPLE 12
Verify stokes theorem for F 5 y 2 zi 1 z 2 xj 1 x 2 yk , where S is the open surface of the cube
formed by the planes x 5 2a, x 5 a, y 5 2a, y 5 a, z 5 2a, z 5 a in which z 5 2a is cut open.

Solution.
H E
Stoke’s theorem is ∫ F . dr = ∫∫ curl F .n ds
S
G
F
Given F = y zi + z x j + x yk
2 2 2 z

y
o
l j k x
D
A
∴ Curl F = ∂ ∂ ∂
∂x ∂y ∂z B C

y z z2 x
2
x2 y Fig. 9.38

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 95 5/19/2016 12:19:50 PM

⎡ ∂ ∂ ⎤ ⎡∂ ∂ ⎤ ⎡∂ ∂ 2 ⎤
= i ⎢ ( x 2 y ) − ( z2 x ) ⎥ − j ⎢ (x2 y ) − ( y 2 z ) ⎥ + k ⎢ ( z 2 x ) − ( y z)⎥
⎣ ∂y ∂z ⎦ ⎣ ∂x ∂z ⎦ ⎣ ∂x ∂y ⎦
= ( x 2− 2zx) i + ( y2− 2xy ) j + (z 2 − 2yz) k
We shall now compute ∫∫ curl F. n dS
S
Given S is the open surface consisting of the ﬁve faces of the cube except face ABCD
∴ ∫∫ curl F . ndS = ∫∫ curl F. ndS + ∫∫ curl F. ndS + ∫∫ curl F. ndS + ∫∫ curl F. ndS + ∫∫ curl F. ndS
S S1 S2 S3 S4 S5

Face Equation Normal n Curl F.n dS

S1 = BCFG x=a i a − 2az
2
dy dz
S2 = ADEH x = −a −i −(a + 2az)
2
dy dz
S3 = CDEF y=a j a − 2ax
2
dz dx
S4 = ABGH y = −a −j −(a + 2ax)
2
dz dx
S5 = EFGH z=a k a2 − 2ay dx dy
a a

∫∫ curl F . ndS = ∫ ∫ (a − 2 az)dydz

S1 −a −a
a
⎡ a ⎤⎡ a ⎤ a ⎡ z2 ⎤
= ⎢ ∫ dy ⎥ ⎢ ∫ a 2 − 2az dydz ⎥ = [ y ]− a ⎢a 2 z − 2 a ⎥ = [ a + a ] ⎡⎣a 2 ( a + a ) −a (a 2 − a 2 )⎤⎦ = 4 a4
⎣−a ⎦ ⎣−a ⎦ ⎣ 2 ⎦−a
a a
⎡ a ⎤⎡ a ⎤
∫∫ curl F. ndS = ∫ ∫ − (a + 2az ) dydz = − ⎢ ∫ dy ⎥ ⎢ ∫ a 2 + 2az⎥ dz
2

S2 −a −a
a
⎣−a ⎦ ⎣−a ⎦
⎡ z2 ⎤
= − [ y ]− a ⎢a 2 z + 2 a ⎥ = − [ a + a ] ⎡⎣a 2 ( a + a ) + a (a 2 − a 2)⎤⎦ = − 4 a4
a

⎣ 2 ⎦−a
a a
Similarly, ∫∫ curl F. ndS = ∫ ∫ (a
2
− 2ax )dzdx = 4a 4
S3 −a −a

a a

∫∫ curl F. ndS = ∫ ∫ −( a + 2 ax) dzdx = −4a4

S4 −a −a

a a

∫∫ curl F. ndS = ∫ ∫ (a − 2ay ) dxdy = 4a

2 4
and
S5 −a −a

∴ ∫∫ curl F. n dS = 4a − 4a4 + 4 a4 − 4 a4 + 4 a4 = 4a4

4
(1)
S

We shall now compute the line integral over the simple closed curve C consisting of the edges AB, BC,
CD, DA. Here z = −a, dz = 0
∴ F. dr = y 2 zdx + z 2 xdy + x 2 ydz = −ay 2dx + a 2 xdy

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 96 5/19/2016 12:19:54 PM

Vector Calculus ■ 9.97

On AB: y = −a ∴dy = 0 A D
(−a, −a) (−a, a)
F. dr = − a3dx and x varies from –a to a.
a
∴ ∫ F. dr = ∫ − a dx = −a [ x ] = −a3 .2a = −2a4
3 3 a
−a
AB −a

On BC: x = a ∴dx = 0 , F. dr = a3dy and y varies from −a to a. B C

(a, −a) (a, a)
a
∴ ∫ ∫ a dy = a [ r]
a
F. d r = 3 3
−a
= a3 .2 a = 2 a4 Fig. 9.39
BC −a

On CD: y = a ∴ dy = 0, F. dr = − a3dx and x varies from a to −a

−a

∫ ∫ − a dx = −a [ x ]
−a
∴ F. dr = 3 3
a
= −a3 ( −2a) = 2a4
CD a

On DA: x = −a ∴dx = 0, F. dr = − a3dy and y varies from a to −a.

−a

∫ ∫ − a dy = −a3 [ y ]a = −a2 ( −2a) = 2a4

−a
∴ F. dr = 3
(2)
DA a

∫ F ⋅ d r = −2 a + 2 a4 + 2 a4 + 2 a4 = 4 a4
4
∴ (3)
c

From (1) and (2), we get ∫∫ curl F. nds = ∫ F. dr

S C
Hence, Stoke’s theorem is veriﬁed.

EXERCISE 9.4

∫∫ F ⋅ n dS, where F = 12 x yi − 3 yz j + 2 zk and S is the portion of the plane x + y + z = 1

2
1. Evaluate
S
included in the ﬁrst octant.

∫∫ F ⋅ n dS, where F = (2 x − 3z )i + 2 y j − 4 xzk , where S is the surface of the solid

2
2. Evaluate
S
bounded by the planes x = 0, y = 0, z = 0 and 2x + 2y + z = 4.
3. Evaluate ∫∫ F ⋅ n dS , where F = zi + x j − y 2 zk and S is the curved surface of the cylinder x2 + y2 = 1
S
included in the ﬁrst octant between the planes z = 0 and z = 2.
4. If F = xy 2 i − yz 2 j + zx 2 k , ﬁnd ∫∫ F ⋅ n dS over the sphere x
S
2
+ y2 + z2 = 1.

∫∫ F ⋅ n dS, where F = 4 xzi − y j + yzk and S is the surface of the cube bounded by
2
5. Evaluate
S
x = 0, x = 1, y = 0, y = 1, z = 0 and z = 1.

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 97 5/19/2016 12:20:00 PM

∫ (x + xy )dx + ( x 2 + y 2 ) dy, where C is the square formed by the lines y = ±1, x = ±1,
2
6. Evaluate
C
by Green’s theorem.
7. Using Green’s theorem evaluate ∫ ( x 2 + y )dx − xy 2 dy taken around the square whose vertices are
C
(0, 0), (1, 0), (1, 1) (0, 1)
8. Using Green’s theorem ﬁnd the value of ∫ ( xy − x 2 )dx + x 2 y dy along the closed curve C formed
C
by y = 0, x = 1 and y = x.

∫ (15 x − 4 y 2 )dx + ( 2 y − 3x ) dy , where C is the curve enclosing the

2
9. Verify Green theorem for
C
area bounded by y = x2, x = y2
10. Verify Green theorem in the plane for ∫ (3 x 2 − 8 y 3 )dx + ( 4 y − 6 xy ) dy, where C is the boundary
C
of the region defined by x = 0, y = 0, x + y = 1.
11. Using Green’s theorem find the area of x 2 / 3 + y 2 / 3 = a 2 / 3 .
1
[Hint: Area = ∫ ( xdy − ydx ), C is the boundary of the curve]
2C
12. Using Green’s theorem in xy plane find the area of the region in the xy plane bounded by
y3 = x2 and y = x.
∫ (2 x − y 2 )dx + ( x 2 + y 2 ) dy , where C is the boundary of the
2
13. Using Green’s theorem evaluate
C
area in the xy plane bounded by x−axis and the semi circle x2 + y2 =1 in the upper half of the plane.
14. Verify Gauss divergence theorem for F = x 2 i + zi + yzk taken over the cube bounded by x = 0,
x = 1, y = 0, y = 1, z = 0, z = 1.
15. Verify Gauss divergence theorem for F = ( x 3 − yz )i − 2 x 2 y j + 2k over the parallelopiped bounded
by the planes x = 0, x = 1, y = 0, y = 2, y = 2, z = 0, z = 3.
16. Verify Gauss divergence theorem for F = x 2 i + z j + yzk over a unit cube.
17. Verify Gauss divergence theorem for F = ( x 3 − yz )i − zx 2 y j + 2k over the cube x = 0, x = a, y = 0,
y = a, z = 0, z = a.
18. Verify the divergence theorem for F = 2 xyi + yz 2 j + xzk , where S is the rectangular
parallelopiped bounded by x = 0, y = 0, z = 0, x = 2, y = 1, z = 3.
19. Using divergence theorem show that
1
∫∫
S
x 2dy + y 2dzdy + 2z ( xy − x − y )dxdy = , where S is the surface of the cube
2
x = y = z = 0, y = z =1.

∫∫ (2 xyi + yz j + xzk ).dS , where S is the surface of the

2
20. Use divergence theorem to evaluate
S
region bounded by x = y = z = 0, y = 3, x + 2z = 6.
21. Prove that ∫∫ [x ( y − z )i + y ( z − x ) j + z ( x − y )k ] ⋅ dS = 0, where S is any closed surface.
S

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 98 5/19/2016 12:20:04 PM

Vector Calculus ■ 9.99

22. Verify Stoke’s theorem for F = 2 zi + x j + y 2 k , where S is the surface of the paraboloid z = 4 − x2 − y2
and C is the simple closed curve in the xy plane.
23. Verify Stoke’s theorem for F = yi + z j + xk , where S is the upper half surface of the sphere
x2 + y2 + z2 = 1 and C its boundary.
24. Verify Stoke’s theorem for F = ( x 2 − y 2 )i + 2 xy j + xyzk over the surface of the box bounded by
the planes x = 0, y = 0, x = a, y = b, z = c above the xy plane.
25. Verify Stoke’s theorem for F = ( x 2 − y 2 )i + 2 xy j in the rectangular region in the xy plane bounded
by x = 0, x = a, y = 0, y = b.
26. Verify Stoke’s theorem for F = − y 3 i + x 3 j and the closed curve C is the boundary of the ellipse
x2 y2
+ = 1.
a2 b2
27. If f is scalar point function, use Stoke’s theorem to prove curl (grad f) = 0.
28. Evaluate ∫∫ ∇ × F ⋅ n dS,
S
where S is the surface x2 + y2 + z2 = a2 above the xy−plane and

F = yi + ( x − 2 xz ) j − xyk .
29. Evaluate ∫ yzdx + zx dy + xy dz , where C is the curve x
C
2
+ y2 = 1, z = y2.

30. Evaluate ∫∫ ∇ × F ⋅ n dS for F = (2 x − y + z )i + ( x + y − z 2 ) j + (3x − 2 y + 4 z )k over the surface of

the cylinder x2 + y2 = 4, bounded by the plane z = 9 and open at the end z = 0.
31. Find the area of a circle of radius a using Green’s theorem.

∫ [(2xy − x )dx + ( x 2 + y 2 )dy ] where C is the closed curve of

2
32. Using Green’s theorem evaluate
C
the region bounded by y = x2 and y2 = x
33. Verify Green’s theorem in a plane for the integral
C
∫ ( x − 2 y)dx + xdy taken around the circle

x2 + y2 = 4.

∫ [(x − xy 3 )dx + ( y 2 − 2xy ) dy ] where C is the square

2
34. Verify Green’s theorem in the plane for
C
with vertices (0, 0) (2, 0), (2, 2), (2, 0).

∫∫∫ ∇ ⋅ F dV if F = x i + y j + z 2 k and V is the volume of the region enclosed by the

2 2
35. Evaluate
V
cube x = 0, x = 1, y = 0, y = 1, z = 0, z = 1.
36. If S is any closed surface enclosing volume V and F = axi + byj + czk prove that
∫∫ F ⋅ n dS = (a + b + c) V
S
37. Verify Gauss divergence theorem for F = ( x 2 − yz )i + ( y 2 − zx ) j + ( z 2 − xy )k taken over the
rectangular parallelopiped bounded by 0 ≤ x ≤ a, 0 ≤ y ≤ b, 0 ≤ z ≤ c.
38. Verify Stoke’s theorem for F = y 2 zi + z 2 x j + x 2 yk where S is the open surface of the cube formed
by the planes x = −a, x = a, y = −a, y = a, z = −a, z = a in which z = −a is cut open.

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 99 5/19/2016 12:20:09 PM

39. Evaluate ∫∫ Curl F ⋅ n dS, where F = ( y − z )i + yz j − xz k and S is the open surface bounded by
S
the planes x = 0, x = 1, y = 0, y = 1, z = 0, z = 1 above the xy plane.

ANSWERS TO EXERCISE 9.4

49 16 4 3 4
1. 2. 3. 3 4. p 5. 6. 0 7. −
120 3 3 2 3
1 3 3 1 4 351
8. − 11. pa 12. 13. 20. 28. 0 29. 0
12 8 10 3 2
30. 8p 31. pa2 32. 0 35. 3 36. (a + b + c) V 39. −1

SHORT ANSWER QUESTIONS

1. If r 5 xi 1 y j 1 z k and r 5 r, then ﬁnd ∇r.
2. Find grad f at the point (1, −2, −1), where f = 3x2y − y3z2.

3. What is the greatest rate of increase of f 5 xyz2 at the point (1, 0, 3)?
4. Find the unit normal vector to the surface x2 + xy + z2 = 4 at the point (1, −1, 2).
5. Find the directional derivative of f 5 xyz at (1, 1, 1) in the direction of i + j + k
6. The temperature at a point (x, y, z) in space is given by T (x, y, z) 5 x2 1 y2 2 z. A mosquito located at
the point (4, 4, 2) desires to fly in such a direction that it gets cooled faster. Find the direction in which
it should fly.
7. Find the normal derivative of f 5 x3 2 y3 1 z at the point (1, 1, 1).
8. Find the angle between the surfaces x2 1 y2 1 z2 5 9 and x2 1 y2 2 z 5 3 at the point (2, −1, 2).
9. Find the equation of the tangent plane to the surface x2 + y2 − z = 0 at the point (2, −1, 5).
10. If F 5 x 3 i 1 y 3 j 1 z 3 k , find div (curl F).
11. Prove that F = (2 x 2 y + yz ) i + ( xy 2 − xz 2 ) j − (6 xy + 2 x 2 y 2 ) k is solenoidal.
12. Find a such that (3 x − 2 y + z ) i + (4 x + ay − z ) j + ( x − y + 2 z ) k is solenoial.
13. If f is a scalar point function, prove that ∇f is solenoidal and irrotational if f is a solution of Laplace
equation.
14. Find the values of a, b, c if F 5 ( x 1 2 y 1 az ) i 1 (bx 2 3 y 2 z ) j 1 (4 x 1 cy 1 2 z ) k is irrotational.

15. If A and B are irrotational, prove that A 3 B is solenoidal.

16. Find the work done, when a force F 5 ( x 2 2 y 2 1 x ) i 2 (2 x 1 y ) j moves a particle from the origin
to the point (1, 1) along y2 5 x.
17. Show that F 5 x 2 i 1 y 2 j 1 z 2 k is a conservative vector ﬁeld.

∫ (x 2 xy )dx 1 ( x 2 1 y 2 )dy , where C is the square formed by the lines y 5 61, x 5 61

2
18. Evaluate
C
using Green’s theorem.

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 100 5/19/2016 12:20:17 PM

Vector Calculus ■ 9.101

19. Using Stoke’s theorem prove that curl (grad f) 5 0.

20. If S any closed surface show that ∫∫ curl F ⋅ n dS 5 0.

OBJECTIVE TYPE QUESTIONS.

A. Fill up the blanks
1. ∇ ⎛ 1 ⎞ = ________
⎜⎝ ⎟⎠
r
2. If f (x, y, z) = x2y + xy2 + z2, then ∇ f at (1, 1, 1) is =________.

3. The directional derivative of f = x3 + y3 + z3 at (1, −1, 2) in the direction of i + 2 j + k is = ________.

4. The unit normal to the surface xy2z3 = 1 at the point (1, 1, 1) is = ________.

5. The greatest rate of increase of f = xyz2 at the point (1, 0, 3) is = ________.

6. Equation of the normal to the surface x2 + y2 + z2 = 25 at the point (1, 0, 3) is = ________.

7. If F = ∇(x3 + y3 + z3 − 3xyz), then curl F = ________.

8. If F = (3x − 2 y + z )i + (4 x + ay − z ) j + ( x − y + 2z )k is solenoidal, then value of a is = ________.

9. If F = (2xy + z 3 )i + x 2 j + 3xz 2 k and the curve c is the line joining the points (1, −2, 1) and (3, 2, 4), then

∫ F ⋅ dr
C
= ________.

10. If F = x 2 i + y 2 j + z 2 k and V is the region bounded by the cube x = 0, x = 1, y = 0, y = 1, z = 0, z = 1,

then ∫∫∫ ∇ ⋅ Fdv = ________.
v

B. Choose the correct answer

1. If f = x2 + y2 + z2 − 8, then grad f at (2, 0, 2) is

(a) i + 4 k (b) i + j + k (c) 4i + k (d) 4i + 4 j + 4 k

⎛ r⎞
2. div ⎜ ⎟ is equal to
⎝ r⎠

(a) 1 (b) 2 (c) 3 (d) 4

r r r r

3. If r = x i + y j + z k , then curl r is equal to

(a) o (b) i (c) j (d) k

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 101 5/19/2016 12:20:23 PM

∂ 2f ∂ 2f
4. If f = x2 − y2, then ∇2f = + is equal to
∂x 2 ∂y 2

(a) 0 (b) 2 (c) −2 (d) 1

5. If ∇f = (6 xy + z 3 )i + (3 x 2 − z ) j + (3 xy 2 − y ) k , then f is equal to
(a) xz − yz + c (b) 3x2y + xz3 (c) xz3 − yz + c (d) 3x2y − yz + c

6. The unit normal at (1, 2, 5) on x2 + y2 = z is

1 1 1 1 1 i + 4 j − 5k 2i + 4 j − 5k
(a) i+ j− k (b) i− k (c) (d)
3 3 3 2 2 42 3 5
7. The equation of the tangent plane to the surface at (2, 0, 2) is

(a) x − y − z = 0 (b) 2x − z = 2 (c) 3x + y − 2z = 2 (d) None of these

8. If F = x 2 i + xy 2 j, then ∫ F ⋅ dr, where c is the segment on y = x from (0, 0) to (1, 1) is

7 7 7
(a) − (b) (c) (d) − 7
6 12 6 12

9. Find the work done when the force F = 5xy i + 2y j displaces a particle from the points corresponding to x
= 1 to x = 2 along y = x3

(a) 24 (b) 64 (c) −84 (d) 94

10. Using Green’s theorem in the plane, evaluate ∫ ( 2x − y )dx + ( x + y )dy , where c is the circle x2+ + y2 = 4 in the
plane c

(a) 2p (b) 4p (c) −4p (d) 8p

ANSWERS
A. Fill up the blanks
1. − r 2. 3i + 3 j + 2k 3. 7 6 4. i + 2 j + 3k 5. 9
r3 2 14
x−4 y z −3
6. = = 7. o 8. −5 9. 21 10. 3
4 0 3

B. Choose the correct answer

1. (c) 2. (c) 3. (a) 4. (a) 5. (b) 6. (d) 7. (b) 8. (b) 9. (d) 10. (d)

M09_ENGINEERING_MATHEMATICS-I _CH09_Part B.indd 102 5/19/2016 12:20:28 PM

1
Curves in Space
1. INTRODUCTION
Differential geometry is that branch of geometry which is treated using the methods of
calculus. In particular, we investigate curves and surfaces in space in differential geometry.
Differential geometry plays an important role in engineering designs, geodesy, geograph and
space travel. Formulae regarding vector algebra and vector calculus are frequently used in
the study of differential geometry.
2. BRANCHES OF DIFFERENTIAL GEOMETRY
There are two branches of differential geometry.
(i) Local Differential Geometry. In this branch of differential geometry, we study
the properties of curves and surfaces in space which depend only upon points close to a particular
point of the geometric figure under consideration.
(ii) Global Differential Geometry. In this branch of differential geometry, we study
the properties of curves and surfaces in space which involve the entire geometric figure under
consideration.
In the present course, we shall study some of the fundamentals of local differential
geometry.
3. FUNCTIONS OF CLASS Cm
A scalar valued (or vector valued) function defined on an interval I belongs to class Cm
on the interval I if the mth order derivative of the function exists and is continuous on I.
The class of continuous functions is denoted by C0. The class of functions having deriva-
tives of all orders is denoted by C∞.
If a function belongs to the class Cm, then that function is called a Cm function.
We know that a vector function is continuous or has a derivative if and only if all com-
ponents of the functions are continuous or have derivatives.
∴ A vector function f(t) = f1(t)i + f2(t)j + f3(t)k belongs to Cm on I if and only if its
components f1(t), f2(t) and f3(t) belong to Cm on I.
Remark 1. We know that a differential function is always continuous.
∴ If a function belongs to Cm, then it belongs to Ck for all k ≤ m.
Remark 2. In printing work, the vector quantity f is depicted by using bold letter. In writing
→
work, the vector f is written as f or f .

3
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

4 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

Example 1. Show that the vector function f(t) = (cos t)i + t3 j + t5/3 k, – ∞ < t < ∞ belongs
to C1 on – ∞ < t < ∞ and not C2 on – ∞ < t < ∞.
Sol. We have f(t) = (cos t)i + t3j + t5/3k
. 5 .FG df IJ
∴ f (t) = (– sin t)i + 3t2j + t2/3k
3
f=
H dt K
5
– sin t, 3t2, t 2/3 are continuous functions of t, where – ∞ < t < ∞.
3
.
∴ f (t) is continuous on – ∞ < t < ∞.
∴ f (t) belongs to C1 on – ∞ < t < ∞.
.. 10
Also, f (t) = (– cos t)i + 6tj + k
9t1/ 3
The function 10 is not continuous at t = 0.
9t1/3
∴ The scalar function t5/3 does not belong to C2 on – ∞ < t < ∞.
∴ f(t) does not belong to C2 on – ∞ < t < ∞.
Remark. f(t) belongs to Cm for all m ≥ 0 on any interval not containing ‘0’.

4. CURVE IN SPACE
A curve in space is defined as the locus of a point whose position vector relative to a
fixed origin may be expressed as a function of a single parameter.
Thus, a curve C in space may be represented by a vector function
r(t) = x(t)i + y(t)j + z(t)k, where t is a parameter. Here
r(t) is the position vector of the point P on the curve C and
x(t), y(t), z(t) are the cartesian coordinates of the point P. To
each value t′ of t there correspond a unique point of the
curve C with position vector r(t′) and cartesian coordinates
(x(t′), y(t′), z(t′)).
As t increases, the direction of travelling along the
curve C is called the positive sense on the curve C. Also
as t decreases, the direction of travelling along the curve C is called the negative sense on
the curve C.
If a curve in space lies wholly in a plane, then it is called a plane curve.
If a curve in space does not lie wholly in a plane then it is called a skew curve or a
tortous curve or a twisted curve.
Example 2. Show that the curve in space r(t) = a cos t i + b sin t j + 0k is a plane curve.
Sol. We have r(t) = a cos ti + b sin tj + 0k.
∴ Let (x, y, z) be the coordinates of the point with position vector r(t).
∴ x = a cos t, y = b sin t, z = 0

x2 y2
∴ + = 1, z = 0 (∵ cos2 t + sin2 t = 1)
a2 b2
This represents an ellipse in the xy-plane.
∴ The given curve is a plane curve.
CURVES IN SPACE 5

5. REGULAR CURVE
A curve r = r(t), a ≤ t ≤ b is called a regular curve if
.
(i) r (t) exists and is continuous on a ≤ t ≤ b i.e., r(t) is of class C1 on a ≤ t ≤ b.
.
(ii) r (t) ≠ 0 for all t in a ≤ t ≤ b.
For example, consider the curve
r = r(t) = 3ti + t4j + 2k, – ∞ < t < ∞.
.
Here r (t) = 3i + 4t3j + 0k
.
r (t) is continuous on – ∞ < t < ∞ and also non-zero.
∴ The given curve is a regular curve.
dx dy dz
Remark. If r = r(t) = x(t)i + y(t)j + z(t)k is a regular curve then, , , are never zero
dt dt dt
simultaneously.

6. SIMPLE CURVE
A curve r = r(t), a ≤ t ≤ b is called a simple curve if
(i) r = r(t), a ≤ t ≤ b is a regular curve.
(ii) t1 ≠ t2 ⇒ r(t1) ≠ r(t2) i.e., the curve is without points at
which the curve intersects or touches itself.
Remark. A point where a curve intersects or touches itself is called
a multiple point.

EXERCISE 1.1
1. Show that the function f(t) = t2 + t5/2, belongs to:
(i) C2 on (– ∞, ∞) (ii) C3 on (1, 4).
2. Show that the function f(t) = 3t i + 6t j + k belongs to C∞ on (– ∞, ∞).
4 9

3. If the vector functions f and g belong to Cm on I, then show that the vector functions f + g , f · g ,
f × g also belong to Cm on I.
4. If a and b are constant vectors, then show that the curve in space r(t) = a + tb is a plane curve.
5. Show that the curve in space r(t) = 4 sin ti + 0j + 3 cos tk, – ∞ < t < ∞ is a plane curve.
6. Show that the curve in space r(t) = 2t2i + (1 + t3)j + 7tk, – ∞ < t < ∞ is a regular curve.

7. ARC OF A CURVE
An arc of a curve is the portion of the curve between any two points of the curve.
For simplicity, we shall say ‘curve’ for curves as well as for arcs.
8. LENGTH OF A CURVE r(t5)

The length of a curve is defined in r(b) = r(tn)

r = r(t)
terms of the lengths of approximating P
r(t4)
polygonal arcs. Let r = r(t), a ≤ t ≤ b be the r(t1)
given curve.
r(t2)
Let a = t0 < t1 < ..... < tn = b r(t3) n=6
be a subdivision of the interval a ≤ t ≤ b. This r(a) = r(t0)
subdivision determines a sequence of points
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

6 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

r0 = r(t0), r1 = r(t1), ....., rn = r(tn).

These points are joined in sequence to form an approximating polygonal arc P as shown
in the figure.
n n
∴ Length of P= ∑
i=1
ri − ri − 1 = ∑
i=1
r (ti ) − r (ti − 1 )

We make subdivisions of the interval arbitrarily small so that the greatest | ti – ti–1 |
approaches 0 as n → ∞.
n
If lim
n→ ∞
∑
i=1
r (ti ) − r (ti − 1 ) exists finitely, then the given curve is said to a rectifiable

curve and the value of this limit is called the length of the given curve.
Theorem. If r = r(t), a ≤ t ≤ b be a regular curve then prove that this curve is rectifiable

and its length is given by the integral

Note. The proof of this theorem is beyond the scope of this book.
za
b .
|r (t)| dt .

WORKING RULES FOR FINDING LENGTH OF THE CURVE

r = r(t) BETWEEN a ≤ t ≤ b
. .
Step I. Find r and then |r|.
Step II. Evaluate
z a
b .
|r| dt . This gives the required length of the curve.

