Chinmoy Taraphdar - The Classical Mechanics (2007)

The Classical Mechanics
Chinmoy Taraphdar

Asian Books Private Limited
r
r
The Classical
Mechanics
Chinmoy Taraphdar
Lecturer, Dept. of Physics
Bankura Christian College
Bankura, West Bengal
/4Slall ?',,,,/U 7JJ'loal,l. t . . l H l l l l . ~
7/28, Mahavir Lane, Vardan House, Ansari Road,
Daryaganj, N. Delhi-02
Registered and Editorial Office
7/28, Mahavir Lane, Vardan House, Ansari Road, Darya Ganj, Delhi-l 10 002
E-mail:asian@asiabbooksindia.com
World Wide Web: http://www.asianbooksindia.com
Phones: 23287577,23282098,23271887,23259161 Fax: 91 1123262021
Sales Offices
Bangalore 103, Swiss Complex No. 33,
Race Course Road, Bangalore-560 001
Ph. : 22200438 Fax: 918022256583 E-mail: asianblr@airtelbroadband.in
Chennai Palani Murugan Building No. 21, West Cott Road,
Royapettah, Chennai-600 014
Ph.: 28486927,28486928 E-mail: asianmds@vsnl.net
Delhi 7/28, Mahavir Lane, Vardan House, Ansari Road,
Darya Gan j, Delhi -110 002
Phones: 23287577,23282098,23271887,23259161
E-mail:asian@asiabbooksindia.com
Guwahati 6, G.N.B. Road, Panbazar, Guwahati, Assam-781 001
Ph.: 0361-2513020, 2635729 E-mail:asianghyl@sancharnet.in
Hyderabad 3-5-11Ol/11B I1nd Floor, Opp. Blood Bank,
Narayanguda, Hyderabad-500 029
Phones: 24754941, 24750951 Fax: 91-40-24751152
Email: hydasian@hd2.vsnl.net.in
Kolkata lOA, Hospital Street, Calcutta-700 072
Ph. : 22153040 Fax: 91-33-22159899 E-mail: calasian@cal.vsnl.com
Mumbai Shop No. 3 & 4, Ground Floor, Shilpin Centre
40, G.D. Ambekar Marg, Sewree Wadala Estate,
Wadala, Mumbai-400031
Ph.: 91-22-22619322, 22623572 Fax: 91-22-24159899
Pune Shop No. 5-8, G.F. Shaan Brahma Complex,
Budhwar Peth, Pune-02
Publisher
Ph.: 020-24497208 Fax: 91-20-24497207
E-mail: asianpune@asianbooksindia.com
I st Published 2007
ISBN 978-81-8412-039-4
All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system,
or transmitted in any form or by any means, electronic, mechanical, photocopying, recording
and/or otherwise, without the prior written permission of the publisher.
Published by Kamal Jagasia for Asian Books Pvt. Ltd., 7128, Mahavir Lane, Vardan House,
Ansari Road, Darya Ganj, New Delhi-I 10 002.
Typeset at : Arpit Printographers, Delhi-32
Printed at .' Gopalji Enterprises, Delhi.
Preface
This book, organised into ten chapters, is written to cover the syllabus of classical
mechanics for the students of physics at the graduate and postgraduate level. I
hope that the clear, lucid and comprehensive coverage of this book will help
students to gain a thorough grounding of the subject.
The beginning part of this book explains several chapters on the basis of
Newtonian mechanics and then the text explains the generalised co-ordinates
and Lagrangian mechanics along with Hamiltonian mechanics basically for the
holonomic system. The concluding chapter deals with the canonical
transformations by which the solution to the particular problem can be obtained
trivially. Some problems have been worked out to aid in understanding the
underlying theory at the end of each chapter.
Finally, the expressions of acknowledgements. I am indebted to my
colleagues, students who have kindly given me valuable comments and
suggestions. I acknowledge with a deep sense of gratitude my indebtedness to
the authors whose standard works in the field I have freely consulted to my
benefit. I also acknowledge my indebtedness to my wife 'Anamika', my daughter
'Sreetama' and my son 'Jyotirmoy' for their help at every stage of the preparation
of the manuscript. My special thanks are due to all concerned of 'Asian Books
Private Limited', especially to Ms. Purobi Biswas, Production Manager. Mr.
Subhadip Khan, the Branch Manager of Kolkata office for their kind help in
bringing out the volume in its admirable form and bearing with me at every
stage with unfailing patience and good humour.
It is fervently hoped that the book will be of value to the students and
teachers alike. Comments and suggestions for improvements to the text will be
thankfully acknowledged.
Chinmoy Taraphdar
"This page is Intentionally Left Blank"
Contents
Preface
Chapter 1. Vector
1.1. Fundamental Concept of Scalar and Vectors, 1
1.2. Unit Vectors and General representation of a vector, 1
1.3. Multiplication and Division of Vectors by Scalar, 2
1.4. Collinear Vectors, 3
1.5. Linear Dependence or Independence of Vectors, 3
1.6. Addition and Subtraction of two Vectors, 4
1.7. Addition of More Than Two Vectors, 5
1.8. Position Vector and Its Representation in Co-ordinate System, 5
1.9. Condition of Co-planarity of Vectors, 7
1.10. Rotational Invarience of Vector in Reference Frame, 8
1.11. Product of Two Vectors, 8
1.12. Scalar Tripple Product, 10
1.13. Vector Tripple Product, 11
1.14. Pseudo Vectors and Pseudo Scalars, 11
1.15. Vector Derivatives (Ordinary), 13
1.16. Vector Derivatives (Partial) and Vector Operators, 13
1.17. Laplacian and D'Alambertian Operator, 18
1.18. Vector Integration, 19
1.19. Gauss's Divergence Theorem, 20
1.20. Green's Theorem, 21
1.21. Stoke's Theorem, 22
1.23. Reciprocal Vectors, 24
1.23. Scalar and Vector Field, 24
1.25. Elementary Idea about Vector Space, 24
1.25. Linear Operator in Vector Space, 25
Summary, 26
Worked Out Examples, 29
Exercises, 34
Chapter 2. Linear Motion
2.1. Introduction, 37
2.2. Kinematics, 37
2.3. Basic Definitions of Required Parameters, 37
2.4. Velocity and Acceleration in Several Co-ordinate System, 40
2.5. Tangential and Normal Component of Velocity and Acceleration, 41
2.6. Radial and Transverse Component of Velocity and Acceleration, 42
2.7. Newton's Laws of Motion, 44
2.8. Accelerated Linear Motion, 45
(iii)
1-36
37-64
(vi)
2.9. Graphical Treatment of Linear Motion, 45
2.10. Conservation of Linear Motion, 46
2.11. Time Integral of Force (Impulse), 46
2.12. Work,47
2.13. Power, 48
2.14. Energy: Kinetic and Potential, 48
2.15. Conservative Force, 49
2.16. Conservation of Energy, 49
2.17. Center of Mass and Its Motion, 50
2.18. The Two Body Problem, 51
2.19. Application of the Principle of Linear Motion, 52
2.20. Mechanics of Variable Mass, 53
Summary, 55
Exercises, 64
Chapter 3. Rotational Motion: Rigid Body Rotation 65-94
3.2. Angular velocity and Angular Momentum, 65
3.3. Angular Acceleration, 66
3.4. Moment of Inertia and Torque, 67
3.5. Centrifugal force, 70
3.6. Rotational Kinetic Energy, 71
3.7. Angular Momentum for Rigid Body Rotation, 72
3.8. Kinetic Energy for Rigid Body Rotation, 74
3.9. Axes theorem for Moment of Inertia, 75
3.10. Calculation of Moment of Inertia in different cases, 76
3.11. Momental Ellipsoid or Ellipsoid of Inertia, 83
3.13. Moment and product of Inertia and Ellipsoid of inertia of some,
symmetrical bodies, 84
3.12. Moment of Inertia Tensor, 87
3.14. Routh's Rule, 87
3.15. Euler's Angles, 88
Summary,89
Exercises, 93
Chapter 4. Reference Frame 95-111
4.2. Non Inertial Frame and Pseudo Force, 95
4.3. Effect of rotation of earth on acceleration due to gravity, 99
4.4. Effect of Coriolis Force on a particle moving on the surface of earth, 101
4.5. Effect of Coriolis force on a particle falling freely under gravity, 103
4.6. Principle of Foucault's Pendulum, 104
4.7. Flow of River on Earth Surface, 106
Summary, 106
Worked out Examples, 107
Exercises, 110
(vii)
Chapter 5. Central Force 112-124
5.2. Definition and Characteristics of Central force, 112
5.3. Conservation of Angular Momentum under Central Force, 113
5.4. Conservation of energy under central force, 113
5.5. Equation of motion under attractive central force, 115
5.6. Application of central force theory to gravitation Deduction of
Keplar's law, 116
5.7. Energy conservation for planetory motion, 118
5.8. Stability of Orbits, 120
Summary, 120
Worked out Examples, 121
Exercises, 124
Chapter 6. Theory of Collision 125-137
6.2. Characteristics of Collision, 125
6.3. Lab Frame and Center of Mass Frame, 126
6.4. Direct or Linear Collision, 127
6.5. Characteristic of Direct Collision, 129
6.6. Maximum Energy transfer due to head on elastic collision, 130
6.7. Oblique Collision, 131
Summary, 132
Exercises, 137
Chapter 7. Conservation Principle and Constrained Motion 138-154
7.1. Characteristics of Conservation Principle, 138
7.2. Mechanics of a single particle and system of particles, 139
7.3. Conservation of linear momentum, 140
7.4. Conservation of Angular Momentum, 141
7.5. Conservation of Energy, 142
7.6. Constrained Motion, 145
7.7. Generalised Co-ordinates and other Generalised Parameters, 146
7.8. Limitation of Newton's Law, 151
Summary, 151
Worked out examples, 153
Exercises, 154
Chapter 8. Variational Principle and Lagrangian Mechanics 155-193
8.2. Forces of Constraint, 155
8.3. Virtual Displacement, 156
8.4. Principle of Virtual Work, 156
8.5. D' Alembert's Principle, 157
8.6. Lagrange's equations for a holonomic System, 158
8.7. Lagrange's equation for a conservative, non-holonomic system, 160
8.8. Introduction to Calculus of variations, 161
8.9. Variational Technique for many independent variables:
Euler-Lagrange's differential equation, 165
( viii)
8.10. Hamilton's Variational Principle, 166
8.11. Derivation of Hamilton's principle from Lagrange's equation, 167
8.12. Derivation of Lagrange's equations from Hamilton's principle, 168
8.13. Derivation of Lagrange's equation from D' Alambert's principle, 169
8.14. Derivation of Hamilton's Principle from D' Alambert's Principle, 171
8.15. Cyclic or Ignorable Co-ordinates, 172
8.16. Conservation Theorems, 172
8.17. Gauge Function for Lagrangian, 175
8.18. Invarience of Lagrange's equations under Generalised
Co-ordinate, transformations, 177
8.19. Concept of Symmetry: Homogeneity and Isotropy, 178
8.20. Invarience of Lagrange's equation under Galilean Transformation, 179
8.21. Application's of Lagrange's equation of motion in several mechanical
systems, 180
Summary, 186
Exercises, 193
Chapter 9. Hamiltonian Formulation in Mechanics 194-209
9.2. Hamiltonian of the System, 194
9.3. Concept of Phase Space, 195
9.5. Hamilton's Canonical Equations in different Co-ordinate System, 197
9.6. Hamilton's Canonical equations from Hamilton's Intergral Principle, 199
9.7. Physical Significance of Hamiltonian of the System, 201
9.8. Advantage of Hamiltonian Approach, 201
9.9. Principle of Least Action, 201
9.10. Difference between Hamilton's Principle and the principle of least,
action, 203
9.11. Application of Hamilton's Canonical Equations, 203
Summary, 206
Exercises, 209
Chapter 10. Canonical Transformations 210-232
10.1. Introduction to Canonical Transformations, 210
10.2. Hamilton-Jocobi Method, 215
10.3. Application of Hamilton-Jacobi method to the particle falling freely, 217
10.4. Han,iIton's Characteristics function, 219
10.5. Action and Angle Variables, 220
10.6. Application of Action Angle Variables to Harmonic Oscillator Problem, 221
10.7. Poisson's Bracket, 222
10.8. Poisson's Theorem, 223
10.9. Jacobi's Identity, 224
10.10. Lagrange's Brackets, 225
10.11. Liouville's Theorem, 226
Summary, 227
Exercises. 231
Chapter-I
Vector
1.1. Fundamental Concept of Scalar and Vectors:
Any physical quantity having both direction and magnitude is called vector
quantity and these quantity must obey some fundamental laws of addition and
subtraction.
But the quantity having only magnitude, but no direction is called scalar
quantity. Basically, the magnitude or measure of a scalar quantity is quite
independent of any co-ordinate system, but the measure of vector quantity
depends on the frame choosen.
All scalar quantities are, however, subject to the ordinary algebraical laws
of addition and multiplication:
viz. (i) a + b = b + a
a x b = b x a (Commutative law)
(ii) (a + b) + c = a + (b + c)
(a x b) x c = a x (b xc) (Associative law)
(iii) a x (b + c) = a x b + a x c
(a + b) x c = a Xc + b x c (Distributive law)
On the other hand vectors are subject to the triangular or parallelogram law
of addition (or subtraction).
1.2. Unit Vectors and General representation of a
vector:
A vector is analytically represented by a letter
with an arrow over it, i.e, A and its magnitude is
p
o
* Basically, Physical quantities are Tensor. Any tensor is that quantity which may
change under co-ordinate transformatIOn. The tensor of rank 'n' is that quantity
which transforms through 'n' no. of cu-efficient matrices. Actually a vector is a
tensor of rank 1 where as a scalar is a tensor of rank O.
2 The Classical Mechanics
denoted by IAI or A. Graphically, a vector is represented by an arrow 0 P
defining the direction, the magnitude of the vector being indicated by the length
of the arrow. Here the end 0' of the arrow is called the origin or initial point of
the vector, and the head p is called the terminal point or terminus.
A vector having unit magnitude is called unit vector. This is in general
denoted by n and so ini = I unit.
The significance of any unit vector is that it is required to construct* any
vector along any direction. Practically, vector can be constructed by arranging
required no. of unit vectors or fraction of it one after another in a choosen
direction. Thus a vector can mathematically be represented by simply the scalar
multiplication of its magnitude with directional unit vector along it.
i.e, A = IAI n
or A = An.
Thus unit vector is in that sense significant
because it gives direction of given vector and if it
be reversed then opposite vector with same
magnitude can be obtanied. Let A = A n is a given
vector. If -A be its opposite vector then
-A = A(-n) when IAI = I-AI = A
Now if a given vector be added with its opposite vector in vector addition
method then resultant vector in called zero vector or null vector. Mathematically.
A+(-A) =6.
This in represented simply by a 'dot' and only in vector algebra, it is
significant.
1.3. Multiplication and Division of Vectors by Scalar
Any vector can simply be multiplied or divided by scalar quantity. In case
of multiplication**, the new vector will have magnitude magnified by that
* Let, A = 5 which has magnitude 5 units and a
specifIC direction. If we now arrange 5 unit vectors
one after another in that specific direction then we
will get
A=5n=An when A = 5 unit
- -+-
Similarly for A = 3.7, A = 3n + (.7n) = 3.7n
** The scalar multiplication of vector by scalar quantity obeys the following rules:
rnA Am (Commutative law).
rn (nA) (rnn)A (Associative law)
(Ill + n)A rnA + nA (Distributive law)
Vector
3
quantity in compare to the magnitude of original vector. Also, in similar way,
for division by scalar quantity, the magnitude of new vector will be reduced by
that quantity. (Here we take that the scalar has magnitude> l). But in both
case, the new vector will have the same direction to original vector. So for a
vector A and scalar p,
H = pA, where, B = pA > A and
- 1 _ 1
also, B'= -A whereB,=-A<A (forp> 1)
p p
1
Since - = q = another scalar, so, we have
p
_ 1 - -
B' =-A = qA
p
and B' = qA.
So, division of vector by scalar is one type of multiplication by another
scalar.
1.4. Collinear Vectors:
The term collinear is synonymous with parallel. So vectors having the same
or parallel lines of action are said to be collinear. So collinear vectors are parallel
to each other and they are 'like vectors'. The new vectors H or H' , as obtained
through multiplication or division by scalar quantity with an original vector A
(as discussed earlier), are collinear vectors with A.
1.5. Linear Dependence or Independence of Vectors :
A set of vectors Ai (i = 1, 2, .... , n) are said to be linearly dependent, if
for a set of scalars a
i
(i = 1, 2, .... , n) where not all ai' . s are zero,
- - -
alAI + a2A2 + .... + anAn = O.
i.e.
L>iAi =0.
But if all 0../ s are zero, the vectors Ai are said to be linearly dependent.
* The cross product of two collinear vector is zero. (see cross product)
**LetthreevectorA =(2,1,1),13 =(1,2,2),(: =(1,1,1).
Then for aA + P13 + y(: = 0,
we get, 20, + 13 + Y = 0
0,+ 213 + y = 0
0,+ 213 + y = O.
By solving, we get a = -k, 13 = -k, y = 3k. Since, k is arbitrary, there exist a, 13, y
which are not all zero. So the vectors are linearly dependent.
In another way, if A, 13, e are three vectors and a, p, "I are three scalar
quantities, then for,
aA + P13 + "Ie =
if a = P = "I = 0, thenA,13, e are independent.
But for a, p, "I, not all zero, these vectors A, 13, e are dependent.
Although the definition of linear dependence or linear independence refers
to a set of vectors, we shall also state that the individual vectors AI> A
2
, .... A
n
are linearly dependent or independent.
1.6. Addition and Subtraction of two Vectors:
Any vector can be added with one or more than one vector and the new
vector obtained by such addition, is called resultant vector. This is denoted by R
and soR = LAi.
For two vector addition, the resultant vector R can be obtained by triangular
or parallelogram law or by component* resolution method.
Let A and 13 are two vectors
which are at an angle a with each
other. If R be their resultant
vector then, R = A + 13 .
This can be obtained from
Triangular or Parallelogram law.
By this law, the resultant has
magnitude
R =IA + 131 = (A2 + B2 + 2AB cos a)1I2.
Also, if, ro = A - 13 = A + ( - B) be
subtraction if two vectors which is
basically the vector addition of 1st vector
with opposite vector of 2nd vector, then
by same law, we get the magnitude of
this subtraction as, ro =IA - 131 = (A2 +B2
- 2AB cos a)I/2**.
I
11t - a
~ I
-8/
/
I
R
~ ~ ~
~
* Here, component means effective part of vector along choosen direction.
** By replacing a by (7t - a) in addition formula, we get corresponding subtraction
formula. Here we should remember that 1-
13
1 = 1
13
1 = B. In classical mechanics, for
the estimation of relative velocity of two moving particles or, for obtaining similar
quantities, the knowledge of vector subtraction is required.
Vector 5
Again, if 8 and ~ be the respective angle made by the resultant and
subtraction vector with given vector A, then,
B sin a Bsina
tan 8 = and
A+B cos a
tan ~ = A - B cos a .
1.7. Addition of More Than Two Vectors:
Polygon Rule and Perpendicular Resolution Method : Any no. of given
vectors can be added by any of two methods. One is Polygon rule and other is
Perpendicular resolution method.
In case of Polygon rule, if a Polygon has 'n' arms with (n - 1) arms
representing (n - 1) vectors in direction and magnitude in the same sense, then
the nth arm of the same polygon gives the resultant vector of all given vector
in direction and magnitude in opposite sense. So by this rule, R = I Ai and
basically this rule follows triangular rule of vector addition. On the other hand,
in case of perpendicular resolution method, if each of given vectors be resolved
into two mutually perpendicular components along x or y direction, (say), then
the magnitude of resultant of these vectors will be:
R =(RT + ~
112
where, Rl = IAi cos 8
i
, R2 = IAi sin 8
i
Here, 8
i
is the angle made by Ai with +ve X axis. Also, if this resultant
makes angle ~ with same +ve X axis, then,
tan =(R2/) = (IAI sin
lIl
1
lRl "A cos8 .
~ I I
1.8. Position Vector and Its Representation in Co-ordinate System:
The position vector of any point P, Y
taken in any frame (2 or 3 dimension or in
any dimension of configuration space) is 1
obtained by the vector joining origin to that
point. This position vector is denoted by r
~
and so for any point P, r = OP = position
vector of P W.r.t. the origin '0'
A
e r
(x, ~ I p(r, 8)
I
I
I
I
I
I
I
I
I
I
I
I
If this point 'P' has cartesian co-
~ L - - - - - - - ~ - - - - ~ - - ~ X
ordinates (x, y) in 2 dim. Thenr = xi + yJ
A
i
where, i and J has unit vector along x and y axis, respectively.
6
Now, in plane polar if
(r, 8) be the co-ordinates or the same point
P then,
x = r cos 8, y = r sin 8 where
i = r cos 8 - e sin 8

J
A
r
) = r sin 8 + e cos 8. __ L-____________
where rand e are respective unit vectors
along increasing rand Q directions.
In this plane polar co-ordinate system
(2-dimension).
;: = r cos 8 (r cos 8 - e sin 8) + r sin 8 (r sin 8 + e cos 8)
= rr where, r is unit vector along ;:
and it has no component in 8- Z
direction .
In 3 -dimensional cartesian
coordinate system, if (x, y, z) be the
coordinate of the point P, thus

position vector
OP
=;:
p (x, y, z)
= xi + y) + zk where, i,), k are
respective unit vectors along x, y,
and z directions.

Here, in spherical polar
coordination system if (r, 8, <p) be
the co-ordination of the same point X
'P' then,
_ A
r = rr
and here, r cos ai + cos p) + cos yk .
when, a, p, yare the respective angles made by;: with x, y and z axis and
cos a, cos p, cos yare the direction cosines.
* Here, S-direction means along the direction of increase of S. Actually, if the
rotation through angle S be small or infinitesimal then it may be regarded as a
vector quantity; for, then they are described by the body in a small interval of
time is more or less a straight line and it is thus represented as a vector.
** The direction cosines are in general represented by (I. m, n)
when, I = cos a, m = cos n = cos y, and f2 + m
2
+ n
2
= 1.
In that case, for;: = rr ,
A r ZA A
r = - = -/ + -; + -k = COSaI + cos p; + cos yk.
r r r r
Vector 7
z
~ - - - - - - - - - - - - - ~ - - ~ y
Also, in 3 dim. cylinderical polar co-ordinator system, if (p, cp, z) be the
co-ordinates of the same point P,
then, ; = p cos cp (p cos cp - ~ sin p) + p sin cp (p sin cp + cos cp) + zk
or, ; = pp + zk .
where, p and are respective unit vectors along p and cp directions. This is
position vector in cylinderical polar coordinate system.
1.9. Condition of Co-planarity of Vectors:
Two vectors are said to be co-planer if they lies in the same plane. If A
and 13 be two co-planer vectors then C = aA + 1313 will be also co-planer with
A and B.
Similarly, for three co-planer- vectors A, 13, C , we have aA + 1313 + yC = 0
for three possible scalars (a, 13, y).
For their components, we have,
a Ax + 13 Bx + y C
x
= O.
a Ay + 13 By + y C
y
= O.
a A
z
+ 13 B
z
+ y C
z
= O.
and by eliminating these a, 13, y, we get the following determinant form
This is the condition of co-planarity of the three vectors.
* For three co-planer vectors A, 13, C , it can be shown that
A.(BxC)=B.(CxA)=C.(AxB)=O
(See scalar tripp Ie product.)
1.10. Rotational Invarience of Vector in Reference Frame:
Consider a vector A = A) + AyJ + Ai- in 3 dimensional rectangular
cartesian system. Now if this frame be Y
rotating through angle '8' keeping the y'
direction of vector A unchanged then
in new frame, the vector is
.\ = + +
where, (Ax', A
v
', A
z
) are components
taken along new co-ordinate axes.
Basiclllly, the mag!litude i.e., the
length of vector A must obviously be Z
quite independent of the orientation of
the co-ordinate system or frame of
reference, we have,
A2 =A2 +A2 +A2 =A,2 +A,2 +A,2
x y z x y z
1
/
z'
or, (A; + A; + A;)"Z = + A/ +

The magnitude of vector will remain invarience under rotation.
1.11. Product of Two Vectors:
The product of two vectors may give scalar or vector quantity. For this case,
the product may be divided into two classes--one is called DOT product or scalar
product and other is called CROSS product or vector product.
(A) Scalar Product (DOT Product)-For two vectors A and B having
angle a between them, the DOT product is given by,
A . B = AB cos a.
This is a scalar quantity and is simply obtained by multiplying the magnitude
of the two vectors and cosine of angle between them.
This DOT product has the following characteristics-
(i) This is a scalar quantity.
(if) For two vectors (A, BJ. if A.l B then A. B = O.
This is condition of perpendicularity.
(iii) DOT product is commutative
i.e., A . B = 13. A .
(iv) For three unit vectors t, J, k along x, y, z axes respectively,
t.t =J.J =k.k = l.
;.J = J. k = k.J = O.
Utilisation of this DOT product are, _
(i) By this product component or a given vector A along any given
Vector 9
direction can be obtained. This is given by component = (A. n)n .
where, iz is unit vector along given direction.
(ii) Angle between two vectors A and B can be obtained and this is given
by,
(
A.B)
e = cos-
i
AB .
(iii) In Physics, several scalar quantities such as work done (W), current
flow (1), .... etc. can be obtained by using DOT product.
Mathematically, these are given by
W= fdW = LF.dr (F = Force)
I = f dI = S} as. (J = Current density)
(b) Vector Product (CROSS Product)-For two
vectors A and B, having angle u between them, the cross
product is given by
C = (A x B) = IA x Bin
when IA x BI == AB sin u.
A x B =(AB sin u)n.
This is a vector quantity and magnitude of it is obtained
by multiplying the magnitude or two vectors with sine of
angle between them.
This direction or this cross product can be obtained by
following right handed screw relation and by this rule, the product vector will
be perpendicular to the plane of two given vectors.
The CROSS product has following characteristics.
(i) This is a vector quantity.
(ii) For two vectors A and B,if A!IB,thenAxB =0.
This is condition or colinearity of two vectors. *
(iii) CROSS product will be anti commutative, is (A x B) =-(B x A) and
the direction of this product vector will follow right handed screw
rule.
(iv) For three unit vectors i, J and k along x, y and z axes respectively.
(i x i) = (J x J) = (k x f) = 0