Example 1. Find the length of the helix r = (a cos t)i + (a sin t)j + btk, 0 ≤ t ≤ 2π.
Sol. We have r = (a cos t)i + (a sin t)j + btk, 0 ≤ t ≤ 2π.
.
∴ r = (– a sin t)i + (a cos t)j + bk
.
∴ |r| = a 2 sin 2 t + a 2 cos 2 t + b 2 = a 2 + b 2

∴ Length of the helix = z 2π

0
.
|r| dt

z
2π
2π
= a 2 + b 2 dt = a 2 + b 2 t = 2π a 2 + b2 .
0
0

Example 2. Find the length of the curve r = (4 cosh 2t)i + (4 sinh 2t)j + 8tk, 0 ≤ t ≤ π.
Sol. We have r = (4 cosh 2t)i + (4 sinh 2t)j + 8tk, 0 ≤ t ≤ π.
.
∴ r = (8 sinh 2t)i + (8 cosh 2t)j + 8k
.
∴ |r| = 64 sinh 2 2t + 64 cosh 2 2t + 64

= 8 2 cosh 2 2t = 8 2 cosh 2t
CURVES IN SPACE 7

∴ Length of the curve = z

π .
|r| dt

z
0
π
= 8 2 cosh 2t dt
0
π
= 4 2 sinh 2t = 4 2 sinh 2π.
0
Example 3. Find the length of the semicubical parabola r = ti + t3/2 j from (0, 0, 0)
to (4, 8, 0).
Sol. We have r = ti + t3/2 j.
The coordinates of the point with position vector r are (t, t3/2, 0).
t = 0, t3/2 = 0 ⇒ t = 0 and t = 4, t3/2 = 8 ⇒ t = 4.
∴ The given points correspond to the values 0 and 4 of t.
. 3 1/2
r=i+ t j
2
. 9 1
∴ |r|= 1 + t = 4 + 9t
4 2
∴ Length of the given curve

= z
0
4 .
|r| dt = z 0
4 1
2
4 + 9t dt

4
1 ( 4 + 9t )3/ 2 1
= . = [( 40) 3/ 2 − ( 4 ) 3/ 2 ]
2 (3/ 2) 9 27
0

1 8
= [80 10 − 8] = [ 1000 − 1] = 9.073.
27 27
Example 4. Show that the length of the curve x = 2a(sin–1 t + t 1 − t 2 ), y = 2at2,
z = 4at between the points t = t1 and t = t2 is 4 2 a(t2 – t1).

Sol. We have x = 2a(sin–1 t + t 1 − t 2 ), y = 2at2, z = 4at.

Let r be the position vector of the point (x, y, z) on the given curve.
∴ r = xi + yj + zk = 2a(sin–1 t + t 1 − t 2 )i + 2at2j + 4atk
F 1 I
∴
.
r = 2a GG 1
+ t . (1 − t ) ( − 2t ) + 1 − t . 1J i + 4atj + 4ak
JK
2 −1/ 2 2

H 1−t 2 2

F 1 I
= 2a G + 1 − t J i + 4atj + 4ak
2
t
GH 1 − t 1 − t JK
2
−
2 2

F I
= 2a H 1 − t + 1 − t K i + 4atj + 4ak
2 2

= 4a 1 − t 2 i + 4atj + 4ak
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

8 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

.
∴ |r| = 16 a 2 (1 − t 2 ) + 16 a 2 t 2 + 16 a 2

= 4a (1 − t 2 + t 2 + 1) = 4 2 a
∴ Length of the given curve

=
z
t1
t2 .
|r| dt =

t2
zt1
t2
4 2 a dt

= 4 2 at = 4 2 a(t2 − t1 ) .
t1
Example 5. Find the arc length as a function of θ along the epicycloid:
FG a + b θIJ , y = (a + b) sin θ – b sin FG a + b θIJ , z = 0.
x = (a + b) cos θ – b cos
H b K H b K
Sol. The given epicycloid is
FG a + b θIJ , y = (a + b) sin θ – b sin FG a + b θIJ , z = 0.
x = (a + b) cos θ – b cos
H b K H b K
x = – (a + b) sin θ + (a + b) sin FG a + b θIJ ,
.
∴
H b K
y = (a + b) cos θ – (a + b) cos FG
.
a+b I .

H b θJK and z = 0
Let r be the position vector of the point (x, y, z) on the epicycloid.
. . . .
∴ r = xi + yj + zk ∴ r = x i + y j + z k
.
|r|2 = x 2 + y 2 + z 2 = (a + b)2 − sin θ + sin
FG
FG a + b θIJ IJ 2
∴
H b KK H
F
+ (a + b) G cos θ − cos FG
a + b II
θJ J + 0
2

H KK
2
H
2
b
L
= (a + b) M2 − 2 sin θ sin
2 a +b
θ − 2 cos θ cos
a +b O
θP
N b b Q
L F a + b θ − θIJ OP = (a + b) LM2 – 2 cosFG a θIJ OP
= (a + b) M2 − 2 cos G
N H b 2
KQ N H b KQ
2

= 4(a + b) sin G
F a θIJ
H 2b K
2 2

F a θIJ
| r | = 2(a + b) sin G
.
∴
H 2b K
2( a + b) cos G
F a θIJ θ

∴ z
s = |r | dθ = 2(a + b) sin G
0
θ . F a I
z
H 2b θJK dθ = –
0
θ

a / 2b
H 2b K
0
4(a + b) b a LM FG
4(a + b)b IJ OP
a LM FG IJ OP .
= −
a
cos
2b
θ −1 =
N H
a
1 − cos
K Q
2b
θ
N H KQ
CURVES IN SPACE 9

x2 y2
Example 6. Find the length of the curve given by the intersection of the surfaces − = 1,
a2 b2
z
x = a cosh from the point (a, 0, 0) to the point (x, y, z).
a

x2 y2
Sol. We have 2
− =1 ...(1)
a b2
z
and x = a cosh ...(2)
a
Let x = a cosh t, y = b sinh t, z = at
∴ (1) and (2) are satisfied.
Let r be the position vector of the point (x, y, z) on the given curve.
∴ r = (a cosh t)i + (b sinh t)j + atk
a cosh t = a, b sinh t = 0, at = 0 ⇒ t = 0
∴ The initial point corresponds to t = 0.
.
r = (a sinh t)i + (b cosh t)j + ak
.
∴ |r| = a2 sinh 2 t + b2 cosh 2 t + a 2
= a 2 (sinh 2 t + 1) + b 2 cosh 2 t

a 2 cosh 2 t + b2 cosh 2 t =
= ( a 2 + b2 ) cosh t
∴ Length of the given curve

= z0
t .
|r|dt = z0
t
a 2 + b2 cosh t dt
t
= a 2 + b 2 sinh t = a 2 + b 2 sinh t
0

a 2 + b2
= y.
b

EXERCISE 1.2
1. Find the length of the helix r = (3 cos t)i + (3 sin t)j + 4tk, 0 ≤ t ≤ 2π.
2. (i) Find the length of one complete turn of the helix :
r = (a cos t, a sin t, bt), – ∞ < t < ∞, a > 0, b > 0.
(ii) Find the length of the helix r = a cos ui + a sin uj + cuk, – ∞ < u < ∞ from the point (a, 0, 0) to
the point (a, 0, 2πc).
3. Find the length of the curve r = (3 cosh 2t)i + (3 sinh 2t)j + 6tk, 0 ≤ t ≤ π.
4. Find the length of the catenary r = ti + cosh tj from t = 0 to t = 1.
5. Find the length of the curve r = (1 + 2t)i + (2 + t)j – k, 3 ≤ t ≤ 7.
6. Find the length of the curve r = (2 + 9t)i + (1 – 3t)j + tk, 8 ≤ t ≤ 15. Also, verify the result by using
the distance formula to find the distance between two given points.
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

10 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

7. Find the length of the semicubical parabola r = ti + t3/2k from (0, 0, 0) to (9, 0, 27).

8. Find the length of the curve r = (sin–1 t + t 1 − t2 )i + t2j + 2tk from t = 1 to t = 3.

9. Find the length of the curve given by the intersection of the surfaces x2 – y2 = 1, x = cosh z from
the point (1, 0, 0) to the point (x, y, z).
10. Find the length of the curve r = et cos ti + et sin tj + etk, 0 ≤ t ≤ π.

Answers

1. 10 π 2. (i) 2π a2 + b2 (ii) 2π a2 + c2 3. 3 2 sinh 2π

4. sinh 1 5. 4 5 6. 7 91 7. 28.73
8. 4 2 9. 2 y 10. 3 (eπ
– 1).
Hint
2. (i) The limits in the definite integral are to be t0 and t0 + 2π, where t0 is any number.

9. ARC LENGTH AS PARAMETER IN REPRESENTATIONS OF CURVES

Let r = r(t) be any regular curve. Let A(t = a) be any arbitrary but fixed point on the
curve. We define a function s of t as

s = s(t) = z
a
t .
|r| dt ...(1)

s(t) is called the arc length function of the curve r = r(t). If t0 > a, then s(t0) is the
length of curve between the points with parametric values a and t0. If t0 < a, then s(t0) is
negative of the length of the curve between the points with parametric values a and t0. Thus
s(a) = 0 and for points on one side of A the value of s will be positive ; for points on other side,
negative. The choice of the fixed point A(t = a) is arbitrary. Changing point A shall mean
changing s by a constant quantity.
For simplicity, the arc length function s is written arc length. The use of arc length s
as parameter in space curves would help us a lot in studying their curvature and torsion.
By the fundamental theorem of calculus, (1) implies
ds . dr ...(2)
= |r | =
dt dt

∴
dr dr dt dr dt
= . =
ds dt ds dt ds

dr ds dr dr
= = =1 (Using (2))
dt dt dt dt
dr
∴ = 1.
ds
If the equation of a curve is given in terms of arc length, then we say that the equation
of the curve is a natural representation of the curve.
If the parameter in the equation of a curve is other than, ‘arc length s′, then the equa-
tion of the curve is called an arbitrary representation of the curve.
CURVES IN SPACE 11

In general, the geometric quantities along a curve are defined in terms of a natural
ds dr
representation of the curve. By using the chain rule and the relation = , these quanti-
dt dt
ties can also be derived in terms of any arbitrary parameter.

WORKING RULES FOR WRITING r = r(t) IN TERMS OF s

.
Step I. Find |r|.

Step II. Solve s = z

0
t .
|r| dt to find s in terms of t.
Step III. Using the relation found in Step II, find t in terms of s.
Step IV. Substitute the value of t in r = r(t) to get the required natural representation of
the given curve.

Example 1. Find the equation of the helix r = a cos ti + a sin tj + ctk, – ∞ < t < ∞ in terms
of arc length s as parameter.
Sol. We have r = a cos ti + a sin tj + ctk , – ∞ < t < ∞ ...(1)
.
∴ r = – a sin ti + a cos tj + ck
.
∴ |r| = a 2 sin 2 t + a 2 cos 2 t + c 2 = a 2 + c 2
Let the point with t = 0 be the fixed point for the arc length parameter s.

∴ s = s(t) =
z t .
|r| dt

z
0
t
= a 2 + c 2 dt = a 2 + c 2 t
0
s
∴ t=
a + c2 2

Substituting the value of t in (1), the equation of the given helix in terms of arc length s
as parameter is
s s cs
r(s) = a cos i + a sin j+ k.
a 2 + c2 a2 + c2 a + c2
2

Example 2. Express the curve r = et cos ti + et sin tj + etk, – ∞ < t < ∞ in terms of arc
length s as parameter.
Sol. We have r = et cos t i + et sin tj + etk. ...(1)
.
∴ r = (et cos t – et sin t)i + (et sin t + et cos t)j + etk
= et(cos t – sin t)i + et (sin t + cos t)j + etk
.
∴ |r |= e 2t (cos t − sin t ) 2 + e 2t (sin t + cos t ) 2 + e 2t

t 2 2 2 2
= e cos t + sin t − 2 cos t sin t + sin t + cos t + 2 sin t cos t + 1

= et 3
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

12 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

Let the point with t = 0 be the fixed point for the arc length parameter s.

∴ s = s(t) =
z t

0
|r|dt
.

= z
0
t
e t 3 dt = 3e t
t

0
= 3 (et – 1)

et – 1 =
s
⇒ et =
s FG s IJ
⇒
3 3
+ 1 ⇒ t = log
H 3
+1
K
Substituting the value of t in (1), the equation of the given curve in terms of arc length
s as parameter is
FG s IJ F F IJ I i + FG s IJ F F s IJ I j + FG s IJ
r(s) =
H 3 K
+ 1 cos log
GH GH s
3
+1
K JK H 3 K
+ 1 sin log GH GH 3
+1
K JK H 3 K
+ 1 k.

EXERCISE 1.3

1. Find the equation of the helix r = cos ti + sin tj + tk, – ∞ < t < ∞ in terms of arc length s as
parameter.
2. Find the equation of the curve r = e2t cos ti + e2t sin tj + e2tk, – ∞ < t < ∞ in terms of arc length s
as parameter.
3. For the helix x = a cos t, y = a sin t, z = at tan α, show that the length of the curve measured
ds
from the point t = 0 is at sec α. Also show that = a sec α.
dt
1 1 1
4. Show that r = ( s + s2 + 1 ) i + j+ log ( s + s2 + 1 ) k is a natural representa-
2 2
2( s + s + 1 ) 2
tion of a curve.

Answers
s s s
1. r = cos i + sin j+ k
2 2 2

FG 2s + 1IJ LMcos FG 1 log FG 2s + 1IJ IJ i + sin FG 1 log FG 2s + 1IJ IJ j + kOP .

2. r=
H 3 K MN H 2 H 3 K K H 2 H 3 K K PQ
Hint

1 1 1
4. u=s+ s2 + 1 implies r = ui + j+ (log u ) k
2 2u 2

1 1 F 1 I FG1 + I
J
dr dr du i− j+ GH kJK G s
+ 1 JK
∴ = . =
ds du ds 2 2u 2 2u
H s2
CURVES IN SPACE 13

F s2 + 1 + s I
∴
dr
=
1
+
1
+
1
. GG JJ
ds 4 4u 4 2u 2 H s2 + 1 K
FG 1 + 1 IJ u u2 + 1
=
H 2 2u K 2
s2 + 1
=
2u s 2 + 1
= 1.

10. TANGENT TO A CURVE

Let C be a curve and P be any point on C. The tP
nta L
tangent at P to the curve C is the limiting position of a nge
straight line L through P and a point Q of C as Q approaches Ta
C
P along C. Q

11. UNIT TANGENT VECTOR

P
Let r = r(t) be the equation of a regular curve C in
terms of an arbitrary parameter t. Let P and Q be the points
on the curve whose position vectors are r and r + δr
corresponding to the values t and t + δt of the parameter
respectively.
→ → →
∴ PQ = OQ − OP = (r + δr) – r = δr
δr
∴ The quotient is a vector parallel to the line
δt
PQ. Since the given curve is regular, r(t) has continuous
non-zero derivative.

∴
δr dr exists and is non-
lim =
δt → 0 δt dt
zero.
dr .
By the definition of a tangent to a curve at a point, the vector i.e., r (t) is parallel to
dt
the tangent at the point P.
.
The vector r (t) is called the tangent vector of C at the point P.

1 .
The corresponding unit vector .
r is called the unit tangent vector of C at the point
|r|
.
P and it is denoted by t. The vectors r and t point in the direction of increasing t. Thus the
.
directions of r and t are same and depend upon the orientation of the curve C.
In particular, if the equation of the curve C is given in terms of the arc length s, then the

tangent vector of C at P is dr .
ds
dr dr
The vector is denoted by r′. We know that i.e., r′ is a unit vector.
ds ds
∴ Unit tangent vector ‘t’ at P = r′.
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

14 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

Example 1. Find the unit tangent vector t and direction cosines of the tangent at a point
on the circular helix x = a cos t, y = a sin t, z = bt, – ∞ < t < ∞.
Sol. The given helix is
x = a cos t, y = a sin t, z = bt, – ∞ < t < ∞.
Let r be the position vector of the point (x, y, z) on the helix.
∴ r = a cos ti + a sin tj + btk

1 .
Unit tangent vector, t= r
.
|r|
.
We have r = – a sin ti + a cos tj + bk
.
∴ |r | = a 2 sin 2 t + a 2 cos 2 t + b 2 = a 2 + b 2

1 . 1
∴ t= r= (– a sin ti + a cos tj + bk)
.
|r| a + b2
2

a a b
=– sin ti + cos tj + k
2 2
a +b 2 2 a +b a + b2
2

Since tangent is parallel to t and t is a unit vector, the d.c.’s of the tangent are
a a b
– sin t, cos t, .
2 2 2 2
a +b a +b a + b2
2

Example 2. Show that the tangent vectors along the curve r = ati + bt2j + t3k, where
2b2 = 3a make a constant angle with the vector i + k.
Sol. Given curve is r = ati + bt2 j + t3k.
.
∴ r = ai + 2btj + 3t2k
.
The tangent vector at point ‘t’ is r i.e., ai + 2btj + 3t2k.
Given vector is i + k, i.e., 1.i + 0.j + 1.k
.
Let θ be the angle between the tangent vector r and the vector i + k.

a(1) + 2bt(0) + 3t 2 (1)

∴ cos θ =
a2 + 4b2 t 2 + 9t 4 1+ 0 + 1

a + 3t 2 a + 3t 2 1
= = =
2
a + 2( 3a )t + 9t 2 4
2 ( a + 3t 2 ) 2 2

∴ θ = π/4, which is a constant angle.

∴ The result holds.
CURVES IN SPACE 15

12. EQUATION OF THE TANGENT AT A POINT ON A CURVE

Let r = r(t) be the equation of a regular curve C
in terms of an arbitrary parameter t. Let P(r) be any
.
point on the curve. We know that r is the tangent
vector at the point P and the tangent at P is parallel
to this vector. Let Q be a general point on the tangent
at P. Let R be the position vector of the point Q.
∴ The equation of the tangent at the point P(r)
.
is R = r + λ r , where λ is a scalar parameter.
Let the coordinates of P and Q be (x, y, z) and
(X, Y, Z) respectively.
∴ r = xi + yj + zk and R = Xi + Yj + Zk
. . . .
Also r = xi + y j + zk
∴ The equation of the tangent at P is
. . .
Xi + Yj + Zk = xi + yj + zk + λ( x i + y j + z k )
. . .
⇒ X = x + λ x , Y = y + λ y, Z = z + λ z
X−x Y−y Z−z
∴ .
= .
= . (= λ).
x y z
. . .
These are the cartesian equations of the tangent at the point P(x, y, z). Here x, y, z are
direction ratios of the tangent at the point P.
In particular, if the equation of the curve C is given in terms of the arc length s then, r′
is the unit tangent vector of C at P.
∴ The tangent at P is parallel to the vector r′.
∴ The equation of the tangent at P(r) is R = r + λr′, where R is the position vector of
the general point Q(R) on the tangent at P.
The cartesian form of the equations of tangent at P are
X−x Y−y Z−z
= = (= λ)
x′ y′ z′
Since r′ = x′i + y′j + z′k is a unit vector, x′, y′, z′ are the direction cosines of the tangent
at P.
Example 3. Show that a curve is a straight line if all tangent lines are parallel.
Sol. Let r = r(s) be the given curve.
∴ Tangent vector = r′
Since tangent line at a point is parallel to the tangent vector at that point and all
tangent lines are parallel, the direction of r′ is fixed. Also r′ is a unit vector.
∴ r′ is a non-zero constant vector, say a.
dr
∴ =a
ds
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

16 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

Integrating, we get
r = as + b, where b is a constant vector.
∴ The curve is a straight line passing through the point with position vector b and
parallel to the vector a.
1 2
Example 4. Find the equation of the tangent to the ellipse x + y2 = 1 at the point
4
FG 1IJ .
H 2,
2K
Sol. The given ellipse is
1 2
x + y2 = 1.
4
The parametric equations of this ellipse are
x = 2 cos t, y = sin t, z = 0
Let r be the position vector of the point (x, y, z).
∴ r = xi + yj + zk = 2 cos ti + sin tj + 0k
1 π
2 cos t = 2 , sin t = ⇒ t=
2 4

∴ The point
FG 1 IJ in xy-plane corresponds to t = π .
H 2,
2K
.
4

r = – 2 sin ti + cos tj + 0k
π . π π
At t = , r = – 2 sin i + cos j + 0k
4 4 4
1
=– 2 i+ j + 0k
2
FG
IJ passes through FG 2 , 1 , 0IJ and has d.r.’s –
1 1 , 0.
∴ The tangent at
H
2K
2,
H 2 K 2,
2
F 1 , 0IJ are
The equations of the tangent at G 2 ,
∴
H 2 K
1
y−
x− 2 z−0
2 x− 2 2y − 1 z.
= = or = =
− 2 1 0 − 2 1 0
2
Example 5. Find the equation of the tangent to the curve x = 1 + t, y = – t2, z = 1 + t2,
– ∞ < t < ∞ at the point for which t = 2.
Sol. The given curve is
x = 1 + t, y = – t2, z = 1 + t2, – ∞ < t < ∞.
t = 2 ⇒ x = 1 + 2 = 3, y = – (2)2 = – 4, z = 1 + (2)2 = 5
∴ t = 2 corresponds to the point (3, – 4, 5) on the curve.
. . .
Also, x = 1, y = – 2t, z = 2t
CURVES IN SPACE 17

. . .
t = 2 ⇒ x = 1, y = – 2(2) = – 4, z = 2(2) = 4
∴ The tangent at (3, – 4, 5) passes through (3, – 4, 5) and has d.r.’s 1, – 4, 4.
∴ The equations of the tangent at (3, – 4, 5) are
x − 3 y − (− 4) z − 5 y +4 z−5
= = or x – 3 = = .
1 −4 4 −4 4
Example 6. Show that the equation of the tangent at any point on the curve whose
equation referred to rectangular axes are x = 3t, y = 3t2, z = 2t3 makes a constant angle with the
line y = z – x = 0.
Sol. Given curve is x = 3t , y = 3t2, z = 2t3.
Let r be the position vector of the point (x, y, z) on the given curve.
∴ r = 3ti + 3t2j + 2t3k
.
∴ r = 3i + 6tj + 6t2k
.
The tangent line at the point with parametric value t is parallel to the vector r .
∴ D.r.’s of the tangent are 3, 6t, 6t2..
Given line is y = z–x = 0
x y z
⇒ = =
1 0 1
∴ D.r.’s of the given line are 1, 0, 1.
Let θ be the angle between the tangent and the given line.
3(1) + 6t(0) + 6t 2 (1)
∴ cos θ =
(3) 2 + (6t) 2 + (6t 2 ) 2 12 + 0 2 + 12
3 + 6t 2 1
= =
(3 + 6t 2 ) 2 2 2
∴ θ = π/4, which is a constant angle.
∴ The result holds.
13. DIRECTION RATIOS OF THE TANGENT AT A POINT ON THE CURVE OF
INTERSECTION OF TWO SURFACES
Let the given curve be the intersection of the surfaces
F(x, y, z) = 0 ...(1)
and G(x, y, z) = 0 ...(2)
Eliminating x, y, z from (1) and (2), let the equation of the given curve be
r = r(t) = x(t)i + y(t)j + z(t)k,
where t is an arbitrary parameter.
. . .
∴ D.r.’s of the tangent to the curve at the point ‘t’ are x, y, z .
Differentiating (1) and (2) w.r.t. t, we get
∂F dx ∂F dy ∂F dz
. + . + . =0
∂x dt ∂y dt ∂z dt
∂G dx ∂G dy ∂G dz
and . + . + . =0
∂x dt ∂y dt ∂z dt
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

18 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

. . .
⇒ (Fx) x + (Fy) y + (Fz) z = 0
. . .
and (Gx) x + (Gy) y + (Gz) z = 0
. . .
Solving these equations for x, y and z , we get
. . .
x y z
= =
Fy G z − Fz G y Fz G x − Fx G z Fx G y − Fy G x
∴ FyGz – FzGy , FzGx – FxGz , FxGy – FyGx
are also d.r.’s of the tangent to the curve given by the equations F(x, y, z) = 0, G(x, y, z) = 0 at
the point t.
Example 7. Show that the equation of the tangent to the curve of intersection of the
x2 y2 z2 x2 y2 z2
ellipsoid + + = 1 and the confocal + + = 1 is
a2 b2 c2 a2 − λ b2 − λ c2 − λ

x( X − x ) y(Y − y) z(Z − z)
= 2 2 2 2
= ,
2 2
a (b − c )(a − λ )2 2 b (c − a ) (b − λ ) c (a − b 2 ) (c 2 − λ )
2 2

where (x, y, z) is an arbitrary point on the curve.

Sol. The given surfaces are
x2 y2 z2
+ + –1=0 ...(1)
a2 b2 c2
x2 y2 z2
and + –1=0 + ...(2)
a 2 − λ b2 − λ c 2 − λ
Let the equation of the curve of intersection of given surfaces be r = r(t), where t is an
arbitrary parameter.
Differentiating the equations (1) and (2) w.r.t. t, we get
2x . 2y . 2z .
2
x+ 2
y+ 2
z=0
a b c
2x . 2y . 2z .
and x+ 2 y+ 2 z =0
a2 − λ b −λ c −λ
. . .
xx y y zz
⇒ + 2 + 2 =0
a2 b c
. . .
xx yy zz
and + + =0
a2 − λ b2 − λ c2 − λ
. . .
Solving these equations for x, y and z , we get
. . .
x y z
= =
yz yz zx zx xy xy
2 2
− 2 2 2 2
− 2 2 2 2
−
b (c − λ ) c (b − λ ) c (a − λ) a (c − λ) a (b − λ ) b (a 2 − λ )
2
CURVES IN SPACE 19

. . .
x y z
⇒ = =
λyz(b2 − c 2 ) λzx (c 2 − a 2 ) λxy(a2 − b 2 )
b 2 c 2 ( c 2 − λ ) (b 2 − λ ) c 2 a 2 (a 2 − λ ) (c 2 − λ ) a2 b2 (b 2 − λ ) (a 2 − λ )
. . .
x z y
⇒ = 2 2 = 2 2
a (b − c ) ( a − λ ) b (c − a ) (b − λ ) c ( a − b 2 ) (c 2 − λ )
2 2 2 2 2 2

x y z
∴ D.r.’s of the tangent at the point (x, y, z) on the curve are
a 2 (b2 − c 2 ) (a 2 − λ) b2 (c 2 − a 2 ) (b2 − λ) c 2 (a2 − b2 ) (c 2 − λ)
, , .
x y z
∴ The equations of the tangent at the point (x, y, z) on the curve are
X−x Y−y Z−z
2 2 2 2
= 2 2 2 2
=
a (b − c ) ( a − λ ) b ( c − a ) (b − λ ) c ( a − b2 ) (c 2 − λ )
2 2

x y z
or x(X − x ) y(Y − y ) z(Z − z ) .
2 2 2 2
= 2 2 2 2
=
a (b − c ) ( a − λ ) b ( c − a ) (b − λ ) c ( a − b2 ) (c 2 − λ )
2 2

14. NORMAL PLANE TO A CURVE

Let C be a curve and P be any point on C. The normal

t
gen
plane at P to the curve C is the plane passing through P and C

Tan
perpendicular to the tangent at P.

P
Normal
plane

15. EQUATION OF THE NORMAL PLANE AT A POINT ON A CURVE

Let r = r (t) be the equation of a regular curve C, where
t is an arbitrary parameter. Let P(r) be any point on the curve.
We know that the tangent at P is parallel to the tangent
t
en

C
ng

.
vector r .
Ta

Let Q be a general point on the normal plane at P. Let

R be the position vector of the point Q.
Q(R)
→ .
∴ PQ and r are perpendicular. P(r) Normal
plane
→ . .
⇒ PQ . r = 0 ⇒ (R – r) . r = 0 ...(1)
This represents the equation of the normal plane at
the point P(r).
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

20 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

.
r
(1) ⇒ (R – r) . = 0 ⇒ (R – r) . t = 0
.
|r|
∴ The equation of the normal plane at the point P(r) can also be written as
(R – r) . t = 0.
Example 8. Find the equation of the normal plane to the curve r = ti + t2j + t3k at t = 1.
Sol. The given curve is
r = ti + t2j + t3k.
∴ r(1) = 1.i + (1)2 j + (1)3 k = i + j + k
∴ t = 1 corresponds to the point (1, 1, 1) on the curve.
.
Also r = i + 2tj + 3t 2 k
.
∴ r (1) = i + 2 j + 3k
.
∴ The equation of the normal plane at (1, 1, 1) is (r – r(1)) . r (1) = 0.
⇒ (r – (i + j + k)) . (i + 2j + 3k) = 0
⇒ r . (i + 2j + 3k) – (1(1) + 1(2) + 1(3)) = 0
⇒ r . (i + 2j + 3k) = 6.

WORKING RULES FOR SOLVING PROBLEMS

.
r
Rule I. t= .
= r′ is the unit tangent vector.
|r|
.
Rule II. The equation of the tangent to the curve r = r(t) at the point P(r) is R = r + λ r ,
where λ is a scalar parameter. If R = Xi + Yj + Zk and r = xi + yj + zk, then
X−x Y−y Z−z
this equation reduces to .
= .
= .
(= λ).
x y z
Rule III. The equation of the tangent to the curve r = r(s) at the point P(r) is R = r + λr′,
where λ is a scalar parameter. If R = Xi + Yj + Zk and r = xi + yj + zk, then this
X −x Y−y Z−z
equation reduces to = = (= λ).
x′ y′ z′
Rule IV. The equation of the normal plane to the curve r = r(t) at the point P(r) is
.
(R – r) . r = 0 or equivalently (R – r) . t = 0.
CURVES IN SPACE 21

EXERCISE 1.4
1. Find the unit tangent vector t and the direction cosines of the tangent to the helix
x = a cos t, y = a sin t, z = bt, – ∞ < t < ∞ at the point, where t = π/4.
2. Find the unit tangent vector t and the direction cosines of the tangent to the helix x = a cos t,
y = a sin t, z = at, – ∞ < t < ∞ at the point, where t = π/3.
3. Find the unit tangent vector t to the curve r = ti + t3j at the point (1, 1, 0).
4. Find the unit tangent vector t to the curve r = cos ti + 2 sin tj at the point (1/2, 3 , 0).
5. Find the unit tangent vector t to the curve r = cosh ti + sinh tj at the point (5/3, 4/3, 0).
6. Find the unit tangent vector t to the curve r = log cos ti + log sin tj + 2 t k at the point ‘t ’.
7. Find the equation of the tangent to the curve x = 1 + t, y = – t2, z = 1 + t2, – ∞ < t < ∞ at the point
for which (i) t = 1 (ii) t = 5.
8. Find the equation of the tangent to the helix r = (a cos t, a sin t, bt), – ∞ < t < ∞ at the point ‘t’.
9. Find the equation of the tangent to the curve r = ti + t3j at the point (1, 1, 0).
FG 1 , 3 , 0IJ .
10. Find the equation of the tangent to the curve r = cos ti + 2 sin tj at the point
H2 K
11.
F5 4 I
Find the equation of the tangent to the curve r = cosh ti + sinh tj at the point GH , , 0JK .
3 3
12. Find the point of intersection of the xy-plane and the tangent line to the curve r = (1 + t) i – t2 j +
(1 + t3) k at t = 1.
13. Show that the tangent at any point on the curve r = ati + bt2j + t3k, 2b2 = 3a makes a constant
angle with the line x – z = 0, y = 0.
14. Find the equation of the normal plane to the curve r = (1 + t)i – t2j + (1 + t3)k at t = 1.
15. Find the point of intersection of the xy-plane and the normal plane to the curve
π
r = (cos t)i + (sin t)j + tk at the point t = .
2
Answers
a a b a a b
1. − i+ j+ k;– , ,
2 2 2 2 2 2 2 2 2 2
2( a + b ) 2( a + b ) a +b 2(a + b ) 2( a + b ) a + b2
2

6 2 2 6 2 2 1 3
2. – i+ j+ k;− , , 3. i+ j + 0k
4 4 2 4 4 2 10 10
3 2 4 5
4. – i+ j + 0k 5. i+ j + 0k
7 7 41 41
6. – sin2 ti + cos2 tj + 2 sin t cos tk
x − 2 y +1 z − 2
7. (i) 1 = − 2 = 2 (ii) x − 6 = y + 25 = z − 26
1 − 10 10
x − a cos t y − a sin t z − bt
8. = = 9. r = i + j + λ(i + 3j)
− a sin t a cos t b
F 3 I 5 4 4 5 FG IJ
10. r =
1
2
i+ 3j+ λ − GH 2
i+ jJK 11. r =
3
i + j+ λ
3 3
i+ j
3 H K
FG 4 , 1 , 0IJ
12. H3 3 K 14. r · (i – 2j + 3k) – 10 = 0
15. (– π/2, k, 0), – ∞ < k < ∞.
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

22 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

16. MOVING TRIHEDRON OF A CURVE

Let r = r(s) be the equation of a regular curve C with arc length s as parameter. We
assume that r″(s) exists and |r″(s)| ≠ 0. We know that r′(s) equals the unit tangent vector t(s)
at the point r(s) on the curve C.
∴ t = r′
t′
We define n= .
|t ′ |
n is meaningful, because |t′| = | r″ | ≠ 0.
Also, t is a unit vector.
⇒ |t|=1 ⇒ | t |2 = 1 ⇒ t·t=1
⇒ t · t′ + t′ · t = 0 ⇒ 2t · t′ = 0 ⇒ t · t′ = 0
⇒ t′ is perpendicular to t ⇒ n is perpendicular to t.
t′ 1
Also, |n|= = |t ′| = 1
|t ′| |t ′ |
∴ n is a unit vector and is perpendicular to the unit tangent vector t.
∴ n lies in the normal plane at the point under consideration. The vector n is called
the unit principal normal vector to the curve C at the point r(s). In terms of r, we have
r ′′
n= . (∵ t = r′)
|r ′′ |
We define b = t × n.
π
∴ | b | = |t × n| = |t||n| sin =1×1×1=1
2
Also by the definition of vector cross product, b
is perpendicular to vectors t and n both and the
Principal normal

vectors t, n and b form a right handed triad.

The vector b is called the unit binormal vector Osculating
to the curve C at the point r(s). plane
∴ We have
C
unit tangent vector, t = r′ Normal n
plane
t′ r ′′ t Tangent
unit principal normal vector n = =
|t ′ | |r ′′| r(s)
b
(∵ | t′ | = | r″ | ≠ 0)
unit binormal vector, b = t × n al Rectifying
orm plane
n
r ′′ r ′ × r ′′ Bi
= r′ × = .
|r ′′| |r ′′|
Thus (t, n, b) forms a right-handed orthonormal triplet as shown in the figure. The
triplet (t, n, b) is called the moving trihedron of the given curve r = r(s).
Remark. Since the unit vectors t, n, b form a right handed triad, we have
(i) t · n = 0 n· b=0 b·t=0
(ii) t × n = b n × b = t b × t = n.
CURVES IN SPACE 23

The straight lines in the directions of t, n and b are respectively called the tangent, the
principal normal and the binormal of the curve C at the point r(s). The equations of the
tangent at the point r(s) is R = r + λt, where R is a general point on the tangent and λ is a
scalar. The equation of the principal normal at the point r(s) is R = r + λn, where R is a
general point on the principal normal and λ is a scalar. The equation of the binormal at the
point r(s) is R = r + λb, where R is a general point on the binormal and λ is a scalar.
We know that the unit vectors n and b are both perpendicular to the unit vector t.
∴ The normal plane of C at r is parallel to the vectors n and b both at the point r and
its equation is (R – r)· t = 0, where R is a general point on the normal plane.
If R is the position vector of a general point on the normal plane at the point r, then the
vectors R – r , n and b lie in the normal plane and are thus coplanar vectors.
∴ [R – r n b] = 0.
This also gives the equation of the normal plane at the point r.
The plane through the point r and parallel to the vectors t and b at the point r is called
the rectifying plane of C at the point r. The rectifying plane is perpendicular to the vector n.
∴ The equation of the rectifying plane at the point r is (R – r) · n = 0, where R is a
general point on the rectifying plane.
If R is the position vector of a general point on the rectifying plane at the point r, then
the vectors R – r, t and b lie in the rectifying plane and are thus coplanar vectors.
∴ [R – r t b] = 0
This also gives the equation of the rectifying plane at the point r.
The plane through the point r and parallel to the vectors t and n at the point r is called
the osculating plane of C at the point r. The osculating plane is perpendicular to the vector b.
∴ The equation of the osculating plane at the point r is (R – r) · b = 0, where R is a
general point on the osculating plane.
If R is the position vector of a general point on the osculating plane at the point r, then
the vectors R – r, t and n lie in the osculating plane and are thus coplanar vectors.
∴ [R – r t n] = 0
This also gives the equation of the osculating plane at the point r.
Thus at each point r on the curve C we have the following three characteristic lines and
three characteristic planes :
Tangent R = r + λt
Principal normal R = r + λn
Binormal R = r + λb
Normal plane (R – r) · t = 0 or [R – r n b] = 0
Rectifying plane (R – r) · n = 0 or [R – r t b] = 0
Osculating plane (R – r) · b = 0 or [R – r t n] = 0.
17. CARTESIAN EQUATIONS OF CHARACTERISTIC LINES AND PLANES
Let r = r(s) be the equation of a regular curve with arc length s as parameter. We
assume that r″(s) exists and |r″(s)| ≠ 0. Let r(s) = x(s)i + y(s)j + z(s)k.
∴ t = r′ = x′i + y′j + z′k
and t′ = x″i + y″j + z″k
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

24 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

∴ n=
t′
|t ′ |
=
1
b
x ′′i + y ′′j + z ′′k g
x ′′ 2 + y ′′ 2 + z ′′ 2
x ′′ y ′′ z ′′
= i+ j+ k
x ′′ 2 + y ′′ 2 + z ′′ 2 x ′′ 2 + y ′′ 2 + z ′′ 2 x ′′ 2 + y ′′ 2 + z ′′ 2
b=t×n
i j k
x′ y′ z′
x ′′ y ′′ z ′′
=
x ′′ 2 + y ′′ 2 + z ′′ 2 x ′′ 2 + y ′′ 2 + z ′′ 2 x ′′ 2 + y ′′ 2 + z ′′ 2

i j k
1
= x′ y′ z′
x ′′ 2 + y ′′ 2 + z ′′ 2 x ′′ y′′ z ′′
1
= ((y′z″ – y″z′)i + (z′x″ – z″x′)j + (x′y″ – x″y′)k)
x ′′ + y′′ 2 + z ′′ 2
2

∴ The equation of the tangent R = r + λt at the point r reduces to

X− x Y− y Z−z
= = .
x′ y′ z′
The equation of the principal normal R = r + λn at the point r reduces to
X−x Y− y Z−z
= = ,
x ′′ y ′′ z ′′
because x″, y″, z″ are d.r.’s of the principal normal at the point (x, y, z).
The equation of the binormal R = r + λb at the point r reduces to
X−x Y−y Z−z
= = ,
y′ z ′′ − y ′′ z ′ z ′x ′′ − z ′′x ′ x ′y ′′ − x ′′y′
because y′z″ – y″z′, z′x″ – z″x′, x′y″ – x″y′ are d.r.’s of the binormal at the point (x, y, z).
The equation of the normal plane (R – r) · t = 0 at the point r reduces to
(X – x)x′ + (Y – y)y′ + (Z – z)z′ = 0.
The equation of the rectifying plane (R – r) · n = 0 at the point r reduces to
(X – x)x″ + (Y – y)y″ + (Z – z)z″ = 0.
The equation of the osculating plane (R – r) · b = 0 at the point r reduces to
(X – x)(y′z″ – y″z′) + (Y – y)(z′x″ – z″x′) + (Z – z)(x′y″ – x″y′) = 0.
In the above equations, the point R with coordinates (X, Y, Z) is a general point on the
corresponding line (or plane).

18. VALUES OF UNIT VECTORS t, n AND b ALONG A CURVE GIVEN IN TERMS OF

AN ARBITRARY PARAMETER
Let r = r(t) be the equation of a regular curve C, where t is an arbitrary parameter. We
.. ..
assume that r (t) exists and | r (t) | ≠ 0.
CURVES IN SPACE 25

In terms of arc length s, we have

t′
t = r′, n = and b = t × n.
|t ′|
FG s =
z IJ
. .
dr dr dt r r t . ds .
(i) t = r′ = = .
ds dt ds
=
ds
= .
|r |
H a
| r|dt ⇒
dt
=| r|
K
dt
. .
dt dt dt t t
(ii) t′ = = . = = .
ds dt ds ds
|r |
dt
. .
t | t|
and | t′ | = .
= .
|r | |r|
. . .
t′ t /|r | t
∴ n= = . .
= .
|t′| |t |/|r | |t |
. . . .
r t r×t
(iii) b = t × n = .
× .
= . .
|r | |t | |r ||t |
∴ In terms of arbitrary parameter t, we have
. . . .
r t r×t
t= , n= .
.
and b= .
.
.
|r| |t | |r ||t |

WORKING RULES FOR SOLVING PROBLEMS

Let r = r(s) be the equation of a regular curve with arc length s as parameter. Let r″(s)
exists and |r″(s)| ≠ 0.
Rule I. (i) t = r′ = x′i + y′j + z′k
t′ 1
(ii) n = = ( x ′′ i + y ′′ j + z ′′ k )
|t ′ | x ′′ + y ′′ 2 + z ′′ 2
2

1
(iii) b = t × n = ((y′z″ – y″z′)i + (z′x″ – z″x′)j + (x′y″ – x″y′)k)
x ′′ + y ′′ 2 + z ′′ 2 2

Rule II. Equation of tangent:

X −x Y −y Z−z
(i) R = r + λt (ii) = =
x′ y′ z′
Rule III. Equation of principal normal:
X −x Y −y Z−z
(i) R = r + λn (ii) = =
x ′′ y ′′ z ′′
Rule IV. Equation of binormal:
X−x Y−y Z−z
(i) R = r + λb (ii) = =
y′ z ′′ − y ′′ z ′ z ′x ′′ − z ′′x ′ x ′y ′′ − x ′′y′
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

26 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

Rule V. Equation of normal plane:

(i) (R – r) · t = 0 (ii) [R – r n b] = 0
(iii) (X – x)x′ + (Y – y)y′ + (Z – z)z′ = 0
Rule VI. Equation of rectifying plane:
(i) (R – r) · n = 0 (ii) [R – r t b] = 0
(iii) (X – x)x″ + (Y – y)y″ + (Z – z)z″ = 0
Rule VII. Equation of osculating plane:
(i) (R – r) · b = 0 (ii) [R – r t n] = 0
(iii) (X – x)(y′z″ – y″z′) + (Y – y)(z′x″ – z″x′) + (Z – z)(x′y″ – x″y′) = 0.

Theorem 1. Let r = r(s) be the equation of a regular curve with arc length s as parameter.
If r″ exists and |r″| ≠ 0 at a point r, prove that the equation of the osculating plane at the point
r is
[R – r r′ r″] = 0.
Proof. We know that the plane through the point r and parallel to the vectors t and n
at the point r is the osculating plane at the point r.
dr
We have r′ = =t
ds
∴ r′ is parallel to the osculating plane.

Also r″ =
d
(t) = t′ = | t′ | n
FG
t′ IJ
ds
∵ n=
H
|t ′| K
∴ r″ is parallel to the osculating plane.
Let R be the position vector of a general point on the osculating plane at the point r.
∴ The vector R – r, r′ and r″ lie in the osculating plane and are thus coplanar vectors.
∴ [R – r r′ r″] = 0.
This is the equation of the required osculating plane.
Corollary. If R = Xi + Yj + Zk and r = xi + yj + zk, then r′ = x′i + y′j + z′k
and r″ = x″i + y″j + z″k.
X−x Y− y Z−z
∴ The equation of the osculating plane is x′ y′ z′ = 0.
x ′′ y ′′ z ′′
Theorem 2. Let r = r(t) be the equation of a regular curve, where t is an arbitrary parameter.
.. ..
If r exists and | r | ≠ 0 at a point r, prove that the equation of the osculating plane at the point r is
. ..
[R – r r r ] = 0.
Proof. We know that the plane through the point r and parallel to the vectors t and n
at the point r is the osculating plane at the point r.
. dr dr ds . .
We have r= = . = t s = st
dt ds dt
.
∴ r is parallel to the osculating plane.
.
.. d . . dt ds
Also, r= ( s t) = s + t
dt dt dt
CURVES IN SPACE 27

F .I
. . .. ..
= s t + s t = s t + s|t |n
. .
GG∵ n=
t JJ
H |t |K
.

..
∴ r lies in the plane of t and n.
..
∴ r is parallel to the osculating plane.
Let R be the position vector of a general point on the osculating plane at the point r.
. ..
∴ The vectors R – r, r and r lie in the osculating plane and are thus coplanar vectors.
. ..
∴ [R – r r r ] = 0.
This is the equation of the required osculating plane.
. . . .
Corollary. If R = Xi + Yj + Zk and r = xi + yj + zk, then r = x i + y j + z k
.. .. .. ..
and r = x i + y j + z k.
X−x Y− y Z−z
. . .
∴ The equation of the osculating plane is x y z = 0.
.. .. ..
x y z
Example 1. For the curve x = 3t, y = z= 3t2, 2t3,
(i) show that any plane meets it in three
points and (ii) find the equation of the osculating plane at the point t1.
Sol. The given curve is
x = 3t, y = 3t2 , z = 2t3.
(i) Let ax + by + cz + d = 0 be any plane in space.
Putting x = 3t, y = 3t2, z = 2t3, we get 3at + 3bt2 + 2ct3 + d = 0
⇒ 2ct3 + 3bt2 + 3at + d = 0
This is a cubic equation in t and gives three values of t.
∴ The plane ax + by + cz + d = 0 meets the given curve in three points.
(ii) We have x = 3t, y = 3t2, z = 2t3
. . . .. .. ..
∴ x = 3, y = 6t, z = 6t2 and x = 0, y = 6, z = 12t
Let (x, y, z) be a general point on the osculating plane at the point t1.
∴ The equation of the osculating plane is
x − 3t1 y − 3t12 z − 2t13
3 6t1 6t12 =0
0 6 12t1

x − 3t1 y − 3t12 z − 2t13

1 2t1 2t12 =0
0 1 2t1
(x – 3t1) (2t12) – (y – 3t12) (2t1) + (z – 2t13) (1) = 0
⇒ 2t12x – 2t1y + z = 2t13.
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

28 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

Example 2. Find the equation of the osculating plane to the curve x = 2 log t, y = 4t,
z = 2t2 + 1 at the point t.
Sol. The given curve is
x = 2 log t, y = 4t, z = 2t2 + 1.
.2 . . .. 2 .. ..
∴ x= , y = 4, z = 4t and x = − 2 , y = 0 , z = 4
t t
Let (X, Y, Z) be a general point on the osculating plane at the point t.
X−x Y− y Z−z
. . .
∴ The equation of the osculating plane is x y z = 0.
.. .. ..
x y z

X − 2 log t Y − 4t Z − (2t 2 + 1)
2
⇒ 4 4t =0
t
2
− 2 0 4
t
X − 2 log t Y − 4t Z − (2t 2 + 1)
⇒ 1 2t 2t 2 =0
1 0 − 2t 2
⇒ (X – 2 log t)(– 4t3) – (Y – 4t)(– 4t2) + (Z – 2t2 – 1)(– 2t) = 0
⇒ 2t2(X – 2 log t) – 2t(Y – 4t) + Z – 2t2 – 1 = 0
⇒ 2t2X – 2tY + Z = 4t2 log t – 6t2 + 1.
Example 3. Let r = r(t) be the equation of a regular curve. By using the equation
. ..
[R – r r′ r″] = 0, show that the equation of the osculating plane at the point r is [R – r r r ] = 0.
Sol. Given equation of the osculating plane at point r is
[R – r r′ r″] = 0 ...(1)
.
dr dr dt r
r′ = = . =
ds dt ds s.
F . I F . I . .. .. . ..

r″ =
dr′
=
d r
=
d r dt
. GG
= JJ
s r − s r 1 1 .. s .
. = 2 r− 3 r GG JJ
H K H K
2
ds .
ds s .
dt s ds s s s s

LM F1 s .. I OP
rG JJ P = 0
1 . .. .
∴ (1) ⇒
MMR − r .
GH s r − s2 3
r
K PQ
N s

LM 1 1 OP LM 1 s .. O
rP = 0
. .. . .
⇒
MNR − r s r s rPQ − MMNR − r s r − s
. 2 . 3 PPQ
1 F s I
r – G – J R – r r r
⇒ 3
s
R – r r
H s K 4
=0
(∵ Determinant with two equal rows is zero)
CURVES IN SPACE 29

..
1 . .. s
⇒ [ R − r r r ]+ .0 = 0
s 3 s 4
1 . ..
⇒ [R − r r r] = 0
s 3
. ..
⇒ [R – r r r ] = 0.
∴ The result holds.
Example 4. For the curve x = 4a cos3 t, y = 4a sin3 t, z = 3c cos 2t, find
(i) the equation of the principal normal at the point t.
(ii) the equation of the osculating plane at the point t.
Sol. (i) Let r be the position vector of the point (x, y, z) on the given curve.
∴ r = xi + yj + zk = 4a cos3 ti + 4a sin3 tj + 3c cos 2t k
dr dt dt
∴ r′ = = (– 12a cos2 t sin ti + 12a sin2 t cos tj – 6c sin 2tk)
dt ds ds
dt
= 12 sin t cos t (– a cos ti + a sin tj – ck) (Using sin 2t = 2 sin t cos t)
ds
dt
∴ | r′ | = 12 sin t cos t a 2 cos 2 t + a 2 sin 2 t + c 2
ds
dt
⇒ 1 = 12 sin t cos t a2 + c 2 (∵ | t | = | r′ | = 1)
ds
ds
∴ = 12 sin t cos t a2 + c2
dt
1
∴ r′ = 12 sin t cos t (– a cos ti + a sin tj – ck) .
12 sin t cos t a 2 + c 2
1
= (– a cos ti + a sin tj – ck)
a + c2
2

1 dt
∴ r″ = (a sin ti + a cos tj)
2
a +c 2 ds

a 1
= (sin ti + cos tj)
2 2
a +c 12 sin t cos t a 2 + c 2

a
= (sec ti + cosec tj)
12(a2 + c 2 )
t′ r ′′ a
∴ n= = = (sec ti + cosec tj)
|t ′| |r ′′| 12(a + c 2 )|r ′′|
2

∴ D.r.’s of principal normal are sec t, cosec t, 0.

∴ The equations of the principal normal at the point t are
x − 4 a cos 3 t y − 4 a sin 3 t z − 3c cos 2t
= = .
sec t cosec t 0
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

30 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

(ii) The equation of the osculating plane at the point t is

x − 4 a cos 3 t y − 4 a sin 3 t z − 3c cos 2t

a a −c =0
− cos t sin t
a + c2
2
a + c2
2 2
a +c 2

a a
sec t cosec t 0
12(a + c 2 )
2
12(a + c 2 )
2

x − 4 a cos 3 t y − 4 a sin 3 t z − 3c cos 2t

⇒ − a cos t a sin t −c =0
sec t cosec t 0
(x – 4a cos3 t)(c cosec t) – (y 4a sin3 t)(c sec t)
+ (z – 3c cos 2t)(– a sin t cos t – a sin t cos t) = 0
⇒ c cosec t · x – c sec t · y – 2 sin t cos t z
= 4a sin t cos3 t – 4a sin3 t cos t – 6ac sin t cos t cos 2t
⇒ c cosec t · x – c sec t · y – 2 sin t cos t z
= 2a sin 2t cos2 t – 2a sin 2t sin2 t – 3ac sin 2t cos 2t.
Example 5. Find the vectors t, n and b along the curve r = (3t – t3)i + 3t2j + (3t + t3)k.
Sol. We have r = (3t – t3)i + 3t2j + (3t + t3)k.
.
∴ r = (3 – 3t2)i + 6tj + (3 + 3t2)k
.
or r = 3[(1 – t2)i + 2tj + (1 + t2)k]
.
| r | = 3 (1 − t 2 ) 2 + 4 t 2 + (1 + t 2 ) 2

= 3 2 + 2t 4 + 4t 2 = 3 2 (1 + t2)

.
r 1
∴ t= .
= . 3[(1 – t2)i + 2tj + (1 + t2)k]
| r| 3 2 (1 + t 2 )

1 − t2 2t 1
= 2
i+ 2
j+ k
2 (1 + t ) 1+ t 2

. (1 + t 2 ) (− 2t) − (1 − t 2 ) 2t 2 ((1 + t 2 ) . 1 − t . 2t)

t= i+ j + 0k
2 (1 + t )2 2
(1 + t 2 ) 2

2
= (– 2t i + (1 – t2) j)
(1 + t 2 ) 2
. 2 2
∴ | t| = 2 2 [4t2 + (1 – t2)2]1/2 =
(1 + t ) 1 + t2
.
t 2 2
2)j) . 1 + t = – 2t 1 − t2
∴ n= .
= (– 2ti + (1 – t i + j
| t| (1 + t 2 ) 2 2 1 + t2 1 + t2
CURVES IN SPACE 31

i j k
1 − t2 2t 1
b=t×n=
2 (1 + t 2 ) 1 + t2 2
2t 1 − t2
− 0
1 + t2 1 + t2
i j k
1
= 1 – t2 2t 1 + t2
2 (1 + t 2 ) (1 + t 2 ) − 2t 1 – t2 0
1
= [– (1 – t4)i – 2t (1 + t2)j + (1 + t2)2k]
2 (1 + t 2 ) 2
1
= [(t2 – 1)i – 2tj + (1 + t2)k].
2 (1 + t 2 )
Example 6. Show that the points on the helix r = a cos ti + a sin tj + btk, a > 0, b ≠ 0 at
which the osculating planes pass through a fixed point are all coplanar.
Sol. The given helix is
r = a cos ti + a sin tj + btk.
.
∴ r = – a sin ti + a cos tj + bk
..
r = – a cos ti – a sin tj + 0k
. ..
The equation of the osculating plane is [R – r r r ] = 0.
Let R = Xi + Yj + Zk
∴ The equation of the osculating plane is
X − a cos t Y − a sin t Z − bt
− a sin t a cos t b =0
− a cos t − a sin t 0
⇒ ab sin t (X – a cos t) – ab cos t(Y – a sin t) + a2(Z – bt) = 0
⇒ b sin t X – b cos t Y + aZ = abt
Let the osculating plane at the point r passes through the fixed point (α, β, γ).
∴ (b sin t)α – (b cos t)β + aγ = abt
⇒ – bβ (a cos t) + bα(a sin t) – a2 (bt) = – a2 γ
⇒ bβ(a cos t) – bα(a sin t) + a2 (bt) = a2 γ
∴ The locus of the point r ( = (a cos t, a sin t, bt)) is
bβx – bαy + a2z = a2γ, which is a plane.
∴ The result holds.
Example 7. Find the equations of characteristic lines and planes to the helix r = cos t i
+ sin t j + tk at the point where t = π/2.
Sol. We have r = cos ti + sin tj + tk.
FG π IJ = cos π i + sin π j + π k = j + π k
∴ r H 2K 2 2 2 2
∴ The point under consideration is (0, 1, π/2).
.
r = – sin ti + cos tj + k
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

32 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

.
∴ |r|= sin 2 t + cos 2 t + 1 = 2
.
r 1
∴ t= .
=
(– sin ti + cos tj + k)
| r| 2
. 1
t= (– cos ti – sin tj + 0k)
2
. 1 1
and | t| = cos 2 t + sin 2 t =
2 2
.
t 1
∴ n= .
= (– cos ti – sin tj) . 2 = – cos ti – sin tj
| t| 2
i j k
sin t cos t 1 sin t cos t 1
−
b=t×n= 2 2 2 = i– j+ k
2 2 2
− cos t − sin t 0
1 1 1
∴ t(π/2) = ( – 1 · i + 0 · j + k) = – i+ k
2 2 2
n(π/2) = – 0 · i – 1· j = – j
1 1 1 1
b(π/2) = i–0·j+ k= i+ k.
2 2 2 2
Tangent. The equation of the tangent is

π 1 1FG IJ
r = r(π/2) + λt(π/2) i.e., r = j +
2
k+λ −
2
i+
2 H
k .
K
x − 0 y − 1 z − π/ 2
Equations in cartesian form are = = .
−1 0 1
Principal normal. The equation of the principal normal is
π
r = r(π/2) + λn(π/2) i.e., r = j + k + λ(– j).
2
x − 0 y − 1 z − π/ 2
Equations in the cartesian form are = = .
0 −1 0
Binormal. The equation of the binormal is
π 1 1 FG IJ
r = r(π/2) + λb(π/2) i.e., r = j + k + λ
2 2
i+
2
k .
H K
x − 0 y − 1 z − π/ 2
Equations in the cartesian form are = = .
1 0 1
Normal plane. The equation of the normal plane is (r – r(π/2)) · t(π/2) = 0
FG FG π IJ IJ FG
1 1 IJ
π FG FG IJ IJ
i.e.,
H H
r − j+ k . −
2 KK H
2
i+
2
k = 0 or r − j + k · (– i + k) = 0
2K H H KK
FG π IJ π
or
H 2 K
r · (– i + k) – 0 (− 1) + 1(0) + (1) = 0 or r · (– i + k) = .
2
CURVES IN SPACE 33

π π
Equation in the cartesian form is – x + z = i.e., x – z + = 0.
2 2
Rectifying plane. The equation of the rectifying plane is (r – r(π/2)) . n(π/2) = 0
FG r – FG j + π kIJ IJ · (– j) = 0
i.e.,
H H 2 KK or – r · j + 1 = 0 or r · j – 1 = 0.