A
and
i xj
=
k'
jxk
-
i
,
k x i
-
j

j
xi
=
-k'
kxj
=
-i ' i
xk
= - j.
* As mentioned earlier.
Utilisation of this CROSS product are:
(i) By this product, co linearity of two vectors can be examined.
(U) Angle between two vectors can also be obtained from it as
(iii) In Physics, several quantities, such as cross section of any surface (5),
moment of force Cr) , angular momentum (1.) .... etc. can be obtained
by using this product. These quantities are mathematically, given by
when d(i and ib are two adjacent elementary vectors.
i = J di = r x F moment of force = torque.
L = r x p = r x (mv) = m (r x v) moment of linear momentum =
angular momentum.
1.12. Scalar Tripple Product :
For three vectors A, 13, C, the scalar product is given by,
S = A .(13 x c)
This is a scalar quantity and is popularly known as box product.
This product has following characteristics.
(i) This is a scalar quantity.
(ii) For any three vectors A, B, C ,
A.(BxC) =B.(CxA) = C.(AxB).
(iii) For three co-planner vectors, A, 13, C
A. (B x c) = O.
Because, in that case (B xC) is perpendicular to A when (B xC) is
also normal to the plane of two vectors Band C .
* Geometrically, the cross product can be interpreted by the twice the vector area
of a triangle formed by the magnitudes of the two vectors as adjacent sides.
1 (_ -)
Area of triangle = "2 a x b .
** This box product is also denoted as
[A 13 c] :; A. (13 xC) such that
[ABC]=[BCA] = [CAB]'
Vector 11
(iv) A. (A x B) = 0 when (A x B) -1 A. This scalar tripple product has
popular application in rmding volume element which is mathematically
given by,
v = fvdv = f f f da . (iG x de).
1.13. Vector Tripple Product:
This is now represented for three vectors A, Band C as
S = A x (B x c)
and it is a vector quantity.
This product has following characteristics:
(i) AX(BxC) = (A.C)B-(A.B)C
(ii) A x (B x c) * (A x B) x C
(iii) AX(BxC)+Bx(CxA)+Cx(AxB) =0.
We can now remember an example of this vector tripp Ie product which is
centrifugal force
Fc =-m[wx(wxr)] = m[wx(rxw)]
1.14. Pseudo Vectors and Pseudo Scalars:
There are several quantities in Physics like velocity, acceleration etc. whose
directions are clearly indicated by the direction of motion of the particle or the
system.
But other types of quantities are angular velocity, angular momentum, ....
etc. whose direction does not indicate the direction of rotation of the body. Here
only the axis of rotation is well defined. The 1st type of vectors as mentioned
above, is called polar vector where as the 2nd type of vectors are axial vector
and these are basically known as Pseudo vectors.
Actually, any ordinary vector or polar vector (like displacement, velocity,
. force, .... etc.) changes sign under
inversion or mirror reflection, but the axial Z
vectors or pseudo vectors which are
obtained by taking cross product or two
polar vectors (like, angular velocity, torque
... etc.) remain invariant under inversion.
To understand it, let w be the angular
velocity which is an axial vector. This is O I C ~ x
* If any axis of a co-ordinate system be reversed by rotating it through angle 180
0
then that improper rotation is called 'inversion', or 'mirror reflection'.
12
given by

v = wxr (for point particle rotation)
where, v is instantaneous velocity at illstantaneous
position r during rotation along circular path.
Here we get,
y
rxv =rx(wxr)

= (r . r) w - (r . w) r
2 (.: r 1- w'\
=r w r.w=OJ
/
I
W=2rxv.
r
-z
Now under inversion or mirror reflection about xy plane (through inversion
of z axis), the direction of r and v in right hand system also remains invarient
in left handed system and the direction ofw remain same under inversion
obeying right hand screw rule. So this axial vectorw is a pseudo vector.
On the other hand, the dot product of two polar vectors is a scalar quantity**
but the dot product of one polar and one axial or pseudo vector is called pseudo
scalar. As for example, the volume 'V' is a pseudo scalar,
because, V = ii . (b x c) (where, ii, b, c are three adjacent vectors)
ii = Polar vector.
(b xc) = Pseudo vector.
Here we should note that pseudo scalar will change sign under a mirror
reflection.
* Inversion can also be made by reversing any axis, say x-axis. In that case also W
has the same direction because of the invariance of sense of rotation.
Here, in 1 st frame.
1
lv=-r
x
v
r2
And in y the 2nd
frame
z
z
w = x (v)]
r-
1
= 2"" r x v . _y
r
** Actually, the dot product of two polar vectors is a scalar quantity which is called
'true scalar'.
y
Vector 13
1.15. Vector Derivatives (Ordinary) :
Let a vector A which is a continuous function of same scalar variabie, say x.
Thus, A = A(x).
It is observed that the differentiation in .vector analysis follows the same
rules of differential calculus. The only difference is that of the non-commutative
property of the vector product of two or more vectors.
The ordinary differentiation of the vector A W.r. 1. the scalar x is
dA = lim l1.A = lim A (x + l1.n) - A ( x )
dx Llx ~ o!1.x Llx ~ 0 !1.x
By using this 1st principle, various rules of differentation* are
dA dA A dAy A dA A
(i) = __ x i + __ j +_Z k.
dx dx dx dx
- -
(iii) .!!..- (A . B) = dA .13 + A. dB .
dx dx dx
- -
d (- -) dA - - dB
(iv) dx A x B = ~ x B + A x dx'
1.16. Vector Derivatives (Partial) and Vector Operators:
Gradient, Divergence and Curl:
Consider a vector A such that
A = A(x,y)
-----------------
* In physics vector differentiation w.r.t. scalar is used in various cases, such as,
for position vector, r = r(t) ,
Velocity,
l1.- dr
- = lim.....!.... =-
v r u ~ l1.t dt'
Acceleration,
__ lim ~ v _ dV
a - ru 0 l1.t - dt .... etc.
Here partial differentiation of A w.r.t. x or y as
aA .
hm
ax
so, we can have, for
A = A(x,y),
. .
dA
- = hm hm

- -
_ aA aA
dA = ax dx+a;d
y
.
Similarly, for A = A( Xi) , i = 1, 2, .... n, the ordinary change of A
"" aA
dA = dx
j
j 'Xi '
Now with partial derivative operators, the vector differential operators
'DEL', written as V, is defined as
A a A a A a
v = i ax + j ay + k az [In Cartesian Coordinates (x, y, z)]
= P [2 + .!. ( sin cp - cp) +
V ap p cos cp sm cp acp
 + (cos cp - sin cp ) + Ie
p acp sin cp cos cp ap az . [In
coordinates (p, cp, z)].
and also,
Cy linderical*
- _ r [3 +.! (2 sine - cose) +.! (sincp - coscp)
V - ar r cose sine as r coscp sincp acp
polar
+9[(2 cose _ cose (sincp _
sine cose ar rae r sine cos cp sin cp acp
+ [_l_(COSCP _ +_l_(COSCP _ + 2
sin e sincp cos cp ar r cose sincp coscp as r sine acp
* In cylinderical polar coordinates, x = p cos <p, y = p sin <p, z = z where i =
(
A A) A (A. A ) a
p cos cp - cp sin p ,j = p sm cp + Cjl cos Cjl ,and a
a 1 8 1 8 'X
1 a a
- ----+---
ay - sin Cjl 8p p cos cp 8cp .
Vector 15
[In Spherical* polar coordinates (r, e, <p)]
With this V operator, three operations can be defined which are,
(i) Gradient (V)
(ii) Divergence (V.)
and (iii) Curl (V x).
Here, gradient operation is only made on a scalar function <p(r) = <p(x,y,z),
defined in a certain region. This operation gives a vector quantity which is defined
as,
_ d<p A
A =V'<p = dnn
w h e n ~ ~ ) =IV<p1 is basically the maximum rate of change of<p(x,y,z) in
normal direction to the surface <p( x, y, z) = 0 or
constant. This gradient operation-- is given by
~ o, y = r sin 0 sin q>, z = r cos 0
where i = (r sinO cosq> + e cosO cosq> - q, sinq
j = (r sinO sinq> + e cosO sinq> + q, cosq, k = r cosS - e sinS
a
a a a
and,
-+
ax
sinO cosq> or rcosO cosq> ae r sinO sinq> oq>
a a a a
- -+ +
Oy sin 0 sin q> or r cosO sinq> ae r sinO cosq> oq>
a a a
-------
=
az cosO or rsinSoS
This gradient operation gives the directional derivatives along a direction (given)
which is given by
Directional Directive == V q> . a
whena is unit vector along given direction w.r.t. the surface q> = o.
. dqJ dqJ
Here, SInce, --;;; is rate of change of qJ along PQ and dn is the maximum
rate or change of qJ along nonnal direction PR then dn = dr cos 8 = dr . n
and
dqJ
dr
dqJ d ~
-cos8<-'I'-
dn dn
and so, ~ : (n.dr) = ~ : dn = dqJ (When VqJ == ~ : n)
The divergence operation on a vector quantity, say, A, gives a scalar
quantity which is mathematically given by
_ _ aA
x
aAy aA
z
S=V.A = fu+ay+8;""'
This physically gives the amount of
outward flux of vector A per unit volume.
aA
x
aAy aA
z
Actually, fu' ay' 8;"" are the rates of
change of these components in respective
x, y, and z axes. Taking a cuboidal box of
volume
dv = dx, dy, dz
the net out flow or the flux of vector A
in the x-direction is
y
z
(
" a ~ ) aA
Ax + __ x dx dy dz - A x dy dz = __ x dx dy dz
ax ax
aAy
Similarly, the flux of same vector.A.. in y and z directions are a; dy dx dz
aA
z
and az dz dx dy resp. so the net outflow per unit volume is
(
aA
x
+ aAy + aAz) __ _
ax ay az - V.A.
Let us now discuss about curl of a vector A which is mathematically given
by another vector field B
as
when
Vector 17
(
VxA) = (OA
X
_ OAz)and(vxA) = (OA
y
_ OAx)
y oz ax z ax oy
Physically every component of curl of any vector is obtained from line
integra1* of that vector per unit area of a surface. From figure, we see that
(v x A) = (! A .dr)!dx dz
y J'o abc 0
_ -- A dz+ A +--" dz dx- A +--dx dz-A dx
1 [ ( OAt) ( OAz) ]
- dx dz z x oz z ax x
= (OA
x
_ OAz)
oz ax'
z
a
~ y
x
The basic formulae involving V for two differentiable vector function A
and :i3 and for two scalar functions (<p, \jf) (also differentiable) of position (x,
y, z), are given below.
(i) V (<p + \jf) = V<p + V\jf .
(ii) V.(A+:i3)= V.A+V.:i3.
(iii) V x (A +:i3) = V x A + V x:i3.
(iv) V.(<pA) = V<p.A+<pV.A.
(v) V x (<pA) = V<p x A + <p V x A .
(vi) V.(Ax:i3) = B.(VxA)-A.(VxB).
* See line integral of a field vector.
18
(vii)
(viii)
(ix)
(.r:)
(xi)
(xii)
VX(AX13) = (13.V)A-13(V.A)-(A.V)13+A(V.13).
V(A.13) = (13.V)A+(A.V)13+13x(VxA)+Ax(Vx13).
V.(V<p) = V2<p .(V
2
= Laplacian operator).
V x (V<p) = O.
V.(VxA)=O.
Vx(VXA) = V(V.A)-V
2
,A..
1.17. Laplacian and D' Alambertian Operator:
In vector analysis, another two important operators are:
(i) Laplacian operator (V2)
(ii) D' Alambertian operator (IT)
which are very important in several branch of Physics. This operators are
mathematically given by,
(cylinderical coordinate)
= -- r - + - sm8- + --
1 0 ( 2 0) 1 0 (. 0 ) 1 0
2
r2 or or r2 sin8 08 09 r2 sin
2
9 0<p2
and
IT = V
2
__ 1 0
2
c
2
0(2
where C = velocity of light or e.m. wave in free space.
* Here this property of curl (VX) operation gives a most popular expression of
pseudo vector was,
Vxv = Vx(wxr) = (r.V)w-r(V.w)+ w(V.r)-(w.V)r.
See, w is constant vector, V opertion on w gives zero. So we
get, Vxv =w(V.r)-(w.V)r
= 3w-w = 2w.
Vector 19
1.18. Vector Integration:
Vector Integeration is basically identical with usual procedures of integral
calculus. There are three types of vector integration can be mentioned which
are
(i) Line integral
(ii) Surface integral
(iii) Volume integral
(I) Line Integral: In physics we usually have to go through several type
of integration like
L q> dr, LA. dr, L A x dr
and all of these are line integral.
Here q> and A are both position dependent or not and they are respective
scalar and vector fields.
This integration are also known as contour integrations and c is a contour,
either close or not
A A A
such
r
xi + yj + zk
A A A
dr dx i + dy j + dz k .
hI b:J
and
i f q> dx + J f q> dy + k f q> dz
hI b:J
LA. dr = fAx dx + fAy dy + f A
z
dz
and LAxdr = i f(Aydz-Azdy)+J f(Azdx-Axdz)+k f(AxdY-Aydx).
Here all the limits of integrations are obtained from the nature of contour,
'C'.
(ii) Surface Integral: This is commonly known as double integration which
is taken over a surface area (given).
The integrals like
Lq> as, LA. as, LA x as are common surface integrals and here d'S is small
surface element which can be written as, d'S = ds n
(n is unit normal to ds).
To evaluate surface integrals, it is convenient to express them as double
integrals taken over the projected area of the surface'S' on one of the coordinate
planes.
Mathematically, the integral over projected region is given by,
I I A . n lit
l
A d!1 or = I I A . n d
l
:, ~ l
R n.k R n.l
or IIA.nlitIA ~ l
R n.]
where, n is mathematically given by
V'cp
n = IVcpl for given surface, cp = cost.
(iii) Volume Integral: This is a tripple integral of a vector point function
over a certain volume element, say V.
This is simple given by,
Iv A dv = i I Ax dv + J I Ay dv + k I A
z
dv.
1.19. Gauss's Divergence Theorem
This theorem states that if A is a continuously differentiable vector point
function and '8' is a closed surface enclosing a volume 'v', then
Is A . n ds = Iv V . A dv .... (1 )
where, n is the unit outward drawn normal vector.
Briefly, this theorem states that the normal surface integral of a function
A over the boundary of a closed region is equal to the volume integral of V . A
taken throughout the region.
We will now suppose that the region V is such that it is possible to choose
rectangular cartesian co-ordinate system such that each line parallel to any co-
ordinate axis which has internal points in common with the region meets the
boundary 8 in two points.
In terms of cartesian co-ordinates, the equation (1) stated as follows,
r (A dy dz + A dz dx + A_ dx dy) = J (a Ax + a Ay + a Az) dxdydz.
J
s
x y v ax ay az
Consider now the volume integral Iv( a z ) litdydz let R3 denote the
projection of the region V, on the OXY plane. Every line through a point
(x, y, 0) of R3 meets the boundary 8 in two points. Let the z co-ordinates of
these points be,
and
z = cp (x, y), z = \If (x, y)
cp (x, y) ~ \If (x, y).
Vector 21
Now, f oA
z
dxdydz = f I [l oA
z
dZ] dxdy
OZ R) \jI OZ
= f f[Az (x,y,<p) -A
z
(x,y, \jf)]dxdy
R)
Let us now denote the parts of the surface S corresponding to z = <p (x, y)
and z = \jf (x, y) by SI and S2 respectively.
If it denotes the outward drawn unit normal vector at any point of S, we
have,
f f Az (x,y, <p) dxdy = 1 A
z
. (it. k) ds
R3 s,
f fAz(x,y,\jf)dxdy = -1 Az.(it.k)ds
R3 s,
for the outward drawn normal at any point of SI makes an acute angle with
positive direction of z axis and then at any point of S2 makes an obtuse angle
with z axis.
From above equation we now have
f
oAz dxdydz = r A
z
it. k ds + r A
z
it. k ds
v oz Js, JS2
fs A
z
(ii ,k) ds
Similarly, it may be proved that
f
oAy r A ~
--dxdydz = J! Ay n.j s
V Oy s
and
r oAx dxdydz = r Ax ii.t ds
Jv ox Js
So we get,
f
(
OAx oAy OAz) f( A A A)
--+--+-- dxdydz = A i +A j' +A k .nds
Vox Oy oz x y z
This is Gauss divergence theorem.
1.20. Green's Theorem
This theorem states that if <p and \jf are two continuously differentiable scalar
point functions such that V<p and V\jf are also continuously differentiable, then
fJ<p\72\jf - \jf\7
2
<p) dv = fJ<p(\7
2
\jf) - \jfV (<p)]. n ds
where V is the volume which is involved by S.
Now, div(<pV\jf) V .(<pV\jf)
V<p . V\jf + <p \7
2
\jf
L<pV\jf.nds = fvV<p.V\jfdv+ fv<P\72\jfdv
Now, D<pV\jf-\jfV<P).nds = fJ<p\72\jf_\jf\72<p)dv
and this result of this theorem can also be written as,
f
(
2 2 ) f' (O\jf o<p ) 1:: o\jf o<p
v <p\7 \jf - \jf\7 <p dv = s\. <p an -\jf on . uS and an and on are called
directional derivatives of \jf and <p respectively along outward drawn normal at
any point of S.
1.21. Stoke's Theorem:
This theorem states that if A be any continuously differentiable vector point
function and S be a surface bounded by a curve 'C'
then, LA. dr = D V x A) . n ds
where the unit normal vector n at any point of S is drawn in the sense in
which a right handed screw would move when rotated in the sense of description
of C. We can now easily prove this Stoke's theorem for plane which is also
often referred to as Green's theorem.
Take a system of cartesian rectangular co-ordinate axes such that the plane
of a given surface S is the OXY plane and the z-axis lies along the direction of
the normal vector ii.
A A A
Let
A
Ax i + Ay j + A
z
k
We have, l
A
.
dr
l
A
.
nds
f
(
A A A) (A oX A oy A oz)
i Ax + jAy + k A
z
. i - + j - + k - x ds
c os os os
f
(
A ox + A a
y
) ds
c x os y os
Vector
dz
for dx = 0, the tangent at any point lying in the OXY plane.
Also, fJv x A). n ds = fJv x A). k ds
= f'(OA
y
_ OAx)dxd
Y
ax Oy .
Thus for the case of plane surface the theorem is equivalent to showing
f
(
A dx + A dy) = f [a Ay _ a A x ) dxdy
C
X
y cox Oy .
We shall now prove it. Firstly suppose y
that the region'S' is such that any line
parallel to either axis meets the boundary
'c' in at the most two points. Let the
region be included between the lines x =
x=a x=b
a and x = b and any line parallel to y axis
with abscissa x meet c in the points given
by
y = \jf (x)
y = <p (x).
Y= \jI(x)
23
where, \jf (x) <p (x). 0 X
Thus the boundary curve 'C' is split up into two arcs C
1
and C
2
as shown
in figure.
We have,
f
OA
x
dxdy
s oy
f
b[fljl (x) a Ax d
Y
] dx
cp(x) oy
a
b
f[Ax (x, \jf(x)) - Ax (x, <p (x))] dx
a
-fAx (x,y)dx- J Ax (x,y)dx
C2 CI
-LAx (x,y)dx.
Similarly, we may show that
J
OA
y
dxdy = JAy(x,y)dy
s ax C
Finally,
f ~
O y OA)
f(
A dx+A dy) == ___ x dxdy
eX Y ox oy
1.23. Reciprocal Vectors:
For every set of vectors, say, (a, b, c) , there is a reciprocal set (a', b', c')
such that
a.a' == b.b' == c.i? == 1.
And for existence of this reciprocal set, the restriction is that the vectors
(a,b,c) mustnotbecoplanner, i.e., a.(bxc) :;t:0.
Now for this case, the vectors of reciprocal set for the given set of vectors
(a, b, c) are given by,
b x c - c x a _, a x b
a' == a. (b x c)' b' == a . (b x c)' c == a . (b x c)"
1.23. Scalar and Vector Field:
Basically, any scalar or vector field means a set or collection of a no. of
scalars or vectors. For example, a vector field V can be expressed as V ==
(a),a2' .... ) where, a, eV i.e., a;'s are component vectors along several
direction in field. Similarly, a scalar field, say F is a collection of a no. of scalars
i.e.,F=(u]>U2'U3' ... ) and uieF.
This scalar or vector fields are very significant to construct vector space,
having finite dimensionality.
1.25. Elementary Idea about Vector Space:
A space G over the vector field V and scalar field F is said to be a vector
space for binary operations EB among vectors and 0 among scalars and vectors,
when the following conditions are satisfied.
(i) a EB h eV for all a, h eV.
(ii) a EB h == h EB a for all a, h e V .
(iii) a EB (h EB c) = (a $ h) $ c for all a, h, c eV.
(iv) There exists an element e in V such that a EB e = a for all a e V .
* For reciprocal set of vectors, we should apply the property of vectors tripple product
a .(b x c) == b .(C x a) == c (a x b)to satisfy the condition a .a' = b .b' = c .c'.
= I
For this set of reciprocal vector, one should remember that
a'.b == a'.c = h'.c = b'.a = c'.a == c'.b =0.
Vector 25
(v) For each {i in V there exist an element (i' in V such that a EEl a' = e.
(vi) a 0 a EV, o.a EV for all a EF,a EV.
(vii) o.0(P0a) = (o.p)0a for all o.,pEF,all aEV.
(viii) a 0 (a EEl b) = (a 0 a) EEl (P 0 a) for all a, p E F, a E V .
(xi) (o.+p)0a = (o.0a) +(P0a) for all o.,p EF,a EV.
(x) 10 a = a, 1 being the identity element in F.
For these characteristics, any vector space*, 'G' is denoted by
(V, F, +,., EEl, 0) .
Two important parameters of a vector space are its dimensionality and basis.
The dimensionality of the space is given by the maximum possible number of
linearly independent vectors in that space. For example, in a three dimensional
space, the maximum possible number of linearly independent vectors can be
three.
Suppose, in three dimensional space, al> a2' a3 are three linearly
independents vectors. Now any vector b in the space can be expressed as,
h = a, a, + 0.2 a2 + 0.3 a3'
when, o.
p
0.
2
, 0.
3
E F and at least one of them is non-zero.
On the other hand, the basis of a vector space are those linearly independent
vectors which are essential to construct any vector of the space. So more clearly,
a set of n linearly independent vector is known as a basis for the space.
One should remember that for a given space, basis is not unique. It can be
selected in an infinite no. of ways. As for example, for a 3-dimensional space,
one of the basis is well known set of three unit vectors i, J and k along x, y
and z axes of cartesian co-ordinate system, such that for any vector p,
A A A
P = Px i + Py j + pz k
1.25. Linear Operator in Vector Space:
A linear mapping T : V V is called a linear mapping on V. T is also
called linear operator on the vector field V of the given space.
A set of all linear operators on a vector space (V, F, +,., $, 0) over V and
F, in its own right, a linear space over F.
* For vector space (V, F, +,., $, 0), the elements of V are vectors, elements of Fare
scalars, and F is called ground field of the vector space. Four symbols. +, ., $, 0
are four different compositions +: F x F F,.: F x F F, EEl: V x V V,
o : F x V V when: stands for mapping.
Summary
1. Except algebraical laws, the scalar quantities also obeys commutative,
associative and distributive laws.
2. Unit vector is essential to represent a vector and mathematically, A = A Ii .
3. Opposite vector has same magnitude, but opposite direction to the original
vector.
i.e., 1- ~ = IAI = A, -A = A (-Ii) .
4. For collinear vectors, all vectors are parallel to each other.
5. For linearly independent vectors A/s
La
i
Ai = 0 when all a, = O.
But for at least one a. '* O.
,
They are dependent vectors.
6. Magnitude of resultant of two vectors A and B is IA + BI = (N + B2 +
2AB cos a)112
Bsina
and if this resultant makes angle 9 with A then, tan 9 = A + B cos a .
7. Magnitude of subtraction of two vectors A and B , is IA - BI = (A' + B2 -
2AB cos a)1/2
and if this subtraction vector makes angle 9 with vector A, then
B sin a
tan9=A B .
- cosa
8. For resultant of any no. of vectors by perpendicular resolution method,
"" - 2
it = L.J Ai , R = (R; + R;)
I
if
and 9 = tan-
1
(RylRx).
when, Rx = LAi cos 9i
, Ry = LAi sin 9i
i i
~
9. Position vector at any point P is r = OP when, '0' is arbitrarily choosen
origin.
This position vector r has following representations,
r = xi + yJ (2 dim cartesian coordinates)
r = r r (2 dim polar coordinates)
and r = xi + yJ + zk (3 dim cartesian coordinates)
r = r r (3 dim spherical polar coordinates)
r = pp + z k (3 dim cylinderical polar coordinates).
Vector 27
10. For three co-planner vectors A,B,C, we have C = cx.A + PB, which will
be co-planner with A and 13 .
11. For rotational invarience of any vector A, IAI will remain unchanged for
rotation of the frame.
- -
12. For dot product of two vectors A and B, A.B = A B cos ex
When (i) A.B
(ii) A.B
(iii) i .i
o for A..L 13
B.A ~ Commutative property.
J.] = k.k = 1,
i.] = J.k = k.i = O.
13. For cross product of two vectors X and 13
X x 13 = IX x
13
1 n = (AB sin a)n
when (i) X x 13 = 6 for X II 13 .
(ii) X x 13 = -13 x X Anti commutative property
(iii) i xi = J x J = k x k = 0
and i x J = k, J x Ie = t., Ie x t = J.
J x i = -k, k x J = -i, i x k = - J.
14. For scalar tripple product of (X, 13, C),
A.(B x C) = B.(C x X) = C.(X x B)
15. For vector tripple product of three vectors A, B, C
X x (13 x C) = (A.C) 13 - (A.B) C.
and A x (13 xC) + 13 x (C x A) + C x (X x B) = 6
16. Pseudo vectors does not change sign under co-ordinate inversion or mirror
reflection.
17. Pseudo scalars changes sign under co-ordinate inversion or mirror reflection.
18. For ordinary derivative of a vector X,
dX ~ d x -:dAy AdA
z
- = z--+}--+k--.
da da da dcx.
and it follows the rule of ordinary differentiation.
19. For partial derivatives of vector X = X(Xi)
28
dA = dx
i
20. For gradiant operator V
_ - dp A 1- IA
A = V P = dn n = V P n
when, dp = V p.d;
21. For divergenee operator V.
"oAa
V.A = LJ OU .
a
22. For Curl operator Vx
- - OA
y
) OAz) A(OAy OAx)
VxA = 1----- +} ----- +k -----
oy oz oz ox ox Oy
0
2
23. Laplacian operator V
2
= V. V = L OU 2
a
1 0
2
D' Alambartian operator IT = V2 -""""2 !:l 2 .
C ut
24. Line integral of a vector is given by
2
fA.d; = f[Axdx+AydY+ Azdzl.
C I
25. Surface integral of a vector A is
f
- - . f r - A dxdy
A.ds = JRA.n-
IA
AI
S n.k
f f A.n = f fAn dxdz
= R n.i R InJI
26. Volume integral of a vector A is given by
fA dv = f f f A dx dy dz
= i J J J Ax dxdydz + J J J J Axdxdydz + k J J J Azdxdydz.
Vector
27. Gauss's Divergence theorem states that
1 A.n ds = Iv div(A) dv
28. According to Green's theorem,
- = 1 -\V :!)as
29. By Stoke's theorem,
30. For reciprocal set (a', b', c') of (a, b, c)
bXc cxa axb
a'
a.(b x c) , b' = a.(b x c) , c' = a.(b x c)
when, a.a' = b.b' = c.c' = 1.
31. A vector space is represented as
(V, F, +, ., EB, 0)
29
32. For basis vector aj (inn dimension), any vector b = Lujaj (i= 1,2,., n)
Worked Out Examples
Example 1. If A = 3/ - J - 4k, 8 = -2f + 4J - 3k, c = / + 2J - k ,
Find a unit vector parallel to 3A - 28 + 4(: .
Ans. Let the vector a = 3A - 28 + 4(:
or
or,
a = 3( 3f - J - 4k) - 2( -2f + 4J - 3k) + 4(f + 2J - k)
a = (9 + 4 + 4)/ + (- 3 - 8 + 8)J + (- 12 + 6 - 4)k
a = 17 f - 3J - 10k
so, the unit vector along a is
_ A A ,..
a 17i - 3j - 10k
n = lal = .J398 .
Example 2. The position vectors of points P and Q are given by
Fr = 2f + 3 J - k, '2 = 4/ - 3 J + 2k.
30
~
Determine PQ and its magnitude.
Ans. Here,
or = ~ = 21 + 3J - k.
00 =- r2 = 41 - 3 J + 2k
PO = OO-OP = r2 ~ = 21 -6J+3k.
So, its magnitude is IpQI = 7 unit.
Example 3. If a and b are non-collinear vectors and A
(x + 4y)a + (2x + y + I)b and 13 = (y - 2x + 2)a + (2x - 3y - I)b, find x
andy
such that 3A = 213
Ans. Here 3A = 213
or, 3(x + 4y)a + 3(2x + y + I)b = 2(y - 2x + 2)a + 2(2x - 3y -I)b
Equating coefficient of a and b from both side,
3x + 12y = 2y - 4x + 4 => 7x + lOy = 4
6x + 3y + 3 = 4x - 6y - 2 => 2x + 9x = -S
Solving this two equations, x = 2, y = -I.
Example 4. Examine whether the vectors Sa + 6b + 7(5 , 7a - 8b + 9(5 and
3a + 20b + S(5 , (a , b ' (5 being non-coplanar vectors) are linearly independent
or dependent.
Ans. If possible, let the given vectors are linearly dependent. Then there
exist scalars ai' a
2
, a
3
not all zero, such that
al(Sa + 6b + 7c) + a2(7a - 8b + 9c) + a3(3a + 20b + Sc) = 0
or, (5a
l
+ 7a
2
+ 3(
3
)a + (6a
l
- 8a
2
+ 20(
3
)b ... 1
+ (7a, + 9a
2
+ 5(
3
)c = O.
As, a, b, (5 non-eo-planner vectors,
5a
l
+ 7a
2
+ 3a.
3
= 0
6a
l
- 8a
2
+ 20('.(.3 = 0
7a
l
+ 9a
2
+ 5a
3
= O.
From 1 st two equations, we get
al a,2 a3
- = - = - =k(say).
2 -1 -1 .
:. a
l
= 2k, a
2
= -k, a
3
= -k
These values also satisfy the third equation. Hence, there exist scalars ai' a
2
,
a
3
such that equation (1) holds. Hence, given vectors are linearly dependent.
Vector 31
Example 5. Consider three vectors A, Band C has respective magnitude
1, 2 and 3 units and these are directed along arms of a equilataral triangle, taken
in the same sense. Find magnitude of resultant of these three vectors by resolu-
tion method.
Ans. By perpendicular resolution method,
let us take X component, Rx = A cos 0 + B cos 120 + C cos 240 = -105 unit.
and take Y component, R = A sin 0 + B sin 120 + C sin 240 = - 0.866 unit.
y
where the vector A is taken along +ve x-axis.
So magnitude of resultant R = .[(R; + Rn = J3 unit.
Example 6. If A = 41 - J + 3k and E = 21 + J - 2k, find a unit vector
perpendicular to both A and E.
Ans. Here, we know that for any two vectors A and E, (A x E) is perpen-
dicular to the plane of A and B so, the unit vector perpendicular to both A and
A (AxE)
B IS n = + 1- -I
AxB'
Now AxE
i j
4 -1
-2 1
k
3 = 1(2-3)+ J(-6+8)+k(4-2)
-2
A A A
= -i + 2j + 2k.
:. unit vector
or,
(
-7 + 2J + 2k)
iz = """'---------"-
3
(
7 - 2J - 2k)
iz=+ .
3
Example 7. Find the magnitude projection of the vector 27 - 3J + 6k on
the vector 7 + 2J + 2k.
Ans. The projection is R = (A.iz)
when A = 2f - 3J + 6k
and n is unit vector along (1 + 2J + 2k)
A A A
i + 2j + 2k
n = ----"'--
3
and R = (2 - 6 + 12) = % unit.
32
Example 8. Simplify: (A + i3).(i3 + c) x (C + A)
Ans. (A + i3).(i3 + c) x (c + A)
= (A+B).[BxC+BxA+CxA] [':CxC=O]
= A.(i3 x c) + A.(i3 x A) + A.(C x A) + i3.(i3 x c) + i3.(i3 x A) + i3.(C x A)
= A. (i3 x C) + 0 + 0 + 0 + 0 + A. (i3 xC)
= 2A.(i3 x C).
- -
dA _ dB
Example 9 Solve - = B -- = A-
. dt 'dt -'
- 2--
dA _ d A dB _
Ans. - = B => -- = - = -A.
dt dt
2
dt
d
2
A -
--+A =0
dt
2