Equation in the cartesian form is y – 1 = 0.

Osculating plane. The equation of the osculating plane is
FG FG π IJ IJ . FG 1 1 IJ
(r – r(π/2)) · b(π/2) = 0 i.e., r − j +
H H 2
k
KK H 2
i+
2 K
k =0

FG r − FG j + π kIJ IJ · (i + k) = 0 π
or
H H 2 KK or r · (i + k) =
2
.

π
Equation in the cartesian form is . x+z=
2
Example 8. For the curve r = (e–t sin t, e–t cos t, e–t), find the following at the point t:
(i) the unit tangent vector t
(ii) the equation of the tangent
(iii) the unit principal normal vector n
(iv) the equation of the normal plane
(v) the unit binormal vector b
(vi) the equation of the binormal.
Sol. (i) The given curve is
r = e–t sin ti + e–t cos tj + e–tk.
.
∴ r = (e–t cos t – e–t sin t)i + (– e–t sin t – e–t cos t)j – e–tk
∴ = e–t [(cos t – sin t)i – (sin t + cos t)j – k]
.
∴ | r | = e–t (cos t − sin t) 2 + (sin t + cos t) 2 + 1 = 3 e–t
.
r 1
∴ t= .
= · e–t [(cos t – sin t)i – (sin t + cos t)j – k]
|r | 3e − t
1
∴ t=
[(cos t – sin t)i – (sin t + cos t)j – k]
3
(ii) Using t, the d.r.’s of the tangent are cos t – sin t, – (sin t + cos t), – 1.
∴ The equations of the tangent at the point t are
x − e − t sin t y − e − t cos t z − e−t
= = .
cos t − sin t − (sin t + cos t) −1
. 1
(iii) t= [(– sin t – cos t)i – (cos t – sin t)j + 0k]
3
1
= [– (sin t + cos t)i + (sin t – cos t)j]
3
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

34 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

. 1 2
∴ | t| = (sin t + cos t) 2 + (sin t − cos t) 2 =
3 3
.
t 1 3
∴ n= .
= [– (sin t + cos t)i + (sin t – cos t)j] .
|t | 3 2
1
∴ n= [ – (sin t + cos t)i + (sin t – cos t)j].
2
(iv) The equation of the normal plane is (R – r) · t = 0.

⇒ [(xi + yj + zk) – (e–t sin t i + e–t cos t j + e–t k)] ·

LM 1 ((cos t − sin t)i − (sin t + cos t) j − k)OP
N3 Q
=0
⇒ x(cos t – sin t) – y(sin t + cos t) – z + e–t (– sin t cos t + sin2 t + sin t cos t + cos2 t + 1)
=0
⇒ (cos t – sin t)x – (sin t + cos t)y – z = – 2e–t
⇒ (sin t – cos t)x + (sin t + cos t)y + z = 2e–t.
(v) b = t × n
i j k
1 1 1
= (cos t − sin t ) − (sin t + cos t ) −
3 3 3
1 1
− (sin t + cos t ) (sin t − cos t ) 0
2 2

i j k
1
= sin t − cos t sin t + cos t 1
6 sin t + cos t cos t − sin t 0
1
= [(sin t – cos t) i + (sin t + cos t) j – 2k].
6
(vi) Using b, the d.r.’s of the binormal are sin t – cos t, sin t + cos t, – 2.
∴ The equation of the binormal at the point t are
x − e − t sin t y − e − t cos t z − e − t
= = .
sin t − cos t sin t + cos t −2
Example 9. Find the equation of the osculating plane at a general point on the curve
r = (t, t2, t3). Show that the osculating planes at three points on this curve meet at a point lying
in the plane determined by these three points.
Sol. The given curve is r = (t, t2, t3). Parametric equations of the curve are
x = t, y = t2, z = t3.
. . . .. .. ..
∴ x = 1, y = 2t, z = 3t2 , x = 0, y = 2, z = 6t
∴ Equation of the osculating plane at point ‘t’ is
X − t Y − t2 Z − t3
1 2t 3t 2 =0
0 2 6t
CURVES IN SPACE 35

⇒ 6t2(X – t) – 6t(Y – t2) + 2(Z – t3) = 0

⇒ 3t2X – 3tY + Z – t3 = 0. ...(1)
Let t1, t2, t3 be the values of the parameter t at any three points on the curve.
∴ The equations of the osculating planes at these points are
3t12X – 3t1Y + Z – t13 = 0
3t22X – 3t2Y + Z – t23 = 0
and 3t32X – 3t3Y + Z – t33 = 0
Let these planes intersect at the point (α, β, γ).
∴ 3t12α – 3t1β + γ – t13 = 0
3t22α – 3t2β + γ – t23 = 0
and 3t32α – 3t3β + γ – t33 = 0
∴ t1, t2, t3 are the roots of the cubic equation
3t2 α – 3tβ + γ – t3 = 0
or t3 – 3αt2 + 3βt – γ = 0 ...(2)
Let the equation of the plane passing through the points t1, t2 and t3 be
ax + by + cz + d = 0 ...(3)
⇒ 2 3
at + bt + ct + d = 0
⇒ ct3 + bt2 + at + d = 0,
where t = t1, t2, t3 ...(4)
∴ Equations (2) and (4) are same.
1 − 3α 3β − γ
∴ = = =
c b a d
∴ a = 3βc, b = – 3αc, d = – γc
∴ (3) ⇒ 3βcx – 3αcy + cz – γc = 0
⇒ 3βx – 3αy + z – γ = 0
(α, β, γ) lies on this plane if
3βα – 3αβ + γ – γ = 0 if 0 = 0, which is true.
∴ The result holds.
Example 10. Show that there are three points on the curve x = at3 + b, y = 3ct2 + 3dt,
z = 3et + f such that their osculating planes pass through the origin and that the three points lie
on the plane 3cex + afy = 0.
Sol. The given curve is
x = at3 + b, y = 3ct2 + 3dt, z = 3et + f.
. . . .. .. ..
∴ x = 3at2, y = 6ct + 3d, z = 3e, x = 6at, y = 6c, z = 0
∴ Equation of the osculating plane at point ‘t’ is
X − (at 3 + b) Y − (3ct 2 + 3dt) Z − (3et + f )
3at 2 6ct + 3d 3e =0
6 at 6c 0
⇒ – 18ce(X – at3 – b) + 18aet(Y – 3ct2 – 3dt)
+ (18act2 – 36act2 – 18adt)(Z – 3et – f) = 0
⇒ ce(X – at – b) – aet(Y – 3ct2 – 3dt) + (act2 + adt)(Z – 3et – f) = 0
3

Let this plane pass through the origin.

SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

36 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

∴ ce(0 – at3 – b) – aet(0 – 3ct2 – 3dt) + (act2 + adt)(0 – 3et – f) = 0

⇒ – acet3 – bce + 3acet3 + 3adet2 – 3acet3 – acft2 – 3adet2 – adft = 0
⇒ acet3 + acft2 + adft + bce = 0 ...(1)
This is a cubic in t. Let the roots of this equation be t1, t2 and t3.
∴ There are three points on the given curve whose osculating planes passes through
the origin. Let x1 = at13 + b, y1 = 3ct12 + 3dt1, z1 = 3et1 + f
x1 − b z1 − f
∴ t13 = , t1 =
a 3e
y1 − 3dt1 1 FG z −f FG IJ IJ = ey − dz1 + df
y1 − 3d 1
H KK
1
t12 =
3c
=
3c H 3e 3ce
(1) ⇒ acet13 + acft12 + adft1 + bce = 0
FG x − bIJ + acf FG ey − dz + df IJ + adf FG z − f IJ + bce = 0
H a K H 3ce K H 3e K
1 1 1 1
⇒ ace

⇒ ce(x – b) +
1
af
3e
b ey – dz + df g +
1 1
adf
3e
( z – f ) + bce = 0
1

⇒ 3ce2(x1 – b) + af(ey1 – dz1 + df) + adf(z1 – f) + 3bce2 = 0

⇒ 3ce2x 1 – 3bce2 + aefy1 – adf z1 + adf 2 + adf z1 – adf2 + 3bce2 = 0
⇒ 3ce2x1 + aefy1 = 0
⇒ 3cex1 + afy1 = 0
∴ (x1, y1, z1) lies in the plane 3cex + afy = 0.
Similarly, the points corresponding to t2 and t3 also lie on the plane 3cex + afy = 0.

EXERCISE 1.5
1. Find the intersection of the xy-plane and the tangent lines to the helix r = cos ti + sin tj + tk,
(t > 0).
2. Find the equation of the osculating plane at any point on the curve r = (t, t2, t3).
3. Find the equation of the osculating plane to the curve r = ti + t2j + t3k at the point for which t = 1.
4. Find the equation of the osculating plane to the curve x = 3t, y = 3t2, z = 2t3 at the points (3, 3, 2),
(– 3, 3, – 2) and (6, 12, 16).
5. Find the equation of the osculating plane at the point ‘ t’ on the helix x = a cos t, y = a sin t, z = ct.
x y z
6. Show that the osculating plane at the point t = 1 of the curve r = (3at, 3bt2, ct3) is − + = 1.
a b c
7.
Find the equation of the osculating plane at the point t of the curve
x = a cosh t, y = a sinh t, z = bt.
8. Find the equation of the osculating plane at the point t of the curve
r = 4a cos3 ti + 4a sin3 tj + 2a cos 2tk.
9. Find the osculating plane at the point t of the curve x = a cos 2t, y = a sin 2t, z = 2a sin t.
10. Find the basic unit vectors t, n and b of the curve r = (t, t2, t3) at the point t = 1. Find also the
equations of the characteristic lines and planes at this point.
CURVES IN SPACE 37

Answers
1. (cos t + t sin t, sin t – t cos t, 0) 2. 3t2x – 3ty + z = t3
3. 3x – 3y + z = 1
4. 2x – 2y + z = 2, 2x + 2y + z = – 2, 8x – 4y + z = 16
5. c(x sin t – y cos t – at) + az = 0 7. bx sinh t – by cosh t + az = abt
8. 2x cos t – 2y sin t – 3z = 2a cos 2t
9. (sin 3t + 3 sin t)x – (cos 3t + 3 cos t)y + 4z = 6a sin t
1 1 1
10. t = (i + 2j + 3k), n = (– 11i – 8j + 9k), b = (3i – 3j + k),
14 266 19

x − 1 y − 1 z − 1, x − 1 y − 1 z − 1, x − 1 y − 1 z − 1,
= = = = = =
1 2 3 11 8 −9 3 −3 1
x + 2y + 3z = 6, 11x + 8y – 9z = 10, 3x – 3y + z = 1.
2
Curvature and Torsion
1. INTRODUCTION
For curves in space, the concepts of curvature and torsion are of fundamental importance.
We know that line segments are uniquely determined by their lengths, circles by their radii,
triangles by side-angle-side etc. In geometry, we look for geometric quantities which distinguish
one figure from another. The importance of curvature and torsion can easily be estimated
from the fact that it can be proved that a curve is uniquely determined (except for its position
in space) if its curvature and torsion are given as continuous functions of arc length ‘s’.
2. CURVATURE OF A CURVE
Let r = r(s) be a regular curve C of class Cm(m ≥ 2), where s is the parameter ‘arc length’.
The vector r″(s) is called the curvature vector
on the curve C at the point r(s) and it is denoted by
t(s)
κ(s) (or by κ). The magnitude of the curvature vector
is called the curvature of the curve C at the point r(s)
r(s) and it is denoted by κ(s) (or by κ). C
∴ κ(s) = |r″(s)| k(s)
Also t(s) = r′(s), so we have
κ(s) = t′(s).
We know that t(s) is a unit vector.
⇒ t(s) · t(s) = 1
⇒ t(s) · t′(s) + t′(s) · t(s) = 0
⇒ 2t′(s) · t(s) = 0
⇒ κ(s) · t(s) = 0
∴ The curvature vector κ(s) is orthogonal to t(s) and hence parallel to the normal
plane at r(s). When κ(s) is non-zero, it is in the direction in which the curve is turning.
The reciprocal of the curvature at a point is called the radius of curvature at that
point and it is denoted by ρ.
1
∴ ρ= (Assuming κ ≠ 0)
κ
A point on the curve C is called a point of inflexion if the curvature κ at that point is
zero.
κ r″ t′
Remark. We have = = = n.
κ |r ″ | |t′ |
κ
∴ n= .
κ

38
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 39

Example 1. Show that: κ = | r′ × r″ |.

π
Sol. | r′ × r″ | = | t × t′ | = | t | · | t′ | sin =1·κ·1=κ
2
∴ κ = | r′ × r″ |.
Example 2. For the helix r = a cos ti + a sin tj + btk, a > 0, b ≠ 0, find the curvature at
the point t.
Sol. We have r = a cos ti + a sin tj + btk.
∴ r = – a sin ti + a cos tj + bk

⇒ | r | = a 2 sin 2 t + a 2 cos 2 t + b2 = a 2 + b2

r 1
∴ t= = (– a sin ti + a cos tj + bk)
|r | a2 + b2

dt dt dt dt ds
∴ κ = t′ = = . =
ds dt ds dt dt
1 ds
= (– a cos ti – a sin tj)
2
a +b 2 dt

a
a2 + b2
FG∵ ds IJ
= −
2
a +b 2
(cos ti + sin tj)
H dt
= |r |
K
a
= − (cos ti + sin tj)
a + b2
2

a a
∴ Curvature, κ=|κ|= ( − cos t)2 + ( − sin t )2 = .
a +b2 2
a + b2
2

1 2 1
Example 3. For the curve r = ti + t j + t3k, find the curvature vector and curvature
2 3
at the point t = 1.
1 2 1
Sol. We have r = ti + t j + t3k.
2 3

∴ r = i + tj + t2k and | r | = 1 + t2 + t4
r 1
∴ t= = (i + tj + t2k)
|r | 1+ t + t 2 4

1 F 2t + 4t 3 I (i + tj + t k)
t =
1+ t + t 2 4
(j + 2tk) + − GH 2 (1 + t 2 + t 4 ) 3 / 2
JK 2
40 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

1
= [(1 + t2 + t4)(j + 2tk) – (t + 2t3)(i + tj + t2k)]
(1 + t 2 + t 4 ) 3 / 2
1
= [– (t + 2t3)i + (1 – t4)j + (2t + t3)k]
(1 + t + t 4 ) 3 / 2
2

dt dt dt ds
∴ κ = t′ = = =t = t |r |
ds dt ds dt
1
= [– (t + 2t3)i + (1 – t4)j + (2t + t3)k]
(1 + t + t 4 ) 2
2

∴ At t = 1,

1 1 1 1 1 2
κ= [− 3i + 0 j + 3k] = − i + k and κ = | κ | = + = .
(3) 2
3 3 9 9 3
Theorem 1. Prove that the curvature of a regular curve at a point is equal to the rate of
change of direction of the tangent with respect to arc length.
Proof. Let r = r(s) be a regular curve of class Cm(m ≥ 2), where s is the parameter ‘arc
length’. Let r(s) be any point P on the given curve. Let r(s + δs), δs > 0 be a neighbouring point
Q of r(s). Let t(s) and t(s + δs) be the unit tangent vectors at the points r(s) and r(s + δs)
respectively.
t(s)
t(s + ds) C
Q t(s + ds) – t(s)
P r(s + ds) t(s)
M
r(s)
dq
A t(s + ds) B

Let δθ denote the angle between the tangent vectors t(s) and t(s + δs).
By definition,
t (s + δs) − t (s) t (s + δs) − t (s)
κ = | t′ | = lim = lim
δs → 0 δs δ s → 0 δs
|t ( s + δs) − t ( s)|
∴ κ = lim ...(1)
δs → 0 δs
Since t′ is a unit vector, we have AC = AB = 1.
In ∆ABC, CB = | t(s + δs) – t(s) |
δθ
Also CB = 2 CM = 2 sin ∠CAM = 2 sin
2

= 2
LM δθ − (δθ/2) 3 OP
+ ......
MN 2 3! PQ
(By using Taylor’s expansion for the sine function.)

( δθ)3
= δθ − + ......
24
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 41

F ( δθ) 2 I
= δθ 1 −GH 24 JK
+ ......

F ( δθ )2 I
GH
δθ 1 −
24 JK
+ ......
∴ (1) ⇒ κ = lim
δ s→ 0 δs
δθ F ( δθ )2 I
= lim
δs→ 0 δs δθ→ 0 GH
. lim 1 −
24
+ ......JK (∵ δθ → 0 as δs → 0)

dθ dθ
= .1= = rate of change of θ w.r.t. s
ds ds
∴ The curvature at a point is equal to the rate of change of the tangent with respect to
the arc length.
Theorem 2. Prove that a regular curve of class Cm(m ≥ 2) is a straight line if and only if
its curvature is identically zero.
Proof. Let r = r(s) be a regular curve of class Cm(m ≥ 2), where s is the parameter ‘arc
length’.
Let the curve be a straight line.
Let the curve passes through the point whose position vector is a and is parallel to
vector b.
∴ r = a + tb, where t is a parameter.

dr dr
⇒ =b and =|b|
dt dt

r b
∴ t= =
| r | | b|

dt dt dt
∴ κ = t′ = . =0. =0
dt ds ds
∴ κ = |0| = 0 i.e., the curvature is identically zero.
Conversely, let the curvature of the curve be identically zero i.e., κ = 0.
⇒ κ=0 ⇒ t′ = 0 ⇒ t = c, a constant vector.
⇒ r′ = c ⇒ r = cs + d, where d is a constant vector.
∴ The curve is a straight line passing through the point whose position vector is d and
is parallel to the vector c.
∴ The result holds.
42 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

Theorem 3. Let r = r(t) be a regular curve of class Cm (m ≥ 2), where t is an arbitrary

parameter. Prove that
| r ×
r|
κ= 3 .

|r |
dr dr ds
Proof. We have r = = = r ′ s = sr ′
dt ds dt
dr d dr ′
and
r= = (sr ′ ) = s sr ′ = s(r ″ s) +
+ sr ′ + s 2 r ″
sr ′ =
dt dt dt
∴ sr′ + s 2 r″ ) = (s
r = sr′ × (
r × s) (r ′ × r′ ) + (s 3 ) (r′ × r ″ )
s 0 + ( s 3 )( r′ × r″ ) = s3 ( r′ × r″ )
= ( s )
3
= r (r ′ × r ″ ) (∵ s = |r |)

∴ | = |r |3 |r ′||r ″|sin θ, where θ is the angle between r′ and r″.
|r × r
Now r′ = t, r″ = t′ and t and t′ are orthogonal.
∴ θ = π/2
Also | r′ | = | t | = 1 and | r″ | = κ
∴ | = |r |3 . 1 . κ . 1
| r × r
| r × r
|
∴ κ= 3
.
|r |
Corollary. If r = x(t)i + y(t)j + z(t)k, then

| r ×
r| = z − y z ) 2
Σ( y

and | r | = Σ x 2 .

|r ×
r| y z ) 2
Σ( y z −
∴ κ= 3 = .
|r| ( Σ x 2 )3/ 2
Remarks 1. If r = r(s), then
κ = | r″ | = | r′ × r″ |.
| r × r
|
2. If r = r(t), then κ= .
| r |3
Example 4. For the circle r = a cos ti + a sin tj, a > 0, find the radius of curvature at
point t.
Sol. We have r = a cos ti + a sin tj, a > 0.
∴ r = – a sin ti + a cos tj
⇒ | r | = a 2 sin 2 t + a 2 cos 2 t = a
r 1
∴ t= = (− a sin ti + a cos tj) = – sin ti + cos tj
|r | a
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 43

dt dt dt
∴ κ = t′ = = .
ds dt ds
ds
= (– cos ti – sin tj)
dt
= – (cos ti + sin tj) | r |
1
=– (cos ti + sin tj)
a

1 1
∴ Curvature, κ=|κ|= ( − cos t ) 2 + ( − sin t ) 2 =
a a
1 1
∴ Radius of curvature = = =a
κ 1/a
∴ Radius of curvature is equal to the radius of the given circle.
Alternative method
We have r = a cos ti + a sin tj, a > 0.
∴ r = – a sin ti + a cos tj
and
r = – a cos ti – a sin tj
i j k
∴ r ×
r = − a sin t a cos t 0 = (a2 sin2 t + a2 cos2 t)k = a2k
− a cos t − a sin t 0

∴ | r ×
r | = a2
Also | r |2 = a2 sin2 t + a2 cos2 t = a2 ∴ | r | = a

1 |r |3 a3
∴ Radius of curvature = = = 2 = a.
κ |r ×
r| a
Example 5. Show that along the plane curve r = x(t)i + y(t)j,
|x y − x y |
κ= .
( x 2 + y 2 )3 /2
Sol. We have r = xi + yj. ...(1)
| r × r
|
Also, κ= 3
...(2)
|r |
(1) ⇒ r = xi + y j and i +
r = x yj
i j k
∴ r = x
r × y 0 = ( x y − x y ) k
x
y 0

∴ | r × e
y − x y | and | r | = x 2 + y 2
r | = |x j 1/2

|x y − x y | |x y − x y |

∴ (2) ⇒ κ= 2 2 1/ 2 3
= 2 .
[( x + y ) ] ( x + y 2 )3/ 2
44 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

Example 6. For the curve

r = (t – sin t)i + (1 – cos t)j + tk,
find the curvature at point t.
Sol. We have r = (t – sin t)i + (1 – cos t)j + tk.
∴ r = (1 – cos t)i + sin tj + k
and
r = sin ti + cos tj + 0k
| r ×
r|
Now κ= ...(1)
|r |3

i j k
r ×
r = 1 − cos t sin t 1 = – cos ti + sin tj + (cos t – 1)k
sin t cos t 0

∴ | r ×
r |= cos 2 t + sin 2 + (cos t − 1) 2

F F
= G 1 + G − 2 sin
tI I
J F 2 1/2
= G 1 + 4 sin
tI
J
1/2

H H J
2K K
2
H 2K
4

| r | = e(1 − cos t) + sin t + 1j

2 2

2 2 1/ 2
= (1 + cos t − 2 cos t + sin t + 1)
F
= G 1 + 4 sin
tI
J
1/2

H 2K
2
= (1 + 2(1 – cos t))1/2

FG 1 + 4 sin t IJ 1/2
FG 1 + 4 sin t IJ 1/2

H K H 2K
4 4
2
∴ (1) ⇒ κ= = .
F FG 1 + 4 sin tI
J
1/2 3I FG 1 + 4 sin t IJ 3/ 2

GH H JK H 2K
2

2K
2

Example 7. For the curve x = 4a cos3 t, y = 4a sin3 t, z = 3c cos 2t, show that
a
κ= .
6(a 2 + c 2 ) sin 2t
Sol. Let r be the position vector of the point (x, y, z) on the curve.
∴ r = xi + yj + zk = 4a cos3 ti + 4a sin3tj + 3c cos 2tk
dr dt
∴ t = r′ =
dt ds
dt
= (– 12a cos2 t sin ti + 12a sin2 t cos tj – 6c sin 2tk)
ds
dt
= (– 6a cos t sin 2ti + 6a sin t sin 2tj – 6c sin 2tk)
ds
dt
∴ t = 6 sin 2t (– a cos ti + a sin tj – ck) ...(1)
ds
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 45

dt
⇒ | t | = 6 sin 2t . a 2 cos 2 t + a2 sin 2 t + c 2
ds

dt
⇒ 1 = 6 sin 2t a2 + c2 (∵ | t | = 1)
ds

ds
⇒ = 6 a 2 + c 2 sin 2t
dt
ds
⇒ = 6 a 2 + c 2 sin 2t (Assuming sin 2t > 0)
dt
1
∴ (1) ⇒ t = 6 sin 2 t (– a cos ti + a sin tj – ck) .
6 a + c 2 sin 2 t
2

1
= (– a cos ti + a sin tj – ck)
a + c2
2

dt dt dt
Now t′ = =
ds dt ds
1 1
= ( a sin ti + a cos tj − 0k)
a2 + c2 6 a 2 + c 2 sin 2t

a
= (sin ti + cos tj)
6(a + c 2 ) sin 2 t
2

a a
∴ κ = | t′ | = sin 2 t + cos 2 t = .
6(a2 + c 2 ) sin 2 t 6( a + c 2 ) sin 2t
2

Example 8. Find the radius of curvature at any point of the curve

x2 + y2 = a2, x2 – y2 = az.
Sol. The given curve is
x 2 + y 2 = a2 ...(1) x2 – y2 = az ...(2)
Let x = a cos t, y = a sin t
∴ (1) is satisfied.
(2) ⇒ a2 cos2 t – a2 sin2 t = az
⇒ z = a cos 2t
∴ The given curve is
x = a cos t, y = a sin t, z = a cos 2 t.
Let r be the position vector of the point (x, y, z) on the curve.
∴ r = xi + yj + zk = a cos ti + a sin tj + a cos 2 t k
∴ r = – a sin ti + a cos tj – 2a sin 2tk
and
r = – a cos ti – a sin tj – 4a cos 2tk
46 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

i j k
∴ r = − a sin t a cos t − 2a sin 2 t
r ×
− a cos t − a sin t − 4 a cos 2 t
i j k
= a2 sin t − cos t 2 sin 2 t
cos t sin t 4 cos 2 t
= a2 [(– 4 cos t cos 2t – 2 sin t sin 2t)i – (4 sin t cos 2t – 2 cos t sin 2t)j
+ (sin2 t + cos2 t)k]
= a2 [(– 2 cos t cos 2t – 2 cos (t – 2t))i
+ (– 2 sin t cos 2t + 2 sin (2t – t))j + k]
2 3 3
= a [– 4 cos ti + 4 sin tj + k]
∴ |r |2 = a2 sin2 t + a2 cos2 t + 4a2 sin2 2t
= a2(1 + 4 sin2 2t) = a2(5 – 4 cos2 2t)
F 4z 2 I = 5a
GH
= a2 5 −
a 2 JK 2 – 4z2

∴ | r | = 5a2 − 4 z 2
r |2 = a4(16 cos6 t + 16 sin6 t + 1)
| r ×
= a4[16{(cos2 t + sin2 t)3 – 3 cos2 t sin2 t (cos2 t + sin2 t)} + 1]
= a4[16{13 – 3 cos2 t sin2 t . 1} + 1] = a4[17 – 48 cos2t sin2t]
= a4[17 – 12 sin2 2t] = a4[5 + 12 cos2 2t]