and solution is A = C
1
sin t - C
2
cos t
similarly, B = C
1
cos t - C
2
sin t . where, C
1
and C
2
are two constant vec-
tors.
Example 10. If A is a constant vector, prove V(r.A) = A.
Ans. Now let A = Axi + Ay} + A/ .
when (A , A , A) all are constants for constant vector A.
x y z
Now
r
- A- =xA + yA + zA
x y z
a ( a ( -) a ( -)-
V(r.A) = - r.A i +- r.A j +- r.A k
rx ry rz
- ( 2 2
Ans. V. 2x z i - xy z j + 3y z k
= + +
rx ry rz'
= 4xz - 2xyz + 3y2.
Example 12. Evaluate V.(r
3
r)
Vector
-(3-) -(3)- 3-- __
Ans. V. r r = V r .r + r V.r = 3rr.r + 3r
3
= 3,-3 + 3,-3 = 6,-3.
Example 13. Prove, (V2 rn) == n(n + - 2.
Ans. We know that V(rn) = nr
n
-
2
r.
V2(rn) = V.V(rn) = V.(nr
n
-
2
r)
= n V.(r
n
-
2
r) = n[V(r
n
-
2
).r + r
n
-
2
V.r]
= n[(n - 2)r
n
-
4
r.r + 3r
n
-
2
]
= n(n - 2) +
= (n2 - 2n + 3n) = (n2 + n)
= n(n + 1)
Example 14. Prove that f n x (a x r) ds = 2aV.
s
Where a is constant vector
ADS. By Gauss's divergence theorem
Now
f n x (a x r) ds = f v x (a x r)dv
s v
_ (_ _)
V x (a x r) = L..,.i x - a x r
ox
(oa __ or)
=L..,.IX -xr+ax-
ox ox
= L i x (0 + a x i)
= L[(i.i)a - (i.a)i]
= L(a -a.i)
= La- Lai
= 3ii - ii = 2ii
f n x (a x r)ds = f 2a dv = 2a v .
s r
33
Example 15. Verify Stoke's theorem for the function F = x(xi + yJ).
integrated round the square in the plane z = whose sides are along the lines
x = 0, y = 0, x = a, y = a.
Ans. We have, V x [x(xi + yi)] = yk = V x F.
s
a a a a
= f f kyo X dx dy = f f y dx dy
00 00
f
a fa a
2
1
= ydy. dx = -a = 2"a
3
.
o 0 2
f (V x F).as
1 3
= -a
2
s
Again,
t
F
.
dr
= f F.dr + f F.dr + f F.dr + f F.dr
C OA AB Be co
Now t
F
.
dr
= fx
2
dX+ f(a
2
i + ay)).)dy+ f(x
2
i + a)x).(-i)dx + fO(-))dY
coo 0 0
13
1
3
1
3
1
3
= -a + -a - -a =-a
3 2 3 2
:. t
F
.
dr
= f(v x F).ds
c s
Hence, Stoke's theorem is verified.
EXERCISES
1. Simplify 2A+13+3e-{A-213-2(2A-313-e)} [ADs.SA-313+e]
2. Show that IA + 13 + el IAI + 1131 + lei
3. A quadrilateral ABeD has masses of 1, 2, 3 and 4 units located respectively
at its vertices A(-I, -2,2), B (3, 2, -1), C(1, -2,4) and D(3, 1,2). Find the
co-ordinates of the centroid. [ADS. (2, 0, 2)]
4. In each case determine whether the vectors are linearly independent or lin-
early dependent.
(a) A = 2i + J - 3k, 13 = i - 4k, e = 4i + 3) - k
(b) A = i - 3 J + 2k, 13 = 2i - 4 J - k, C = 3i + 2) - k
[ADS. (a) Linearly dependent. (b) Linearly independent]
Vector 35
5. For two vectors A = iF - 3) + k, B = -i +2) - k.
Find the projection of A on B.

..[6.
6. If 111 and 112 be the unit vectors and e is the angle between them, show that
. ellA A I
sm - = - nl - n2 .
2 2
7. For three vectors A = i - 2) - 3k. B = 2i + J - k, (: = i + 3) - 2k
find (A x B) (B. C) .
8. Prove that (A x 13).(C x D) + (13 x C).(A x D) + (c x A).(13 x D) = 0.
9. If (:1 and (:2 are constant ve\;Lors and I.. is a constant scalar, show that B =
e -Ax (C
1
sin I..y + (:2 cos I..y) satisfies the partial differential equation
(iB a
2
B
ax2 + 8y2 =0.
10. Show that V.( cpA) = Vcp.A + cpV.A
11. Evaluate V2(2n r) [ADS. r;]
12.
Evaluate v.[r vL; ) 1
13. If A = V(V.A)
[ADS.-2r-
3
F]
14. Prove V
2
(cp\jf) = cpV2\jf + 2Vcp. V\jf + \jfV2cp .
15. Prove that F = r2F is conservative and find the scalar potential.
[ADS. cp = r: + constant]
16. Evaluate f A.dF along a curve x
2
+ y = 1, Z = 1 in the positive direction
C
from (0,1, I) to (I, 0, I) if A = (yz + 2x)i +xz)+(xy+2z)k [ADS. 1]
17. Evaluate f f x
2
+ l dxdy over the region R in the xy plane bounded by
R
x
2
+ y2 = 36. [ADS. 144n]
18. Evaluate f r.ds over the surface ofasphere of radius a with centres at (0,0,0).
s
[Ans.41ta
3
]
19. Evaluate f F dv where V is the region bounded by the surfaces x = 0, y = 0,
v
y = 6, Z = X2, Z = 4. and F = 2xzt - xJ + ifc [Ans.128t - 24J + 384 fc]
20. A fluid of density p(x, y, z) moves with velocity vex, y, z, t). If there are no
- - op
sources or sinks, prove that V.J + - = where, j = pv.
rt
21. Verify Stoke's theorem for A = (2x - y)t - yz2 J - iz fc, where S is the
upper half surface of the sphere x
2
+ y2 + Z2 = 1 and c is its boundary.
f
dV fr.n
22. Prove ? = -;Ids.
v s
23. Prove f nds = for any close surface S.
s
24. Prove f r x as = 6 for any close surface S.
s
25'. Prove ~ p dr = f as x Vcp .
c s
-:0:-
Chapter-2
Linear Motion
2.1. Introduction:
Classical mechanics basically deals with motion and cause of motion. At the
very beginning, it is important to have a clear conception about linear motion and
its characteristics to study whole mechanics when a particle is displaced in a
particular direction then it is said to be in linear motion. Such change in position
along a given direction may occur under external force or not, but the fact is that
characteristics of linear motion in several situation is very significant at several
situation of classical mechanics.
Linear motion is thus, a starting branch of mechanics and we will actually go
through this chapter with kinematic view of study.
2.2. Kinematics:
A part of mechanics which deals with forces causing the motion is called
Dynamics. Kinematics is that branch of dynamics which deals with motion alone
without considering its cause.
In this branch, we shall discuss the path described by a particle and determine
the velocity and acceleration at any point along its path.
At every instant of motion, it is required to find the instantaneous position of
particle about an arbitrarily choosen origin and then taking variation of it w.r.t.
time, one can deal with the nature of particles motions. We will now deal with the
kinematical treatment for the particle, moving linearly.
2.3. Basic Definitions of Required Parameters:
It is very urgent to define some basic parameters required in dealing any
mechanical problem. These parameters are given below:
(i) Displacement: For motion of particle in a given direction, its change of
position in that direction is called displacement. It is a vector quantity and is
denoted by S.
37
So mathematically, S = t.;i = r2 - r) = change in position.
where rj and Y2 are instanteneous positions at initial and final position ofthe
moving particle.
(ii) Velocity: For a moving particle, its rate of change of position or rate of
displacement is its velocity. Here, this velocity is a vector quantity and is denoted
by v.
So,
But we should always remember the instantaneous velocity of the moving
particle at any instant of its motion is
fly dy
v= Lt -=-
M dt
which is now estimated by the slope* of a tangent, drawn at any instantane-
ous point of "position-time" graph.
(iii) Acceleration: It is another significant parameter for linear motion which
is defined by the rate of change of velocity and is mathematically given by, J =
flv v - u
where, v and ii are respective final and initial velocities of this moving particle.
Now
and since,

iv - iii
f= =-
t
I
iv-iii = (v
2
+u
2
-2uv cos e)2
I
where, e is the angle between ii and v. So,/= + u
2
- 2uv cos
and in case, v II ii in same direction, e = 0
and f=
v-u
f= -- (where, v> or < u)
t
* Practically, this slope of position-time graph at any instantaneous point gives the
magnitude of velocity at that instant and is precisely given by,
Ivl = I = = tan e = Slope of "position-time" curve at instantaneous
point.
Linear Motion
Also for particles motion, the instantaneous acceleration* is given by,
when,
1
= Lt .ilv = dV
i\t -+ 0 .ilt dt
I
ii L I.ilVI IdVl
1 = J I = L\/-! 0 dt = dt I
39
which is estimated by the slope of tangent drawn at instantaneous point of
velocity-time graph.
(iv) Average velocity and acceleration: For motion of particle, average
velocity is estimated by taking ratio of total displacement and total time taken for
the motion of particle, provided that the particle traverse several displacement in
several interval of time. So mathematically
Total displacement
Average velocity =
Total time taken
:. < v >
SI + S2 + '"
.iltl + Llt2 + ...
n
n
where, SI' 8
2
, .... are respective displacement occurs in respective time
intervals .ilt I' .ilt
2
, etc.
We can also have, for average velocity,
I il2 -
< V > = -t--t- v dt when we study the variation of v(t) in a particular
2 - 1 1\
interval of time from t\ to t
2
Similarly, the average acceleration is the ratio of
total change in velocity to total time taken and it is then given by

.ilvI + .ilv2 + '"
Llt) + .ilt
2
+ '"
I 112 -
and also, < ~ > = -- 1 dt.
t2 - t) 1\
Momentum: It is such a parameter which is the sample product of velocity
with mass of particle and is given by, Momentum, p = mv.
where p is directed along v and thus for the instantaneous momentum, the
instantaneous velocity v is now, v = v(t)
* The direction of acceleration is along the direction for the change of velocity,
occured in a certain time interval. The retardation or negative acceleration is taken
in reversed direction of acceleration.
Again, the average value of p is given by,
1 il2 - m r
l2
-
 = -- pdt = --J, vdt
t2 - tl II t2 - tl II
 m<v>
(vi) Relative velocity and acceleration : For two moving particle, the ve-
locity of one w.r.t. another is the relative velocity of 1st particle w.r.t. 2nd parti-
cle.
If VI and V2 be the instantaneous velocity of particle -1 and 2 respectively
then,
Relative velocity of particle -1 w.r.t. particle 2 is vl2 = VI -- v2
Similarly, the relative acceleration is the acceleration of one particle w.r.t.
another particle which is given by,
- --
fi2 = fi - 12'
we should now remember that
also,
when
and
!VI2! = !V2I! and V12 = -V21
I
V
12
= V
21
= (vf + vi - 2vlv2 cos a)2
I
J..2 =1;1 = (fi2 + Il- 2fih cos a)2
2.4. Velocity and Acceleration in Several Co-ordinate System :
We have already discussed in our earlier section that the instantaneous veloc-
ity and acceleration are given by,
di _ dV d
2
;
V = dt ' 1 = dt = dt 2
So we see that, these parameters entirely depends on position vector ; and
its variation W.r.t. time, thus it is important to express them in several coordinate
system.
In certesian co-ordinate system (3 dim)
v
or,
v
and
1
di dx A dy A dz A
- = -i +-j+-k
dt dt dt dt
Linear Motion
xi + yJ +zk.
In cy1inderical polar co-ordinates (p, <p, z) it can be shown that
v = vpP + v<p<l> + v) = PP + pq, + ik
I Ii) + l<p<l> + I)
(15 - pq, 2)p + (PiP + 2pq,) + z"k
In spherical polar coordinates (r, e, <p)
We have, v = vrr + vee + v<p<l>
or v = f r + r e e + r sin 9 q, .
and similarly for acceleration,
J = (r - r 8
2
- r sin
2
9 q, 2) r + (r e + 2f 8 - r sin 9 cos 9q, 2) e +
41
(r sine <p + 2f q, sine + 2r6q, cose)
2.5. Tangential and Normal Component of Velocity and Acceleration:
Consider that a particle moves along a curve
AB from P to pi during interval of time from t Y
to t + 8t. The position vectors of the points p
and pi are respectively rand r + 8r and the
8r
velocity of the particle at P is fu'
8r dr dr ds
Now - = Lt - = - = -.-
V c S t ~ O 8t dt ds dt
when, in limiting case, as P approaches pi,
A dr
dr = ds i and - = :-
ds I
_ :- ds A
v = I dt = vi
~ A ~ t
A
x
Here, v = dt and i is the unit vector along the tangent at P. So we see that
the velocity of the moving particle has a direction, tangent to the path and a
ds
magnitude dt'
Let i + 8i be the unit vector along the tangent at pl. If <p and <p + 8<p be the
inclinations of the vectors and + 8 to OX and J is unit vector at P normal
to then, from figure, we have
. 8m
Slll-'t'
2
= 1 for, 8<p 0
2
1
8i
I 8i A
. - = 1 and I:
m
= J ..
.. 8<p u".
di A dj
Similarly, d<p = j, d<p
A
-i .
Hence we get the acceleration,
_ dV d(-:dS) .d
2
s ds d -:d
2
s ds ii d<p
f = dt = dt I dt = I dt2 + dt dt = I dt
2
+ dt d<p dt
- = i
d2s
+ ds J d<p ds = d
2
s + Jd<P(dS)2
f dt
2
dt ds dt dt
2
ds dt
ds
;(:7" :fit;; at P is p = dp .
dv
So the tangential component of acceleration is dt and the normal
v
2
component of acceleration is -.
p
2.6. Radial and Transverse Component of Velocity and Acceleration:
Consider instantaneous position of a moving particle is given by per, 8) by
using plane polar co-ordinates. If v be instantaneous velocity at that position,
then, v = vxi + vyj ... 1
Now the component of this instantaneous velocity in
radial direction is called radial component of velocity whereas
the component, perpendicular to the radial direction is called
cross-radial or transverse component of velocity. These two
components are denoted by vr and va by which, we may
write
v = Vr r + va e ... 2
Now from fig. i = r cos 8 - e sin 8
Linear Motion 43
and J = r sin S + e cos S.
So, if (x, y) be the instantaneous position in Cartesian Co-ordinate system,
then -:-
J
x = r cos S
A
y = r sin S
r
and we now have,
dx .
v
x
= dt =rcosS-rsinS.S.
dy .
v
y
= dt = rsinS+rcosS.8.
Using these equations in expression 1, we get
v = (r cos S - r sinS. 8)(r cos S - e sin S) + (r sin S + r cosS. 8)(r sin 8 + e cos 8)
=> v = ; cos
2
9 r -; cos 9 sin 9 e -r sin 9 cos 9 8 r
+ 8 r sin
2
9 e + r sin
2
9 r + r sin9 cos9 e + r cos 9 sin 9 8 r + r cos
2
9.8.
v = ;r+r8e
Comparing this equation with equation (2), we get
dr . dS
v = . = - v = r 9 = r-
r r dt' e dt
... 3
which are respective radial and transverse components of velocity.
Similarly, for acceleration having radial and cross radial component f,. and
fa respectively,
we have,
and,
When,
and
J=f,.r+/ee
J = f) + fyJ
dv x d (. S S 9)
!, = - = - r cos - r sm
x dt dt
= r cosS - 2; sin9 8 - r e sin9 - r e
2
cose
J: = dvl... =
y dt dt
= r sin S + 2; 8 cos S + r e cos S - r 8
2
sin 9
Thus from equation 5, we get,
J = (r cosS - 2; 8 sinS - resinS - r 8
2
cosS)(r cosS - e sinS)
.. .4
... 5
+ (r sin S + 2; 8 cos S + r e cos S - r 8
2
sin S) (r sin 8 + e cos 9 )
On solving this equation, we finally get,
j = (r - dF)P + (r 8 + 2f e)8 . . .. 6
Comparing this equation with equation (4)
We get,
and
_.. . . _ r d
2
e + 2(dr) (de)
Ie - r e + 2 r e - dt 2 dt dt
which are respective r ~ d i l and transverse components of acceleration.
2.7. Newton's Laws of Motion:
In kinematics, Newton gave his three famous laws which are essential for
any mechanical problem. These laws are,
(i) Law of Inertia: This law states that ifno force be applied from outside,
then rest particle will always remain at rest and moving particle will always
remain in motion along a straight line. That is, every particle will try to retain
their self inertia.
(ii) Law of Force : This law states that if forces be applied on a moving
system then its change in momentum will occur in the direction of resultant
force and the resultant force will be directly proportional to the rate of change
of momentum of that system as a whole. By this law, mathematically, we have,
_ " _ d (" _ .J djJ
F = L..J Fi = dt ~ pl = dt'
where, P = L Pi = momentum of the system as a whole.
i
(iii) Law of Interaction: This 3rd law of Newton states that to every action,
there is an equal and opposite reaction, when these two are acting on two separate
bodies. So if F and R be the respective action and reaction then
F = - R when, F = R.
One should now remember that application of Newton's laws has some
limitations.
(i) These must be valid in inertial frame
(ii) This should be applied on a moving particle having velocity much
less than that of light in free space. This is because of the fact that
for non-inertial frame, a pseudo force* will act on system which can
not be detected from outside and for particle velocity tend to that of
* See the chapter 'Reference Frame'.
Linear Motion 45
light in free space, we should pass through relativistic zone* in which
particles mass will inverse with increase of velocity. But Newton's laws
are only valid for fixed mass only.
2.S. Accelerated Linear Motion :
When a force or a no. of force be applied on a moving particle, then the
motion will be accelerated and that unbalance force (resultant) creates acceleration
of the particle or system of particles.
The acceleration is then given by
LFi
J = _i _ by Newton's law and for such accelerated motion, the equation
M
d
2
; --
of motion will be M-
2
=" F. = F effective.
dt .L..J I
i
2.9. Graphical Treatment of Linear Motion :
When a particle or system of particle is in linear motion, either accelerated
or not, the motion can also be studied graphically. Since, during its motion,
position is given by, ; = ;(t)
which changes with time, the position-time and t
velocity-time graph gives r
(i) Magnitude of instantaneous velocity which is
estimated by the slope of tangent drawn at
any instantaneous position of positive-time
graph.
/).r dr
i.e., Ivl = t t ~ 0 /).t = dt = tan O.
*
-.;>/
1M 1M
I I
I I
I I
, I
I I
I I
I I
o t ~
(ii) Magnitude of displacement which is estimated by the area of velocity-
I
t2
time graph and it is given by, lsi = v dt .
t,
= Area of velocity time graph.
(iii) Average velocity (magnitude only) can also be estimated from the area
of velocity-time graph divided by the interval of time. It is now given
by
* In relativistic region, mass varies with velocity as m =
where, p = vic, m = moving mass, mo = rest mass.
46
(iv)
1 rt2
<Ivl> = --J, vdt.
t2 - t) II i
u
Magnitude of instantaneous acceleration can
be estimated by the slope of tangent drawn
at instantaneous point of velocity-time
graph,
L'1v dv
Lt - = - = tan 8.
0 L'1t dt
i.e.,
o

-->\
1M
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I

(v) Nature of motion can also be studied from position-time and velocity-
time graph.
2.10. Conservation of Linear Motion:
If no force be applied on a moving particle or system or if the resultant of
all applied force be two, then the momentum of that particle or system will be
constant. This is principle of conservation* of linear momentum.
Now from law
d
O(f
S
L.li = - "p.
mce, . dt I
I i
So for IFi = 0, !!'-(Ipi) = 0
i dt i
or, I Pi = constant
This is conservation of Linear Motion.
2.11. Time Integral of Force (Impulse) :
The time integral of force is basically the impulse of force and it is denoted
by j.
Mathematically j = jdJ = r FCt) dt
II
If F , the force be time independent, then
or, J = F M = Force x Time of action
* This conservation of linear momentum also can be obtained from Newton's 3rd
law and so for collision of two particles, mutually colliding or mutually interacting
without any applied force, the conservation of linear momentum holds.
See the chapter, 'Collision Theory'.
Linear Motion
But from Newton's law,
F
m(v - u)
M
- -
mv-mu
!!.t
J = F M = mv - mu = change of momentum.
Similarly, also for time dependent force,
J = r F dt = r dj; dt
II II dt
or, J = r
l2
dj; = (p\ - (p) I = P2 - PI
~ 2 I
= !J.p = change in momentum
47
Thus the time integral* of force or impulse of force is estimated by the
change in momentum. Also for time varying force, the average of force in given
time interval can also be obtained from this time integral of force, and is given
by,
1 112 -
F = -- Fdt
av t2 - tl II .
2.12. Work:
The work done by the force F causing displacement Si of a particle is
given by,
!J. W = F.!J.i = F!J.r cos 8 = F S cos 8 where, 8 is the angle between the
vectors F and SF and S is the magnitude of !J.i which also denotes an element
of length along the path of the particle and F cos 8 denotes the projection of F
on the tangent to the path.
Generally, mathematical representation of work done is represented as
W = f dW = f F dr cos8 = f F.di.
where, F may not be constant and may depend on the co-ordinates of the
particles.
The work done has particular sign convention that it is taken when work
is done against force, and is considered -ve when work is done in favour of
force.
* Time integral of force or impulse is also estimated by the area of "Force-time"
graph.
2.13. Power:
The power* of force F or its activity is the time rate at which work is
done by the force and is given by
p = ~ = :t (F.r).
In case, F is time independent,
- dr
p = F'd( = F.v
Thus the power is equal to the scalar product of the force and the velocity
of the point of application of the force.
2.14. Energy: Kinetic and Potential:
The energy of a particle or system of particles is the ability of doing work.
When work is done on the system, its energy will increase and also if work is
done by the system, then the energy will decrease.
So, normally, the work done is related with change of energy, and is
mathematically given by,
f
- fE2
W = F.dr = E
J
dE = E2 - El .
=> W = ~ = E2 - E\
or, work done = change in energy.
Now for any system, the energy or total energy is the sum of kinetic and
potential energy and thus it is given by,
E = EK + Ep = K.E + P.E.
Now for system of particle or motion, its kinetic energy is
,,1 2,,1_-
E = K.E = L..J -2 mivi = L..J -2 mivi' vi
K . .
I I
On the other hand, the potential energy is the energy which a body possesses
by virtue of its position or configuration and is measured by the work done by
the body in passing from its present position or configuration to some standard
position or configuration. The potential energy in the standard position is taken
to be zero.
It can be shown that for particle in linear motion, if the power P remain constant,
then the distance'S' traversed by the particle along straight path will be
3
proportional to '2 the power of time taken. i.e., for P = constant, S a(tj312 .
.. This sum of kinetic and potential energy is taken in Newtonian Mechanics. But
in relativistic zone, the total energy F 0 is the sum of Kinetic -energy and rest energy
of particle. --+ See any book on Relativity.
Linear Motion 49
Thus the potential energy is now given by
where, F denotes the force acting on the particle whose position vector is '1;
r2 denotes the position vector of the fixed point.
2.15. Conservative Force:
A force is said to be conservative if the work done by the force along a
close path be zero. So for this conservative force, if energy loss occurs in one
part of the path, then it will overcome in another remaining part of the path
and total work done along close path will be zero, and ultimately no change in
energy will occur.
So for conservative* force, F
f
F
.
dr
= o.
c
Now by Stokes theorem, fF.dr
f(\7 x F).as and then for conservative
c s
force.
\7 x F = O.
Since, for any scalar <p, \7 x (\7 <p) = 0
we can write F = \7<p for \7 x F = 0
But conventionally, conservative force, F is mathematically represented as
F = -\7<p .**
Here, the scalar function <p is known as scalar potential, where
<p = <p(r) = <p(x, y, t).
2.16. Conservation of Energy:
For conservative force F = -\7<j) , where <p is the potential energy, we can
easily show the conservation energy.
Since, F.dr = - \7<p.dr =-d<p
* Examples of conservative force are gravitational force, coulomb force,
magneto static force. etc, all of which are negative gravient of gravitational
potential energy, coulomb potential, magnetostatic potential energy ... respectively.
** Here, -ve gradient is taken for requirement of energy conservation.
we have,
Also,
f
V
2
m (') 2)
= m vdv = - vi - VI
VI 2
or,
I 2 1 2
-
mv
2 + <P2 = - mVI + <PI
2 2
This shows that in a conservative field, the sum of the kinetic and potential
energy of a particle at any point remains constant. This is called conservation
of energy.
2.17. Center of Mass and Its Motion:
A fixed point with respect to a body at which the total mass of the body is
supposed to be concentrated, is called center of mass of the body.
If r is the position of center of mass of the body w.r.t. an arbitrarily choosen
origin, '0'.
then, mathematically,
r
M
when, M = L mi = Total mass of the body and mj is mass of ith constituent
of the body.
Here, L ~ = Mr is the total moment of mass* of that rigid body.
Now in actual case, for every constituent particle of a rigid body, two types
of force will act on every particle. One is external force and another is internal
force, one to mutual interaction with one constituent by other remaining
constituent particles.
So for ith constituent particle,
* For rotation of body about an axis passing through its center of mass, total moment
n
of mass will be two. i.e., L mi r; = 0 for this rotation.
i= I
Linear Motion
n
Fj + IFij
j=l
(j,. j)
51
when, body contains total 'n' constituents and here, F
j
is external force on
ith particle.
We can now write down,
n 11 n d
Z
-
IFj + I IFij = Im; dt;j
j= 1 j= 1 i= 1
(J" i)
But here, 2nd term of L.H.S. is zero since the sum of pair of forces Fij
and Fji is zero.
ov,
or, F
This is equation of motion for center of mass. Thus the motion of center
of mass of a system of particles relative to an inertial frame is the same as that
of a single particle of mass equal to the total mass of the system under the action
or some interval force F which is equal to the vector sum of the external forces
acting on the system.
2.1S. The Two Body Problem:
We now consider two particles of mass m 1 and m
2
separated by a distance
r and acted on by external forces FI> F
z
and internal forces F
1Z
and FZl
respectively.
Now, by Newton's 3rd law, F
1Z
= FZl . y
So the equation of motion for the two particles are
d
z
-
rl
ml dt
Z
= F12 + Fl
x
where, il and r2 are instantaneous position vectors of two masses
respectively and r = il - rz .
Now,
- -
d
2
i). = FI2 +!L
dt
2
ml ml
and
If the motion of two particles is only due to internal forces arising from
the interaction between the two particles, then FI = F2 = 6
and the above equation is reduced to
d
2
(i).- r2) _
dt 2 = FI2
mlm2
where, 11 = is reduced mass of the system such r = rl - r2 is the
ml +m2
d (- - ) di). dr2
relative position vector of mass m
l
W.r.t. mass nl2 and dt rl -r2 = Tt-Tt
= VI - v2 = v12 = relative velocity of m
l
w.r.t. m
2
, this two body motion is
thus possible to replace by n single particle motion for the particle of mass 11,
having velocity, equal to the relative velocity of one particle w.r.t. another. Here
one should remember, that, for two body problem, six to internal force only,
center of mass or the system remains stationary, only for this case, such
replacement of two body motion by one body motion is possible otherwise, two
body motion can be considered as motion of reduced mass '11' and center of
mass* (nil + m
2
)
2.19. Application of the Principle of Linear Motion:
Principle of linear motion means principle of linear momentum
conservation. This principle can be applied in several dynamical case, such as
* For two body motion with center or mass Chanj'ng its position, it can energetically
1 2 1 2 1 ( ml m2 ( )2 1 ( ) 2
be shown that -mini + -m2n2 = - VI - v2 + - ml + m2 xU
2 2 2 ml + m2 2
(mlvl + m2
v
2)
where, U = (ml + m2)
Linear Motion 53
collision theory, Firing of gun, ... etc. In many cases another type of motion
may be replaced by equivalant linear motion and then in that case, all axioms
will hold for such kind of motion. A study of linear motion thus helps to have
basic knowledge of mechanics, and when required principle of linear motion
can be applied to deal that problem.
2.20. Mechanics of Variable Mass :
Principle of Rocket Motion: Uptill now we have considered the equation
of motion and the laws of conservation in such cases when the mass of the system
was constant during the motion of the system. We will now consider the motion
of a system when the mass varies with time. There are so many examples of
such system in nature and also in technology. As for example, when a drop of
rain falls through a cloud, it will gain in mass as it descends. Rocket motion is
another example of such system when mass varies with time for linkage of
exhaust material or fuel after burning from it.
A rocket fired from the earth will always be affected by the gravitational
pull of the earth. We will at first consider rocket motion in horizontal direction
and then will modify it for its vertical motion against gravitational pulling.
M
......------...> -4 I

Let M be the instantaneous mass of a rocket with its fuel, while moving
along horizontal direction. V is its instantaneous velocity and we also consider
that due to burning of fuel, the exhaust material or gases leaves from rocket
with constant velocity Il W.r.t. rocket in backward direction. Hence, if at instant,
(t + Ot), the mass of rocket with its remaining fuel be (M - and velocity
be (v - Ov) (obeying momentum conservation) then in that said interval M,
mass of exhaust gas leaves from the rocket.
Hence, from momentum conservation,
Mv = (M - (v - - v)
M = u (.,' = small)
Dividing throughout by and taking the limit as M 0, we get
dv dM
M- = -u- (1)
dt dt ...
Here -ve sign is added on the right hand side to indicate that velocity v
increases as mass M decreases.
So integrating equation (1). w.r.t. time, we get
J
v 1M dM
dv = -u -
Vo Mo M
=>
v = Vo - U lOge( )
=>
v = Vo ... (2)
when Vo and Mo are respective initial velocity and mass at instant t = O.
This equation 2, thus gives an expression for instantaneous velocity of rocket
motion.
dM
Let us suppose that the fuel is burnt at constant rate K =Tt and it lasts
for time T. If the mass of the vehicle be Mv and that of the fuel at t = 0 be
MFo then Mo = Mv + MFo .
The mass of the vehicle-fuel system at any instant t can be written as
M = Mv +MFo (1-;)
t
= M +M1\ -M1\ -
v 0 0 T
t
=> M = Mo - M1\ - for 0 < t < T
o T --
and M = Mv for t T. '
Then for equation 2,
_ dx _ Vo -ulog. (1- MFo .!....)
v - dt - I Mo T
Again, integrating this equation with respect to time we get.
S
t ( MFO t)
x = Xo + vot - U 0 loge 1 - Mo T dt.
But we have,
Thus,
x = Xo + vol -{(I - }Og. (1 -'::: -I]
... (3)
... (4)
This is the distance covered by the rocket in time t. Since the rocket attains
maximum velocity at t = T when all its fuel is burnt out. The maximum velocity
calculated from equation (3) is given by,
Linear Motion
55
v = Vo -ulog (1- MFO)*
max e Mo
V = Vo + u loge (Mo )
max Mv
=> ... (5)
If the rocket is moving vertically upward and if the gravitational pull of
the earth on it is assumed to be constant, then the equation of motion of the
rocket can be written as
dv dM
M- = -u--Mg
dt dt
... (6)
Integrating this equation (1) w.r.t. time, we get,
v = ) -gt (taking, Vo = 0 at t = 0)
and similarly, x = ut-'!gt
2
-(t- MoT)lOge(l_ MFo .!.-). (taking 11 =
2 MFo Mo T 0
o att = 0)
This is all about rocket motion.
SUMMARY
1. The displacement of a moving particle is change in position in a given
direction.
i. e. S = & = r2 - rj .
2. The velocity of a moving particle is the rate of change of position. i.e.
!1r dr
-= Lt-=-
v M dt .
3. Acceleration of a moving particle is the rate of change of velocity.
- = Lt !1v = dV
i.e. f dt .
MF
* Here the ratio M 0 has a practical limit and it can not be increased indefinitely.
v
A single rocket therefore will not attain high velocity that is required.
4. Average velocity is the ratio of total displacement and total time taken by
the moving particle.
LSi 1 r/L
< v > == " == --J, v dt .
L../l t2 - tl /,
i.e.
5. Average acceleration is the ratio of total change in velocity and total time
taken by the accelerated particle.
i.e.
_ ~ v i 1 1/2-
< f > == ,,- == -- fdt
L.ji t2 - tl /, .
6. Relative velocity is the velocity of one moving particle w.r.t. another particle
in motion or at rest.
i.e. Vru = V2 - vI
Similarly, relative acceleration is the acceleration of one accelerated particle
w.r.t. another accelerated particle.
- - -
i.e. fru == i2 - fi .
A A A
7. For particle moving in three dimension, the velocity v = vxi + vyi + vzk
== xi + yJ + iu (cartesian)
or, v == v
r
; + vee + vee
== i-; + ree + r sin Oe (spherical polar)
or, v == vpP + v<pcP + v/c
== PP + pci>cP + ik (cylinderical polar).
8. For any moving particle in three dimension, the acceleration,
J == f) + fyJ + fJ
== xi + yJ + zk (cartesian)
or, J == fr.p+fee+/z.z
== (r - re
2
); + (re + 08) e + zz . (cylinderical polar)
or, J == 1,.; + fee + flp
== (i- - re
2
- rsin
2
94>2); + (re + 2;.8 - r sin 9 cos 94>2) e
+ (r sin 9 + 2i-4> sin 9 + 20r4> cos9)cP (spherical polar).
9. Tangential component of velocity and acceleration
ds dv d
2
s
VI == dt ,1; == dt == dt2
Linear Motion
Nonnal component of velocity and acceleration
v
2
vN=O,jN= -.
P
10. Radial and Transverse component of velocity are
vr = r, va = ,e
11. Radial and Transverse component of acceleration are
. 2 .. .
J,. = r - rB ,fa = rB + 2Ye.
12. For accelerated motion of particle, equation of motion is
d
2
-
M_r = "F. = -
dt 2 I Feffective
I
13. For conservation of linear momentum,
L Pi = constant when Fapplied = O.
14. Time integral of force or impulse, is
j = r
t2
F dt = P2 - PI = /).P = change in momentum.
J
tl
15. Work done by the force is
W= f
F
.
dr
W = -ve for work done by the force
= +ve for workdone against the force.
16. Power of a system is mathematically given by,
dw
P = dt = F.v
17. Energy of a system is given by
E = K.E + P.E
K.E = L mi(Vj'V
j
)
j
P.E = W = f F.dr .
18 For conservative force F, fl
dr
= 0
and V x F = 0 when, F = -V<p .
<p = secular potential.
19. For center of mass motion.
57
m i ~ "m.
when, ; =" M = L.-J I
L.-Jmi i
20. For two body motion, with stationary center of motion,
d
2
;
/..1.-- -
dt2 - F
when, F12 = F21 = IFI = mutual interaction
mlm2
/..I. = = Reduced mass.
ml +m2
and
dv dM
21. For Rocket motion M- = -U-.
'dt dt
But for motion in vertical direction the equation of motion is given by,
dv dM
M dt = -uTt - Mg .
Worked Out Examples
1. A car covers is quarter of its journey at a speed 30 kmIhr, second quarter
at 40 kmlhr and the rest at 60 kmIhr. What is the average speed of the car during
its journey?
Ans. Let the total distance covered by the car is S.
So if v be the average velocity then
Total distance travelled
v=
Total time taken
S
-S---S---S- = 43.63 kmIhr
--+--+--
4 x 30 4 x 40 2 x 60
2. A motor car moves 40 km due east and then 30 km due north in 6 hr.
Calculate its average speed.
Ans. Here, the total distance covered
1
S = (40
2
+ 30
2
)"2 = 50 km.
50
so the average speed, v = 6' = 8.34 kJnlhr.
3. A car moves due north at 50 kmlhr, while a wind blows from the north
east at 30 kmIhr. Find the speed and direction of the wind as they appear to a
man in the car.
Let, VI = velocity of man, v
2
= velocity of wind
Here, if Vo be the relative velocity as appear to a man, then
Linear Motion
I
Vo = [1-l\12 + IV212 + 21-vlllv21 cos 45)2
I
= (50
2
+ 30
2
+ 2 x 50 x 30 x ~ y
= 74 kmlhr.
If it makes angle 8, with south of west,
30 sin 45
then, tan 8 = 50 + 30 cos 45
59
N
~ E
4. To a cyclist riding towards north at 10 kmlhr the wind appears to be
blowing from the north west at 12 kmIhr. Find the true velocity of the wind.
Let, vI = velocity of cyclist, vI = 10 kmlhr N
V2 = True velocity of wind, which makes
angle 8 with east or north.
Vo = Relative velocity of wind w.r.t.
the cyclist.
r ~ L ~ ~ E
and Vo = 12 kmIhr
Now from fig, 122 = 10
2
+ v/ + 20 v
2
cos(90 + 8)
=> 44 = v/- 20 v
2
sin 8.
V2 sin (90 + 8)
Now tan 45 = 1 = => 10 - vI 88 = v
2
cos 8.
10 + vI cos(90 + 8)
But
kmlhr
1
cos 45 = Ji
10
2
+ 122 _ v
2
_____ =-z => V2
2
2 x 10 x 12
and Sin 9 = (vi - 44)/20 v2 , 9 ~ 10 20'.
244 - 120Ji , v
2
= 8.38
5. A man is walking on a horizontal road at 3 k.rnItr. The rain appears to
him to come down vertically at the rate of 3 kmIhr. Calculate the actual velocity
and the direction of the rainfall.
Here, VI = velocity of man, VI = 3 kmIhr
Vo = Relative velocity of rain w.r.t. man,
Vo = 3 kmIhr
v2 = True or actual velocity of rain fall,
which makes angle 8 with vertical.
. . 3
2
= 3
2
+ v/ + 2 x 3 x v
2
cos(90 + 9)
=>v
2
- 6 sin 8 = 0 :. 6 sin 8 = v
2
v2 sin(90 + 9)
Again, tan 90 = 00 = 3 + v2 cos(90 + 8)
-vt
v+
2
:. v
2
sin 8 = 3
:. sin
2
8 = 112 :::::> 8 = 45, v
2
= 6 sin 45 = 3.J2 kmIhr.
6. A projectile shot from the ground has range R and the maximum height
it reaches is it. Find the magnitude and the direction of its initial velocity.
Ans. Let v be the initial velocity for projection and 8 is the angle of
projection. I
I
For maximum height reached H, I
v
2
sin
2
8
H = ---. If T be the total
2g
time of flight, v 08 = g T/2.
. . T = 2 v sin 8/g.
(
2v sin8
So for range, R, R = v cos 8.T = v cos 8. 2
g
I
IH
I
I
I
R ------+)
v sin 28
g
e cos-
I
( ). This is angle ofprojection.
2gH 2gH
Again, v2 = = 1- (R2g/8H) .. v =
16gH2
8H _ R 2 g velocity
of projection
7. A particle has total energy E and the force on it is due to potential field
v(x). Show that the time taken by the particle to go from x\ to x
2
is
t2 - t\ = [2E - v(x)t
/2
dx.
1
Ans. We have, total energy E = "2 mv2 + v(x).
.. v = [E - v(x)] = g [E -
\
: = [E - V(X)]Y
:::::> 5:
2
dx \ = {2 r
l2
dt
I [E - v(x)]2 V-;;; JI1
Linear Motion
61
8. A particle is projected vertically with velocity v. What height it will reach
from ground?
If it reaches upto height 'h' then
1 2 rR+h aMm
"2
mu
= W = J
R
--;:z-dr where, R is the radius of earth.
1 2 aMmh 2aMh
"2mv = R(R+h) R(R+h)'
aM 2gRh
Brg= R2 v2= (R+h)'
R + h _ 2gR _ R _ (2
g
R _ 1)
h - v
2
- h - v2 .
- h = 1 ( - 1) This is the height reached
9. A particle is allowed to fall under gravity from height h. If the air
damping on it be proportional to instantaneous velocity of the particle, find
the velocity attend at time t.
Here, for instanteneous velocity 'v', this air damping F a. v F = -Rv.
Here, R = resistive constant.
dv
Since, the equator of motion of the falling particle is m dt = mg - Rv.
f = f dt = s: g v = lotdt
g--v
m m
(
R J
g--v
loge gm = _
R
y--v = -Rt/m
mg e
This is instantaneous velocity of falling particle.
10. Given force F = xy i - i J ' find the work done in moving a particle
from (0. 0) to (2, 1).
For displacement from (0, 0) to (2, 1), ut, Y = t, x = 2t
So, t runs from 0 to 1
and dy = dt, dx = 2dt.
workdone = JF.dr = J xydx - J idy
rl 2 f 2 r) 2
= J
o
2t .2dt - t dt = J
o
3t dt = 1 unit
11. Two particles of mass m) and m
2
are moving respective with velocity
v) and v
2
Show that the total kinetic energy of the system is
when, v =
m)v) + m2v2
(m) + m2)
Here, total energy of the system is
1 2 1 2
E
= - m)v) + -m2v
2 //)
2 2 1'
m)v) + m2v2
But let, U = v) - v
2
' V = +
m) m2
m1v
1
+ m
2
v
2
= (m
l
+ m
2
)v
m)(u + v
2
) + m
2
v
2
= (m
l
+ m
2
)v.
m)u
v = v---:...-
2 m) + m2
m2
u
similarly,v
1
= v +
m) +m2
So, from equation (l), the total energy is
E
1 ( ) 2 1 m)m2 2
=- In) + m2 v + - ( ) u
22m) + m2
12. Two particles of mass m and 2m approaches to each other due to their
mutual attractive force. Find velocity of their center of mass.
Ans. Since, no external force acts on the system, are from momentum
conservation principle,
mV
I
= 2mv
2
. where, VI and v
2
are their resp. velocities.
Linear Motion
mixi - m2
x
2
and for, center of mass position, Xc =
ml +m2
the velocity of center of mass,
dx
c
ml vI - m2 v2 mVI - 2mv2
v = - = =0
c dt ml + m2 ml + 1n2
13. Find acceleration and tension of string for
At wood's machine.
63
Ans. For tension of string T and acceleration of the system 'f, m
2
g - T =
m2f
T-m2g= mJ
(m2 - md2
f= (ml +m2) T=m2g-
m
2
f
=> T = 2m
l
m
2
g
(m
l
+ m
2
)
Now, we can hence have,
[(ml + m2)2 - 4mlm2 Jg2
2
(ml + m2)
This is the relation between acceleration f and tension T.
Here, are all that for At wood machine,! increases for T decreases.
~ f
T
14. A rocket rises from Launching pad with an exhaust speed of 2kg sec-I.
For what value of the ratio MJM
y
can the rocket escape the gravitational pull
of the earth? Here Mo is the initial total mars of the rocket and My is the mars
of the vehicle.
Ans. For rocket velocity, v = Vo + u loge ( : ) .
Now, Vo = 0 (say).
v = escape velocity = 12 km/sec.
u = 2 km/sec.
12 = 210
g
e ( ~ ~ . where, Mo = initial mars
M = My final mars.
Mo
M = (e)6 = 403.43.
y
This is the ratio of ( ~ ~ ) .
EXERCISES
1. Define the following parameters.
Displacement, Speed, Velocity, Average velocity, Instantaneous velocity,
Instantaneous acceleration, Average acceleration, Momentum.
2. What is relative velocity? Find its limit of magnitude.
3. Two bodies are allowed to fall freely from same height in 5 sec interval.
Find their average velocity and acceleration after further 10 sec.
4. Draw velocity-time and acceleration-time curve in the following cases:
(i) Particle is moving with uniform velocity.
(ii) Particle is moving with uniform acceleration.
(iii) Acceleration of particle increases uniformly with time.
(iv) Acceleration of particle decreases uniformly with time.
S. Give the significance of velocity time graph.
6. Rain is falling vertically in downward direction and water is accumulating
in a container. Now, if wind flows horizontally, then do you think that water
will accumulate in that container at the same rate.
7. A train starts from a station at rest and then moves with uniform
acceleration./; along time tl. After that it moves with uniform velocity along
time t2 and after that it fmally moves with retardation/
2
along time t3 and
comes at rest in another station. Draw velocity-time and acceleration-time
graph and also find out the distance between that two stations.
8. A man is walking on a horizontal road at 3kglhr. The rain appears to him
to come down vertically at the rate of 3 kmIhr. Calculate the actual velocity
and the direction of the rain fall.
9. Three masses 2 gm, 3 gm and 5 gm are situated at the corners of an
equilateral triangle with sides 5 cm long. Find the position of center of
mars.
10. A rocket has a mars of 1100 kg and contains 900 kg of fuel. The maximum
exhaust speed of gases equated from it is 104 km/sec. If the rocket rises
vertically, then what should be the minimum rate of fuel consumption for
just lifting it off the launching pad.
-:0:-
Chapter-3
Rotational Motion Rigid Body Rotation
3.1 Introduction
For particle motion along any curved path, the instantaneous position vector
of particle will change its direction and magnitude at every instant of motion.
Now for circular motion the magnitude of position will remain invarient and
due to rotation of particle along any circular path, the instantaneous velocity
will be directed tangentially at every position of particle and for any curved
motion, except circular motion, this instantaneous velocity will be tangential. *
In this chapter, we will deal the motion through several circular or angular
parameters, like, angular velocity, angular momentum ... etc and then we will
extend the whole mechanism to the rotation of system of particle, i.e., to rigid
body rotation.
So our main objectivity is that in this chapter we will try to study the whole
rotational theory in general view and try to draw the several conclusions to make
the chapter easy.
3.2 Angular velocity and Angular Momentum:
For point particle rotation along any circular path, the angular velocity is
defined by the rate of change of angular position, i.e. the rate of angular
displacement. This is basically a vector quantity and is denoted by w
Lt ~ = de
tl.t -t 0 !1.t dt .
Mathematically, it is given by, w
* For any curved motion, the instantaneous velocity has two components. One is
radial component vr =r and other is transverse or cross radial component va = r e .
So, v = v/ + veS = rr + res. When radial means along r and cross radial
direction means normal to r, i.e along transverse direction. Now for circular
motion. r = constant, r = 0, and velocity is only tangential, v = va = r e = r w.
65
66
This is actually, an axial vector* and is
directed along the axis of rotation. For point
particle rotation, this angular velocity w is
normal to the plane of circle, i.e. the plane
of v and r when, v is instanteneous linear
velocity and r is instantaneous position
vector.
It can be shown that v =
because, the instantaneous
displacement for circular motion is
wxr;
small
Now,
Hence
s = ex r.
v = w x r when, w
de
dt
r xv = rx(wxF) = (r.r)w-(r.w)r
= r
2
w (:r1-w,;:.w=O).
1 ~ ~ )
w=zrxv.
r
So, w is normal to the plane of r and v and it is along the axis of rotation.
On the other hand, for rotation of point mass. 'm', the angular momentum
is the moment oflinear momentum. This is also an axial vector, which is directed
along the axis of rotation, and it is now given by,
[, = r x (mv) = m(r x v) = m[r x ( w x r)]
Now we have
[, = m[(r.r)w - (r. w)F] = mr2 w
So, [, = Iw. (some, r.w = 0 for r 1- w)
where, for a point mass rotating along a circle of radius r, I = mr2 = moment
of inertia** of that point mass about axis of rotation.
3.3 Angular Acceleration:
For rotational motion of the particle, the angular acceleration is defined by
the rate of change of angular velocity. This is a vector quantity and is denoted
* In our previous chapter, we have discussed about axial vector which is actually
pseudo vector and is obtained by the cross product of two polar vectors. Basically,
this axial vector does not change sign under coordinate inversion (i.e. mirror
reflection).
** This moment of inertia gives inertia of rotation. For point particle rotation, it is
treated as scalar quantity, but we will see very soon that for rigid body rotation it
will become a 2nd rank tensor having 9 components.
Rotational Motion : Rigid Body Rotation 67
~ dw
by ,-; which is mathematically given by, ,-; = Lt ~
VI. VI. !'J.t ~ 0 ut - dt .
When magnitude of velocity of rotating particle along tangential direction
changes with time, then body will have a linear acceleration which for circular
motion is given by j = ii x r .
and j has magnitude f = ar
for ii .1 rand ii is directed in perpendicular direction to the plane of circle,
i.e. parallel to the axis of rotation.
_ de
Since we have, W = dt
_ d
2
e d (de)
we can write, a = dt
2
= dt dt
which is another mathematical representation for angular acceleration.
3.4 Moment of Inertia and Torque:
Radius of Gyration : When a particle rotates along a circular path, its
rotational inertia is represented by a rotational parameter, called, moment of
inertia. This is very similar with 'mass' for linear motion and gives rotational
inertia of the rotating body. It is a scalar quantity and is in general denoted by I.
When particle of mass m (point mass) rotates along circle of radius r, this
moment of inertia is estimated by the product of mass and square of radius of
the circular path, w M
i.e. I = mr2.
But for a rigid body rotation or for the rotation of a
system of particles about an axis of rotation, the moment
of inertia will be the sum of all moment of inertias for
all point particles, existing within the system.
Mathematically,
N
1= Im;r/
;=1
N
r,
If M = I mi be the total mass of the whole system then for rigid body
i= 1
rotation, this moment of inertia can be written as
I = Imi
r
/ = M.!c2.
i
Where, k is another rotational parameter, called 'swing radius' or 'radius
of gyration'. *
Here,
k ~ =
This radius of gyration for rigid body rotation is defined by such normal
distance of an imaginary point from axis of rotation, when the total mass of the
system is supposed to be concentrated at that imaginary point. With respect to
that radius of gyration, we have the moment of inertia.
Since for continuous mass distribution.
I = Lm/i
2
= tr2dm,
i a
When limits of integration depends on the geometric structure of body and
mode of rotation, the radius of gyration is now given by
1
k = [ ~ f: r
2
dm y .
Now for point particle or rigid body rotation about some axis of rotation,
the 'torque' is such parameter, which is rotational homolog of 'force'. This is
defined by moment of force and is that parameter, which, on application on the
body, gives angular acceleration mathematically, it is given by
1 = r x F = r x (mJ)
= r x m (a x r) = m[r x (a x r)]
For point particle rotation,
1 = m[(r.r)a - (r.a)r] = mr
2
a
t = lei ** (when, r.l..<i, r.a
-
't
d
2
e
1-
dt
2
.
= 0) and it is also given by,
* One should remember that the imaginary point taken to define radius of gyration
is no longer the center of mass of the body. It is purely an imaging point whose
position depends on the axis of rotation choosen.
** This expression for torque corresponds to the conservation of angular momentum,
since,
dw d _ tiL
i = lei = I dt = dt (Iw) = dt
dL -
So for i = 0, - = 0, L = constant.
dt
Rotational Motion : Rigid Body Rotation
Here, since, angular momentum is given
L
= r x p . are can have,
for
L
=rxp = r x (mv) m(r x v)
the torque is
-
= r x F
- dP
't
= rx-
dt
dL
dr _ _ dP
- dP
Hence,
=-xp+rx- = rx-
dt dt dt dt
_ _ dr _
(because, since, v lip, v x p = - x p = 0).
dt
dL
i = dt and for i = 0,
dL
dt = 0, L = constant.
This is conservation of angular momentum.
Similarly, for system of particle,
L
-*
't
dL
dt
";:: x dPi (_ _ ~ - )
~ I dt .: vi x Pi = dt x Pi =
I
69
which also leads to conservation of angular momentum for i = 0, L = constant.
dL
* Actually the relation t' = is taken with respect to a fixed frame or body frame
dt
of reference. If we take the time derivatives in a rotating frame of reference then
from the chapter 'Reference Frame' we will see that
(
d
1 (d) _
-I - - +wX
dt J FIX - dt Rot
(
dl") (tiL) _-
.. t = dt Fix = dt Rot + w x L
For rigid body rotation we will see later
that L= I w = j w :. t = j .t + w x L
(where, I is inertia tensor) This is Euler's equation for rigid body rotation.
3.5 Centrifugal force:
It is basically a pseudo force* which
appears within a rotating body and acts in
radially outward direction. It originates in a
rotating frame and is balanced by centripetal
force, required for rotation.
Since we know that for rotation of a
particle along any circular path, the direction
of its instantaneous velocity changes
continuously, so, the centripetal force can now
be given by
F = m
dv
= m ~ v i i )
dt dt'
dii
mVdi . (ii = unit vector along velocity direction)
since, Ivl = v remains stationary for uniform circular motion.
Now from figure, ii = J cose - i sine
where, e is instantaneous angular position.
So,
dii (A A ) de (A A)
di = - j sin e - i cos e di = -ro i cos e + j sin e
F = -m u ro(i cose + J sine).
If Fe be the centrifugal force developed, then Fe = -F'
and Fe = m u ro(i cose + J sine)
So, this centrifugal force has magnitude
u
= m u ro = mv-
r
= mro
2
r.
r
(.: u = ror)
x
If we want to write down the expression of this centrifugal force in vector
notation, then we can write it as
Fe = m if> X (r X (0) = - mif> X (if> X r)
* This centrifugal force only appears in rotating forms, i.e. the frame on rotating
body. It has no essence in inertial frame. It will not appear for real interaction and
thus it is a pseudo for::e, We will also discuss the appearance of it in next chapter
'Reference Frame',
Rotational Motion .' Rigid Body Rotation
= -mro x v = m(v x ro)
Because, in that case,
since,
We get,
Since,
v
v cos(90 + e) i + v cose J
A A
= -vsinei+vcosej
vx * = -v sin e, Vy = v cos e.
ro = rok' we can write down
i j k
-v sinO v cose 0
vxro
0 0 ro
A A
= i rov cose + j rov sinO
and finally,
Fe = m(v x ro)
= m ro v( cosO i + sinO J)
which is identical with equation (1).
71
x
x
So, in vector notation, the centrifugal force, appeared in rotating particle,
is expressed as
Fe = m(v x ro) = mro X (F x ro)
and for point particle rotation, since, F 1. ffi .
Fe = m(ffi.ro)F = mro
2
F. which is directed in radially outward direction.
3.6 Rotational Kinetic Energy
For rotation of a particle along any circular path, the rotational kinetic
energy is the energy necessary for rotation of particle.
For single particle rotation, this energy is given by,
1 1-1
2
1 __
KE
=-mv =-mv.v
.. 2 2
or,
1 (_ _) (_ _)
K.E = 2
m
roxr . ro xr
1 _ {_ (- _)}
= -m ro. r x ro x r
2
* Here, the components of linear velocity along x and y directions are Vx = v cos
(90 + 8) = -v sin 8 and, Vy = v cos 8.
1 _ {( __ )_}
= "2
mro
. r.r w (.or l.ro)
1 (- -) ( __ ) 1 2 2
= - m ro.ro r.r = - mr ro
2 2
K E
= !Iro
2
.. 2
where, I is the moment of inertia for particle's rotation about said axis. Similarly,
for rotation of a system of particles, this kinetic energy is the sum of kinetic
energies for all constituent particles. and is given by,
K.E.=T=
i i ...
( .o ro has some magnitude for all constituent particle)
K.E. H )m2
when, I = I mi
r
? = moment of inertia for system of particles.
3.7 Angular Momentum for Rigid Body Rotation:
Rigid body means collection of a number of point particles, interacting*
mutually and rigidly such that their mutual separation
Ir) = - r) = constant throughout the rotation about any axis.
For rigid body rotation, the angular velocity for rotation will remain constant
for all constituent particles and hence, the angular momentum is given by,
* This condition - 01 = constant for rigid body rotation is called constraint and
thus it is constrained motion. We will discuss it in other chapter.
(x;i' + Yi +Zi u)}
-xl)())x + (-XiY,)())y + (-XiZ,)())Z}
so we get,
1= i(lxx())x +Ixy ())y + Ixz())z) + J(lyx())x +Iyy ())y +Iyz ())z)+
k(lzx()) x + Izy()) y + I
zz
()) z)
L = Lx i + Ly J + LJ ... 1
When, Lx = Ixx ())x + IXY ())y + Ixz ())z ... 2
Ly = Iyx ())x + Iyy ())y + I
yz
())z ... 3
L
z
= Izx ())x + I
zy
())y + I
zz
())z .. .4
These are the relations among components of angular momentum and
components of angular velocities for rigid body rotation.
Here, Ixx = Imi(r/ - x;) = Imi (Y; + z;)
i
I = Imi(r/ - Yl) = Imi(x; +zl)
>y , ,
, ,
and similarly for Izz' which are called principal moment of inertia about x,
Y and z axis respectively.
On the other hand,
I
=" -m,x,y' I =" -m,y,z,
.xy L.J ""yz L.J 111
i i
and similarly for Izx' which are called product of inertia. *
Here, equations (2), (3) and (4) can be arranged together as,
* We will see in the next section of this chapter, how can this be obtained in several
symmetrical case.
or, L I ro .
This is angular momentum for rigid body rotation. Here, I is called moment
of inertia tensor* of rank-2. This has 9 components, 3 of which are principal
moment of inertia and remaining 6 are product of inertias. This 2nd rank tensor
(
Ixx Ixy Ixz)
is given by, I = Iyx Iw Iyz
Izx Izy I
zz
3.8 Kinetic Energy for Rigid Body Rotation:
When rigid body rotation about a given axis is taken, its kinetic energy is
the sum of kinetic energies of all point particles. This kinetic energy is given
by
= L x (ro
I
= L mj {(r/) ( ro; + ro; + w;) - (Xjro x + yjro y + Zjro z )
2
}
I
"" 1 {2 (2 2 2) (2 2 2 2 2 2
= L.. - mj ri ro x + ro y + ro z - Xj ro x + Yj ro y + Zj ro z + Xj Yj ro x
ro
y +
. 2
I
+ YjXjro yro x + YjZjro yro z + ZjYjro zro y + XjZjro x
ro
z + ZjXjro zro x)}
So, we get, the kinetic energy for rotating body
"" 1 (2 2) 2 "" 1 (2 2) 2 "" 1 (2 2) 2
T = L.. - mj ri - Xj ro x + L.. - mj rj - Yj ro y + L.. - mj rj - Zj ro z +
. 2 . 2 . 2
I I I
* The characteristics of this tensor will be discussed in next section.
where, lxx' Iyy' I
zz
are principal moment of inertias and I
xy
' Iyz' ... are product
of inertias.
3.9 Axes theorem for Moment ~ Inertia:
The moment of inertia for rotation of rigid body about any axis parallel or
perpendicular to the axis passing through center of gravity, can be obtained by
two axes theorem, which are
(i) Perpendicular Axis theorem
(ii) Parallel Axis theorem
Let us discuss this two theorems.
(i) Perpendicular Axis theorem :
(a) For two dimensional body: For a two
dimensional laminated body, if lx, Iy and I
z
be the
respective moment of inertia for rotation about x,
y and z axis respectively, then by the perpendicular
axis theorem.
I
z
= Ix + Iy when the body initially exist
inxy plane
Ix = Iy + I
z
when the body initially exist
inyz plane
Iy = Ix + I
z
when the body initially exist
is xz plane
Because, Let the body exist in xy plane.
Here, Ix = Emy2 (Em = total mass)
I = Emx
2
and 1 = Em?
y z
z
y
when the point P has co-ordinates (x, y) and the mass is supposed to be
concentrated at P.
Now since,? = x
2
+ y2
I
z
= Em(x
2
+ y2) = Emx
2
+ Emy
1=1+1 =1+1.
z y x x y
This is perpendicular axis theorem.
(b) For three dimensional body: In this case, by perpendicular axis theorem,
if I , I , I. be principal moment of inertia of a three dimensional body, when
x y
rotating about x, y and z axis respectively, then z p
Ix + Iv + I
z
= 2 Lm(Op)2.
wheri P is imaginary point at which total mass
is supposed to be concentrated, and '0' is arbitrarily
choosen origin.
Here, by definition,
y
It = Lm(y2 + z2).
I = Lm(x
2
+ z2).
Y
I, = Lm(x2 + y2). 0 x
Because, Ix == ~ m P C 2 ) = Lm(AC2 + PA2) == Lm(y2 + z2) and similarly for
Iv' I
z