= a4 5 +
LM 12 z 2 OP = a (5a
2 2 + 12z2)
N a2 Q
2 2
∴ r | = a 5a + 12 z
| r ×
1 |r |3 (5a 2 − 4 z 2 )3/ 2
∴ ρ= = = .
κ |r ×
r| a 5a 2 + 12 z 2
Example 9. Find the equation of the osculating plane and curvature at point t of the
curve x = a cos 2t, y = a sin 2t, z = 2a sin t.
Sol. Let r be the position vector of the point (x, y, z) on the curve.
∴ r = xi + yj + zk = a cos 2ti + a sin 2tj + 2a sin tk
∴ r = – 2a sin 2ti + 2a cos 2tj + 2a cos tk
and
r = – 4a cos 2ti – 4a sin 2tj – 2a sin t k
i j k
∴ r = − 2a sin 2t 2a cos 2t
r × 2a cos t
− 4 a cos 2t − 4 a sin 2t − 2a sin t
i j k
= – 4a2 − sin 2t cos 2t cos t
2 cos 2t 2 sin 2t sin t
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 47

= – 4a2[(cos 2t sin t – 2 sin 2t cos t)i

– (– sin 2t sin t – 2 cos 2t cos t)j + (– 2 sin2 2t – 2 cos2 2t)k]
= – 4a2 [(sin (t – 2t) – sin 2t cos t)i + (cos (2t – t) + cos 2t cos t)j – 2k]
= 4a2[(sin t + sin 2t cos t)i – (cos t + cos 2t cos t)j + 2k]
Equation of the osculating plane is [R – r r r ] = 0.
⇒ (R – r) . ( r ×
r) = 0 ...(1)
Let R = xi + yj + zk
∴ (1) ⇒ [(x – a cos 2t)i + (y – a sin 2t)j + (z – 2a sin t)k] · 4a2[(sin t + sin 2t cos t)i
– (cos t + cos 2t cos t)j + 2k] = 0
⇒ (x – a cos 2t) (sin t + sin 2t cos t) – (y – a sin 2t) (cos t + cos 2t cos t)
+ (z – 2a sin t)2 = 0
⇒ 2
(sin t + sin 2t cos t)x – (cos t · 2 cos t)y + 2z
= a cos 2t sin t + a cos 2t sin 2t cos t – a sin 2t cos t – a sin 2t cos 2t cos t + 4a sin t
= a sin (t – 2t) + 4a sin t
= 3a sin t
∴ (sin t + sin 2t cos t)x – 2 cos3 t y + 2z = 3a sin t
This is the equation of the osculating plane.
|2 = 16a4[(sin t + sin 2t cos t)2 + (cos t + cos 2t cos t)2 + 4]
|r × r
= 16a4[sin2 t + sin2 2t cos2 t + 2 sin t sin 2t cos t + cos2 t + cos2 2t cos2 t
+ 2 cos t cos 2t cos t + 4]
4 2 2 2
= 16a [1 + (sin 2t + cos 2t) cos t + 2 cos t (sin t sin 2t + cos t cos 2t) + 4]
= 16a4[1 + cos2 t + 2 cos t cos (t – 2t) + 4]
= 16a4[5 + 3 cos2 t]
2
∴ r | = 4a
| r × 5 + 3 cos 2 t
Also, |r |2 = 4a2 sin2 2t + 4a2 cos2 2t + 4a2 cos2 t
= 4a2 + 4a2 cos2 t = 4a2(1 + cos2 t)
∴ |r |3 = 8a3(1 + cos2 t)3/2
| 4a 2 5 + 3 cos 2 t
| r × r 5 + 3 cos 2 t
∴ κ= 3
= = .
|r | 8a 3 (1 + cos2 t ) 3/ 2 2a (1 + cos 2 t) 3 / 2
Example 10. Show that a curve r = r(s) of class Cm(m ≥ 2) is a straight line if all tangent
lines are concurrent.
Sol. The equation of the tangent line at the point r(s) is
R(s) = r(s) + λ(s)t(s),
where R(s) is a general point on the tangent line and λ(s) is a parameter.
Let all tangent lines intersect at the point r0(s).
∴ r0(s) = r(s) + λ0(s)t(s) for some value λ0(s) of λ(s)
Differentiating w.r.t. s, we get
0 = r′(s) + λ0(s)t′(s) + λ0′(s)t(s)
⇒ 0 = t(s) + λ0(s)t′(s) + λ0′(s)t(s)
⇒ 0 = (1 + λ0′(s))t(s) + λ0(s)t′(s)
48 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

Multiplying by t′(s), we get

⇒ 0 = (1 + λ0′(s)) (t(s) · t′(s)) + λ0(s)(t′(s) . t′(s))
⇒ 0 = (1 + λ0′(s)) · 0 + λ0(s) |t′(s)|2
⇒ λ0(s) |t′(s) |2 = 0 ⇒ | t′(s) | = 0 ⇒ t′(s) = 0 (Assuming λ0(s) ≠ 0)
⇒ t(s) = c, a constant vector
⇒ r′(s) = c ⇒ r(s) = sc + d, where d is a constant vector
∴ The curve r = r(s) is a straight line passing through the point whose position
vector is d and is parallel to the vector c.

WORKING RULES FOR SOLVING PROBLEMS

Rule I. (i) κ = r″ = t′ (ii) κ = | r″| = | t′|
1
Rule II. Radius of curvature, ρ =
κ
|r × |
r
Rule III. (i) κ = | r′ × r″| (ii) κ = .
|r |3
Rule IV. Curve is a straight line if and only if its curvature is identically zero.

EXERCISE 2.1

1. For the curve r = a cos ti + b sin tj, a, b > 0, find the curvature at point t.
2. For the curve r = cosh ti + sinh tj, find the curvature at point t′.
3. For the curve r = ti + t3/2j, t > 0, find the curvature at point t.
4. Show that a curve r = r(t) of class Cm (m ≥ 2), where t is an arbitrary parameter, is a straight line
if r (t) and r
(t) are linearly dependent for all t.
5. For the curve r = ti + t2j + t3k, find the curvature at the point (0, 0, 0).
6. Show that the curvature of a circle of radius a is equal to 1/a.
7. Let r = r(t) be a regular curve of class Cm(m ≥ 2), where t is an arbitrary parameter. Show that:

r ) − ( r . r)2
( r . r ) ( r .
κ= .
( r . r )3/ 2
8. Show that for a curve y = y(x) in the xy-plane:
| y″ |
κ(x) = .
(1 + y ′2 )3 / 2
9. For the following curves in the xy-plane, find curvature: (i) y = x2 (ii) xy = λ.
10. For the curve x = a(3t – t3), y = 3at2, z = a(3t + t3), show that:
1
κ= .
3a ( 1 + t 2 ) 2
11. For the curve x = t, y = t2, z = t3, show that:
4(9t4 + 9t 2 + 1)
κ2 = .
(9t4 + 4t 2 + 1)3
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 49

1
12. For the helix x = a cos t, y = a sin t, z = at cot α, show that: κ = sin2 α.
a
13. Find the curvature of the curve given by r = a(t – sin t)i + a(1 – cos t)j + btk.

3
14. For the curve x = 3t, y = 3t2, z = 2t3, show that: ρ = (1 + 2t2)2.
2

Answers
ab 1 6
1. 2 2 2 2 3/ 2 2. 2 3.
(a sin t + b cos t) cosh 2t t (4 + 9t)3 / 2

2 2λ x3
5. 2 9. (i) (ii)
(1 + 4 x ) 2 3/ 2 ( x + λ2 )3 / 2
4

FG tIJ 1/2

H
a b2 + 4 a2 sin4
2 K .
13.
FG b tI
J
3/ 2

H 2K
2
+ 4 a2 sin2

Hints
4. Let r (t) = λ r
(t) .
7. ( r . r ) (
r . r)2 =|r |2 |
r) − (r . r|2 − (|r ||
r|cos θ)2 = | r |2 | |2 .
r |2 (1 − cos2 θ) = | r × r

3. TORSION OF A CURVE
Let r = r(s) be a regular curve C of class Cm(m ≥ 3) and r″(s) ≠ 0, where s is the param-
eter ‘arc length’.

t ′( s ) t ′( s )
r″(s) ≠ 0 ⇒ n(s) = = ⇒ n(s) is defined.
|t ′( s )| |r ′′( s )|

r ″ (s) r ′ (s) × r″ (s)

Also, b(s) = t(s) × n(s) = r′(s) × = ⇒ b(s) is defined.
|r ″ (s)| |r ″ (s)|
∴ n(s) and b(s) exist at the point r(s).
Since r″′(s) exists, the binormal b(s) is differentiable w.r.t. s.
∴ b′(s) exists.
The scalar quantity – n(s) . b′(s) is called the torsion of the curve C at the point r(s) and
it is denoted by τ(s) (or by τ).
The reciprocal of the torsion is called the radius of torsion at that point and it is
denoted by σ.
1
∴ σ= . (Assuming τ ≠ 0)
τ
50 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

Remark. It can be proved that a curve is uniquely determined (except for its position in space) if
we are given its curvature κ(≠ 0) and torsion τ as continuous functions of arc length s. This result shows
the importance of curvature and torsion in the study of differential geometry of space curves.
Example 1. For the helix r = a cos ti + a sin tj + btk, a > 0, b ≠ 0, find the torsion at the
point t.
Sol. We have r = a cos ti + a sin tj + btk
∴ r = – a sin ti + a cos tj + bk

⇒ | r | = a 2 sin 2 t + a 2 cos 2 t + b2 = a 2 + b2

r 1
∴ t= = (– a sin ti + a cos tj + bk)
| r | a + b2
2

dt dt dt dt ds
t′ = = =
ds dt ds dt dt
1
= (– a cos ti – a sin tj) |r|
a + b2
2

a
= − (cos ti + sin tj) a 2 + b2
2 2
a +b

a
=– (cos ti + sin tj)
a + b2
2

a a
∴ |t′| = (cos2 t + sin2 t)1/2 =
a 2 + b2 a 2 + b2

t′ a a 2 + b2
∴ n= =− 2 (cos t i + sin t j) . = – (cos ti + sin tj)
|t′| a + b2 a

i j k
a a b
∴ b=t×n= − sin t cos t
a 2 + b2 a 2 + b2 a 2 + b2
− cos t − sin t 0
i j k
1
=– − a sin t a cos t b
a 2 + b2 cos t sin t 0
1
=– [– b sin ti + b cos tj – ak]
a + b2
2

db db dt db ds
∴ b′ = = =
ds dt ds dt dt
1
=– (– b cos ti – b sin tj – 0k) |r|
a + b2
2
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 51

1
= (b cos ti + b sin tj) a 2 + b2
a + b2
2

b
= (cos ti + sin tj)
a + b2
2

∴ Torsion, τ= – n · b′
b
= (cos ti + sin tj) . 2 (cos ti + sin tj)
a + b2
b b
= 2 2
. (cos2 t + sin2 t) = 2 .
a +b a + b2
Note. Torsion at each point of a helix is always constant.
Theorem 1. (Serret-Frenet formulae). Let r = r(s) be a regular curve of class Cm(m ≥ 3),
where s is the parameter ‘arc length’ and r″(s) ≠ 0. Prove that
(i) t′ = κn (ii) n′ = – κt + τb (iii) b′ = – τn.
FG t′ IJ = t′
Proof. (i) κn = |r″|n = | t′ |
H|t′|K
∴ t′ = κn.
(iii) (b · t)′ = b · t′ + b′ · t = b · (κn) + b′ · t
= κ(b · n) + b′ · t = 0 + b′ · t = b′ · t
Also, (b . t)′ = 0′ = 0
⇒ b′ . t = 0 ⇒ b′ is perpendicular to t.
Also, |b|=1 ⇒ b · b = 1 ⇒ b · b′ + b′ · b = 0 ⇒ b′ · b = 0
⇒ b′ is perpendicular to b.
∴ b′ is perpendicular to the plane determined by t and b.
∴ b′ is parallel to n.
Let b′ = λn.
⇒ n . b′ = n · (λn) = λ(n · n) = λ · 1 = λ
⇒ –τ=λ (∵ τ = – n . b′)
∴ b′ = – τn.
(ii) We have n = b × t.
∴ n′ = (b × t)′ = b × t′ + b′ × t = b × (κn) + (– τn) × t
(Using (i) and (iii))
= κ(b × n) – τ(n × t) = κ(– t) – τ(– b) = – κt + τb
∴ n′ = – κt + τb.
Remark 1. Serret-Frenet equations shows that we can express the vectors t′, n′, b′ as linear
combinations of the vectors t, n, b.
Remark 2. We have proved equation (iii) before proving equation (ii) because the result of (iii) is
used in proving (ii).
Remark 3. The Serret-Frenet equations can also be written as
t′ = 0t + κn + 0b
n′ = – κt + 0n + τb
b′ = 0t – τn + 0b.
LM 0 κ 0OP
In the above equations, the coefficients of t, n and b form the matrix
MN−0κ PQ
0 τ .
−τ 0
52 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

Remark 4. In the above discussion, t, n, b, t′, n′, b′, κ, τ are all functions of the parameter s. For
the sake of simplicity, we have written t(s) as t etc.
Example 2. Show that along the curve r = r(s), κτ = | t′ · b′ |.
Sol. We have t′ · b′ = (κn) · (– τn) = – κτ(n . n) = – κτ(1) = – κτ
∴ | t′ · b′ | = | – κτ | = κτ.
Example 3. Show that along the curve r = r(s), τ = [t n n′], provided κ ≠ 0.
Sol. n × n′ = n × (– κt + τb) = – κ(n × t) + τ(n × b)
= – κ(– b) + τ(t) = κb + τt
∴ [t n n′] = t · (n × n′) = t · (κb + τt)
= κ(t · b) + τ(t . t) = κ(0) + τ(1) = τ
∴ τ = [t n n′].
[ r′ r″ r′ ″ ]
Example 4. Show that along the curve r = r(s), τ = , provided κ ≠ 0.
κ2
Sol. We have r′ = t and r″ = t′ = κn
dr ″ dt ′ d
∴ r′″ = = = (κn) = κn′ + κ′n
ds ds ds
= κ(– κt + τb) + κ′n (Using n′ = – κt + τb)
= – κ2t + κτb + κ′n
∴ r″ × r′″ = κn × (– κ2t + κτb + κ′n)
= – κ3(n × t) + κ2τ(n × b) + κκ′(n × n)
= – κ3(– b) + κ2τt + κκ′0 = κ3b + κ2τt
∴ [r′ r″ r′″] = r′ · (r″ × r′″) = t · (κ3b + κ2τt)
= κ3(t · b) + κ2τ(t · t) = 0 + κ2τ · 1 = κ2τ
[ r ′ r ″ r ′″ ]
∴ τ= .
κ2
Theorem 2. Let r = r(t) be a regular curve of class Cm(m ≥ 3), where t is an arbitrary
parameter. Prove that
[r
r
r]
τ= , provided | r × r | ≠ 0.
|r × r|2
dr dr ds
Proof. We have r = = = r ′ s = st
dt ds dt
dr d dt

r= = (st) = s +
st = s(t ′ s) +
st
dt dt dt
2
st = st + κs 2 n
= s ( κn) +

∴ r × s( t × t ) + κ s 3 ( t × n)
r = ( s t ) × ( s t + κs 2 n) = s

( 0) + κ s 3 b = κs 3 b
= ss
Differentiating w.r.t. t, we get
3 2
r ×
r +
r × s b + κs3 ( b′ s )
r = κ ′s b + κ3s
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 53

3 2
⇒ r × s b + κs 4 ( − τ n)
r + 0 = ( κ ′ s + 3κs )
3 2 4
⇒ r = ( κ ′ s + 3κs s)b − κ τ s n
r ×

∴ . (r ×
r st + κs 2 n) . (κ ′ s 3 + 3κs 2
r) = ( s) b − κ τ s 4 n
2 6
= − κ s τ (n . n) (Using t · b = 0, t · n = 0, n · b = 0)

⇒ [r r ] = – κ 2 s 6 τ
r
2 6
⇒ – [ r r ] = – κ s τ
r

⇒ [ r r ] = κ 2 s 6 τ
r ...(1)

Also r | = κs 3 |b|= κs 3

| r ×

∴ (1) ⇒ [ r
r |2 τ
r ] = |r × r
[r r
r]
∴ τ= .
|2
| r × r
Corollary. If r = x(t)i + y(t)j + z(t)k, then

x y z
[ r r

r ] =
x y z |2 = Σ( yz
and |r × r yz 2 .
− )

x
y
z

x y z
[ r
r
r]
∴ τ= = x y yz) 2 .
−
z ÷ Σ( yz
|r × r|2
x y
z
Remarks 1. If r = r(s), then
[ r′ r″ r′′′ ]
τ= . (Using κ = | r′ × r ″ |)
|r′ × r″ |2
2. If r = r(t), then
[ r
r r]
τ= .
r |2
| r ×
Theorem 3. Prove that a regular curve of class Cm(m ≥ 3) is a plane curve if and only if
its torsion is identically zero.
Proof. Let r = r(s) be a regular curve of class Cm(m ≥ 3), where s is the parameter ‘arc
length’.
Let the curve be a plane curve.
Let the plane of the curve be normal to the vector a.
∴ r · a = λ, where λ is some constant and r = r(s).
d
⇒ (r . a ) = 0
ds
⇒ r′ · a = 0 ⇒ t · a = 0 ...(1)
54 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

d
(1) ⇒ ( t . a) = 0 ⇒ t′ . a = 0
ds
t′
⇒ .a = 0 ⇒ n.a=0 ...( 2)
|t ′|
(1) and (2) imply that a is perpendicular to the plane of t and n.
∴ a is parallel to the unit binormal vector b.
Let b = µ a.
db d
∴ b′ = = (µ a) = 0
ds ds
∴ τ = – n . b′ = – n . 0 = 0 i.e., the torsion is identically zero.
Conversely, let the torsion of the curve be identically zero.
By Serret-Frenet equation, b′ = – τn.
∴ b′ = 0n = 0 ∴ b = b0, a constant vector.
Let r be any point on the curve.
d
∴ (r . b 0 ) = r′. b0 = t . b0 = 0 (∵ t . b = 0)
ds
∴ r . b0 = constant.
This equation represents a plane.
∴ The curve r = r(s) lies on a plane.
∴ The result holds.
Remark. For a curve to lie in a plane it is sufficient to show that its unit binormal vector b is a
constant vector.
Example 5. Show that the curve r = a cos t i + a sin t j + bk, a > 0, b ≠ 0 is a plane curve.
Sol. We have r = a cos t i + a sin t j + bk
∴ r = – a sin t i + a cos t j + 0k

| r | = a 2 sin 2 t + a 2 cos 2 t = a
r 1
∴ t = = (– a sin t i + a cos t j)
| r| a
= – sin t i + cos t j

∴ t = – cos t i – sin t j and | t | = cos 2 t + sin 2 t = 1

t
∴ n=
|t |
1
b g
= − cos t i + sin t j = – (cos t i + sin t j)
1

i j k
∴ b = t × n = − sin t cos t 0 = k
− cos t − sin t 0

∴ b is a constant vector.
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 55

⇒ b′ = 0 ⇒ τ = – n . b′ = – n . 0 = 0.
∴ The given curve is a plane curve.
Alternative method
− a sin t a cos t 0
[ r r r ] = − a cos t − a sin t 0 = 0

a sin t − a cos t 0
[ r
r
r] 0
∴ τ= 2
= = 0.
|r × r |2
| |r × r
∴ The given curve is a plane curve.
F 1 + t 1 − t2 I lies on a plane.
Example 6. Show that the curve r = t, GH t
,
t JK
Sol. The given curve is
1+ t 1 − t2
r = ti + j+ k
t t
FG
1 1 IJ FG IJ
∴
H
r = i + − 2 j + − 2 − 1 k
t t K H K
2 2
r= j+ 3 k
3
t t
6 6
and r= −
4
j− k
t t4
1 1
1 − 2
− −1
t t2
2 2 12 12
∴ [ r r r] = 0
= − + =0
t3 t3 t 7
t7
6 6
0 − 4 − 4
t t
[ r
r
r] 0
∴ τ= 2
= = 0.
|r × r |2
| |r × r
∴ The given curve lies on a plane.
Example 7. Find the torsion of the curve r = ti + t2j + t3k at the point ‘t’.
Sol. We have r = ti + t2j + t3 k.
∴ r = i + 2tj + 3t2k,

r = 2j + 6tk and r = 6k
[ r
r
r]
Now, τ= ...(1)
|r × r|2
i j k
r ×
r = 1 2 t 3t2 = 6t2i – 6tj + 2k
0 2 6t
56 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

∴ r |2 = 36t4 + 36t2 + 4 = 4(1 + 9t2 + 9t4)

| r ×
1 2t 3t 2
[ r r

r ] = 0 2 6t = 12
0 0 6
12 3
∴ (1) ⇒ τ= 2 . 4
=
4 (1 + 9t + 9t ) 1 + 9t + 9t 4 2

Example 8. For the curve r = (at – a sin t)i + (a – a cos t)j + btk, find the torsion at the
point ‘t’.
Sol. We have r = (at – a sin t)i + (a – a cos t)j + btk
∴ r = (a – a cos t)i + a sin tj + bk

r = a sin t i + a cos tj + 0k and r = a cos ti – a sin tj + 0k
[ r
r
r]
Now, τ= ...(1)
|r × r|2
i j k
r = a − a cos t a sin t b
r ×
a sin t a cos t 0
= – ab cos t i + ab sin t j + (a2 cos t – a2)k
∴ r |2 = a2b2 cos2 t + a2b2 sin2 t + a 4 (cos t – 1)2
| r ×

FG t IJ 2
2 2LM 2 4 t OP
= a2b2 + a4 − 2 sin 2
H 2 K = a b + 4a sin
N 2 Q
a − a cos t a sin t b
Also, [ r r

r] = a sin t a cos t 0 = b[– a2sin2 t – a2cos2 t] = – a2b
a cos t − a sin t 0
− a 2b b
∴ (1) ⇒ τ=
FG b t IJ =–
t
.
a2
H K
2
+ 4a 2 sin 4 b2 + 4a 2 sin 4
2 2
Example 9. For the curve x = a tan t, y = a cot t, z = 2 a log tan t, prove that
2 2a
ρ=σ= .
sin 2 2t
Sol. The given curve is
x = a tan t, y = a cot t, z = 2 a log tan t.
Let r be the position vector of the point (x, y, z) on the curve.
∴ r = xi + yj + zk = a tan ti + a cot tj + 2 a log tan tk
dt
∴ t = r′ = r
ds
F sec 2 t Idt
GH2 2
= a sec ti − a cosec tj + 2 a
tan t
kJK
ds
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 57

F 2 dt I
∴ GH 2 2
t = a sec ti − cosec tj +
sin t cos t
k
ds JK ...(1)

F
| t | = a G sec
2 I 1/ 2
dt
⇒
H
4 4
t + cosec t + 2
sin t cos 2 J
tK ds

ds F sin
.1 = a G
4
t + cos 4 t + 2 sin 2 t cos 2 tI
1/2
a
⇒
dt H sin 4 t cos 4 t
JK =
sin t cos 2 t
2

∴
dt sin 2 t cos 2 t
=
ds a

F 2 2 I 2
∴ (1) ⇒ GH
t = a sec 2 ti − cosec 2 tj +
sin t cos t JK
k . sin t cos t
a

or t = (sin2 ti – cos2 tj + 2 sin t cos tk)

dt sin 2 t cos 2 t
∴ t′ = t = (2 sin t cos ti + 2 cos t sin tj + 2 cos 2tk) .
ds a

sin 2 t cos 2 t
⇒ κn = (sin 2ti + sin 2tj + 2 cos 2tk) ...(2)
a
(Using t′ = κn)

sin 2 t cos 2 t 2 sin 2 t cos 2 t

⇒ κ| n | = (sin2 2t + sin2 2t + 2 cos2 2t)1/2 =
a a

2 sin 2 t cos 2 t 1 a
∴ κ= and ρ = =
a κ 2 sin t cos 2 t
2

1
∴ (2) ⇒ n= (sin 2ti + sin 2tj + 2 cos 2tk)
2

dt 1 sin 2 t cos 2 t
⇒ n′ = n = (2 cos 2ti + 2 cos 2tj – 2 2 sin 2tk)
ds 2 a
2 sin 2 t cos 2 t
⇒ τb – κt = (cos 2ti + cos 2tj – 2 sin 2tk)
a
(Using n′ = τb – κt)
⇒ (τb – κt) · (τb – κt)
2 sin 4 t cos 4 t
= (cos2 2t + cos2 2t + 2 sin2 2t)
a2
4 sin 4 t cos 4 t
⇒ τ2 + κ2 =
a2
4 sin 4 t cos 4 t 2 sin 4 t cos 4 t 2 sin 4 t cos 4 t
⇒ τ2 = − =
a2 a2 a2
58 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

2 sin 2 t cos 2 t 1 a
∴ τ= and σ = =
a τ 2 sin t cos 2 t
2

a 2 2a
∴ ρ=σ= = .
2
2 sin t cos t 2
sin 2 2t
Example 10. Prove that for the curve of intersection of the surfaces x2 + y2 = z2 and
y
z = a tan–1 :
x
a (2 + θ 2 )3/2 a (8 + 5θ 2 + θ 4 )
ρ= and σ = , where y = x tan θ.
(8 + 5 θ 2 + θ 4 )1/2 6 + θ2
Sol. Given surfaces are
y
x2 + y2 = z2 ...(1) and z = a tan–1 ...(2)
x
Let y = x tan θ ∴ (2) ⇒ z = a tan–1 (tan θ) = aθ
(1) ⇒ x2 + x2 tan2 θ = a2θ2 ⇒ x2 sec2 θ = a2θ2 ⇒ x = aθ cos θ
∴ y = (aθ cos θ) tan θ = aθ sin θ
∴ The parametric equations of the given curve are
x = aθ cos θ, y = aθ sin θ, z = aθ.
Let r be the position vector of the point (x, y, z) on the curve.
∴ r = xi + yj + zk = aθ cos θi + aθ sin θj + aθk
∴ r = a [(cos θ – θ sin θ)i + (sin θ + θ cos θ)j + k]
r = a [(– 2 sin θ – θ cos θ)i + (2 cos θ – θ sin θ)j]

r = a [(– 3 cos θ + θ sin θ) i + (– 3 sin θ – θ cos θ)j]

i j k
cos θ − θ sin θ sin θ + θ cos θ 1
∴ r = a2
r ×
− 2 sin θ − θ cos θ 2 cos θ − θ sin θ 0

= a2 [(– 2 cos θ + θ sin θ)i + (– 2 sin θ – θ cos θ)j

+ (2 cos2 θ – θ sin θ cos θ – 2 θ sin θ cos θ + θ2 sin2 θ
+ 2 sin2 θ + 2 θ sin θ cos θ + θ sin θ cos θ + θ2 cos2 θ)k]
= a2 [(– 2 cos θ + θ sin θ)i – (2 sin θ + θ cos θ)j + (2 + θ2)k]
[ r r ] = r . (r
r ×
r ) = (r ×
r ) .
r
= a2 [(– 2 cos θ + θ sin θ)i – (2 sin θ + θ cos θ)j
+ (2 + θ2)k] . a [(– 3 cos θ + θ sin θ)i – (3 sin θ + θ cos θ)j]
= a3 [(6 cos2 θ + θ2 sin2 θ – 5θ sin θ cos θ)
+ (6 sin2 θ + θ2 cos2 θ + 5θ sin θ cos θ)]
= a3(6 + θ2)
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 59

| r |2 = a2 [(cos2 θ + θ2 sin2 θ – 2θ cos θ sin θ)

+ sin2 θ + θ2 cos2 θ + 2θ sin θ cos θ + 1]
= a2 (2 + θ2)
2
|r × r|
= a4[(4 cos2 θ + θ2 sin2 θ – 4θ cos θ sin θ)
+ (4 sin2θ + θ2 cos2 θ + 4θ sin θ cos θ) + (2 + θ2)2]
= a4 [4 + θ2 + 4 + θ4 + 4θ2] = a4 [8 + 5θ2 + θ4]

1 |r |3 a 3 ( 2 + θ 2 )3 / 2 a( 2 + θ 2 )3/ 2
∴ ρ= = = 2 =
κ |r × r
| a ( 8 + 5θ 2 + θ 4 )1/ 2 ( 8 + 5θ 2 + θ 4 )1/ 2

1 |r × r|2 a 4 ( 8 + 5θ 2 + θ 4 ) a( 8 + 5θ 2 + θ 4 )
and σ= = = = .
τ [ r
r
r] a3 (6 + θ2 ) 6 + θ2
Example 11. For a point on the curve of intersection of the surfaces x2 – y2 = c2 and
z 2x 2
y = x tanh , show that ρ = σ = .
c c
Sol. Given curve is
z
x2 – y2 = c2 ...(1) y = x tanh ...(2)
c
Let x = c cosh t, y = c sinh t.
∴ (1) is satisfied.
z
(2) ⇒ c sinh t = c cosh t tanh
c
z
⇒ tanh = tanh t ⇒ z = ct
c
∴ The parametric equations of the given curve are
x = c cosh t, y = c sinh t, z = ct.
Let r be the position vector of the point (x, y, z) on the curve.
∴ r = xi + yj + zk = c cosh t i + c sinh t j + ctk
∴ r = c sinh t i + c cosh t j + ck
r = c cosh t i + c sinh t j

and r = c sinh t i + c cosh t j

∴ | r | = c(sinh2 t + cosh2 t + 1)1/2 = c(2 cosh2 t)1/2 = 2 c cosh t
i j k i j k
r =
r × c sinh t c cosh t c = c2 sinh t cosh t 1
c cosh t c sinh t 0 cosh t sinh t 0
= c2 [– sinh ti + cosh tj – k]
∴ r | = c2 (sinh2 t + cosh2 t + 1)1/2 = c2(2 cosh2 t)1/2 =
| r × 2 c2 cosh t

|r ×
r| 2 c 2 cosh t 1
∴ κ= = =
|r |3 ( 2 c cosh t) 3 2c cosh 2 t
60 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

1 x 2 x2FG IJ 2
∴ ρ=
κ
= 2c cosh2 t = 2c
c
=
c H K
.

c sinh t c cosh t c
[ r r r] = c cosh t c sinh t 0 = c (c2 cosh2 t – c2 sinh2t) = c3

c sinh t c cosh t 0
[ r r

r] c3 1
∴ τ= = =
|r × r| 2 2
( 2 c cosh t) 2
2c cosh 2 t
1 x FG IJ 2
2x 2
∴ σ=
τ
= 2c cosh2 t = 2c
c H K =
c
.