. Since, OP2 == x
2
+ y2 + z2
.. Ix + Iy + I
z
= 2Lm(x
2
+ r + z2) = 2Lm(OP)2.
(ii) Parallel Axis theorem :
This theorem states that if AB and CD be two parallel axes such that the
former axis passes through the center of gravity of the body then for moment
of inertia lAB and ICD about AB and CD axis respectively,
ICD = lAB + Ma
2
(M = mass of the body). C
Where, 'a' is the normal separation of two
parallel axis AB and CD.
Here, for point P on the body, if it has normal
distance x from the axis AB, then
I = Lmx
2
AB
and ICD = Lm(x + a)2
Now, ICD = Lm(x + a)2
= Lm(x
2
+ 2ax + a
2
)
= Lm(x
2
) + 2aLmx + a
2
Lm
ICD == lAB + 2a Lmx + Ma
2
.
But in rotational equilibrium*, Lmx = O.
So, ICD = lAB + Ma
2
.
This is parallel axis theorem.
D
3.10 Calculation of Moment of Inertia in different cases:
A
a
8
(a) Circular Ring: Consider a circular ring of radius 'a' and mass M. It
is rotating about an axis passing through its center and perpendicular to the plane
of circle. If A be mass per unit length of the ring, then
'L.nlX
* Here, we have the position of center of mass x == 'L.m .; Now if the axis passes
through the center of mass then x = O.
For that case, 'L.nlX = O.
i.e. the moment of mass will be zero.
M
A =-
21ta'
Now, for a small elementary segment 'dl',
having mass dm = Adl, the moment of inertia is elI
= dm.a
2
.
So, for whole ring, the moment of inertia* I =
f dI = f dm.a
2
= fA dl.a
2
= 'A.a
2
f dl
M 2
= --.a .2n:a = Ma
2

21ta
I = Ma
2

(b) Circular Disc: Here, we take a circular disc
of mass 'M' and radius 'a'. It is rotating about an
axis passing though center of it and normal to the
plane of the disc. If eJ be the mass per unit area of
the disc, then eJ = Ml1ta
2

Here the disc can be divided into a large no. of
co-axial circular rings, one of which has radius x and
width dx.
If dm be the mass of this elementary disc, then
dm = 21tXdx. eJ.
w
So the moment of inertia of that elementary ring about the said axis, is
elI = dm.x
2
= 21tXdx. eJ. x
2
= 21teJ .x
3
dx.
77
So, the moment of inertia for the whole disc about the axis passing though
its center and normal to its plane is
I = f elI = 1: 21teJ.x
3
dx
a
1
1 M 4 1 2
= 21teJ.- = 2-1t--
2
a = - Ma
4 4 1ta 2
1 2
1= -Ma
2
This is the moment of inertia of circular disc, rotating about the given axis.
(c) Hollow Sphere: Here we take the rotation of a hollow sphere of mass
* Here, if we want to find out the moment of inertia of the ring about its diameter
then taking the ring on xy plane, we have from perpendicular axis theorem, I
z
= Ix
~ .
But for symmetry of rotation about x or y axis
1=1 . 1 =.!..I =.!..Ma
2
x y'" x 2 z 2
78
M and radius 'a', about its any diametric
axis. The mass per unit area of the sphere
M
is 0" = -4 2 If this sphere be divided
7ta
into a large no. of co-axial circular rings.
One of which has radius a sin e and width
a d8, the mass of this elementary ring is
dm = 27t a sin 8. and 8.a
= 27ta
2
a sin 8 d8.
Now, the moment of inertia of this elementary ring is dI = dm.(a sin 8)2
= 27ta
2
.a=sin8 dv. a
2
sn
2
8
dI = 27ta
4
a.sin
3
8 d8.
So, the moment of inertia of the whole sphere is now given by a
I = fdI = 27ta
4
0". rer sin
2
8 d8
J9 =0
= 27ta
4
a. Jolt (1- cos
2
8) sine de
4 f+l( 2)
= 27ta a -I 1- z dz (when, z = cos 9)
_ 4 M (1) 4 M 4
- 27ta .--2.2 1- - = 27ta .--2 .-.
47ta 3 47ta 3
2 2
1= -Ma
3
This is the required result.
(d) Solid Sphere : For a solid sphere of
mass M, radius 'a', and density p = 1 i
7ta3
.
If we take the rotation about its any
diametric axis, then dividing the whole sphere
into a large no. of co-axial circular disc, we get
an elementary disc of mass dm = 7tYdx.p.
when, y = radius of elementary disc, and dx is its width.
Now from the figure,
y2 = (a
2
- x
2
) where, x is normal position of the elementary disc w.r.t. the
center of sphere. So, we get,
dm = 7t(a
2
- x
2
) dx.p.
Now, the moment of inertia of this disc about the said axis is
Rotational Motion : Rigid/Body Rotation
1 2 1 (2 2)
dI = 2,.dm.
y
= 2,dm a -x .
1 (2 2)2
=2,.7ta -x p.dx.
So the moment of inertia of the whole sphere, is
f
1 f+a( 2 2)2
I = dI = 2,7tP -a a - x dx
[
]
+a
1 4 2 X
3
x
5
I - -7tp. a x-2a .-+-
- 2 3 5
-a
(
15 -10 + 3) 5 3M
I = 15 tta . 47ta3
2 2
1= -Ma
5
(e) Rectangular Lamina: We take a Y
rectangular laminated disc of length 'a' and
breadth 'b'. So ifM be its mass, then mass
M
per unit length cr = db.
If we now want to calculate its moment
of inertia about any of its two adjacent
sides, then placing it in xy plane with x and
x
y axis along its two adjacent sides, ( a
Let us now find out its moment of inertia about y-axis.
79
1
b
)
x
Taking a rectangular narrow strip of mass dm = bdxcr, its moment of inertia
dl = dm.x
2
= bx2dxcr.
So the moment of inertia of the whole foil about y-axis, is
Iy = fdI = f;bx
2
.dx.cr. = bcri
a3
.
1 M 3 1 2
I = -b.-.a = -Ma
y 3 ab 3
Similarly, the moment of inertia* of the same foil about x-axis is
I = 1Mb
2
x 3
(j) Elliptic Foil : Consider an
elliptic foil, having semimajor axis 'a'
and semimirror axis 'b' and mass M.
We can simply find out the moment
of inertia of this foil about any of its
axes. If cr be the mass per unit area
of the foil, then
M
cr =
nab
P(x, y)
x
Let us now find out moment of inertia of this foil about its minor axis,
along which y-axis is taken. If we now take a rectangular narrow strip, at a
distance x from y axis corresponding to the terminal point P(x, y) on the ellipse,
then mass of this strip is dm = (2y.dx).cr.
g
2 bl2 2
But for ellipse, y = b 1 - 2" = - '\j a - x
a a
2bcr 2 2
dm = -- a - x .dx
a
Now the moment of inertia of this linear strip about minor axis, is
2bcr x
2
2 2
dl = dm.x
2
= a - x .dx
a
So the moment of inertia of the whole foil about that said axis is
Let,
for
f
2bcr f+a 2 I 2 2
1= dl=-- x'\ja -x dx
a -a
X = a sin 8. :. dx = a cose de.
x =a, sine = l,e=nI2.
x = -a sine = -1, e = -nI2.
2bcr J+1t/2
I = --. a
2
sin
2
e.acose.acose de
a -1t/2
* Here, if we take the rotation of foil about z-axis, then I, + Iy = M( a
2
+ b
2
)
(By perpendicular axis theorem)
Here, if we want to find moment of inertia of the same foil about axis passing
though center of foil and parallel to z-axis, we get,
a
2
1 (2 2) 1 2
I = I =I+M- => I=-M a +b --Ma
z Z 4 3 4
= 2ber.a
3
-.!.. f+1t/2(4 sin
2
8 cos
2
8)d8
4 -1t12
1 1 f+1t/2
I = - bera
3
. - (1 - cos 8 48) d8
2 2 -1t/2
(I). = = -.!..Ma
2
mmor 4 nab 4
This is moment of inertia of this elliptic foil about its minor axis.
Similarly, the moment of inertia of this foil about its major axis*, is
= -.!..Mb
2
(I)major 4
(g) Right Circular Cylinder: (A) About its own axis:
Here, if we divide the whole cylinder into a large no. of circular
disc, one of which has mass 'm' when M = Lm = mass of whole
cylinder,
then, the moment of inertia of the whole cylinder about
its own axis is
,,1 2 I 2
I = LJ 2. rna = 2. Ma .
(B) About an axis perpendicular to its own axis and
passing through center o/gravity o/the cylinder:
A c
81
Here, we take an axis AB which passes
though the center of gravity of the cylinder.
Ifwe now consider a small elementary
circular disc at a distance x from the center
of gravity and parallel to the axis AB, then
for another parallel axis CD, passing through
the center of gravity of this elementary disc,
\
I
I
1
1
I
I
1\ ta
J.l___ _ j) __
x
B
1
I
I
the moment of inertia of this disc about axis CD, is dIeD = dm.a
2
. **
When, dm is the mass of the elementary disc. and 'a' is radius of the
cylinder.
* Similarly, from perpendicular axis theorem, if z axis be taken perpendicular to the
foil, passing though the center of foil, then
I 2 2
I
- I + I - - M(a + b )
major mmor 4
** It is followed from perpendicular axis theorem.
M
Now if P = --2- be the density of the cylinder (where, L is total Length
7ta L
of cylinder) then, dm = 7ta
2
.dx. p.
1 2 2
dl
co
= "41ta pdx.a
= 1.1ta
4
pdx
4 .
Now by parallel axis theorem, the moment of inertia of this elementary disc
about AB axis is
dI
AB
= dI
co
+ dm.x
2

1 4 2 2
dI = -1ta .pdx + 1ta pdx.x
AB 4
So, the moment of inertia of the whole cylinder about the said axis is
I = JdI = 1.1tpa
4
J+
L
/
2
dx +1ta
2
p J+
L
/
2
x
2
dx
AB AB 4 -L!2 -L!2
1 4 2 1 [L3)
I = -1tpa .L + 1ta p.-. -
AB 4 3 4 .
1 M 4 1
2
3 M
= -7t--a L + -7ta L .--
4 7ta
2
L 12 1ta
2
L
1 2 1 2
I = -Ma +-ML
AB 4 12
I =1. M (a
2
+.!. L2)
AB 4 3
This is the moment of inertia of a right circular
cylinder about an axis passing through its center of gravity
and normal to the axis of cylinder.
(h) For a Right Circular Cone about its own axis:
We now consider a right circular cone of mass M, base
radius a and height
M 3M
p= -
.!. 7ta2L - 1ta
2
L
3
Here we take an elementary circular disc of radius
'y', at a distance x from the vertex of the cone, then
moment of inertia of that elementary disc about axis of
cone is
Rotational Motion " Rigid Body Rotation
1 z
dI = -dm.y
2
where, dm is mass of the elementary disc
and dm = (nYdx).p.
But
1 z z
dI = -7tPY dx.y
2
1 4
= 27tPY dx.
x L
=-::::}
Y a
1 a
4
4
dI = -np.-.x dx
2 L4 .
So the moment of inertia of the whole cone about its own axis is
f
1 a
4
lL
I = dl = -np.-. x
4
dx
2 L4 0
1 a
4
L
5
143M
= -np.-.- = -na L.--
z
-
2 L4 5 10 na L
3 z
1= -.Ma
10
3.11 Momental Ellipsoid or Ellipsoid of Inertia:
Let us now consider the rotation of any
rigid body about any axis 'OA' having
direction cosines (t, m, n).
If P be a point within the body at which
the mass of the body is supposed to be
concentrated, then for perpendicular PM,
drawn from P on the axis OA, we get, Op2
= x2 + y2 + z2.
and, OM = Ix my + nz.
where, (x, y, z) are coordinates ofP.
Now from the fig, the moment of inertia x
of this rotating body for rotation about axis OA, is
lOA = I = :Lm'(PM)2. where, :Lm' = Total mass
1= :Lm' (OP2 - OM2)
= :Lm' [(xl + Y + z2) - (Ix + my + nzf]
z
83
y
1= Lm'[x2(1 - Z2) + y(1- m
2
) + z2(1 - n
2
) -21m xy - 2mnyz - 2/nxz]
= Lm'[x
2
(m
2
+ n
2
) + y(n2 + [2) + z2([2 + m
2
) -2 Imxy - 2mnyz-
2nlxz]
= Lm'(I + z2) [2 + Lm'(x
2
+ z2)m
2
+ Lm'(x
2
' + y)n
2
+ 2 1m L(-mxy)
+ 2mn L(-myz) + 2 nl'ir--m'zx
I = I z2 + I m
2
+ I n
2
+ 21m I + 2 mn I + 2nl I .
xx yy zz' xy yz zx
Since for several axis, (t, m, n) are variables, so let such axis for which,
body has constant moment of inertia for rotation, I = X, m = Y, n = Z
We now get
I = I X2 + I y2 + I Z2 + 21 XY + 21 YZ + 21 ZX
xx yy zz xy yz zx
:::::> Ax2 + By + Cz2 + Dxy + Eyz + Fzx = 1
Ixx Iyy
When, A= -1-' B = -1- ... 1
I
zz
2Ixy 2Iyz 2Izx
C=-I ,D=-I- E=-I-F=-I-'
and all of these are constants.
Now the equations (1) is an equation of an ellipsoid, * which is an ellipsoid
of inertia or momental ellipsoid. More clearly, we can say that if several axes
of rotation are possible to rigid body rotation, for which moment of inertia of
the body remains invariant, then in this condition, the locus of all direction cosines
for several axes, will be ellipsoid. This elliposid is known as ellipsoid of inertia.
3.13 Moment and product of Inertia and Ellipsoid of inertia of some
symmetrical bodies :
We will now find out
principal moment of inertia and
product of inertia and also
ellipsoid of inertia for some
symmetrical bodies.
(i) A Thin Rectangular
Foil: Here we take a rectangular
foil or lamina in xy plane having
length a, breadth b and mass per
z
M
unit axis cr = ab'
~ - - - - a - - - - ~ )
* We have the general equation of the ellipse,
x
2
i 2xy
-+-+-cosS = sin
2
8 => A'x2 + BY + C'xy = D'
a
2
b
2
ab
Similarly, for ellipsoid, the equation should be,
A'x2 + BY + C'z2 + D'xy + Eyz + F'zx + a'.
=> Ax2 + By + Cz2 + Dxy + Eyz + Fzx = 1.
This is equation of an ellipsoid.
Assuming uniform mass distribution,
We have, the principal moment of inertia
Ixx = I mdyl + zl) = Sf (l + z2) dm
i
= Sf (l + z2) crdxdy = Sf lcrdxdy (.,' z = 0)
J
+
a12
2 J+b/2 Ma
2
I = cr. y dy dx = --
xx -a12 -b/2 12
Similarly,
Mb
2
M (2 2)
1=--.1 =-a +b
Y.Y 12 zz 12
(By perpendicular axis theorem).
Also the product of inertia
IX)! = IyX = I -mjXjYi = - JJ cr xy dxdy
i
f
+b/2 f+al2
= -cr. xdx ydy =
-b/2 -a12
and some z = 0, I
yz
= Izy = 0,
I = I = 0.
zx xz
85
So for this thin rectangular foil with '0' at the centre, the ellipsoid of inertia
is given by,
I x2+I y2+I z2= 1
xx Y.Y zz
+b
2
)z2 = 1
=> 12 12 12
=> = [a
2
x
2
+ b
2
l + (a
2
+ b
2
)z2] = 1
which is the equation of a spheroid.
(ii) Circular disc : Here, we consider a uniform
circular disc of radius' a', mass per unit area cr = M!
1ta
2
.
Assuming uniform mass distribution and using
cylinderical co-ordinates x = r cos e, y = r sin e, z =
z and taking any elementary area dA = dx dy = dr.rde
which has mass dm = rdrde.cr
We may write
I
zz
= Sf (x
2
+ l) rdrdecr = cr f: r3 dr fo;;e
a
4
Ma
2
= cr.-.21t = --
4 2
which is our expected result
Also,
a
4
Ma
2
= (j-'lt = -- =1
4 4 y.y
We can also show that, in this case of circular disc, Ixy = I
yz
= Izx = O.
So ellipsoid of inertia is Ixx x
2
+ Iyy y2 + I
zz
z2 = 1
Ma
2
(2 2) Ma
2
2
=> -- x +y +--z = 1 =>x2+y2+zz2=4
a 2
which is equation of an oblate spheroid.
(iii) Thin spherical shell :
We now consider a thin spherical shell of
radius 'r', thickness dr, mass per unit volume
',---------- ,-'
--- -
M
p=
4'ltr
2
dr
Assuming uniform mass distribution and
using spherical polar coordinates,
We now have,
x = r sin S cos S
y = r sin S sin S
z = r cos S.
--------- \
,- \
".,,- r \ \ ..........
\ \ ,
\ I
, 0 I:".'
.... __ I 1.--_ .....
-------------j--,
, I
, I
I I
I I
/"./
V
I
zz
= f(x2 + i)pdv = H(r sinS COS<p)2 + (r sinS sin<p )2].
prZ sinS drdSd<p
I = pr
4
dr r
7t
sin
3
SdS = pr
4
dr . .i.2'lt =
zz J
o
J
o
3 3
Similarly, we can show that
J(
2 2) 2 2
Iy'y = z + x pdv = "3 Mr . = Ixx.
All of these are moment of inertia of the shell about any of its diameter.
Again,
Ixy = - J xypdv = - J r sin S cos <p r sin S sin <p pr2 sin SdrdS d<p
I = -pr
4
dr rx sin
3
SdS cos<p d<p = 0
xy J
o
J
o
and Iyz = Izx = ... = O.
Hence the ellipsoid of inertia will be
%Mr2(x
2
+ i +z2) = 1
or, x
2
+ y2 + z2 = % J.
This is equation of a sphere.
And, similarly, we can get all of these for other symmetric bodies.
3.12 Moment of Inertia Tensor:
87
From our previous discussion we have seen that due to rigid body rotation
about a fixed axis, the angular momentum is given by,
L = i ro and kinetic energy T = j
, J
Here, I is a 2nd rank tensor, called inertia tensor. It has 9 components and
has the form. I
where,
and
Ixz]
Iyz = I..
I 1J
zz
I =" m (,} -
xx L...." ,
I = - "m.xy. I = - " m,Yi
z
,
.\Y L.... 111,yz L....
Izx = - L mizixi ... etc.
In tensor notation we can now write down
I
=" m(r
2
o .. - r.r.)
ij L.... 1J IJ'
where, oij = 1 for i = j
= 0 for i:t; j
Here, I.. = I .. which is called symmetric tensor, and it has rank-2, with three
1J Jl
diagonal elements, principal moment of inertia where as, the off diagonal elements
are product of inertia.
3.14 Routh's Rule:
By this rule, we can also find out the moment of inertia of symmetrical
bodies. According to this rule, the moment of inertia of a body about any of
the three mutually perpendicular axes of symmetry passing through the center
of mass ofthe body is given by,
mass of body x sum of the squares of the other two perpendicular semi axes
n
where, n = 3, 4 or 5 according as the body is rectangle (or parallelopiped),
a circle (or ellipse) or a sphere (or ellipsoid).
As for example, the moment of inertia of a rectangular lamina about the
required axis is given by,
I ~ , ( ~ ) + (%)']
3
= ~ (a
2
+ b
2
)
where, M = mass of lamina and a, b are respectively the length and breadth
of the lamina
For uniform circular disc,
I = M(r2 + r
2
) = 1. Mr2
4 2
M(r2 + r2)
For solid sphere, 1= -"'-----'-
5
... etc.
3.15 Euler's Angles
Uptill now, we have taken the motion of a
rigid body to a body frame of reference, the axes
z
z
y,
y
x
x,
Rotational Motion: Rigid Body Rotation 89
of which are coincident with the principal axes of the body. These axes rotates
along with the body. We will now introduce axes fixed in space which is actually
the inertial frame of reference.
Now the orientation of a rotating body can be completely specified by giving
three angles called Euler's angles.
Let us first take OXYZ frame on the body and the body is at first rotated
through angle <p about z axis. In this case OX axis takes the orientation oX
I
which
is called the line of nodes. We then rotate the xy plane about line of nodes oX
I
through angle e where oz axis takes position to OZ2' This we get a new co-ordinate
system (XI Y 2Z2)' Then a rotation through angle \jI is carried out about the axis
OZ2 to bring the axes of coordinates to the final configuration (123). In this
configuration these axes are coincident with the axes of the body frame of
reference. Here, the angles (<p, e, \jI) are then called Euler's angles which are
used to locate the body with reference to the fixed axes.
SUMMARY
1. For point particle rotation, v = w x r
So angular velocity
and angular momentum
1 (_ _)
w=-Zrxv
r
1= m(rxv)=m{rx(wxv)}
2. For point particle rotation angular acceleration ii = i-(r x 1)
r
3. Moment of inertia I = mr (for point particle rotation)
I = 'Lmr = Mk2 (for rigid body rotation)
4. Radius of gyration
k = ~ = ~ L ; ~ 2
5. For point particle rotation,
t = Torque = Moment of force = r x F = Iii
dl (-I - - --)
where, t = dt = r x p = mr x v
6. For rotation of particle,
Centrifugal force developed Fc = m w x (w x r)
7. Rotational kinetic energy for point particle rotation
1 2
K.E = "2
1
0) , where, I = 'Lmr
8. Angular Momentum for Rigid Body Rotation is given by
9. Kinetic energy for Rigid Body Rotation
K.E
! LLlij.coi'CO
j
2 i k
10. For Rigid Body Rotation.
Parallel Axis Theorem : I = (I)CM + Ma
2
Perpendicular Axis Theorem (2 dim) = I = I I ..... etc.
z x y
Perpendicular Axis Theorem (3 dim) : I + I + I = 2 L.m (OP)2.
x y z
11. Moment of Inertia for rotation about symmetric axis in several cases -
(i) Circular Ring: I = mil
(ii) Circular Disc: I = 1 mr2
(iii) Circular Disc (about diameter) : 1= mr
2
2 2
(iv) Spherical Shell : I = 3 mr
2 2
(v) Solid Sphere : I = "'5 mr
1
(vi) Right Circular Cylinder (about axis) : 1= 2mr2
(vii) Right Circular Cylinder: 1= m(r
2
+ ~ b 2 (about axis .1 to the
axis of Cylinder)
(viii) Rectangular lamina (about axis passing through comer and .1 to the
1 (2 2)
plane on lamina) : I = 3m a + b
(ix) Elliptic lamina: I = m( af + bf)
(x) Circular cone (about own axis) = I = 1 ~ mr2
Rotational Motion.' Rigid Body Rotation
1 2
(xi) Uniform rod : 1= 12
mL
12. Equation for ellipsoid of inertia:
Ax2 + By2 + Cz
2
+ Dxy + Eyz + Fzx = 1
13. Moment of Inertia Tensor:
I = l.m (ro - r r)
IJ I) I J
mass of body x Sum of the squares of the
other two perpendicular semi axes
14. Routh's Rule: 1=------"--''---------
n
n = 3 (rectangle)
= 4 (Circle or ellipse)
= 5 (Sphere or ellipsoid)
15. Euler's angles : (<p, e, \jI)
Worked Out Examples
91
1. Find the moment of inertia about the diagonal of a rectangle of side 'a'
and 'b' respectively.
Ans : Here from the equation for ellipsoid of inertia, the moment of inertia
about the diagonal is,
I = I COS2 a + I Cos
2
(90 - a) + I Cos
2
90
xx yy zz
= I Cos
2
a + I Sin
2
a .
. u yy
But
and,
1 2 1 2
I = -mb I =-ma
xx 3 'yy 3
2 a
2
b
2
= -m---
3 a
2
+ b
2
2. A body of mass hangs from one end of a light rope which is wound on the
surface of the horizontal cylinder of radius 'a', which is free to rotate about its
axis.
Prove that the body descends with an acceleration
Ans : Here, from energy conservation,
!mv
2
+!IW2 = mgh
2 2
1 2 1 v
2
-mv +-1- = mgh
2 2 a
2
(: v = wa)
1 v
2
1
"21:; = "2m(2gh-i)
m(2gh - i)a
2
m(2gh - 2fh)a
2
v
2
2fh
I
m(n- f)a
2
f (': v2 = 2fh,J= linear acceleration)
3. For a thin spherical shell of radius or', thickness 'dr' and mass per unit
volume p, Show that the equation of ellipsoid of inertia is the equation of a
sphere,
Ans : Here, x = r sin a cos cp, y = r sin a sin cp, z = r cos S
for spherical polar co-ordinates (r, a, cp),
So, we have,
I
zz
J(x
2
+ i)pdv
J[(r sin S COScp)2 + (r sinS sin cp)2 ]p,r
2
sin a drdSdcp
1t 21< 4
pr
4
dr J sin
3
ada J dcp = pr
4
dr'3.27t
o 0
2 2
Simiiarly, I = I
zz
= -Mr
yy 3
But the produce or inertia
(': M = 47trdrp)
IXY - J xypdv= - J rsinScoscprsinasincp.p.r
2
sinSdrdadcp
1t 21t
- pr
4
dr f sin
3
adS f sincp coscp dcp = 0 = lyx
o 0
Rotational Motion " Rigid Body Rotation 93
Similarly, I = I = 0, I = 0
yz zy xz
But, I x
2
+ I ,,2 + I Z2 = 1
xx ff zz
= jMr2(x2 + i +z2) = 1
3 1 2
=> x
2
+ y2 + Z2 = ---= a (say)
2Mr2
This is equation of a sphere.
4. Find the moment of inertia of a uniform vertical rod of mass m and length
1
L about a horizontal axis passing through a point '3 L from one end.
Ans : From parallel axis theorem,
(
L L)2 1 1 .
I = (I) +M --- = -ML2 +-ML
2
eM 2 3 12 36
1 _ -
5. For rigid body rotation, show that T = 2"(J).L
where the symbols has usual significance.
II
- Iw-w
Ans : We have, T = 2 lJ' I )
i,j
~ ~ ~
But L = Lxi + Lyj + Lzk
= (Ixxwx +Ixywy + I x z w z ~ +
(Iyxwx + Iy'ywy + Iyzw
z
)] + (Izxwx + Izyw
y
+ Izzwz)i
(i = x, y, Z, j = x, y, z)
EXERCISES
1. State and prove the theorem of parallel axis for moment of inertia.
2. Find an expression for the moment of inertia about any axis of a rigid body.
3. Considering expression for kinetic energy of a rigid body, find its angular
aT -
momentum and also show that aw = L
4. Find the principal moment of inertia of a cylindrical tube of length L, mass
M and inner and outer radii R
J
and ~ respectively.
5. A rigid body is rotating with an angular velocity w about an axis lying in the
y-z plane and making an angle of 45 with the y-axis. Obtain the K.E of the
body in terms of the moments and products of inertia. Also obtain the
components of angular momentum of the rigid body.
6. A rectangular thin plate of dimension a spins about an axis along the diagonal
at a constant angular velocity w. Given Iu == ~ ma
2
, Iyv == lma
2
(i) What
is the direction and magnitude of the angular momentum? (ii) Calculate the
torque about the axis of rotation.
7. A meter stick is held vertically with one end on the floor and is then allowed
to fall. Find the velocity of the other and when it hits the floor, assuming
that the end on the floor does not slip.
8. Find out a relation between the linear velocity and the distance travelled, for
a sphere of radius a and mass m rolling down an inclined plane having
inclination e.
9. Consider two masses m
J
and m
2
separated by a distance ro when connected
with a massless rigid rod. Show that the moment of inertia of the two masses
about an axis through their center of mass in a plane perpendicular to the
"d d' m)m2,.,2
rlgl ro lS 0 .
m) +m2
10. The products of inertia in the principal axis are zero - how is it realised in a
symmetrical body?
-:0:-
Chapter-4
Reference Frame
4.1 Introduction
When we deal with any physical system either in rest or in motion, we
should take everything with respect to a certain region, called frame of reference.
This is very important and significant to consider such a reference frame with
respect to which the physical system and its interaction can be viewed. The
frame or the region of interest can be classified into two types in respect of its
nature. This two are inertial and non inertial frame. The frame which is at
absolute rest or in uniform motion, is called "inertial frame." But the frame
which has some acceleration or retardation is called "non inertial frame".
Basically, earth is itself a non inertial frame and any frame associated with earth
is non inertial. So the idea about inertial frame is absolute and it is physically
impossible to obtain inertial frame in reality.
On the other hand, the frame can be classified into 3 classes in respect of
its geometrical construction, when considered in three dimension. These are
cartesian, cylinderical polar and spherical polar system. Similarly, in 2 dimension,
the frame will have two classes which are cartesian and plane polar system.
The other side about the nature of a frame which is essential to keep in
mind that the law of physics will not alter in any frame but its form of
representation will change from frame to frame. More precisely, if we move
our interest from inertial frame to non inertial frame, we will see that the nature
of the action of force effective on particle motion will change because of the
appearance of some 'so called' pseudo force. So it is a fact that every interaction
with a physical system when considered in a frame, will occur in an organised
way with the frame by obeying the law of physics in every respect.
4.2 Non Inertial Frame and Pseudo Force
When we consider particle motion or the motion of a system in an
accelerated frame or non inertial frame, the particle or the system as a whole,
95
will experience a force in opposite to the direction of acceleration of the frame.
This force appeared in non inertial frame is called 'pseudo force'.
For the presence of such 'pseudo force', the effect of force acting on the
particle will change than that in inertial frame. It can easily be shown that if F
be the force applied on the particle in non inertial frame, the effective force
acting on it will be
Feff = F+(-Fp) = F+Fo
Where, Fo = - Fp is the pseudo force appearing in non inertial frame. The
nature of this pseudo force entirely depends on the mode of acceleration of the
frame. We can clear this point in the following
cases-
(i) Case - I : Pseudo Force in non inertial
frames having translational accelerated motion:
Consider two frames S and S' when S is an
inertial frame and S' is non inertial or accelerated
frame having acceleration J in its translational
motion with respect to S. R is the instantaneous
position vector of the point 0' of S' with respect
to the origin '0' ofS frame.
So if rand r' be the respective position
vector of the instantaneous position P of a moving
particle then from fundamental idea of vector
algebra.
r
R+r'
r' r-R
d
2
r' d
2
r d
2
R
-----
dt
2
dt
2
dt
2
d
2
r'
d', (d'R 1
=:> m--
m--+m ---
dt
2
dt
2
dt
2
=:>
Feff
F' = F + m( - J) = F + (-m J)
Feff
- - (- d
2
r)
F+(-Fp). F=m dt
2
actual force applied from outside.
p
where, - Fp = - mJ is the pseudo force acting on that particle considered in non-
inertial frame.
Reference Frame
97
This pseudo force sharply acts in direction opposite to the acceleration of
the frame.
(ii) Case-II : Pseudo Force in
non inertial rotating frame: ,
Here we consider that'S' (oxyz) is Z \
an inertial frame where as'S" (0' X' Y'
Z') is non inertial rotating frame having
origin coinciding with that of'S' frame.
It is obvious that the basis unit vectors
(i,},k) of S-frame are fixed where as
the other set of unit vectors (i'),k')
changes their orientation w.r.t. time t
during rotation of S'-frame.
So if rand r' be the respective
positions of same instanteneous* posi-
tion 'P' of the moving particle relative
to Sand S' frame then
A A A
r xi +yj+zk
A A A
r' x'i' + y'j' + z'k'
Also for this two frames Sand S',
Z
k
P(x, y, z)
(x', y', z')
A
j
(
dr)
dt fixed
dx ~ dy,: dz A
-1 +-} +-k in S frame or fixed frame.
dt dt dt
(
dr')
d x ~ , dy' ':, dz' k
A
,
and also dt rot = -1 +-} +-
dt dt dt
y'
y
When (t',}',k') are itself fixed w.r.t. rotational frame S' and these are rotating
unit vectors when viewed in S-frame.
(
dr')
dt fixed
(
dr')
dt fixed
d x ~ , dy' ':, dz' A, dt' d}' die'
-1 +-} +-k +x'-+y'-+z'-
~ ~ ~ ~ ~ ~
- +x'-+y,JL+z'-
(
dr') dt' i' die'
dt rot dt dt dt
* Here, in Newtonian machines, the position when measured w.r.t. the frame Sand
S' will be the same in t' = t. So, the position which is instantaneous in one frame,
will be also instantaneous in another frame. But it is not possible in relativistic
case.
98
so,
But we have'" v
di'
dt
(
dr')
dt Fixed
(
dr')
dt Fixed -
dr __
-=wxr
dt
A dj' -. ':, dk' - k
A
,
wxi"-dt =WX) -=WX
, dt
(
dr') +X'(w x j,) + Y'(w xl') +z'(w x k')
dt rot
(
dr') A A A
- +wx(x'i'+Y'J'+z'k')
dt rot
(
dr') __,
- +wxr
dt rot
and we usually have,"'''' (!!..) = (!!..) + wx .
dt Fixed dt rot
This is the relation between time variation operators in fixed frame S(inertial)
and rotating frame S' (non-inertial).