2x 2
∴ ρ=σ= .
c
Example 12. Determine the function f(u) so that the curve given by r = (a cos u, a sin u, f(u))
should be a plane curve.
Sol. The given curve is
r = a cos ui + a sin uj + f(u)k.
Given curve is a plane curve iff τ = 0
[ r
r
r]
iff = 0 iff [ r
r
r ] = 0.
|r × r|2
∴ Given curve is a plane curve iff [ r
r
r ] = 0.
We shall choose f(u) so that we may have [ r r
r ] = 0.
r = – a sin ui + a cos uj + f (u)k
r = – a cos ui – a sin uj + f (u)k

r = a sin ui – a cos uj +
f (u)k
− a sin u a cos u f(u )
[ r r r] = − a cos u
− a sin u f(u )
a sin u − a cos u
f (u )

0 0 f(u ) +
f (u )
= − a cos u − a sin u f(u ) (Operating R1 → R1 + R3)
a sin u − a cos u
f (u )

= ( f (u) +
f (u)) (a2 cos2 u + a2 sin2 u) = a2 ( f (u) +
f (u))
∴ a 2 ( f(u) +
f (u)) = 0 (∵ [ r
r
r ] = 0)
f (u) +
f (u) = 0
⇒
Integrating, we get f(u) + f(u) = c1
⇒ f(u) = c1 – f(u) ⇒ 2 f(u) f(u) = 2(c1 – f(u)) f (u)
Integrating, we get
( f(u)) 2 = – (c1 – f(u))2 + c2

⇒ f (u) = c2 − ( c1 − f (u)) 2
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 61

d( f (u))
⇒ = du
c2 − ( c1 − f (u)) 2
Integrating, we get
c1 − f (u)
– sin–1 = u + c3
c2
c1 − f (u)
⇒ sin (– (u + c3)) =
c2
⇒ – c2 sin (u + c3) = c1 – f (u)

⇒ f (u) = c2 sin (u + c3) + c1

= c2 (sin u cos c3 + cos u sin c3) + c1

= c2 cos c3 sin u + c2 sin c3 cos u + c1

∴ f (u) = A sin u + B cos u + C,
where A = c2 cos c3, B = c2 sin c3, C = c1 are arbitrary constants.
Example 13. If the tangent and binormal at a point of a curve make angles θ and φ with
sin θ dθ κ
a fixed direction, show that: =− .
sin φ dφ τ
Sol. Let the equation of the curve be r = r(s), where the parameter ‘s’ is arc length.
Let the tangent and binormal at the point P of the curve make angles θ and φ with
vector a which is along the given fixed direction.
∴ Angle between t and a is θ and angle between b and a is φ.
∴ t · a = a cos θ ...(1) (∵ | t | = 1)
and b · a = a cos φ ...(2) (∵ | b | = 1)
Differentiating (1) w.r.t. s, we get
dθ
t′ · a = – a sin θ
ds
dθ
⇒ κn · a = – a sin θ ...(3)
ds
Differentiating (2) w.r.t. s, we get
dφ
b′ · a = – a sin φ
ds
dφ
⇒ – τn · a = – a sin φ ...(4)
ds
62 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

Dividing (3) by (4), we get

dθ
− a sin θ
κ (n . a) ds
=
− τ (n . a) dφ
− a sin φ
ds

κ sin θ dθ
⇒ –= .
τ sin φ dφ
Example 14. For the curve r = r(s), if
dt dn db
= w × t, = w × n and = w × b, find the vector w.
ds ds ds
Sol. Given equations are
dt dn db
=w×t ...(1) = w × n ...(2) =w×b ...(3)
ds ds ds
dt
By Frenet formula, = κn .
ds
dt
⇒ = 0 + κn = τ(t × t) + κ(b × t) = (τt + κb) × t
ds
dt
∴ = (τt + κb) × t ...(4)
ds
dn
By Frenet formula, = – κt + τb.
ds
dn
⇒ = – κ(n × b) + τ(t × n) = κ(b × n) + τ(t × n)
ds
= (κb + τt) × n = (τt + κb) × n
dn
∴ = (τt + κb) × n ...(5)
ds
db
By Frenet formula, = – τn.
ds
db
⇒ = – τ(b × t) + 0 = τ(t × b) + κ( b × b ) = (τt + κb) × b
ds
db
∴ = (τt + κb) × b ...(6)
ds
If w = τt + κb, then given equations (1), (2) and (3) are satisfied.
Note. The vector w = τt + κb is called the Darboux vector for the curve r = r(s).
Example 15. Using Serret-Frenet formula, find the direction cosines of the unit principal
normal vector and the unit binormal vector at the point ‘s’ for the curve r = r(s).
Sol. Given curve is r = r(s).
Let r be the position vector of the point (x, y, z) on the curve.
∴ r = xi + yj + zk
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 63

We have t′ = κn.
t ′ ( r′ )′ r″ 1 x″ y″ z″
⇒ n= = = = (x″i + y″j + z″k) = i+ j+ k
κ κ κ κ κ κ κ
F x″ 2
y″ 2 z″ 2 I 1/ 2

|n|=1 ⇒ GH κ 2
+
κ2
+
κ2
JK =1 ⇒ κ2 = x″2 + y″2 + z″2

∴ κ= x ″ 2 + y ″ 2 + z″ 2
x ′′ y ′′ z ′′
∴ n= i+ j+ k
2 2 2 2 2 2
x ″ + y″ + z″ x ″ + y″ + z″ x ″ + y″ 2 + z″ 2
2

Since n is a unit vector, the d.c.’s of n are

x ′′ y ′′ z ′′
, , .
2
x ″ + y″ + z″ 2 2 2
x ″ + y″ + z″ 2 2 x ″ + y″ 2 + z″ 2
2

i j k
t′ 1 1
b = t × n = r′ × = ( r′ × r″ ) = x′ y′ z′
κ κ κ x″ y″ z″
FG κ = t′ t′ t′ IJ
H ⇒ κ = |t ′ | ⇒ n = =
|t ′ | κ K
1
= [(y′z″ – y″z′)i + (z′x″ – z″x′)j + (x′y″ – x″y′)k]
κ
y ′z ″ − y ″z ′ z ′x ″ − z ″x ′ x ′y ″ − x ″y ′
= i+ j+ k
x ″ 2 + y″ 2 + z″ 2 x ″ 2 + y″ 2 + z ″ 2 x ″ 2 + y″ 2 + z ″ 2
Since b is a unit vector, the d.c.’s of b are
y ′z ″ − y ″z ′ z ′x ″ − z ″x ′ x ′y ″ − x ″ y ′
, , .
x ″ 2 + y″ 2 + z″ 2 x ″ 2 + y″ 2 + z″ 2 x ″ 2 + y″ 2 + z″ 2
Example 16. Let r = r(t) be a curve. Prove that:
(i) r = st r = st + κ s 2 n
(ii)
(iii) s − κ 2 s 3 ) t + s (3 κs − κ s)
r = ( n + κτs 3 b .
Hence deduce that:
r −
s s r r × r
(a) n = 3 (b) b =
κs κs 3
2 2
|
r | −
s [
r
r
r]
(c) κ2 = 4 (d) τ = 2 6 .
s κ s
dr dr ds
Sol. (i) r = = = r′ s = st .
dt ds dt
d d dt
(ii)
r= (r ) = ) = s
(st +
st
dt dt dt
2
= s (t ′ s) + st = s ( κn) + s t + κs 2 n .
s t =
64 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

d d
(iii)
r= (
r) = st + κs 2 n)
(
dt dt
FG dt IJ FG dn IJ
=
s
H dt
+
K H
s t + κ s 2 n + κ 2s
s n + κs 2
dt K
s t + κ s 2 n + 2κss
= s ( t′ s ) + n + κs 2 n′ s

s t + κ s 2 n + 2κss
( κn) +
= ss n + κs 3 ( − κt + τb)
s − κ 2 s 3 )t + s(3κ
= ( s + κ s)n + κτ s3 b .
(a) r − sr = s( st + κs 2 n) − s ( st ) = κs 3 n
s

r −
s sr
∴ n= 3 .
κs
(b) r × st + κs 2 n) = s
r = ( st ) × ( s ( t × t ) + κs 3 ( t × n)
= s s ( 0) + κs 3 b = κs3 b
r × r
∴ b= .
κs 3
(c) |2 = r . r = ( st + κs 2 n) . ( st + κs 2 n)
|r

= s2 ( t . t ) + κs 2
s ( t . n) + κs 2
s ( n . t ) + κ 2 s 4 ( n . n)
s 2 (1) + 0 + 0 + κ 2 s 4 (1) =
= s 2 + κ 2 s 4
∴ |r|2 –
s 2 = κ 2s 4

r |2 −
| s2
∴ κ2 = .
s 4
(d) [r
r r × )
r ] = r . ( r = (r ×
r ) .
r
= (κs 3 b) . [(
s − κ 2 s 3 )t + s (3κs + κ s)n + κτ s 3 b] (Using (iii) and (b))
2 6 2 6
= κ s τ ( b . b) = κ s τ (Using b . t = 0, b . n = 0)
[ r
r
r]
∴ τ= 2 6
.
κ s
Example 17. Let r = r(s) be a curve. Prove that:
(i) r′ . r″ = 0 (ii) r′″ = – κ2t + κ′n + κτb
(iii) r′ . r′″ = – κ2 (iv) r″ . r′″ = κκ′
3 2
(v) r″″ = – 3κκ′t + (κ″ – κ – κτ )n + (2κ′τ + τ′κ)b
(vi) r′ . r″″ = – 3κκ′ (vii) r″ . r″″ = κ(κ″ – κ3 – κτ2)
(viii) r′″ . r″″ = κ′κ″ + 2κ3κ′ + κ2ττ′ + κκ′τ2

d FG τ IJ d FG κ IJ .
(ix) [t′ t″ t′″] = κ3(κτ′ – κ′τ) = κ5
ds H κK (x) [b′ b″ b′″] = τ3 (κ′τ – κτ′) = τ5
ds H τK
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 65

Sol. (i) r′· r″ = t · t′ = t · (κn) = κ (t · n) = κ · 0 = 0. (∵ r′ = t and t′ = κn)

(ii) r′″ = (r′)″ = t″ = (t′)′ = (κn)′ = κn′ + κ′n = κ(– κt + τb) + κ′n
= – κ2t + κ′n + κτb. (∵ n′ = – κt + τb)
(iii) r′ · r′″ = t · (– κ2t + κ′n + κτb) (Using (iii))
= – κ (t · t) + κ′(t · n) + κτ(t · b) = – κ · 1 + 0 + 0 = – κ .
2 2 2

(iv) r″ · r′″ = (r′)′ · r′″ = t′ · r′″

= (κn) · ( – κ2t + κ′n + κτb) (Using (iii))
3 2
= – κ (n · t) + κκ′(n · n) + κ τ(n · b) = 0 + κκ′ · 1 + 0 = κκ′.
(v) r″″ = (r′″)′ = (– κ2t + κ′n + κτb)′ (Using (iii))
2
= – (κ t′ + 2κκ′t) + (κ′n′ + κ″n) + (κ′τb + κτ′b + κτb′)
= – κ2 (κn) – 2κκ′t + κ′ (– κt + τb) + κ″n + κ′τb + κτ′b + κτ (– τn)
= – κ3n – 2κκ′t – κκ′t + κ′τb + κ″n + κ′τb + κτ′b – κτ2n.
= – 3κκ′t + (κ″ – κ3 – κτ2)n + (2κ′τ + τ′κ) b.
(vi) r′ · r″″ = t . [– 3κκ′t + (κ″ – κ3 – κτ2)n + (2κ′τ + τ′κ)b] (Using (v))
= – 3κκ′ (t · t) + (κ″ – κ3 – κτ2)(t . n) + (2κ′τ + τ′κ) (t . b)
= – 3κκ′ · 1 + 0 + 0 = – 3κκ′.
(vii) r″ · r″″ = t′ · r″″
= (κn) · (– 3κκ′t + (κ″ – κ3 – κτ2)n + (2κ′τ + τ′ κ)b]
= – 3κ2κ′ (n · t) + κ(κ″ – κ3 – κτ2) (n · n) + κ(2κ′τ + τ′κ) (n · b)
= 0 + κ(κ″ – κ3 – κτ2) · 1 + 0 = κ(κ″ – κ3 – κτ2).
(viii) r′″ · r″″ = (– κ2t + κ′n + κτb). (– 3κκ′t + (κ″ – κ3 – κτ2) n + (2κ′τ + τ′κ)b)
(Using (ii) and (v))
= 3κ3κ′ (t · t) + κ′ (κ″ – κ3 – κτ2) (n · n) + κτ(2κ′τ + τ′κ) (b · b)
= 3κ3κ′ + κ′κ″ – κ′κ3 – κκ′τ2 + 2κκ′τ2 + κ2ττ′
= κ′κ″ + 2κ3κ′ + κ2ττ′ + κκ′τ2.
(ix) [t′ t″ t′″] = [r″ r′″ r″″]
0 κ 0
= − κ2 κ′ κτ
− 3κκ ′ κ ″ − κ 3 − κτ 2 2κ ′ τ + τ ′ κ
(∵ r″ = t′ = κn and the vectors t, n, b form a right handed triad)
= – κ[– κ2(2κ′τ + τ′κ) + 3κκ′(κτ)] = 2κ3κ′τ + κ4τ′ – 3κ3κ′τ
= κ4 τ′ – κ3 κ′ τ = κ3(κτ′ – κ′ τ)
FG κτ′ − κ ′ τ IJ
= κ5 d τ . FG IJ
= κ5 ·
H κ K 2
ds κ H K
(x) We have b′ = – τn = 0t – τn + 0b
b″ = (b′)′ = (– τn)′ = – τn′ – τ′n
= – τ(– κt + τb) – τ′n = τκt – τ′n – τ2b
b′″ = (b″)′ = (τκt – τ′n – τ2b)′
= (τ′κt + τκ′t + τκt′) – (τ″n + τ′n′) – (2ττ′b + τ2b′)
= τ′κt + τκ′t + τk(κn) – τ″n – τ′(– κt + τb) – 2ττ′b – τ2 (– τn)
= τ′κt + τκ′t + τκ2n – τ″n + τ′κt – τ′τ b – 2ττ′b + τ3n
= (2τ′κ + τκ′) t + (τκ2 – τ″ + τ3) n – 3ττ′b
66 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

0 −τ 0
∴ [b′ b″ b′″] = τκ − τ′ − τ2
2τ′ κ + τκ ′ τκ 2 − τ ″ + τ 3 − 3 ττ′
(∵ t, n, b form a right handed triad)
= τ [τκ (– 3ττ′) + (2τ′κ + τκ′)τ2]
= τ [– 3τ2τ′κ + 2τ2τ′κ + τ3κ′]
= τ[ – τ2τ′κ + τ3κ′] = τ3(κ′τ – κτ′)

FG τκ′ − κτ′ IJ FG IJ
d κ
H τ K H K
5
= τ5 2
= τ .
ds τ

WORKING RULES FOR SOLVING PROBLEMS

Rule I. τ = – n · b′
1
Rule II. Radius of torsion, σ =
τ
Rule III. Serret-Frenet Formula:
(i) t′ = κn (ii) n′ = – κt + τb (iii) b′ = – τn
[r ′ r ″ r ′ ″ ] [r
r
r]
Rule IV. (i) τ = 2 (ii) τ =
|r ′ × r ″ | |r × r|2

Rule V. Curve is a plane curve if and only if its torsion is identically zero.

EXERCISE 2.2

1. Find the torsion of the curve r = ti + t2j + t3k at the point where t = 2.
2. Show that the torsion of a plane curve (with κ > 0) is identically zero.
3. Find the torsion of the curve r = (3t – t3)i + 3t2j + (3t + t3)k at point t.
4. Find the torsion of the curve r = (t – sin t)i + (1 – cos t)j + tk at point t.
5. For the curve x = a(3t – t3), y = 3at2, z = a(3t + t3), show that curvature κ and torsion τ each is
1
equal to .
3a(1 + t2 )2
6. Find the torsion of the helix x = a cos t, y = a sin t, z = at tan α at point t.
7. For the curve x = 3t, y = 3t2, z = 2t3, show that:
2
κ=τ= .
3(1 + 2 t2 )2
8. For a point on the curve of intersection of the surfaces x2 + y2 = a2, x2 – y2 = az, find the torsion.
9. Find the torsion at any point t of the curve x = a cos 2t, y = a sin 2t, z = 2a sin t.
10. Let r = r(t) be a regular curve of class Cm (m ≥ 3), where t is an arbitrary parameter. Prove that
[ r r

r]
τ= , provided κ ≠ 0.
(r . r ) (r
. r r)2
) − (r .
11. Show that the curve r = r(t) is a plane curve if and only if [r r

r] = 0.
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 67

12. Is x = a cos t, y = a sin t, z = ct, a > 0, c ≠ 0 a plane curve ? Calculate the curvature and torsion of
the above curve at point ‘t’.
13. Show that the position vector of the current point on the curve r = r(s) satisfies the differential
equation:

d LMd d 2rF I OP + d LM σ dr OP + ρ d r = 0. 2

ds
σ
MN
ds
ρ 2
ds
GH JK PQ ds N ρ ds Q σ ds 2

Answers
3 2 1
1. 3. 4. –
181 3(1 + t2 )2 1 + 4 sin 4
t
2
6 a2 − z2 3
6. a sec2 α 8. 2 2
9.
5a + 12 z a (5 sec t + 3 cos t)
a c
12. , .
a2 + c2 a2 + c2
Hint
2
1 1 dr d r
13. Using ρ = ,σ= , = r′′ = t and = r″ = t′ = κn, we get
κ τ ds ds2
d LM F
d 2r I OP + d L σ dr O + ρ d r 2

ds
σ
d
MN
ds
ρ 2
ds
GH JK PQ ds MN ρ ds PQ σ ds 2

LM 1 F 1 I ′ OP′ L κ O′ τ L1 O′ L κ O′
= G . κnJ + M tP + ( κn) = M ( − κt + τb)P + M tP + τn
MN τ H κ K PQ N τ Q κ Nτ Q Nτ Q
L κ O′ L κ O′
= – M tP + b′ + M t P + τn = – τn + τn = 0.
Nτ Q Nτ Q
4. CONTACT OF A CURVE WITH A SURFACE
In this section, we shall study the degree of contact of a curve with a surface.
A curve r = x(t)i + y(t)j + z(t)k of sufficiently high class is said to have n-point contact
(or contact of (n – 1)th order) with a surface F(x, y, z) = 0 at the point corresponding to t0 if
the function f(t) = F(x(t), y(t), z(t)) satisfies :

f(t0) = f( t0 ) = f(t0 ) = ...... = f

( n − 1)
(t0 ) = 0 and f ( n ) ( t0 ) ≠ 0 .

Example 1. Show that the curve r = ti + t2j + t3k has 6-point contact with the paraboloid
x2 + z2 – y = 0 at the origin.
Sol. Given curve is r = ti + t2j + t3k.
t = 0, t2 = 0, t3 = 0 ⇒ t = 0
∴ The origin corresponds to t = 0.
68 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

Given surface is x2 + z2 – y = 0.
Let f(t) = t2 + (t3)2 – t2 = t6

∴ f( t ) = 6t5, f( t ) = 30t4,

f ( t ) = 120t3, f (4)(t) = 360t2, f (5)(t) = 720t, f (6)(t) = 720
∴ f( 0) = 0, f( 0) = 0,
f ( 0) = 0, f (4) (0) = 0, f (5)(0) =0 and f (6) (0) ≠ 0.
∴ The given curve has 6-point contact with the given paraboloid at the origin.
Example 2. If the circle lx + my + nz = 0, x2 + y2 + z2 = 2cz has 3-point contact with the
l2 + m2
paraboloid ax2 + by2 = 2z at the origin then show that: c = .
bl 2 + am 2
Sol. Given circle is lx + my + nz = 0 ...(1) x2 + y2 + z2 = 2cz ...(2)
Let the parametric equations of this circle be x = φ1(t), y = φ2(t), z = φ3(t).
Substituting the values of x, y, z in (1) and (2) and differentiating w.r.t. t, we get
lx + my + nz = 0 ...(3)
and 2 xx + 2 yy + 2 zz = 2 cz
or xx + yy + zz = cz ...(4)
The circle passes through the origin.
∴ (4) ⇒ 0 x + 0 y + 0 z = cz ⇒ cz = 0 ⇒ z = 0
∴ (3) ⇒ lx + my + n(0) = 0 ⇒ lx + my = 0
x y
⇒ = = λ, say ...(5)
m –l
The paraboloid is ax2 + by2 = 2z. After substituting the values of x, y, z in terms of t, let
f(t) = ax2 + by2 – 2z ...(6)
∴ f (t) = 2axx + 2byy − 2 z ...(7)

f(t) = 2 a[ x 2 + xx] + 2b [ y 2 + yy
] − 2
z
or f(t) = 2ax 2 + 2axx + 2by 2 + 2byy − 2 z ...(8)
Since the circle has 3-point contact with the paraboloid at the origin, we have
f(t) = f (t) = f(t) = 0 ,
f (t) ≠ 0 at the origin.
∴ (6) ⇒ a(0) + b(0)2 – 2(0) = 0
2 ...(9)
(7) ⇒ 2a(0) x + 2b(0) y – 2 z = 0 ...(10)
2
(8) ⇒ 2a x + 2a(0) x + 2b y + 2b(0)
2 y – 2
z=0 ...(11)
(9) ⇒ 0 = 0, which is always true.
(10) ⇒ z = 0, which is also true.
(11) ⇒ 2ax 2 + 2by 2 − 2 z = 0 ⇒ ax 2 + by 2 =
z ...(12)
Differentiating (4) w.r.t. t, we get
x 2 + xx + y 2 + yy
+ z 2 + zz = cz .
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 69

Since the circle passes through the origin, we have

x 2 + 0 x + y 2 + 0 y + z 2 + 0
z = cz.

⇒ x 2 + y 2 + z 2 = cz

⇒ x 2 + y 2 = cz ...(13)(∵ z = 0 )
Dividing (13) by (12), we get

x 2 + y 2 m 2 λ2 + l 2 λ2 m2 + l 2
c= = 2 2 2 2 = . (Using (5))
2
ax + by 2
am λ + bl λ am2 + bl 2

l2 + m2
∴ c= .
bl 2 + am2
Example 3. Find the equation of the plane that has 3-point contact with the curve
x = t4 – 1, y = t3 – 1, z = t2 – 1 at the origin.
Sol. Given curve is
x = t4 – 1, y = t3 – 1, z = t2 – 1.
t4 – 1 = 0, t3 – 1 = 0, t2 – 1 = 0 ⇒ t = 1
∴ The origin corresponds to t = 1.
Let the equation of the required plane through the origin be ax + by + cz = 0.
Let f(t) = a(t4 – 1) + b(t3 – 1) + c(t2 – 1)
∴ f(t) = 4at3 + 3bt2 + 2ct

f(t) = 12at2 + 6bt + 2c

f (t) = 24at + 6b
Since the plane has 3-point contact at t = 1, we have

f(1) = 0, f(1) = 0, f(1) = 0 and

f (1) ≠ 0 .
f(1) = 0 ⇒ 0=0

f(1) = 0 ⇒ 4a + 3b + 2c = 0 ...(1)

f(1) = 0 ⇒ 12a + 6b + 2c = 0 ...(2)

a b c
(1) and (2) ⇒ = =
6 − 12 24 − 8 24 − 36

a b c a b c
⇒ = = ⇒ = =
−6 16 − 12 3 −8 6
Let a = 3, b = – 8, c = 6.
Also
f (1) = 24a + 6b = 24(3) + 6(– 8) = 24 ≠ 0
∴ The equation of the required plane is 3x – 8y + 6z = 0.
70 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

Example 4. Find the lines that have 4-point contact with the surface x4 + 3xyz + x2 – y2
– z2 + 2yz – 3xy – 2y + 2z – 1 = 0 at the point (0, 0, 1).
x−0 y−0 z−1
Sol. Let = = =t ...(1)
a b c
be a line passing through (0, 0, 1).
∴ x = at, y = bt, z = ct + 1
∴ The point (0, 0, 1) corresponds to the value t = 0.
Given surface is
x4 + 3xyz + x2 – y2 – z2 + 2yz – 3xy – 2y + 2z – 1 = 0.
Let f (t) = (at)4 + 3(at)(bt)(ct + 1) + (at)2 – (bt)2 – (ct + 1)2 + 2(bt)(ct + 1)
– 3(at)(bt) – 2(bt) + 2(ct + 1) – 1
∴ f (t) = a4t4 + 3abt2(ct + 1) + a2t2 – b2t2 – (ct + 1)2 + 2bt(ct + 1) – 3abt2 – 2bt + 2ct + 1
∴ f (t) = 4a3t3 + 9abct2 + 6abt + 2a2t – 2b2t – 2(ct + 1) c + 4bct + 2b – 6abt – 2b + 2c

f(t) = 12a3t2 + 18abct + 6ab + 2a2 – 2b2 – 2c2 + 4bc – 6ab

f (t) = 24a3t + 18abc
....
f (t) = 24a3.
Let the line (1) has 4-point contact with the given surface at (0, 0, 1).
....
∴ f (0) = 0, f(0) = 0, f(0) = 0,
f (0) = 0, f (0) ≠ 0.
f (0) = 0 ⇒ – 1 + 1 = 0, which is true.
f(0) = 0 ⇒ – 2c + 2b – 2b + 2c = 0, which is true.

f(0) = 0 ⇒ 6ab + 2a2 – 2b2 – 2c2 + 4bc – 6ab = 0

⇒ a2 – b2 – c2 + 2bc = 0 ...(2)

f (0) = 0 ⇒ 18abc = 0 ⇒ abc = 0 ⇒ a = 0 or b = 0 or c = 0
Case I. a = 0
(2) ⇒ b2 + c2 – 2bc = 0 ⇒ b = c
x y z−1 x y z −1
∴ The line is = = or = = .
0 c c 0 1 1
Case II. b = 0
(2) ⇒ a2 – c2 = 0 ⇒ a=±c
x y z −1 x y z−1
∴ The lines are = = or = = .
±c 0 c ±1 0 1
Case III. c = 0
(2) ⇒ a2 – b 2 = 0 ⇒ a = ± b

∴ The lines are x y z −1 or x y z −1.

= = = =
±b b 0 ±1 1 0
∴ There are five possible lines.
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 71

Theorem 1. Let r = r(s) be any curve and P(s0) be any point on the curve. Prove that the
curve r = r(s) has at least 2-point contact with a plane through P at the point P iff the plane
contains the tangent line at P.
Proof. Let the equation of a plane through P(s0) be (r – r0) . N = 0, where r0 = r(s0)
i.e., the position vector of P and N is a unit vector perpendicular to the plane.
Let f(s) = (r(s) – r0) . N
∴ f ′(s) = r′(s) . N = t(s) . N
Now f(s0) = (r0 – r0) . N = 0 and f ′(s0) = t(s0) . N
∴ f ′(s0) = 0 iff t(s0) . N = 0
iff N is orthogonal to t(s0) iff the plane (r – r0) . N = 0 contains the tangent line at P.
∴ The curve r = r(s) has at least 2-point contact with a plane through P at the point P
on the curve iff the plane contains the tangent line at P.
Theorem 2. Let r = r(s) be any curve and P(s0) be any non-inflexional point on the curve.
Prove that the curve r = r(s) has at least 3-point contact with a plane through P at the point P
iff the plane is the osculating plane at P.
Proof. Let the equation of a plane through P be (r – r0) . N = 0, where r0 = r(s0)
i.e., the position vector of P and N is a unit vector perpendicular to the plane.
Let f(s) = (r(s) – r0) . N
∴ f ′(s) = r′(s) . N = t(s) . N
and f ″(s) = t′(s) . N = κ(s) n(s) . N
Now f(s0) = (r0 – r0) . N = 0, f ′ (s0) = t(s0) . N and f ″(s0) = κ(s0) n(s0) . N.
∴ f ′(s0) = 0, f ″(s0) = 0 iff t(s0) . N = 0, κ(s0) n(s0) . N = 0
iff t(s0) . N = 0 n(s0) . N = 0 (∵ κ(s0) ≠ 0)
iff N is orthogonal to t(s0) and n(s0) iff the plane (r – r0) . N = 0 is the osculating plane at P.
∴ The curve r = r(s) has at least 3-point contact with a plane through P at the point P
on the curve iff the plane is the osculating plane at P.
Remark. If P is an inflexional point, then f ″(s0) = κ(s0) n(s0) . N = 0 even if the plane contains
only the tangent line at P and is not the osculating plane at P.
Example 5. Show that the osculating plane has at least 4-point contact with a curve at
P iff either the curvature or the torsion vanishes at P.
Sol. Let the equation of the curve be r = r(s). Let r0 = r(s0) be the position vector of the
point P on the curve.
The equation of the osculating plane at P is (r – r0) . b0 = 0.
Let f(s) = (r(s) – r0) . b0
∴ f ′(s) = r′(s) . b0 = t . b0,
f ″(s) = t′ . b0 = (κn) . b0 = κ(n . b0)
72 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

and f ′″(s) = κ′(n . b0) + κ(n′ . b0)

= κ′(n . b0) + κ(– κt + τb) . b0
= κ′ (n . b0) – κ2 (t . b0) + κτ (b .b0)
∴ f(s0) = (r0 – r0) . b0 = 0,
f ′(s0) = t0 . b0 = 0,
f ″(s0) = κ0 (n0 . b0) = 0
and f ′″(s0) = κ0′(n0 . b0) – κ02(t0 . b0) + κ0τ0 (b0 . b0) = κ0τ0
∴ f ′″(s0) = 0 iff κ0 = 0 or τ0 = 0.
∴ The osculating plane at P has at least 4-point contact with the curve at P iff either
the curvature or the torsion vanishes at P.

EXERCISE 2.3

1. Show that the curve r = ti + t2j + t3k has 2-point contact with the paraboloid x2 + y2 = z at the
origin.
2. Find the equation of the plane that has 3-point contact with the curve x = 2t + 1, y = 3t2 + 2,
z = 4t3 + 3 at the point (3, 5, 7).
3. Let r = r(s) be any curve and P(s0) be any point of inflexion on the curve. Prove that the curve
r = r(s) has at least 3-point contact with a plane through P at the point P iff the plane contains
the tangent line at P.
4. Show that the osculating plane at P has 3-point contact with a curve at P iff neither the curva-
ture nor the torsion vanishes at P.
5. Show that the osculating plane at P has at least 3-point contact with a curve at P.

Answer
2. 6x – 4y + z = 5.

5. CONTACT OF A CURVE WITH A CURVE

A curve r = x(t)i + y(t)j + z(t)k of sufficiently high class is said to have n-point contact
(or contact of (n – 1)th order) with a curve F(x, y, z) = 0, G(x, y, z) = 0 at the point correspond-
ing to t0 if the functions f(t) = F(x(t), y(t), z(t)) and g(t) = G(x(t), y(t), z(t)) satisfy:
f (t0) = f (t0) = f (t0) = ... = f (n–1)(t0) = 0
g(t0) = g (t0) = g
(t0) = ... = g(n–1) (t0) = 0
and either f (n)(t0) ≠ 0 or g(n)(t0) ≠ 0.
Thus, the curve r = x(t)i + y(t)j + z(t)k has n-point contact with the curve F(x, y, z) = 0,
G(x, y, z) = 0 if and only if the curve r = x(t)i + y(t)j + z(t)k has n-point contact with one of the
surfaces F(x, y, z) = 0 and G(x, y, z) = 0 and at least n-point contact with the other surface.

6. OSCULATING CIRCLE TO A CURVE

A circle having at least 3-point contact with a given curve C at a point P on the curve is
called the osculating circle to the curve C at the point P.
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 73

The centre of the osculating circle at point P is called the centre of curvature of the
curve C at the point P.
By the definition of contact between curves, the osculating circle to the curve C at point
P can be considered as the intersection of a sphere with at least 3-point contact with the curve
C at point P and a plane with at least 3-point contact with C at P. If κ ≠ 0 at P, then the
osculating plane at P is the unique plane having at least 3-point contact with the curve C at P.
In particular, if τ ≠ 0 in addition to κ ≠ 0 at P, then the osculating plane is the unique plane
having exactly 3-point contact with C at P.
Therefore the osculating circle to a curve at a point always lies on the osculating plane
to the curve at that point, provided κ ≠ 0 at the point under consideration.
Thus, the osculating circle to a curve at a point can be considered as the intersection of
a sphere with at least 3-point contact with the curve at that point and the osculating plane to
the curve at the point under consideration, provided κ ≠ 0.

7. EQUATION OF OSCULATING CIRCLE

Let r = r(s) be the equation of a curve C, where s
is the parameter ‘arc length’. Let P be any point on the
curve C for the value s0 of s. Let r0 = r(s0). Let curva- Normal
plane
ture κ0(= κ(s0)) be non-zero at P.
The equation of the osculating plane at P is Q(c)
(r –r0) . b0 = 0, where b0 = b(s0). b0

r0
–
c
Let the osculating circle at P be the intersection
P(r0) M
of the osculating plane at P and the sphere r0n0
| r – c |2 = a 2
t0
with centre at Q(c) and passing through P and having Osculating
plane
at least 3-point contact with the curve r = r(s) at P.
∴ | r0 – c |2 = a2
Let f (s) = | r(s) – c |2 – a2.
∴ f (s) = (r(s) – c) . (r(s) – c) – a2
∴ f ′(s) = (r(s) – c) . (r′(s)) + r′(s) . (r(s) – c) – 0 = 2(r(s) – c) . t(s)
f ″(s) = 2(r(s) – c) . t′(s) + 2(t(s)) . t(s)
= 2(r(s) – c) . κ(s) n(s) + 2(1) = 2κ(s) (r(s) – c) . n(s) + 2
Since the sphere has at least 3-point of contact at P, we have f (s0) = f ′(s0) = f ″(s0) = 0.
f (s0) = 0 ⇒ (r0 – c) . (r0 – c) – a2 = 0 ⇒ | r0 – c |2 – a2 = 0.
⇒ a2 – a2 = 0, which is true.
f ′(s0) = 0 ⇒ 2(r0 – c) . t0 = 0 ⇒ (c – r0) . t0 = 0
⇒ c – r0 lies in the normal plane at P
⇒ centre (Q) of the sphere is in the normal plane at P.
f ″(s0) = 0 ⇒ 2κ0(r0 – c) . n0 + 2 = 0
1
⇒ (r0 – c) . n0 = – = – ρ0 (Using κ0 ≠ 0)
κ0
74 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

⇒ (c – r0) . n0 = ρ0
∴ Projection of c – r0 on n0 = ρ0 ...(1)
Let QM be perpendicular to the principal normal at P.
∴ M is the centre of the osculating circle at P.
Also, (1) ⇒ PM = ρ0n0 ⇒ PM = ρ0 ⇒ M is the centre of curvature at P.
Also, the centre of curvature (M) lies on the principal normal at P and at a distance ρ0
from P.
∴ The radius of the osculating circle at P is ρ0 which is also equal to the radius of
curvature of the given curve at P.
Also, the position vector of the centre of the osculating circle at P(r0)
= P.V. of M = r0 + ρ0n0
and it lies on the principal normal of the given curve at P.
Remark. If κ0 = 0, then f ″(s0) = 0 ⇒ 0 + 2 = 0, which is impossible.
∴ If κ0 = 0, then there does not exist any sphere having at least 3-point of contact with the given
curve at P.

8. LOCUS OF CENTRE OF CURVATURE

Let r = r(s) be the equation of a curve C. For each point P with non-zero curvature, on
the curve C, there exists an osculating circle. Let C1 denote the locus of the centre of curvature
i.e., the centre of osculating circle as the point P moves along the curve C. We shall prove two
properties regarding the curve C1, the locus of centre of curvature.
Property I. The tangent to the locus of centre of curvature lies in the normal plane of the
original curve.
Proof. Let P(r) be any point on a curve C given
by r = r(s). Let κ ≠ 0 at P. Let c be the position vector of
the centre of curvature Q at the point P.
∴ c = r + ρn ...(1)
We shall use suffix ‘1’ with quantities corre-
sponding to the curve C1 of the locus of centre of curva-
ture.
Differentiating (1) w.r.t. s1, we get
dc d ds
= (r + ρn) .
ds1 ds ds1
ds
⇒ t1 = (r′ + ρn′ + ρ′n)
ds1
ds
⇒ t1 = (r′ + ρ(– κt + τb) + ρ′n)
ds1
ds
⇒ t1 = (t – t + ρτb + ρ′n)
ds1
ds
⇒ t1 = (ρτb + ρ′n)
ds1
∴ t1 lies in the plane of b and n.
∴ The tangent at a point to the curve C1 lies in the corresponding normal plane of
curve C.
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 75

∴ The tangent to the locus of the centre of curvature lies in the normal plane of the
original curve.
Remark. Let α be the angle between the tangent to the locus of the centre of curvature at the
centre of curvature at point P and the principal normal at P.
∴ t1 . n = 1 . 1. cos α = cos α
ds n
⇒ (ρτb + ρ′n) . n = cos α
ds1

ds ds
⇒ 0 + ρ′ = cos α, i.e., cos α = ρ′ . t1
ds1 ds1
a p
π –a
Also, angle between t1 and b = –α 2
2 O
b
FG π − αIJ = sin α
∴ t1 . b = 1 . 1. cos
H2 K
ds
⇒ (ρτb + ρ′n) . b = sin α
ds1

ds ds
⇒ ρτ + 0 = sin α, i.e., sin α = ρτ .
ds1 ds1

ρτ ρ
Dividing, we get tan α = =
ρ′ ρ′ σ

FG ρ IJ .
∴ α = tan–1
H ρ′ σ K
Property II. If the original curve C has a constant curvature κ, then the curvature of the
locus C1 of centre of curvature is also constant and the torsion of C1 varies inversely as that
of C.
Proof. Let P(r) be any point on the curve C given
by r = r(s). Let c be the position vector of the centre of
curvature at the point P.
∴ c = r + ρn ...(1)
We shall use suffix ‘1’ with quantities correspond-
ing to the curve C1 of the locus of centre of curvature.
Differentiating (1) w.r.t. s1, we get
dc d ds
= (r + ρn) .
ds1 ds ds1
ds
⇒ t1 = (r′ + ρn′ + ρ′n) ...(2)
ds1
1
′
FG IJ
Since κ is constant, we have ρ′ =
κ H K
= 0.
ds
∴ (2) ⇒ t1 = (r′ + ρ(– κt + τb))
ds1
76 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

ds
⇒ t1 = (t – t + ρτb)
ds1

ds
⇒ t1 = ρτb ...(3)
ds1

FG ds IJ FGds IJ
⇒
H
t1 . t1 = ρτb
ds1 KH
. ρτb
ds1 K
FG ds IJ 2
ds 1
⇒ 1= ρ2τ2
H ds K
1
⇒ =
ds1 ρτ

1
∴ (3) ⇒ t1 = (ρτb) =b ...(4)
ρτ
Differentiating w.r.t. s1, we get
dt 1 db ds
= .
ds1 ds ds1
1 1
⇒ t1′ = b′ ρτ ⇒ ρτt1′ = – τn ⇒ t1′ = – n
ρ
⇒ t1′ = – κn
⇒ κ1n1= – κn
∴ Vectors n1 and n are parallel. Choosing the direction of n1 opposite to that of n, we
have n1 = – n.
∴ κ1 = κ
∴ The curvature of the curve C1 is also constant. (∵ κ is constant)
Also b1 = t1 × n1 = b × (– n) = – b × n = n × b = t (Using (4))
Differentiating w.r.t. s1, we get
db 1 dt ds
=
ds1 ds ds1
ds 1
⇒ – τ1 n1 = t′ ⇒ – τ1 n1 = (κn)
ds1 ρτ
κ κ 1
= κ2 .
1 FG IJ
⇒ τ1 n =
ρτ
n ⇒ τ1 =
ρ τ τ H K (∵ n1 = – n)

∴ Torsion of curve C1 varies inversely as that of C.

Example 1. Show that the principal normal to a curve is perpendicular to the locus of
the centre of curvature at points, where curvature κ is constant.
Sol. Let P(r) be any point on a curve C given by r = r(s). Let c be the position vector of
the centre of curvature at the point P.
∴ c = r + ρn ...(1)
We shall use suffix ‘1’ with quantities corresponding to the curve C1 of the locus of
centre of curvature.
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 77

Differentiating (1) w.r.t. s1, we get

dc d ds
= ( r + ρn)
ds1 ds ds1
ds
⇒ t1 = (r′ + ρn′ + ρ′n)
ds1
ds
⇒ t1 = (t + ρ(– κt + τb) + 0n)
ds1
FG ρ′ = 0 because ρ = 1 is constantIJ
H κ K
F ds IJ b
= G ρτ
⇒ t1
H ds K 1
(∵ ρκ = 1)

F ds IJ n . b = 0
= G ρτ
⇒ n. t1
H ds K 1
⇒ n is perpendicular to the tangent vector to the curve C1 at the point with position
vector c.
∴ Principal normal to the curve C is perpendicular to the curve C1 i.e., the locus of
centre of curvature.
Example 2. If s1 is the arc length of the locus of centre of curvature, show that

ds1 κ 2τ 2 + κ′ 2 FG ρ IJ 2

ds
=
κ2
=
H σK + ρ′ 2 .

Sol. Let the given curve be r = r(s). Let suffix ‘1’ be used for quantities corresponding to
the locus of centre of curvature.
Let r1 be the position vector of the centre of curvature corresponding to the point r on
the curve r = r(s).
∴ r1 = r + ρn
1
⇒ r1 = r +n
κ
Differentiating w.r.t. s1, we get
dr1 d 1 FG
ds IJ
=
ds1 ds
r+ n
κ H
ds1 K
F 1 F κ′ IJ nIJ ds
t = G r ′ + n′ + G −
⇒ 1
H κ H κ K K ds 2
1
κ ′ I ds
t = FG t + (− κt + τb) − nJ
1
⇒ 1
H κ κ K ds 2
1
F κ′ n + τ bIJ ds
t = G−
⇒ 1 H κ κ K ds2
1

F κ′ n + τ bIJ . FG − κ′ n + τ bIJ FG ds IJ
. t = G−
2

⇒ t1 1 H κ κ K H κ κ K H ds K
2 2
1
78 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

F κ′ 2
τ2 I F ds I = F κ′
2 2
+ κ 2τ2 I F ds I 2

⇒ 1=G
Hκ 4
+ 2
κ
JK GH ds JK GH
1 κ4
JK GH ds JK
1

FG ds IJ 2
κ 2τ2 + κ′2
H ds K
1
⇒ = ...(1)
κ4

ds1 κ 2τ2 + κ′2

⇒ =
ds κ2

FG ds IJ 2
τ2 κ′2 ρ2 FF 1 I I
+ GG J J
′
2
ρ2
H ds K HH κK K
1
(1) ⇒ = + = = + ρ′2
κ2 κ4 σ2 σ2

ds1 FG ρ IJ 2

∴ ds
=
H σK + ρ′ 2 .

∴ The result holds.

9. OSCULATING SPHERE TO A CURVE
A sphere having at least 4-point contact with a given curve C at a point P on the curve
is called the osculating sphere to the curve C at the point P.
The centre of the osculating sphere at point P is called the centre of spherical
curvature of the curve C at the point P.
Remark. The radius of osculating sphere is also referred as the radius of spherical curvature.

10. EQUATION OF OSCULATING SPHERE

Let r = r(s) be the equation of a curve C, where
s is the parameter ‘arc length’. Let P be any point on
the curve C for the value s0 of s. Let r0 = r(s0). Let
curvature κ0 (= κ(s)) and torsion τ0 (= τ(s0)) be non-
zero at P.
Let the equation of the osculating sphere be
| r – c |2 = a2
with centre at Q(c) and passing through P and having
at least 4-point contact with the curve r = r(s) at P.
∴ | r0 – c |2 = a2
Let f (s) = | r(s) – c |2 – a2.
∴ f (s) = (r(s) – c) . (r(s) – c) – a2
f ′(s) = (r(s) – c) . r′(s)) + r′(s) . (r(s) – c) – 0 = 2 (r(s) – c) . t(s)
f ″(s) = 2(r(s) – c) . t′(s) + 2 (t(s)) . t(s)
= 2(r(s) – c) . κ(s) n(s) + 2(1) = 2κ(s) (r(s) – c) . n(s) + 2
f ″′(s) = 2κ′(s) (r(s) – c) . n(s) + 2κ(s) (t(s) – 0) . n(s) + 2κ(s) (r(s) – c) . n′(s) + 0
= 2κ′(s) (r(s) – c) . n(s) + 2κ(s) (0) + 2κ(s) (r(s) – c) . (– κ(s) t(s) + τ(s) b(s))
= 2κ′(s) (r(s) – c) . n(s) – 2κ2(s) (r(s) – c) . t(s) + 2κ(s) τ(s) (r(s) – c) . b(s)
Since the sphere has at least 4-point contact at P, we have
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 79

f (s0) = f ′(s0) = f ″(s0) = f ″′(s0) = 0.

f (s0) = 0 ⇒ (r0 – c) . (r0 – c) – a2 = 0 ⇒ | r0 – c |2 – a2 = 0
⇒ a2 – a2 = 0, which is true.
f ′(s0) = 0 ⇒ 2(r0 – c). t0 = 0 ⇒ (c – r0) . t0 = 0 ...(1)
⇒ c – r0 lies in the normal plane at P
⇒ centre (Q) of the sphere is in the normal plane at P.
f ″(s0) = 0 ⇒ 2κ0(r0 – c) . n0 + 2 = 0
1
⇒ (r0 – c) . n0 = – = – ρ0 ...(2) (Using κ0 ≠ 0)
κ0
⇒ (c – r0) . n0 = ρ0
⇒ Projection of c – r0 on n0 = ρ0
f ″′(s0) = 0 ⇒ 2κ0′ (r0 – c) . n0 – 2κ02 (r0 – c) . t0 + 2κ0τ0 (r0 – c) . b0 = 0

⇒ 2κ0′ −
FG 1 IJ – 2κ 2
0 (0) + 2κ0τ0(r0 – c) . b0 = 0
H κ0 K
(Using (1) and (2))
2κ 0 ′
⇒ – + 2κ0τ0 (r0 – c) . b0 = 0
κ0
κ0 ′
⇒ (r0 – c) . b0 = (Using τ0 ≠ 0)
κ 02τ0

κ0 ′ F d F 1 II σ
⇒ (c – r0) . b0 = – 2
κ 0 τ0
= GH ds GH κ JK JK
0
0 = ρ0′σ0

⇒ projection of c – r0 on b0 = ρ0′σ0
∴ The components of the vector c – r0 along the vectors t0, n0 and b0 are 0, ρ0 and
ρ0′ σ0 respectively.
∴ c – r0 = 0t0 + ρ0n + ρ0′σ0b
2 2
∴ | c – r0 | = ρ 0 + ρ0 ′ 2 σ 0
and c = r0 + ρ0 n + ρ0′σ0 b0.
The centre Q(c) of the osculating sphere is the centre of spherical curvature of the curve
C at P.
2 2
Radius of the osculating sphere at P = PQ = | c – r0 | = ρ 0 + ρ0 ′ 2 σ 0 .
Also, the position vector of the centre of spherical curvature of the curve C at
P(r0) = P.V. of Q = c = r0 + ρ0n0 + ρ0′σ0b0 and it lies on the normal plane of the given curve
at P.
Remark 1. In terms of κ0 and τ0, we have

F 1 I +F κ ′ I
2 2

(i) radius of osculating sphere at P(r0) = GH κ JK GH κ τ JK

0 0
2
0

0
and

1 κ0 ′
(ii) p.v. of centre of spherical curvature = r0 + n0 – b0.
κ0 κ 02 τ0
80 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

Remark 2. If curvature of r = r(s) at P is constant, then radius of osculating sphere at

P(r0) = ρ02 + 0. σ 02 = ρ0 and p.v. of centre of spherical curvature at

P(r0) = r0 + ρ0n0 + (0)σ0b = r0 + ρ0n0.
∴ Centre of the osculating sphere coincides with the centre of osculating circle at points where
curvature vanishes.

11. LOCUS OF CENTRE OF SPHERICAL CURVATURE

Let r = r(s) be the equation of a curve C. For each point P with non-zero curvature and
torsion on the curve C there exists an osculating sphere. Let C1 denote the locus of the centre
of spherical curvature i.e., the centre of osculating sphere as the point P moves along the
curve C. We shall prove some properties regarding the curve C1, the locus of centre of spheri-
cal curvature.
Property I. The tangent to the locus of centre of spherical curvature is parallel to the
corresponding binormal to the original curve.
Proof. Let P(r) be any point on a curve C given by r = r(s). Let κ ≠ 0, τ ≠ 0 at P. Let r1 be
the position vector of the centre of spherical curvature at the point P.
∴ r1 = r + ρn + ρ′σb ...(1)
We shall use suffix ‘1’ with quantities corresponding to the curve C1 of the locus of
centre of spherical curvature.
Differentiating (1) w.r.t. s1, we get
dr1 d ds
= (r + ρn + ρ′σb)
ds1 ds ds 1
ds
⇒ t1 = (t + ρn′ + ρ′n + ρ″σb + ρ′σ′b + ρ′σb′)
ds1
ds
⇒ t1 = (t + ρ( – κt + τb) + ρ′n + ρ″σb + ρ′σ′b – ρ′στn)
ds1
ds
⇒ t1 = ((1 – ρκ)t + ρ′(1 – στ)n + (ρτ + ρ″σ + ρ′σ′)b)
ds1
ds
⇒ t1 = (ρτ + ρ″σ + ρ′σ′) b (∵ κρ = 1, τσ = 1)
ds1
∴ t1 is parallel to b.
∴ The tangent to C1 is parallel to the corresponding binormal to C.
Property II. The principal normal to the locus of centre of spherical curvature is parallel
to the corresponding principal normal to the original curve.
Proof. Let P(r) be any point on a curve C given by r = r(s). Let κ ≠ 0, τ ≠ 0 at P. Let r1 be
the position vector of the centre of spherical curvature at the point P.
∴ r1 = r + ρn + ρ′σb ...(1)
We shall use suffix ‘1’ with quantities corresponding to the curve C1 of the locus of
centre of spherical curvature.
Differentiating (1) w.r.t. s1, we get
dr1 d ds
= (r + ρn + ρ′σb)
ds1 ds ds1
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 81

ds
⇒ t1 = (ρτ + ρ″σ + ρ′σ′) b ...(2)
ds1
(For detail see property I)
ds ds
⇒ |t1| = (ρτ + ρ″σ + ρ′σ′) |b| ⇒ 1 = (ρτ + ρ″σ + ρ′σ′) .1
ds1 ds1

ds 1
∴ =
ds1 ρτ + ρ″ σ + ρ′ σ ′
∴ (2) ⇒ t1 = b ...(3)
Differentiating (3) w.r.t. s1, we get
dt 1 db ds
=
ds1 ds ds1

ds
⇒ κ1n1 = – τn
ds1
∴ n1 is parallel to n.
∴ The principal normal to C1 is parallel to the corresponding principal normal to C.
Property III. The binormal to the locus of centre of spherical curvature is parallel to
the corresponding tangent to the original curve.
Proof. Let P(r) be any point on a curve C given by r = r(s). Let κ ≠ 0, τ ≠ 0 at P. Let r1 be
the position vector of the centre of spherical curvature at the point P.
∴ r1 = r + ρn + ρ′σb ...(1)
We shall use suffix ‘1’ with quantities corresponding to the curve C1 of the locus of
centre of spherical curvature.
Differentiating (1) w.r.t. s1, we get
dr1 d ds
= ( r + ρn + ρ′ σb)
ds1 ds ds1
ds
⇒ t1 = (ρτ + ρ″σ + ρ′ σ′ ) b ...(2)
ds1
(For detail see property I)
ds
⇒ |t1| = (ρτ + ρ″ σ + ρ′ σ ′ ) |b|
ds1
ds 1
⇒ =
ds1 ρτ + ρ″ σ + ρ′ σ ′
∴ (2) ⇒ t1 = b ...(3)
Differentiating (3) w.r.t. s1, we get
dt 1 db ds
=
ds1 ds ds1
ds
⇒ κ1n1 = – τn ...(4)
ds1
82 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

ds
⇒ κ1|n1| = |– 1| τ |n|
ds1

ds ds
⇒ κ1 . 1 = 1 . τ . 1 ⇒ κ1 = τ
ds1 ds1
∴ 4) ⇒ n1 = – n
⇒ t1 × n1 = b × (– n) (Using (3))
⇒ b1 = t
∴ b1 is parallel to t.
∴ The binormal to C1 is parallel to the corresponding tangent to C.
Property IV. The product of curvatures at the corresponding points on the locus of
centre of spherical curvature and the original curve is equal to the product of their torsions.
Proof. Let P(r) be any point on a curve C given by r = r(s). Let κ ≠ 0, τ ≠ 0 at P. Let r1 be
the position vector of the centre of spherical curvature at the point P.
∴ r1 = r + ρn + ρ′σ b ...(1)
We shall use suffix ‘1’ with quantities corresponding to the curve C1 of locus of centre of
spherical curvature.
Differentiating (1) w.r.t. s1, we get
dr1 d ds
= (r + ρn + ρ′σb)
ds1 ds ds1

ds
⇒ t1 = (ρτ + ρ″σ + ρ′σ′) b ...(2)
ds1
(For detail see property I)
ds
⇒ | t1 | = (ρτ + ρ″σ + ρ′σ′) |b|
ds1
ds 1
⇒ =
ds1 ρτ + ρ″ σ + ρ′ σ ′
∴ (2) ⇒ t1 = b ...(3)
Differentiating (3) w.r.t. s1, we get
dt 1 db ds
=
ds1 ds ds1
ds
⇒ κ1n1 = – τn ...(4)
ds1
ds
⇒ κ1| n1 | = |– 1| τ |n|
ds1
ds ds
⇒ κ1 . 1 = 1 . τ . 1 ⇒ κ1 = τ
ds1 ds1
∴ (4) ⇒ n1 = – n
⇒ t1 × n1 = b × (– n) (Using (3)
⇒ b1 = t
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 83