Since, in our case, r' = r = OP = O'P
so, we should have,
(
d2;] = (!!..) (!!..) (r)
dt Fixed dt Fixed dt Fixed
= [(!!..) + w x][( dr) + w X r]
dt rot dt rot
(
d
2
r] _ (dr) _ (dr) _ _ _
= -2- +wx - +wx - +wx(wxr)
dt dt rot dr rot
rot
... from fundamental rotational dynamics, v = wr and v = W x F. Because, ii = ,
d (e- -) de - e- dF dF 0 - - -
s = ex F; :. ii = dt x r = dt x r + x dt' But dt = :. v = w x r when
_ de
W=-
dt
(
dW) (dW) __
...... Here, - = - + w x w; But W x W = 0 .
dt Fixed dt rot
(
dW) (dW)
:. dt = dt i.e. angular acceleration will be the same in any frame.
FIxed rot
Reference Frame
Where W, the rotational velocity is time
:. m(d
2
r) = +2mwx(dr) +mwx(wr)
dt
2
dt dt rot
Fixed rot
(
dr)
But if V = dt = velocity of moving particle in rotating frame
rot
(
d
2
r)
then, m dt
2
Fixed
(
d
2
r) _ _ _ _ _
m -2- + 2mw X V + mw X (w X r)
dt rot
m(d
2
r)
dt
2
rot
+(-2mwx v)+[-mwx(wxr)]
dt Fixed
99
where Feff
(
d
2
r) ffi fi . I . .. I
m -2- = e ectlve orce on partlc e m rotatmg non-mertla
dt rot
frame.
(
d
2
r)
F = m -2- = Actual force acting on particle in ftxed inertial frame.
dt Fixed
FI = Fc = -2m( w x V) = Pseudo force which is commonly known as
'coriolis force'. '" and F2 = Fc = -mw x (w x r) = Pseudo force which is
popularly known as 'centrifugal force'.
So we see that in a rotating non inertial frame, two pseudo forces will
appear, one of which is called corio lis force and other is centrifugal force. The
action of 1st one is conditional and it is not effective for a rest particle w.r.t.
rotating frame, where as the 2nd force always acts in rotating frame.
4.3 Effect of rotation of earth on acceleration due to gravity
Neglecting earth's translational motion around sun, if we only consider
earth's rotation then we can show that the acceleration due to gravity at any
place on earth surface except at pole, will change effectively due to earth's
rotation.
'" This force -2m( W X v) will act on a moving particle in non-inertial frame. As for
example, since earth is it self non inertial frame, this coriolis force will act on
any moving particle on earth surface. But this force will be zero for any rest
particle.
Let us now consider two frames Sand S' when the frame S is inertial and
is taken at the center of earth and the non-inertial frame S' at any point on
earth surface. Now if for any particle motion with velocity v=(dr) as
dt Tot
measured from that frame S' on earth surface, we take its weight mg directed
towards the center of earth and any real force (actual) F acting on it, then the
effective force on it as seen from the surface of earth due to earth's rotation,
will become,
m ~ ~ ~ J
Tot
m d 2 ~ ) + [-mw x (w x r)] + [-2mw x v]
dt Fixed
Feff (F + mg) + [-mw X (w X r)] + [-2m(w X v)]
F + m[ g - w X ( W X r)] - 2m( w X v)
F + mgeff - 2m( w X v)
where geff = g - w X (w X r) is the effective gravitational acceleration of
the place on earth surface. This is the resultant of actual gravitational acceleration
g and centrifugal acceleration due to earth's rotation -w X (w X r) and as a
result we have the nature of this effective gravitational acceleration as,
geff g at pole
= g-wx(wxr)
at any place on earth surface other than at equator (not directed towards the
center of earth).
= g - w X (w X Ii) at equator (directed
towards the center of earth.)
If we now be interested to calculate
the maximum effect of this centrifugal
acceleration on actual gravitational
acceleration, then we will see that for earth
radius R = 6400 km and angular velocity
21t
for earth rotation W = --rad/ sec =
86400
7.29 x 10-
5
rod/sec, the maximum value
of centrifugal acceleration at equator is
Reference Frame
a
2
= w2.R = 3.4 x 10-
2
mlsec
2
which is about 0.2% of earths
gravitational acceleration. So at
equator the reduction of actual
gravitational acceleration due to
centrifugal acceleration will be
maximum and the effective value of
the gravitational acceleration at
equator will be.
gefT= g - w2R = 9.81 - 0.034
= 9.776 rnlsec
2
It is a fact that for rotation of
frame, the centripetal acceleration
will act in opposite to the centrifugal
acceleration towards the center at
which inertial rest frame exist. This
fact is also true for any particle
rotation above a fixed point.
101
o
4.4 Effect of Coriolis Force on a particle moving on the surface of earth
Here we are now
interested about the
effect of corio lis force
on a particle moving on
earth surface. We have
from our earlier
discussion that due to
earth rotation, the
coriolis force acting on
particle moving with
velocity v w.r.t. earth
surface is
Fe = - 2m( W x v)
which will be at right angle to the plane of v and w .
Now we consider particle motion at P on earth surface having a latitude e,
with velocity v along y-axis.
Here for the system OXYZ ofS' frame taken at P, the plane XY is horizontal
plane w.r.t. earth surface at P and Z is vertical axis to the same point P, the
effective component of earth's angular velocity along Z axis is
W z = w cos( - e) = w sin e .
So for particles velocity v along
y-axis the coriolis acceleration will be
-2{ W
z
x v) and its magnitude is 2w
z
v
= 2w sin 0 v.
This is a fact that for such
acceleration in transverse direction to
the particle motion, the particle will
deviate from its original path (along
y-axis) in transverse direction (in XY
plane) and will finally move along
the deflected path in XY plane as
shown in figure.
The maximum effect of this x
z
A
k
---
transverse corio lis acceleration will occur at any pole and it is
......................
Deflected'" ,
path ",
1t
a
c
= 2 w sin "2.v = 2wv = 2 x 7.29 x 1O-
5
v
a = 1.458 x 10-4v.
c
When, v is the velocity in horizontal plane. If we now take a typical example
of a particle moving horizontally on earth surface with velocity lkm!sec or 3600
kmIhr, the magnitude of corio lis acceleration will become a
c
= 1.458 x 10-4 x
10
5
= 14.58 cmlhr.2
or, a
c
Ri 0.15 mlsec
2
which is about 0.015 g.
We can now conclude that although the magnitude of coriolis acceleration
is small, it plays an important role in many phenomenon in the earth. It is
important to take into consideration the effects of the corio lis acceleration in
the flight of missiles, the velocity and the time of flight which are considerably
large.
Here for particle moving with velocity v in the horizontal plane on earth
surface, we can immediately find out the angle of deflection for particle motion
along the duration t as
distance travelled in time t in the deflected direction
a=
(for sufficiently small deflection)
distance travelled in time t in the direction of projection
1 (2wsinO. v)t
2
.
-'------'- = w sm O. t
2 vt
Thus, on the north pole (9 = 90), a = wt which is maximum and for any
t = 180 sec.
a = 7 x 10-
3
radian = 0.04 which is found to be quite small, but it assumes
considerably important in guided missiles.
Reference Frame 103
4.5 Effect of Coriolis force on a particle falling freely under gravity
We consider a particle which is falling freely (neglecting air resistance) from
height 'h' on earth surface and in this case, due to corioUs acceleration only
the acceleration of particle will become
a =g-2wx v
When w is angular velocity for earth rotation, a and v both are measured
in rotating frame on earth surface.
Now from our previous topic of discussion, we have, at latitude 8 on earth
surface, where we have considered the falling of particle in -k direction (along
-ve z direction),
Wx = v, Wy cos 8, W
z
= W sin 8
Also, for particle motion along -ve z direction,
i 0, y 0, i = - gt
Again, for quite small deflection produced by coriolis force the components
of the acceleration will be
ax x = 1-2( w x v)1 = 1-2(-0) cos8 gt)1
2w gt cos 8
a
y
ji = 0
a
z
z = - g
Here the acceleration ax along X direction is due to corioUs force and we
thus get
d
2
x 1 3
-2- = 2wgt cos8 => x = -wgt cos8
dt 3
d
2
z 1 2
z = -=-g => z=zo --gt
dt
2
2
when, t = 0, z = zO' x = 0, y = 0 is the initial position, with initial velocities
i( 0) = 0 = i( 0) .
Since, the time of fall from height 'h' is
t = f ' the deflection of a particle towards the east when it is dropped
from rest is
1 8h
(
3 )1/2
8 = "3wcos82
This is called 'eastward effect' due to the effect of coriolis force on a freely
falling body. As for example, if the particle be dropped from a height of 100
m from rest at latitude e = 45, it will be deflected by about 1.55 x 1O-2m
toward east in northen hemisphere.
4.6 Principle of Foucault's Pendulum
The French scientist Leon Foucault showed that the small effect of coriolis
force may be remarkable for the oscillation of a pendulum. He noticed that the
right angled coriolis deflection on one swing of the pendulum could not be
undone in the return swing. The effect of corio lis force of terrestrial origin
moved from the domain of theory and outdoor observations to that of observation
in a laboratory experiment.
The equation of motion for a pendulum* including the coriolis term is
2-
d r 2- (
-2 + k r = -2 W x v)
dt
where, k = = stiffness constant for pendulum and' l' is the effective
length of pendulum.
Now taking cartesian components of this equation, we get
x+k2x 2(Ywz
For i and z negligible in compare to X, y and X, y . Thus the above equations
are the coupled equations of motion in x and y.
To solve the above coupled equations, let us choose a complex variable
u = x + iy (i = .J=l) .
Then, after some algebraic manipulation, we finally get,
or, it + 2iw
z
u + eu 0
The general solution of this equation is now given by,
* For pendulum motion in simple harmonic manner under restoring force only, the
equation of motion becomes,
d
2
- d2- k-
- _ r _ r or
m/=-k r => m-=-kor => -+-=0
o dt
2
dP m
ko g d
2
;: = 0 d
2
;: 2-
But -;;; = I for pendulum oscillation:. dt
2
I => dt
2
+ k r = 0 when no
other force act
Reference Frame
where,
,
/
",;r"-
,
/ \
/ \
I \
I \
I \
I \
J J
I I
I I
I I
I J
I J
\ ,
\ ,
\ I
\ I
\ /
\ /
\ ,
" "
~
rotation of plane of
oscillation In clock wise
sense for Foucaults pendulum.
y
u exp(-iwzt)[c\e
ik
'
t
+C2e-
ik
'
t
]
~ X +/.
k'
+
. ( .) -iw t
u = X ly = Xo + lyO e Z
105
x
where, Xo and Yo are the solutions when Corio lis force is absent and u' :::; k for
k> > w
z
'
So from this solution, we can conclude that the plane of oscillation of the
pendulum rotates with an angular velocity -w), in opposite to the sense of
rotation of the earth. The period for a complete rotation of the plane of oscillation
is
21t 21t
T w
z
= wsinEl at latitude El.
So at the poles, T 24 hr, and at the equator
So that no rotation of the plane of oscillation is observed at the equator.
Foucault had demonstrated the truth of the above facts through his historic
pendulum experiment in 1851. By measuring the period of oscillation, T,
Foucault measured the period of earth rotation. This was the first experimental
proof that the earth is in fact rotating w.r.t. the inertial frame in which Newton's
Laws are valid, with an angular velocity which is precisely the same as that
inferred frame the apparent durinal rotation of the sun, moon and the star sphere.
4.7 Flow of River on Earth Surface
Basically, rivers flow approximately in a horizontal plane for which a slight
downward slope in the direction of flow is required to maintain the speed of
the flow. So the component of gravitational force along the direction of river
flow is primarily important depending on the magnitude of the slope of the
down stream. But there is absolutely no component of geff (including the effect
of centrifugal acceleration) acting along the breadth of a river. Hence, the
component of corio lis acceleration, however small, can act freely on the moving
water across the direction of the stream.
To find the effect of such transverse coriolis acceleration on the flow of
river, let us choose the direction of river flow along x-axis. y-axis is taken in
transverse direction to x-axis on horizontal plane and z -axis in vertical direction.
The flow is now considered at latitude 'e' of earth surface, making angle e (in
the anti cloth wise sense) w.r.t. the geographical north direction. Therefore, the
earth's angular velocity vector w w.r.t. the said reference frame is now given
by,

w = wsinek + wcose coscpi - wcose sincp}
.. w = w(sinek+cos9coscpt -cos9sincp}). and the velocity of flow is
v = vi along x-direction. So the coriolis acceleration is
lie = -2(wxv) = -2vw(sin9}+cos9sincpk)
Thus, a corio lis force will be experienced by the water in the rivers flowing
in any direction, causing a direction towards the right of the flow direction in
the northen hemisphere and to the left of the flow direction in the southern
hemisphere. As a result, the corresponding banks of the river will be denuded,
which is actually observed.
SUMMARY
1. Frame is a certain region w.r.t. rest or motion of particle or system of
particles is considered.
2. Frame has two types which are inertial and noninertial frame. Inertial frame
is at rest or in uniform motion where as non inertial frame is accelerated
frame.
Reference Frame 107
3. Pseudo force will appear in non inertial frame and in that case, the effective
force in non inertial S '-frame is
F' = F + ( - Fp )., F = Real or actual force.
- Fp = Pseudo force appeared.
4. For rotation of the frame with angular velocity W, (;). =(;) +wX
FIxed rot
5. Centrifugal force developed in rotating frame is
-w x (w x r).
For earth's rotation, it reduces the effective gravitational acceleration resulting
its effective value as
geff =g-wx(wxr)
6. Coriolis force developed in a moving body w.r.t. the rotating frame on earth
surface is -2m( w x v).
7. The maximum value of centrifugal acceleration at equator is ~ = w
2
R =
3.4 x 10-
2
mls
2

= 0.3% of g.
8. The maximum effect of transverse coriolis acceleration at any pole of earth
is
a
c
= 2 w sin 90 v = 2 w v = 1.458 X lO-4v.
9. Angular deflection of particle moving in time t at latitude e of earth surface
due to transverse coriolis acceleration is a = w sin e. t
10. If any latitude e on northen hemisphere of earth, the components of angular
speed 'w' of earth rotation are, Wx = Wy = W cos e, W
z
= w sin e.
11. The deflection of particle when dropped from certain height 'h' above earth
1 8h
(
3 )1/
2
surface, due to coriolis acceleration is 8 = "3wcose g and this
deflection occurs towards east.
12. For Foucault's pendulum, the time period for each oscillation at any latitude
e on earth surface is T = ~ .
wsine
\Vorked Out Examples
Example 1 : An object is thrown downward with initial speed vo. Prove
that after time t, it is deflected east of the vertical by an amount
2
wvot2 cose +-wgt
3
cose
3
Ans : Here, the effective component of angular velocity w for earth rotation
along x, )', z axis of the frame S' taken on earth surface, are
Wx = 0, Wy = W cos 9, W
z
= W sin 9
Now, for velocity in time t, if X, y, z be the components then x = 0 , y = 0,
z = Vo + gt.
Again the instantaneous transverse coriolis acceleration
x =
x = d ~ = 12(w x v)1 = 2wcos9( Vo + gt)
dt
2wvo cos 9 + 2wg cos 9 t.
wv
o
t
2
cose + w g t 3 cose
3
This is the deflection towards east of vertical.
Example 2 : Show that if the particle moves near the earth's surface, then
the equation of motion are given by
x 2wsin9y
ji -2(wsin9 x + wcos9z)
z - g + 2w cos 9 y
for latitude '9' on earth surface.
Ans : We have for any frame S' (xyz) on the earth surface,
A A A
W = wxi +wy}+wzk
A A A
= w) +wy}+wzk
wx v = (-wsin9y)i +(wsinex+wcos9z)}-(wcoseY)k
But the equation of motion is given by,
d
2
r
- = g-2(wx v)
dt
2
= -gu + 2wsine yi - 2(w sine x + xcose z)} + 2wcosey k
x = 2wsin9 y
ji = -2(wsinex+wcosi)
i = - g + 2w cos 9 y
Example 3 : An object is thrown vertically upward at latitude 9 with speed
va- Prove that when it returns, it will be at a distance west ward from its starting
4 3 cos9
point equal to 3
WVo
--2-
g
Reference Frame
Ans : Here, by following Ex. 1. the deflection is
x = WVo cose t
2
+.!.wgt
3
cose.
3
But t is the time of ascending which is actually the time of fall.
9 3
so, Vo = gt ... x = "3
wgt
cose
Vo
Also, t = -, and then,
g
4 v
3
the total deflection x = -. wg-i} cos e
3 g
4 3 cose
x= 3WVO-2-'
g
109
Example 4 : An object at the equator is thrown vertically upward with a
speed of 60 mph. How far from its orginal position will it land?
Ans : Here the deflection is given by
4 3 cose
x= 3
WVO
-
2
-
g
But w = 7.29 x 10-
5
rad/sec
Vo = 60 mph = 2.682 x 10
3
cm/sec
q = 0 at equator
g = 980 cm/sec
x = l.952 Ri 2 cm.
Example 5 : How long would it take for the plane of oscillation of a
Foucault pendulum to make one complete revolution if the pendulum is located
at a plane of colatitude 45.
Ans : Here latitude e = 90 - 45 = 45.
So for Foucault's pendulum, the time taken for one complete revolution of
the plane of oscillation .
T = = 2n = 33.84 hr.
w sine 7.29 x 10-
5
x sin45
Example 6 : Prove that if an object is thrown downward from height h
with initial velocity Vo above the earth surface then it will hit the earth at a
point east of the vertical at a distance. (at latitude e)
w;;;e ( V + 2gh _ Vo )2( v + 2gh +
2v
o).
Ans : Here by following the similar problem of Ex. 1. we have the east
ward deflection
x = WVo COSe(2 + .!wg cose (3
3
But for instant T, the downward instanteneous velocity
v = Vo + gt
1
(= -(v- vo)
g
But v2 = v5 + 2gh
1 = ~ [ ~ v J + 2gh - Va]
and
WCOSe 12 ( 3)
X = gl + Vo
3
w ; ~ e ( v5 + 2gh - Vo r ( vJ + 2gh + 2vo )
This is eastward linear deflection of that falling body.
Example 7 : A train having a maximum speed equal moving Vo is moving
round a curve with radius of curvature 'p'. Prove that if there is no lateral
deflection through out the outer tract, then the track should be at a height above
the inner track is given by aV5/ v + p2g2 where 'a' is the distance between
tracks.
ADS : Here, if x be the hight difference between outer and inner track,
then
x v
2
tane= _=_0_
a pletT
But fetT = effective accn. = ~ IJ + /f
2
Vo
IN = normal accn. = -
p
IT = Tangential accn;. = g
EXERCISES
1. What do you mean by a frame? Give its significance.
Reference Frame 111
2. Discuss the appearance of Pseudo force in a non inertial accelerated frame
in translational motion.
3. Discuss the appearance of Pseudo force in a non inertial rotating frame.
4. Discuss the effect of transverse coriolis force on a particle when falling
from a certain height above earth surface.
S. Discuss the effect of centrifugal acceleration on the gravitational acceleration
of earth surface.
6. Discuss the appeamce of cyclone.
7. Discuss the effect of Coriolis force on riverflow.
8. Di8cuss the priaciple of foucaults pendulum.
9. Show that the angular deviation of a plumb-line frame through vertical at
a point on the earth surface at a latitude e is
ro w
2
sin e cos e
2 2 where, r 0 is the radius of the earth.
g-row cos e
10. Suppose that the mass m of a conical pendulum of length I moves in a
I
1/4
horizontal circle of radius a. Prove that (a) Tht speed is ali (P - a
2
)
mg
l
and (b) The tension in the string is I
vP _a
2
11. If an object is dropped to the earth's surface prove that its path is a
semicubical parabola.
12. How long would it take the plane of oscillation of a Foucault pendulum
to make one complete revolution if the pendulum is located at 45 Latitude
or at the north pole.
13. Explain physically why the true vertical and apparent vertical would
coincide at the equator and also the north and south poles.
-:0:-
Chapter-S
Central Force
5.1 Introduction
The discussion of 'Central force' is very richest discussion in classical
mechanics, since the knowledge about central force is important for dealing
with several inverse square force, specially, in case of plane tory motion. Several
forces, like Newton's force of gravitation, Van' dar Waal' s force between' atoms
and molecules in a gas, or even the Yukawa force between the nucleous in the
nucleons of atoms are examples of central forces. The idea about it is in general,
applicable in discussion on tides, dynamical manifulations of the orbits of
spaceships in the space age, the geometry of the orbits of plannets ... etc. In this
chapter, we will restrict ourself in discussion of central force characteristics and
in several applications of it. Basically, the chapter will go through the depth of
central force phenomena.
5.2 Definition and Characteristics of Central force
The force field, either attractive or repulsive which is always in radial
direction towards or away from the center (origin) of the force, is called central
force.
Mathematically, it is represented as
FCentral = F = f(r).c. = f(r)r .
r
Where, '+' sign indicates repulsive force and '-' sign gives the attractive
nature of force, and the force is always directed along (r) radial direction.
The basic characteristics of the force are,
(i) The force may be repulsive or attractive, which is directed towards
or away from the centre of force.
(ii) This force obeys inverse square law of force, and this kind of force
field is derivable from a scalar potential field, cp(r) such that
112
Central Force 113
f(r)r = - v<p(r).
where <p (r) is some scalar function.
(iii) The force which does not depend on time explicitly will be always
conservative in nature.
(iv) Energy and momentum conservation will always hold under action
of central force.
(v) For conservation of central force, angular momentum m(r x v) will
be the constant of motion.
(vi) The magnitude of central force f(r) could be uniquely represented by
a power in r, i.e.
+00
" knr-n
f(r) =
n=-oo
where, are either constants including zeros or at most functions
of time.
(vii) Only radial dependence in f(r) implies that the isotropy of space is
preserved by f(r) about the origin.
5.3 Conservation of Angular Momentum under Central Force:
We now consider particle motion in a plane under central force.
So, the torque acting on it is
't rxF But F = f(r)r
rX[f(r)r] = f(r)(rxr)
But r x r = 0 :. i = 0
Since we have for effective angular momentum L
- -
_ dL dL -
't = dt :. dt = 0 =:> L = constant.
So for particle motion in a plane under conservative central force, its
angular momentum will remain conserved.
5.4 Conservation of energy under central force
Consider the motion of a particle under central force
F(r) = f(r) r along radial direction.
Now if Cr, 9) be the instanteneous plane polar co-ordinates of the particle
then the energy of the particle will be
E = Ek +Ep +r
2
e
2
)+Ep
r
where Ep = - f f(r) dr = Potential energy.
So the expression for total energy will be
E =
Now for our radial central force having no transverse component,
IF(r)1 = mIr
o = mfe
where, f,. and fa are respectively radial and cross radial components of the
acceleration.
m(r - re
Z
) = f(r)
m(re + 2iB) = 0
... Z"
But from equation (2) re + 2re = 0 => r 8 + 2rre = 0
we have
; (rZe)
o
rZe constant = h (say).
This is an outcome for conservative central force.
So, we have from equation (1)
. f(r)
;: _re
2
= --
m
1 (.2 h
2
) f
"2 m r + ? = f(r )dr + constant
... (1)
... (2)
Central Force
=> ~ m(,:2 + r
2
(
2
) + {-f j(r) dr} = constant
=> E = constant.
This gives energy conservation under central force.
5.5 Equation of motion under attractive central force
115
For attractive central force, we have the radial form of central force
F(r) = - j(r)r which gives the equations of motion for particles motion under
such central force as
m(r _re
2
)
m(,B+ ue)
- f(r)
o
Then equation (2) gives the constant of motion
r
2
e = Constant = h
... (1)
... (2)
But to simplify the form of 1 sl equation, let us now make a change in
1
variable as, r = - .
U
Also,
But
Also,
,:
dr
dt
1 .
--u
u
2
dr du e 1 e du
du . de . = --;; de
~ = h U 2
r2
d (dr) _ d ( h dU)
dt dt dt de
So we have from equation (1),
- ~ ~
m
This is equation of motion under attractive central force.
5.6 Application of central force theory to gravitation: Deduction of
Keplar's law
Actually, the gravitational force between two mass points m] and m
2
obeys
inverse square law of force and in that case the force is attractive central force
in nature.
So the gravitational central force is
F{r) =
Gm]m2
where, f(r) = 2
r
Now for planetory motion under such attractive central force for a planet
of mass m arround the sun of mass M, the equation of motion will be
d
2
u 1 ~
--+u = +---
de
2
mh
2
u
2
1
where, u = - ,(r, e) = position of planet
r
and
2
'j 2
re=eu=h
So, for planetory motion under gravitational force, the equation of motion
should be,
d
2
u
GMmu
2
GM
--+u +
mh
2
u
2
=+--=+f.!
de
2
h
2
GM
where,
f.!
-2- = Constant
h
d
2
u'
..
--+u'
0 ... (1)
de
2
for u' u - f.!.
We can now apply this equation of motion (1) to deduce Keplar's law of
planetory motion.
For planetory motion, Keplar gave three basic laws which are
Central Force 117
(i) Law-I : Every planet will rotate arround sun in elliptic orbit with
sun at any of its focie.
(ii) Law-II : For orbital motion of planet, its areal velocity will remain
constant.
(iii) Law-III : For revolution of planet around sun, the square of time
period of revolution will be proportional to the cube if semi major
axis of that elliptic orbit.
We now obtain the solution of equation (1),
u' A cos e => u - 11 = A cos e. (A = Const.)
u 11 + A cos e
GM
=> -2-+Acose
r h
r
1+( ~ c o s e
I = 1 + E cos e
... (2)
r
This is polar equation of the planet orbit where 'l' is semi latus rectum
and E is eccentricity of the orbit.
It is now obvious that the eccentricity
E =
Ah
2
--<1
GM
for GM ~ 10+
19
SI unit
So the orbit is elliptic in nature
and hence 1 st law of Keplar is now
established.
But for planetory motion, the area
swept out by the planet w.r.t. sun'S'
1
in time is dA = -r.{rde)
2
1 2
dA=-r de
2
s
o
So the areal velocity for such planetory motion is now given by
dA 1 2 de 1 2 1
v = -=-r -=-r e=-h=Constant
A dt 2 dt 2 2
P,
Thus 2
nd
law of Keplar is also established. Since, the orbit of planet is
elliptic, if its semi major axis and semi minor axis be 'a' and 'b' respectively
then,
b
2
h
2
Semi latus rectum, I = -;; = GM .
[From equation (2)]
b
2
=ah
2
/GM
So the time period of revolution of the planet along elliptic orbit is
area of ellipse nab
T =
h/2 area of velocity
2
T2
4n a
2
b
2
h
2
T2
4n
2
2 ah
2
4n
2
3
=>
--a .-- = --a
h
2
GM GM
4n
2
T2 =
__ .a
3
=> T2 ex. a
3
GM
i.e. the square of time period will be proportional to cube of semi major axis
of the elliptic orbit. Thus Keplar' s 3
rd
law is established.
5.7 Energy conservation for planetory motion
For planetory motion, the total energy is now given by
E Ek + Ep
But (m = mass of planet)
Now in presence of gravitational force field offered by sun to the planet,
the potential energy is
Now considering
we have,
W = + S' GMm dr = _ GMm
r2 r
00
E
1 (.2 2e 2
) GMm
-mr +r ---
2 r
1 e
r = - and r
2
e = - = h
u u
2
lIE
U = - = - + - cos e
r I I
Where, I is semi latus rectum of the planetory orbit and E is eccentricity.
1 E .
--r --sinee
..
r2
I
=>
r
(.: r
2
e = h)
I
Central Force 119
(
2 2) 1 E 2. 2 h GMm
E - -m -h SIn 8 + - ---
- 2 P r2 r
{
2 2 ( )2} ( )
1 E'"h.2 21 E 1 E
E = -m --SIn 8+h -+-cos8 -GMm -+-cos8
2 P I I I I
E
l E'" h . 2 8 h 2 Eh cos h E 2 8
= -m --SIn +-+ +--cos
{
22 2 2822 }
2 P P P P
GMm
---(1+ E cos8)
I
= -m --+-(1+2 Ecos8) ---(1+ Ecos8)
1 {E2 h
2
h
2
} GMm
2 P P I
= -m - 1+ E +--cos8 ------Ecos8
1 {h2 ( 2) 2h2 E } GMm GMm
2 P P I I
But taking our previous notations
GMm
and
= --2 - GM =
h
h
2
1
I = GM = ;- , we can write down
GMmE
I
and then we have, the total energy of planet is
mh
2
( 2) mh
2
E mh2 E 2
E = -- 1+ E'" +--cos8------cos 8
2P PIP
mh
2
( 2)
E= 2P l+E --1-
1
But also, = I
mh
2
mh
2
E=
2P P
mh
2
E = = constant (E < 1)
2P
Thus energy conservation holds for planetory motion. This energy is also
-ve which implies that the planetory system is a bound system.
5.8 Stability of Orbits
The orbit is the path of an object moving under central force. This orbit is
taken as stable orbit if after giving small disturbance to the radial co-ordinate
keeping energy or angular momentum unchanged, the orbit will be disturbed
small.
For effective potential energy for central force, ~ r ) = vefr<r) the condition
o2v
e
ff(r)
for equilibrium, 2 > 0
or
[
02
Vefr
(r) 1
and for particular r = r
o
' or2 '='0 = 0
For central force, potential energy v( r) = a T'+
1
, a = const.
and centrifugal potential energy verI r) = ~ , b = const. > 0
r
S
'l'b . ~ ov(r) C 'f I ~
lllce at eqm 1 num, lorce Tr = entn uga lorce
.. (n + 1) aT' = 12b r-
3
1
=> n + 1) a = + 2b rn - 3
Now,
2b -(n+3)
a = --r
n+1
v(r)
v -(n+3) n+1 2b_
2
= --r r = --r
n+ 1 n+ 1
[
OVeff ]
-- -4-4(
or2 r=ro = 6bro = 2bro 3 + n)
Now for any circular orbit with r = ro the stability condition is
o2V
e
ff( v)
2 >0
or
n + 3 > 0 => n > -3.
and for circular orbit n = -3.
SUMMARY
1. Central force has the standard form F(r) = j(r); .
Central Force 121
2. This force can be derivable from a scal:!r potential energy =
i.e., F = - \7lj)(r)
3. Magnitude of central force f( r) could be uniquely represented as a power
in r is,
+C()
f(r) = I knr-n
n=-C()
4. Angular momentum will be conserved for planner motion of particle under
central force.
5. Only radial dependence of force preserves the isotropy of space.
6. Conservative nature of force restricts the time independence of the force.
7. Energy conservation holds for conservative central force.
S. The equation of motion of particle under attractive central force is
d2u
--+u = ---
de
2
mh
2
u
2
1 2"j 2
where, u = -, h = r e = e u .
r
9. The equation of planetory orbit under central force is
I
- = 1 + e cos e
r
h
2
Ah
2
where, I = --, e = --; A = constant
GM GM
and e < 1
10. The areal velocity of planet under gravitational central force is
I 2'
V
- - r e = h/2 = constant
A - 2 .
11. For planetory motion, T2 a a
3
, T = Time period. a = Semi major axis of
planetory elliptic orbit
mh
2
12. Total energy of planet E = - -2-(1- e
2
) = const ( - ve) .
21
13. For stability of orbit under effective central force potential a 1"'+1, n > -3
and for circular orbit, n = -3.
Worked out Examples
e
1. A particle describes the parabola r = asec
2
'2 such that the cross radial
d
2
r
component of the velocity is constant. Show that -2 is constant.
dt
Ans : Here for the cross radial component, r
2
e = constant = k (say)
But
..
o
r = asec
2
-
2
i-
F
d
2
r
dt
2
o 0 0 . 1
a.2 sec- sec - tan -0.-
2 2 2 2
asec
2
= r8
2 2 2
k
2
2
d r =
dt
2
2 2 2 a
k
2
- = constant
2a
2. A particle describes a circular orbit given by r = 2a cos 0 under the
influence of an attractive central force directed towards a point on the circle.
Show that the force varies as the inverse fifth power of the distance.
ADS : Here the equation of the circle r = 2a cos 0
1 secO
Putting r = - we have u = --
u 2a
du secO tanO
dO 2a
d
2
u sec
2
0+secOtan
2
0
d0
2
2a
Central Force
f(r) =
(
' 1
:, f r)a
s
r
i. e., the force varies as the inverse fifth power of the distance.
3. A particle of mass m moves in a central force field by F:;,; - r
r
4
123
Show that if E be the total energy supplied to the particle, then its speed is
given by
1
v =
(
_k + 2E)2
mr2 m
ADs: Here, F
IFI

r3
v(r)