Differentiating w.r.t. s1, we get

db 1 dt ds
=
ds1 ds ds1
κ1 FG Using κ ds IJ
⇒ – τ1n1 = (κn)
τ H 1 =τ
ds1 K
κκ 1
⇒ – τ1(– n) = n (Using n1 = – n)
τ
⇒ ττ1n = κκ1n ⇒ κκ1 = ττ1.
∴ The product of curvatures at the corresponding points is equal to the product of the
torsions.
Property V. If curvature κ of a curve C is constant, then the curvature κ1 of the curve C1
of the locus of centre of spherical curvature is also constant.
Proof. Let P(r) be any point on a curve C given by r = r(s). Let κ ≠ 0, τ ≠ 0 at P. Let r1 be
the position vector of the centre of spherical curvature at the point P.
∴ r1 = r + ρn + ρ′σb ...(1)
We shall use suffix ‘1’ with quantities corresponding to the curve C1 of the locus of
centre of spherical curvature.
Differentiating (1) w.r.t. s1, we get
dr1 d ds
= (r + ρn + ρ′σb)
ds1 ds ds1
ds
⇒ t1 = (ρτ + ρ″σ + ρ′σ′) b ...(2)
ds1
(For detail see property I)
ds
⇒ | t1 | = (ρτ + ρ″σ + ρ′σ′) |b|
ds1
ds 1
⇒ =
ds1 ρτ + ρ″ σ + ρ′ σ ′
∴ (2) ⇒ t1 = b ...(3)
Differentiating (3) w.r.t. s1, we get
dt 1 db ds
=
ds1 ds ds1
ds
⇒ κ1n1 = – τn
ds1

FG κ IJ n F F 1I ′ I
⇒ κ1n1 = – τ
H τK GG ρ′ = GH κ JK = 0, ρ′′ = 0 ⇒ ds
=
1
=
κ
JJ
H ds1 ρτ + 0σ + 0. σ ′ τ K
⇒ κ1n1 = – κn
⇒ κ1 | n1 | = | – 1 | κ | n | ⇒ κ1 = κ
⇒ κ1 is constant because κ is constant.
∴ The curvature of the curve C1 of the locus of centre of spherical curvature is also
constant.
84 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

Example 3. If R is the radius of osculating sphere to a curve r = r(s) at point ‘s’, then
show that
t × t″
R= .
κ 2τ
Sol. We have t′ = κn
and t″ = κn′ + κ′n = κ(– κt + τb) + κ′n = – κ2t + κ′n + κτb
∴ t × t″ = t × (– κ2t + κ′n + κτb)
= – κ2 (t × t) + κ′(t × n) + κτ(t × b)
= – κ2(0) + κ′b + κτ (– n) = – κτn + κ′b

t × t″ κτ κ′ FG 1IJ ′ σ b = – ρn – ρ′σb
∴
κ2τ
=–
κ 2τ
n+
κ 2τ
b = – ρn –
H κK
t × t″
⇒ – = ρn + ρ′σb
κ2τ
t × t″ t × t ″
⇒ (– 1)2 . 2 = (ρn + ρ′σb) · (ρn + ρ′σb)
κ 2τ κ τ
2
t × t″
⇒ = ρ2 n· n + ρ′2σ2 b· b = ρ2 + ρ′2σ2
κ 2τ
= R2 (∵ R2 = ρ2 + ρ′2σ2)
t × t″
∴ R= .
κ2τ
Example 4. Show that the radius of osculating sphere of the circular helix
x = a cos θ, y = a sin θ, z = aθ cot α
is equal to the radius of curvature at each point on the helix.
Sol. Given helix is x = a cos θ, y = a sin θ, z = a θ cot α.
Let r be the position vector of the point P(x, y, z) on the helix.
∴ r = xi + yj + zk = a cos θi + a sin θj + aθ cot αk
| r × r
|
We know that κ= 3 (A standard formula)
|r |
Now r = – a sin θi + a cos θj + a cot αk

r = – a cos θi – a sin θj + 0k
i j k
∴ r = − a sin θ a cos θ a cot α
r ×
− a cos θ − a sin θ 0
i j k
= – a 2 – sin θ cos θ cot α
cos θ sin θ 0
= – a2 [– cot α sin θi + cot α cos θj – k]
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 85

∴ r | = a2 cot 2 α sin 2 θ + cot 2 α cos 2 θ + 1

| r ×

= a2 cot 2 α + 1 = a2 cosec α
Also | r | = (a2 sin2 θ + a2 cos2 θ + a2 cot2 α)1/2
= (a2 + a2 cot2 α)1/2 = a cosec α
| r ×
r| a 2 cosec α sin 2 α
∴ κ= = =
|r |3 a3 cosec 3 α a
∴ κ is a constant quantity.

FG 1IJ ′ = 0
∴ ρ′ =
H κK
Let R be the radius of osculating sphere at the point P on the curve.

∴ R= ρ 2 + ρ′ 2 σ 2 = ρ2 + ( 0)2 σ 2 = ρ
Also the radius of curvature at P is ρ.
∴ The result holds.
Example 5. Find the equation of the osculating sphere to the curve x = 2t + 1,
y = 3t2 + 2, z = 4t3 + 3 at the point (1, 2, 3).
Sol. The given curve is
x = 2t + 1, y = 3t2 + 2, z = 4t3 + 3.
The point P(1, 2, 3) corresponds to the value 0 of t.
Let r be the position vector of the point (x, y, z) on the given curve.
∴ r = xi + yj + zk = (2t + 1)i + (3t2 + 2)j + (4t3 + 3)k
Let (α, β, γ) and R be the centre and the radius of the osculating sphere at P(1, 2, 3)
respectively.
∴ The equation of the osculating sphere is |r – c|2 = R2, where c = αi + βj + γk.
This sphere passes through (1, 2, 3) and has at least 4-point contact with the given
curve at P.
∴ |(i + 2j + 3k) – (αi + βj + γk)|2 = R2
⇒ (1 – α)2 + (2 – β)2 + ( 3 – γ)2 = R2 ...(1)
Let f (t) = | (2t + 1)i + (3t2
+ 2)j + (4t3
+ 3)k – (αi + βj + γk) |2 – R2
= (2t + 1 – α) + (3t + 2 – β) + (4t3 + 3 – γ)2 – R2
2 2 2

∴ f ′(t) = 2(2t + 1 – α)2 + 2(3t2 + 2 – β)6t + 2(4t3 + 3 – γ)12t2

f ″(t) = 8 + 108t2 + 12(2 – β) + 480t4 + 48(3 – γ)t
f ′″(t) = 216t + 1920t3 + 48(3 – γ)
f iv(t) = 216 + 5760t2
Since the sphere has at least 4-point contact at t = 0, we have
f (0) = f ′(0) = f ″(0) = f ′″(0) = 0 and f iv(0) ≠ 0
f (0) = 0 ⇒ (1 – α)2 + (2 – β)2 + (3 – γ)2 – R2 = 0, which is true.
86 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

f ′(0) = 0 ⇒ 4(1 – α) = 0 ⇒ α = 1 (By using (1))

f ″(0) = 0 ⇒ 8 + 12 (2 – β) = 0 ⇒ β = 8/3
f ′″(0) = 0 ⇒ 48(3 – γ) = 0 ⇒ γ = 3
8
∴ c = αi + βj + γk = i + j + 3k
3

F 8I
+ G2 − J
2
4 2
(1) ⇒ R2 = (1 – 1)2
H 3K + (3 – 3)2 =
9
⇒ R=
3
∴ Centre and radius of the osculating sphere at the point P are (1, 8/3, 3) and 2/3
respectively.
∴ The equation of the osculating sphere is

FG 8 IJ = FG 2 IJ 2
( xi + yj + zk) − i +
H 3
j + 3k
K H 3K .

FG 8 IJ 2
4
⇒ (x – 1)2 + y −
H 3 K + (z – 3)2 =
9
⇒ 3x2 + 3y2 + 3z2 – 6x – 16y – 18z + 50 = 0.
Example 6. If the radius of the osculating sphere of a curve is constant, prove that the
curve lies on a sphere or has constant curvature.
Sol. Let the given curve be r = r(s).
Let R be the radius of the osculating sphere at each point on the curve r = r(s).
Let P(r) be any point on this curve.

∴ R= ρ 2 + ρ′ 2 σ 2
⇒ R2 = ρ2 + ρ′2σ2 ...(1)
Differentiating (1) w.r.t. s, we get
0 = 2ρρ′ + ρ′2(2σσ′) + (2ρ′ρ″)σ2
⇒ 0 = 2ρ′(ρ + ρ′σσ′ + ρ″σ2)
⇒ Either ρ′ = 0 ...(2) or ρ + ρ′σσ′ + ρ″σ2 = 0 ...(3)
(2) ⇒ ρ is constant ⇒ κ is constant.
Let (3) hold. Let r1 be the position vector of the centre of spherical curvature at the
point P(r).
∴ r1 = r + ρn + ρ′σb
Differentiating w.r.t. s, we get
dr1
= r′ + (ρn′ + ρ′n) + (ρ″σb + ρ′σ′b + ρ′σb′)
ds
= t + ρ(– κt + τb) + ρ′n + ρ″σb + ρ′σ′b + ρ′σ(– τn)
= (1 – ρκ)t + (ρ′ – ρ′στ)n + (ρτ + ρ″σ + ρ′σ′)b
= 0t + 0n + (ρτ + ρ″σ + ρ′σ′)b (∵ ρκ = 1, στ = 1)
FG ρ + ρ″ σ + ρ′ σ′IJ b
=
Hσ K
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 87

1
= (ρ + ρ″σ2 + ρ′σ′σ)b
σ
1
= (0)b = 0 (Using (3))
σ
dr1
⇒ = 0 ⇒ r1 is constant.
ds
∴ The centre of the osculating sphere is independent of the point P(r) on the curve
r = r(s).
∴ The centre of osculating sphere is same at every point on the curve r = r(s).
Also, the radius of the osculating sphere is same for each point on the curve r = r(s).
∴ The osculating sphere is a fixed sphere for each point on the curve r = r(s).
∴ The curve r = r(s) lies itself on this sphere.
Example 7. Show that the necessary and sufficient condition for a curve r = r(s) to lie on
ρ d
a sphere is s + (ρ′ σ) = 0 at every point on the curve.
σ ds
Sol. Necessity. Let the curve r = r(s) lie on a sphere.
∴ This sphere is the osculating sphere to the curve r = r(s) at every point of the curve.
∴ The radius R of the osculating sphere at each point is constant.
We have R2 = ρ2 + ρ′2σ2
Differentiating w.r.t. s, we get
0 = 2ρρ′ + ρ′2(2σσ′) + (2ρ′ρ″)σ2
FG ρ + ρ′ σ′ + ρ″ σIJ
⇒ 0 = 2ρ′σ
Hσ K
ρ
⇒ + ρ′ σ′ + ρ″ σ = 0 (Assuming ρ′ ≠ 0, σ ≠ 0)
σ
ρ d
⇒ + (ρ′ σ) = 0.
σ ds
ρ d
Sufficiency. Let + (ρ′ σ) = 0.
σ ds
ρ
⇒ + ρ′ σ′ + ρ″ σ = 0
σ
FG
ρ IJ
⇒ 2ρ′σ
σH K
+ ρ′ σ′ + ρ″ σ = 0
⇒ 2ρρ′ + ρ′2(2σσ′) + (2ρ′ρ″)σ2 = 0
d
⇒ (ρ2 + ρ′2σ2) = 0
ds
⇒ ρ2 + ρ′2 σ2 = λ, a constant.
⇒ R2 = λ i.e., R is constant.
∴ Radius of osculating sphere is independent of the point on the curve r = r(s).
Let r1 be the position vector of the centre of osculating sphere at the point s.
88 DIFFERENTIAL GEOMETRY AND CALCULUS OF VARIATIONS

∴ r1 = r + ρn + ρ′σb
Differentiating w.r.t. s, we get
dr1
= r′ + (ρn′ + ρ′n) + (ρ″σb + ρ′σ′b + ρ′σb′)
ds
= t + ρ( – κt + τb) + ρ′n + ρ″σb + ρ′σ′b + ρ′σ(– τn)
= (1 – ρκ) t + (ρ′ – ρ′στ) n + (ρτ + ρ″σ + ρ′σ′) b
ρ FG
d IJ
= 0t + 0n + +
σ ds H (ρ′ σ) b = 0 b = 0
K
dr1
⇒ = 0 ⇒ r1 is constant.
ds
The centre of the osculating sphere is independent of the point on the curve r = r(s).
∴ The centre of the osculating sphere is same at every point on the curve r = r(s).
Also, the radius of the osculating sphere is same for each point on the curve r = r(s).
∴ The osculating sphere is a fixed sphere for each point on the curve r = r(s).
∴ The curve r = r(s) lies itself on this sphere.

WORKING RULES FOR SOLVING PROBLEMS

Rule I. A curve r = x(t)i + y(t)j + z(t)k of sufficiently high class is said to have n-point
contact with a surface F(x, y, z) = 0 at the point t 0 if the function
f(t) = F(x(t), y(t), z(t)) satisfies:
f(t0 ) = f (t0 ) = f(t0 ) = ... = f (n–1) (t0 ) = 0 and f (n)(t
0 ) ≠ 0.
Rule II. A curve r = x(t)i + y(t)j + z(t)k of sufficiently high class is said to have n-point
contact with a curve F(x, y, z) = 0, G(x, y, z) = 0 at the point t0 if the functions
f(t) = F(x(t), y(t), z(t)) and g(t) = G(x(t), y(t), z(t)) satisfy:
f(t0 ) = f (t0 ) = f(t0 ) = ... = f (n–1) (t0 ) = 0
g(t0) = g (t0 ) = g
(t0 ) = ... = g(n–1) (t0) = 0
and either f (t0 ) ≠ 0 or g(n)(t0 ) ≠ 0.
(n)

Rule III. A circle having at least 3-point contact with a given curve C at a point P on the
curve is called the osculating circle to the curve C at the point P. The centre of
the osculating circle at the point P is called the centre of curvature of the
curve C at the point P.
If r1 be the position vector of the centre of curvature at point r then,
r1 = r + ρn and radius of osculating circle = ρ.
Rule IV. A sphere having at least 4-point contact with a given curve C at a point P on the
curve is called the osculating sphere to the curve C at the point P. The centre
of the osculating sphere at the point P is called the centre of spherical
curvature of the curve C at the point P.
If r1 be the position vector of the centre of spherical curvature at point r, then

r1 = r + ρn + ρ′σb and radius of osculating sphere = ρ2 + ρ′ 2 σ 2 .

SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

CURVATURE AND TORSION 89

EXERCISE 2.4
1. Show that the tangent to the locus of the centre of curvature lies in the normal plane of the
ρ
original curve and makes an angle tan–1 with the principal normal of the original curve.
σρ′
2. If C is a curve of constant curvature κ, show that the locus C1 of its centre of curvature is also a
κ2
curve of constant curvature κ1 such that κ1 = κ and its torsion τ1 is given by the relation τ1 = .
τ
3. For a curve of constant curvature, show that the centre of spherical curvature coincides with the
centre of circular curvature.
4. If R is the radius of the osculating sphere to a curve r = r(s) at point s, then show that:
R2 = ρ4σ2 |r′″|3 – σ2.
5. For the curve r = x(s)i + y(s)j + z (s)k, show that:
1 1 + ρ′ 2
x′″2 + y′″2 + z′″2 = 2 2 + .
ρ σ ρ4
6. For the curve r = x(s)i + y(s)j + z(s)k, show that:

R2
ρ4(x′″2 + y′″2 + z′″2) = 1 + ,
σ2
where R is the radius of spherical curvature at the point (x, y, z).
7. Show that the radius of spherical curvature of a circular helix is equal to the radius of its circular
curvature.

Hint
4. r′″ = (r″)′ = (t′)′ = (κn)′ = κn′ + κ′n = κ(– κt + τb) + κ′n = – κ2t + κ′n + κτ b
∴ |r′″|2 = (– κ2)2 + (κ′)2 + (κτ)2

1 F ρ′ I
+ G−
2
1 σ 2 + ρ′ 2 σ 2 + ρ2 σ2 + R2
=
ρ4 H ρ JK
2
+
ρ2σ 2
=
ρ4 σ 2
=
ρ4 σ 2
.
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE
SUCCESSCLAP : Best Coaching for UPSC Maths : Visit https://SuccessClap.com
VIDEOS, QUESTION BANK, LIVE TEST SERIES, ASSIGNMENTS, PERSONAL GUIDANCE

Kiran One Liner General Science Hindi
100% (1)
Kiran One Liner General Science Hindi
761 pages
What Is Philosophy For by Mary Midgley
80% (5)
What Is Philosophy For by Mary Midgley
233 pages
BS en 61685-2002
No ratings yet
BS en 61685-2002
42 pages
CHM131 - Chapter 2 - Structure of Atom - Periodicity
No ratings yet
CHM131 - Chapter 2 - Structure of Atom - Periodicity
97 pages
N1234
No ratings yet
N1234
60 pages
Vector Calculus Notes
No ratings yet
Vector Calculus Notes
10 pages
05 Vector Analysis - SuccessClap
No ratings yet
05 Vector Analysis - SuccessClap
353 pages
MA111 Lec8 D3D4
No ratings yet
MA111 Lec8 D3D4
33 pages
Integrals of Vector Fields. (Sect. 16.2) : The Gradient Field of A Scalar-Valued Function
No ratings yet
Integrals of Vector Fields. (Sect. 16.2) : The Gradient Field of A Scalar-Valued Function
10 pages
Line Integral
No ratings yet
Line Integral
8 pages
Test 123
No ratings yet
Test 123
7 pages
Topic 1 - Vector Calculus
No ratings yet
Topic 1 - Vector Calculus
26 pages
Study Sheet (13.1-13.4)-4
No ratings yet
Study Sheet (13.1-13.4)-4
19 pages
2 - Tangent - Spaces (Errata)
No ratings yet
2 - Tangent - Spaces (Errata)
12 pages
MATH 4A - Linear Algebra With Applications: Lecture 17: Subspaces From Linear Transformations/matrices, & Bases
No ratings yet
MATH 4A - Linear Algebra With Applications: Lecture 17: Subspaces From Linear Transformations/matrices, & Bases
35 pages
MA231 - Vector Analysis
No ratings yet
MA231 - Vector Analysis
12 pages
PHYS 419: Classical Mechanics Lecture Notes: I. Work and Energy
No ratings yet
PHYS 419: Classical Mechanics Lecture Notes: I. Work and Energy
8 pages
Final Exam Review
No ratings yet
Final Exam Review
5 pages
Chapter 2. Applications of The Integral
No ratings yet
Chapter 2. Applications of The Integral
16 pages
Vector Valued Functions
No ratings yet
Vector Valued Functions
75 pages
3-1 Space Curves and Their Tangents(113)
No ratings yet
3-1 Space Curves and Their Tangents(113)
8 pages
SIM1003 Chapter 12 Notes
No ratings yet
SIM1003 Chapter 12 Notes
5 pages
Calculus 2 Project Project Official
No ratings yet
Calculus 2 Project Project Official
37 pages
Foundations of Geometry
No ratings yet
Foundations of Geometry
20 pages
EGM 211 - Vector Valued Functions-PPP
No ratings yet
EGM 211 - Vector Valued Functions-PPP
195 pages
Line Integrals
No ratings yet
Line Integrals
7 pages
Mth-301 File Lec 23-45 by Orange Monkey
No ratings yet
Mth-301 File Lec 23-45 by Orange Monkey
23 pages
One Dimension Again: Chapter Fourteen
No ratings yet
One Dimension Again: Chapter Fourteen
13 pages
Lecture Notes For MA5NO Cohomology, Connections, Curvature and Characteristic Classes
No ratings yet
Lecture Notes For MA5NO Cohomology, Connections, Curvature and Characteristic Classes
66 pages
4.integration Along Curves: N N I I I B A T (B A) T 0
No ratings yet
4.integration Along Curves: N N I I I B A T (B A) T 0
12 pages
Integral Differensial Vektor PDF
No ratings yet
Integral Differensial Vektor PDF
9 pages
Arapura, Introduction To Differential Forms
No ratings yet
Arapura, Introduction To Differential Forms
30 pages
Phy306 Probs
No ratings yet
Phy306 Probs
25 pages
MATH CHAP-9 ZILL BITS
No ratings yet
MATH CHAP-9 ZILL BITS
37 pages
Calculus BC Ch12
No ratings yet
Calculus BC Ch12
29 pages
Screenshot 2023-11-09 at 10.57.08 AM
No ratings yet
Screenshot 2023-11-09 at 10.57.08 AM
18 pages
Calc3 Chapter3
No ratings yet
Calc3 Chapter3
11 pages
Chapter 3 - Vector Differentiation, 2010
No ratings yet
Chapter 3 - Vector Differentiation, 2010
11 pages
Session 27_ ConservartiveFields
No ratings yet
Session 27_ ConservartiveFields
23 pages
Sy - Integral Calculus
No ratings yet
Sy - Integral Calculus
12 pages
cartan.Dirac.Good
No ratings yet
cartan.Dirac.Good
54 pages
ENGR 2422 Engineering Mathematics 2 Brief Notes On Chapter 1 1.1 Lines and Planes
No ratings yet
ENGR 2422 Engineering Mathematics 2 Brief Notes On Chapter 1 1.1 Lines and Planes
10 pages
Oct 12
No ratings yet
Oct 12
13 pages
Unit-4, Vector Differential Calculus
No ratings yet
Unit-4, Vector Differential Calculus
31 pages
Stokes Theorem Ii Maths21a, O. Knill: y Z Z X X y X y Z
No ratings yet
Stokes Theorem Ii Maths21a, O. Knill: y Z Z X X y X y Z
1 page
Vector Integration
No ratings yet
Vector Integration
32 pages
A Cosmological Basis For: Fulvio Melia
No ratings yet
A Cosmological Basis For: Fulvio Melia
5 pages
Differential Geometry of Curves: Definition 2.1. A Regular Parametrization of Class C
No ratings yet
Differential Geometry of Curves: Definition 2.1. A Regular Parametrization of Class C
28 pages
MIT18 02SC MNotes v11CDW
No ratings yet
MIT18 02SC MNotes v11CDW
5 pages
Unit2
No ratings yet
Unit2
26 pages
EM-I Mod-III
No ratings yet
EM-I Mod-III
27 pages
Unit 5 Sessionwise Problems
No ratings yet
Unit 5 Sessionwise Problems
6 pages
M PDF
No ratings yet
M PDF
38 pages
MA4006 Notes
No ratings yet
MA4006 Notes
112 pages
1 Topic: Vector Calculus
No ratings yet
1 Topic: Vector Calculus
30 pages
StewartCalcET8 13 01
No ratings yet
StewartCalcET8 13 01
19 pages
Co Vectors and Forms
100% (1)
Co Vectors and Forms
54 pages
vector calculus cambridge000
No ratings yet
vector calculus cambridge000
23 pages
Fundamental Theorem of Line Integral
No ratings yet
Fundamental Theorem of Line Integral
9 pages
Line and Surface Integral-1
No ratings yet
Line and Surface Integral-1
39 pages
UNIT-2 Differential Calculus Cluster C
No ratings yet
UNIT-2 Differential Calculus Cluster C
38 pages
Topology and Geometry for Physicists
From Everand
Topology and Geometry for Physicists
Charles Nash
3.5/5 (1)
Complex Integration and Cauchy's Theorem
From Everand
Complex Integration and Cauchy's Theorem
G. N. Watson
No ratings yet
Multiple Integrals, A Collection of Solved Problems
From Everand
Multiple Integrals, A Collection of Solved Problems
Steven Tan
No ratings yet
Budget_2024-25-Presentation
No ratings yet
Budget_2024-25-Presentation
29 pages
Europe_Water_Bodies,_Peninsulas_&_Other_Features_Part_1 (1)
No ratings yet
Europe_Water_Bodies,_Peninsulas_&_Other_Features_Part_1 (1)
17 pages
PMKVY
No ratings yet
PMKVY
1 page
Dsc 2024 Grl
No ratings yet
Dsc 2024 Grl
7,218 pages
5_6303306909986328567
No ratings yet
5_6303306909986328567
19 pages
5_6314512578445841062
No ratings yet
5_6314512578445841062
26 pages
Embankments On Organic Soils Surcharge
No ratings yet
Embankments On Organic Soils Surcharge
9 pages
Presentation 2 of 3: "The Building Blocks of Nature - The Standard Model"
No ratings yet
Presentation 2 of 3: "The Building Blocks of Nature - The Standard Model"
47 pages
X Class - SSC Andhra Pradesh Physics Differences Answers
100% (1)
X Class - SSC Andhra Pradesh Physics Differences Answers
4 pages
Strain Gauge and Rosettes PDF
No ratings yet
Strain Gauge and Rosettes PDF
10 pages
Vector Calculus2 1714325335885
No ratings yet
Vector Calculus2 1714325335885
17 pages
Progressive GR 9 2nd Q A
No ratings yet
Progressive GR 9 2nd Q A
13 pages
Electric Circuits by Peter Silvester-1
100% (1)
Electric Circuits by Peter Silvester-1
458 pages
Catalog Reductor Armonic Cu Rulmenti
No ratings yet
Catalog Reductor Armonic Cu Rulmenti
20 pages
Voltage Drop Calculation For Transmission and Distribution System
No ratings yet
Voltage Drop Calculation For Transmission and Distribution System
4 pages
FEMEngineering Units
No ratings yet
FEMEngineering Units
6 pages
Electric Machine Design (Module-3)
No ratings yet
Electric Machine Design (Module-3)
19 pages
W1 G12 Physics (2020-2021)
No ratings yet
W1 G12 Physics (2020-2021)
7 pages
Short Circuit
No ratings yet
Short Circuit
9 pages
Industrial Exhaust Fans As Source of Power: Archit Patnaik, S.M.Ali
No ratings yet
Industrial Exhaust Fans As Source of Power: Archit Patnaik, S.M.Ali
4 pages
Polymers PDF
No ratings yet
Polymers PDF
39 pages
FG
No ratings yet
FG
54 pages
Physics F3
No ratings yet
Physics F3
4 pages
Experiment No. 4 Shear Modulus: Course Title: Lab Report #: Submitted To: Submitted By: Date of Submission
No ratings yet
Experiment No. 4 Shear Modulus: Course Title: Lab Report #: Submitted To: Submitted By: Date of Submission
4 pages
Science 6
100% (1)
Science 6
6 pages
Mechanical Design of Overhead Lines PDF
No ratings yet
Mechanical Design of Overhead Lines PDF
70 pages
X-Ray Fundamentals Part 3 V1.7 Theory Training
100% (1)
X-Ray Fundamentals Part 3 V1.7 Theory Training
432 pages
f2 Physics Series One Jan 2024
No ratings yet
f2 Physics Series One Jan 2024
10 pages
12th Phy ch-1 Rotational Dynamics
96% (142)
12th Phy ch-1 Rotational Dynamics
53 pages
Modeling of Clinker Cooler
No ratings yet
Modeling of Clinker Cooler
18 pages
2025-03-20-NEP-B.Sc.(Prog) GROUP-13 MAY-JUNE 2025
No ratings yet
2025-03-20-NEP-B.Sc.(Prog) GROUP-13 MAY-JUNE 2025
5 pages