2r2
So the energy will be E Ek + vCr)
E
v =
.!. m(r2 + r2{)2) _
2 2r2
12k
-my --
2 2r2
I
e:+ m:
2
Y
So the central forc.e is a conservative force.
ADs: We know that if a force F be conservative, then V x F = 0
or,
fF.dr = 0
c
Now for central force, F f(r)!'"
r
V x F vx( V x[ (xi + yJ +Zk)]
124
But
i j k
0 0 0
-
ox By oz
x
--j(r)
r
l.. f(r)
r
:.. f(r)
r
= of _l.. of) + of _:.. 8f)+k(l.. of 81)
rBy roz roz rox rox rBy
of of or x of
- = ---=--
ox or ox r or
of _ l.. of of =:.. of
By - ror' oz ror
VXF =
= i 1. of (zy - yz) + j 1. of (xz - zx) + k 1. of (yx - xy)
r or r or r or
=0
:. The central force is conservative
EXERCISES
1. What is central force? Give its characteristics.
2. Consider the motion of a particle under central force _.!.. Show that its
2
areal velocity is constant. r
3. Find the differential equation of orbit for motion of particle under central
force.
4. Find the equivalent one dimensional problem for the motion of particle
under central force.
S. Discuss Keplar's problem under inverse square law of force.
6. Discuss the condition for stability of orbit.
7. A particle of mass m moves under the action of central force whose potential
v(r) = kmrZ (r> 0), then,
(i) For what energy and angular momentum, will the orbit become circle
of radius 'a' about the origin.
(ii) Calculate the period for circular motion.
(iii) If the problem be slightly disturbed from circular. motion what will
be the period for small radial oscillation about r = a.
8. Show that the gravitational force is conservative.
9. A particle moves in a plane under the force F(r) = ar, where r is the
distance from the center of force. Find the path of the particle.
-:0:-
Chapter-6
Theory of Collision
6.1 Introduction
Collision is a process through which two moving particle or one moving
and one rest particle comes in contact and also transfer their momentum
mutually keeping total momentum conserve. For two elastic bodies, this collision
is commonly known as 'impact'. Since, u ~ i n g collision between two bodies,
no external force acts on the system from outside, the total momentum of the
system will remain conserve through the impact.
Basically, during collision or impact between two elastic bodies, the property
of elasticity plays its role and thus the time* during which the deformation
(slight) in shape takes place through collision, is called 'resistitution time' and
within this time the shape is restored.
During collision between two objects, say two spherical bodies, they exerts
mutual forces to each other along the common normal to their point of contact
and this common normal is called 'line of impact'. This line is the line joining
the centers of two impinging spheres. For such line of impact, we can classify
wheather the collision is direct or oblique collision because, collision takes place
such that the direction of motion of both colliding objects before and after
collision be along the line of impact, then it is direct collision. Otherwise, for
oblique collision, the direction of motion is not along the line of impact. But
for any collision, particle comes in closer and they exchange their momentum
with each other keeping total momentum constant.
6.2 Characteristics of Collision
Newton's Collision Rule
Any type of collision has the following basic characters which are-
(j) Momentum Conservation takes place during collision.
(ii) Kinetic energy (total) may be conserved or energy loss may occur
during collision.
125
(iii) Momentum transfer takes place during collision and through such
transfer impulsive action takes place during collision.
(iv) Collision may be perfectly elastic, or perfectly inelastic energy
conservation takes place along with momentum conservation. For
perfectly semielastic collision, kinetic energy loss occurs and the
particle moves together after collison, and for semi elastic collision,
although energy loss takes place, both particles moves separately after
collision.
During any collision, either direct or oblique collision, Newton's collision
rule is that the magnitude of relative velocity of two colliding object after
collision will be proportional to that before collision.
This constant of proportionality is called coefficient of resistitution* (e)
and mathematically it is given by
where, UI and U2 are respective velocities of two particles before collision and
VI and V2 are that after collision.
Here, by conservation principle, e 1
and, for e = 1, collision is said to be elastic, for e = 0, collision is inelastic.
and, for < e < 1, collision is semielastic.
So by this rule, !VI - V2! = e!ul -U2!' and this is in general true for oblique
collision where as for direct collision.
If particles moves in the same direction before and after collision, (VI - v
2
)
= e (u
2
- u
I
) and if they move opposite to each other before and after collision,
then (VI + v
2
) = e (u
l
+ u
2
).
6.3 Lab Frame and Center of Mass Frame :
In general, the collision between two
particles is oblique collision where both moving
particles approaches to each other in a non linear
manner with respect to their line of impact. After
collision, they move away from each other in
m,

u,

v,
m,
"GENERAL CASE"
e = VI - V2 = 1
* For iii' u2' Vl> lI2 in the same direction. m hour for elastic collision
u2 -ul
Because, ml(u
l
- v
2
) = m
2
(v
2
--v
2
) (Conservation of mometum)
(
2 2) (2 2) (2 2) .
ml UI - V) ",""ml U) - VI = m2 U2 - U2 (Conservation of energy)
Theory of Collision
m,
~
U,
127
m,
the same fashion. So in general case,
collision takes place in a plane and
obviously, the direct or linear collision
is a special case of oblique collision in
which particles moves along their line
of impact before and after collision.
----+------..---------------
Experiments are often carried out
in which one of the particles is at rest
in the laboratory and the other
approaches it and collision takes place.
Such a set up in which a particle
collides with another particle at rest is
called the laboratory frame of reference.
But the discussion of collision between
two particles becomes very simple and
symmetric for motions of particles before
and after collision which are taken such that
the center of mass of the system be at rest
throughout the collision. Such setup is thus
called center of mass frame, and obviously
in this system the collision between two
~
U,=o
"LABORATORY FRAME"
~ v : ~
m, u, /-;t
_ ..... ~ . _____ ' U,
- . .
x / m,
ym,
"CENTER OF MASS FRAME"
particles is treated as they have equal and opposite momenta initially.
6.4 Direct or Linear Collision
Consider the collision between two
particles when they are in motion along
their line of impact before and after
collision. So it is direct or linear
collision.
m, t: m, u:
- - - - - - - - . - ~ - - - - - - - . - ~ - - - - - -
~
R
m, m,
~
F
- - - - - - - - - ~ - - - - - ~ - - - - - - - - - - - - - - -
If m\ and m
2
be the respective
masses of two impinging particles
~ ~
m, v, m, v,
having initial velocities u\ and U2 - - - - - - - - - - - - - - - - - - - - . - ~ - - - - . - ~
before collision (in the same direction) and final velocities 1\ and v2 after
collision (in the same direction along the same line of impact) then for such
direct or head on collision,
we have from momentum conservation,
m\u\ + m
2
u
2
= m\v\ + m
2
v
2
... (1)
But since for u\ > u
2
' v
2
> vi' we have from Newton's collision rule,
v
2
- v\ = e(u\ - u
2
) ... (2)
Multiplying equation (2) by m
2
and then substracting from (1) we get,
m\v\ + m
2
v\ m\u\ + m
2
u
2
- em
2
u\ + em
2
u
2

=> v\(m\ + m
2
) = u\(m\ - em
2
) + m
2
u
2
(1 + e)
v =
I
1 [(ml-
em
2)ul +(1 + e)m2u2]
ml +m2
... (3)
Similarly, multiplying equation (2) m
l
and then adding with (1), we get,
(m
l
+ m
2
) v
2
m
l
u
l
(1 + e) + (m2 - em
l
)u
2
v
2
= 1 [(m2 - e
m
d
u
2 + (1 + e)mlud
nil +m2 .
... (4)
This two equations (3) and (4) respectively gives the velocities VI and v
2
of two impinging particles after direct impact.
For this general direct impact, if we want to find out the loss of kinetic
energy due to impact, then
We have
~ = loss in K.E
~
-[(mIVI + m2
v
2)2 + m\m2(vl - V2)2J}
1 1 [ 2 2 2J
( ) mjm2(uIU2) -m\m2(UI-u2) e
2 m\ +m2
[Useing equation (2) and eauation (1)1
1 mlm2 ( 2)( )2
-' l-e U\ -U2
2 ml +m2
... (5)
This is the general expression for loss in K.E for direct collision and we
should have,
for perfectly in elastic collision, e = 0
nllUI + f1'12
U
2 1 mlm2 ( )2
V = = V2, and loss, ~ = -- --- Uj - u2 and this two
I nil + m2 2 nil + nil
Theory of Collision 129
particles will move together after collision also, for perfectly elastic collision,
e=l
then,
VI ( 1 )[(ml-m2)ul+2m2u2]
m
1
+m2
v
2
= ( 1 ) [2m)u) + (m2 - md
u
2]
m) +mz
and then the loss, ilE = O.
i.e. for elastic collision, particles will move separately along the same line
of impact after collision without any loss in kinetic energy.
6.5 Characteristic of Direct Collision
We have from our previous discussion that for direct collision between two
bodies moving in the same direction before and after collision along their line
of impact,
the velocities after collision.
V =
I
V =
2
Where symbols has their usual meaning and also, the loss in K.E for such
collision. ilE =..!.. m)mZ (l-eZ)(u) -uZ)2. (O:S; e:S; 1)
2 m) +mz
With this results, we can have the following characteristics for direct impact,
which are,
(i) For m) = m
2
, VI = u
2
' v
2
= u
l
for e = 1 i.e. for elastic collision
between two bodies having same mass, the velocity exchange takes
place between them.
(ii) For elastic collision, e = 1, ilE = 0
i.e. no loss in K.E occurs for elastic collision between two particles.
(iii) For elastic collision, e = 1, if u
2
= 0 i.e. if 2
nd
particle is at rest
initially, then
V =
I
1- mz
m) -mz m)
u) =---u)
m) +mz 1+ mz
m)
(a) for m) = m
2
VI = 0, v
2
= u)
2m)u) 2u)
and Vz = ---'-...!....- = ---
m) +mz l+_m_z
m)
i.e. the 1 st body is stopped and the 2
nd
one takes off with the velocity
which the 1 st one originally had. Both the momentum and the kinetic
130
energy of the first are completely transferred to the 2nd.
(b) for m
2
mp VI ::::; - u
l
and v
2
::::; o.
i.e. when a light body collides with a much heavier body at rest, the
velocity of light body is approximately reversed and the heavier body
remains approximately at rest. viz, a ball dropped on earth rebounds
with reverse velocity when the collision is elastic.
(c) for m
2
mp vI ::::; u
l
and u
2
::::; 2u
I
i.e. when heavy body collide elastically which remains practically
unchanged, but the light body rebounds approximately with twice the
velocity of the heavy body. By these considerations, we can explain
the 'slow down' of the fast nutrons in a nuclear reactor, taking proton
of hydrogeneous material (paraffin) has nearly the same mass as that
of a neutron in a good moderator.
6.6 Maximum Energy transfer due to head on elastic collision
Consider a particle of mass m
l
which collides with another mass m
2
, initially
at rest. For such head on collision, if the initial velocity u
l
of mass m
l
changes
to the velocity VI then, for u
2
= o.
IS
We have,
and the fractional loss in kinetic energy of the mass m
l
for such collision
Now let
L\E
E
4x
Here the transfer in energy will be maximum if
I 2
E
f
= "2mlVI =0
is L\E = I 4x = I
E => (l+x)2
Theory of Collision 131
This is the condition for maximum energy transfer for head on collision.
6.7 Oblique Collision
Here we consider the oblique
collision of two masses m
l
and
m
2
with initial velocities
UI and U2 before collision at
resp. angles (II and (I2 with line
of impact and the velocities after
collision are vI and v2 with the
same line of impact.
Now from momentum conservation along the horizontal line of impact,
we get,
mlu
l
cos (II + m
2
u
2
cos (I2 = miv
i
cos PI + m
2
v
2
cos P
2
... (1)
But since the transverse components of momenta remains unchanged, we
have,
mlu
l
sin (II = miv
i
sin PI; m
2
u
2
sin (I2 = m
2
v
2
sin P
2
Also, from Newton's collision rule,
(v2 cos P
2
- vI cos PI) = e(u
l
cos (II - u
2
cos (I2)
By solving these equations, we now get,
(ml - em2)uI COS(II + m2(1 + e)u2 COS(I2
vI cos PI = (ml + m2)
(ml - em2)uI COS(I2 + m2(1 + e)u2 COS(I2
or,
vI (ml + +m2)cosPI
and,
(m2 - emdu2 COS(I2 + ml(l+e)uI COS(II
v =
2 (ml + m2)cosP2
... (2)
... (3)
... (4)
... (5)
So for such oblique impact, if we want to obtain the loss of kinetic energy,
we get
1 [{ 2( 2 . 2) 2( 2 . 2 )}
= 2" mlul cos (II + sm (II + mlu2 cos (I2 + sm (I2
{ml V?( cos
2
PI + sin
2
PI) + m 2 V ~ cos
2
P2 + sin
2
P2)} ]
1 [ 2 2 2 2 ( 2 2 A 2 2 A )]
~ E = 2" mlul cos (II + m2
u
2 cos (I2 - mivi cos 1-'1 + m2
v
2 cos 1-'2
(Using equation (2))
1 [22 2 22 2
= ( ) mlul cos UI +m2u2 cos u2
2 ml +m2
+mlm2(ufcoS
2
UI +uiCOS2(2)
+mlm2(vf cos
2
+ vi cos
2

=
=
E= (m
l
m
2
)(1-e2)(UICOSUI-U2COSU2)2
2 ml +m2
This is the energy loss for oblique impact.
So for perfect inelastic oblique collision, this loss will be
1 mlm2 ( )2
ilE= UICOSUI-U2COSU2 (fore =0)
2 ml +m2
SUMMARY
1. For any collision, momentum conservation takes place.
... (6)
2. The ratio of magnitude of relative velocity after collision to that before
collision is called coefficient of resistitution (e) which is 1 or 0 or 0 < e < 1
according as the collision is elastic, inelastic or sernielastic.
3. For any collision between two particles having initial velocities iii and ii2
before collision and vI and v2 after collision,
liil -ii21 = IV2 - vII
4. For direct collision, we have from momentum conservation
mlii
l
+ m2
ii
2 = ml VI + m2 v2 .
5. The velocities of particles after direct collision; are
Theory of Collision
6. The loss in kinetic energy for direct collision,
1 mjm2 ( 2)( )2
llE = 1- e Uj - U2 .
2 mj +m2
7. For u
2
= 0, Vj = 0, v
2
= u
j
for m
l
= m
2
vI ~ -u
I
, v
2
= for m
2
mj
vI ~ up V
z
2u
I
for m
2
m
l
8. For U
z
= 0, the maximum energy transfer takes place for m
l
= m
2
9. For oblique collision,
mlu
l
cos a
l
+ m
2
u
Z
cos a
z
= miv
i
cos ~ I + m
2
v
2
cos ~ z
(u
2
cos ~ 2 - vI cos ~ I = e (u
l
cos a
l
- U
z
cos a
z
)
when,
(mj - em2)Uj cosaj - U2 cosa2
vI cos ~ I = (mj + m2)
(m2 - emdU2 cosa2 + (1 + e)mjUj cosaj
V2 cos ~ 2 = (mj + m2)
Energy loss for oblique collision
llE = M mj
m
2/(mj + m2)](1- e
2
)(u2 cosa2 - Uj cosaj)2
Worked Out Examples
133
1. Two masses 5 gm and 10 gm approaches to each other with respectively
velocity 5 rnIs and 10 rnIsec. They makes head on collision and after collision,
the 1 st particle comes at rest. What will be the velocity of 2
nd
particle after
collision.
ADS: Here, given that m
l
= 5 gm, m
2
= 10 gm
u
l
= 5 rnIsec u
2
= -10 rnIsec
After collision, Vj = so from momentum conservation
mju
l
+ m
2
v
2
= miv
i
+ m
2
v
2
=> 5 x 5 - 10 x 10 = 5 x + 10 v
2
75
.. vI = -10 = -7.5 rnIsec.
So, after collision the 2
nd
particle moves with velocity 7.5 rnIsec in its
original direction.
2. A particle collides elastically with another particle of same mass initially
at rest. So that after collision, they moves at right angles to each other.
134
Ans : Let both particles has the same mass 'm'.
Here, u
l
= U, u
2
= 0, e = 1
So from momentum conservation,
mu mVI +mv2
=:} U vI + V2
But, if after collision, this two particles be scattered at an angle a with
each other, then
I
_ IV2+ vll
e - 1 = = -'-----------'--
U U
2+22 -u
2
vI v2 - vlv2 cosa -
Also, for elastic collision, we have from energy conServation
I 2
-mu
2
I 2 I 2
-mvl +-mv
2
2 2
... (1)
v? + u
2
... (2)
So from this two equations, u
2
- 2vI v
2
cos a = u
2
2v
l
v
2
cos a =
or, cos a = 0, a = '!t/2
i.e. they will move at right angle to each other after collision.
3. Show that if two equal masses collides elastically, they will exchange
velocities.
Ans : Let, both has same mass m and they has respective velocities u
l
and
u
2
before collision and has velocities vI and v
2
after collision.
so from momentum conservation,
mU
I
+ mU
2
= mV
I
+ mV
2
u
l
+ u
2
= vI + v
2
Again for elastic collision, e = I
. . v
2
- vI = u
l
- v
2
By solving this two equations (1) and (2), we get
vI = u
2
' v
2
= u
l
So they exchange velocity through collision.
... (1)
... (2)
4. Show that for perfectly elastic collision of particles, the total kinetic energy
will remain conserved.
Ans : Let m
l
and m
2
are two masses and u
l
and u
2
are the respective
velocities before collision and VI and v
2
are that after collision.
So for elastic collision, (taken to be direct or head on collision)
v
2
- vI = u
l
- u
2
... (2)(,' e = 1)
and from momentum conservation,
... (2)
Theory of Collision
we can have,
(ml - m2 )UI + 2m2u2
ml +m2
v =
2
ml +m2
so the total kinetic energy after collision,
1 2 1 2
E = -mivi +-m2
v
2
'f 2 2
1 2 1 2
= -mlul +-m2u
2
2 2
= E
j
= Total kinetic energy before collision.
135
5. Consider the direct collision of two particles of mass m
l
and m
2
having
initial velocities U I and u
2
before collision. If e be the coefficient of resistitution,
then show that the momentum which is transferred from the 1 st particle to the
2
nd
particle is
mlm2 (l+e)(ul -U2)
ml +m2
Ans : For this direct collision,
mlu
l
+ m
2
u
2
= mlv} + m
2
v
2
... (1)
and v
2
- VI = e(u
l
- u
2
) ... (2)
Where VI and v
2
are their respective velocities after collision. So the change
in momentum due to such impact is
I'!.p = m
2
v
2
- m
2
u
2
= m
2
(v
2
- u
2
)
[Because by solving equation (1) and (2), we get
m}(l+e) (m2 -em
l
)
v = u\ + U
2
2 (m\ + m
2
) (m
l
+ m
2
)
(m\ -em
2
) m2(1+e)
v = ~ U
2
I (m\ + m2) (ml + m
2
)
So, this momentum transfer
:. f'o..p = mlm2 (1 + e)(ul - U2)
ml +m2
6. A ball which is dropped from a height H onto a floor rebounds to a
height h < H. Determine the coefficient of resistitution
Ans : If the particle after falling from height H, collides with the floor
with velocity u and rebounds with instantaneous velocity v after collision then
v
the cofficient of resistitution, e = -
u
Here u
2
= 2gH, y2 = 2gh
e = ~
7. A billiard ball strikes another identical billiard ball obliquely at an angle
45 with their line of centers at the time of impact. If the coefficient of
1
resistitution is "2' find the angle at which the 1 sl ball will 'bounce off'.
Ans : For such oblique impact,
initially, u
l
= U, a
l
= a = 45.
u
2
= 0
If vI and v
2
be their respective velocities after impact then for their angles
13
1
and 13
2
with the line of impact,
mu cos a = mV
I
cos 13
1
+ mV
2
cos 13
2
U
vI cos 13
1
+ v
2
cos 13
2
= U cos 45 = ..fi
Again, v
2
cos 13
2
- vI cos 13
1
= e(u cos 45) = .!.u. ~
2 v2
u
=2J2
2vI cos 13
1
= ~ 1 - ~ ) = 2:n
Again for normal momentum about the line of impact,
mu sin a = mV
I
sin 13
1
... (1)
Theory of Collision
U
vI sin PI = u sin 45 = fi
tan PI = 2. => PI = tan-
I
(2)
137
... (2)
This is the angle at which the 1 sl ball will 'bounce off after collision.
EXERCISES
1. What are the basic characteristics of collision?
2. State and explain Newton's collision rule.
3. Show that for inelastic direct collision between two particles, the loss in
1
-mlm2
kinetic energy is ~ E = 2 (VI - V2)2 .
ml +m2
4. Find the velocities of two particles after oblique collision between this two
particles. Also find loss in kinetic energy in this regard.
5. Discuss collision between two particles in center of mass frame.
6. A gun fires a bullet of mass m with horizontal velocity v into a block of
mass m which rests on a horizontal frictionless plane. If the bullet becomes
embedded in the block, find the loss in kinetic energy in this impact.
7. Repeat the previous problem (6) if the block is moving away from the
gun with velocity vO'
8. A mass m
l
travelling with speed v on a horizontal plane hits another mass
m
2
which is at rest. If the coefficient of resistitution is E, prove that there
mlm2(1- ~ v 2
is a loss of kinetic energy equal to 2( )
ml +m2
9. A ball is dropped from a height' h' above a horizontal plane on to an
inclined plane or angle a which is resting on the horizontal plane. Prove
that if the cofficient of resistitution is e, then the ball will next hit the
inclined at a point which is at a distance 4e(1 + e) sin a below the original
point of impact.
-:0:-
Chapter-7
Conservation Principle and
Constrained Motion
7.1 Characteristics of Conservation Principle
In classical mechanics, we deal several problems with motion of a single
particle or a system of particles. For such motion, we have to face a term
'conservation' many times. The conservation of a parameter, involve with the
motion, is basically the constantness of that parameter with respect to time.
Clearly, the parameter which is a conserved quantity, will not change with time
throughout. We also observe that the parameter conservation demands one or
more than one conditions which are basically constrained of motion and these
conditions restricts the conservation of that parameter.
With the dealing with several conserved parameter, we can now say that,
(i) Conservation principles are independent on the trajectories of the
particles or system, but it depends on some restrictions.
(ii) Conservation of any parameter means its constantness and all
conserved quantities are constant of motion.
(iii) Conservation laws have an intimate relation with invariance. Their
failure in certain cases may result in the discovery of new and not
yet understood phenomenon.
(iv) From conservation law we can find equation of motion even when
the effective force action on the system is not totally known.
(v) Conservation laws assure us many times that some aspects of motion
are impossible and must be left out.
(vi) Conservation phenomenon tells us a great deal about the motion even
if forces affecting motion are not known in advance.
In our physical world, there exist a no. of conservation principles or laws,
some exact and some approximate. These principles are related with energy,
linear momentum, angular momentum change ... etc. and various other quantities.
We will discuss all such conservations in our present chapter which are very
138
Conservation Principle and Constrained Motion 139
powerful tools in our present chapter which are very powerful tools in solving
several mechanical problems.
7.2 Mechanics of a single particle and system of particles
The motion of a single particle or a system of particles can primarily be
studied by applying Newton's 2
nd
law of motion which estimate the effective
force acting on that particle or system. To study Newtonian mechanics for the
motion of particle, it is mandatory to realize the nature and interaction of force
on it. To get a sharp knowledge about this for the motion of particle or for the
motion of system of particles, we should have the following discussion-
For a single particle motion, we have from Newton's 2
nd
law, the applied
. . - djJ d _ dV d
2
;
force or the effective force 1S F = - = -(my) = m-
d
= m---:x
dt dt t dt
Where, p is instantaneous momentum of the particle 'm' is its mass and
v is instantaneous velocity which has magnitude much less than that of light
wave in vacuum. So motion is restricted in the region of Newtonian mechanics
which is far apart from relativistic region. Here, ; , is instantaneous position
of particle during its motion w.r.t some stationary point.
But if we now consider the system ofN particles, then from Newton's law,
- d
2
;, -(e) " -
the effective force on ith particle, F
j
= mj -+ = F
j
+ Fij .
dt j*j
Where, is instantaneous position of ith particle having mass 'm;'. Here,
F}e) is external force acting on ith particle and L Fij is the sum of internal
forces on ith particle by all Jth particles (j #- i). So for any constituent of the
system, the effective force is the resultant of external force and all possible
internal forces given by the neighbours particles.
But here considering all particles together for the whole system, the net
force,
F = LFj = LFj(e) + LLFij
j j*i
and F = LFj(e) + L LFij = Lmj = d:
j j (j*j) i dt dt j
- -
But we must have, Fij = -Fji
- 1 L(- -)
""p.. = - P.+F =0
lj 2 I} Jl
j j ij
i. e., the net internal force for motion of all particles in the system is zero,
and as a result, for motion of a system of particles,
F
IF/e) = d
2
2
lIm,r;].
j dt j
If we now take the average of all radius vectors for all constituents particles,
then we get the instantaneous position of center of mass of the system,
Mrc where, M = I mj = Total mass,
F ~ F / e =M( ~ ]
Which is a single particle equation. So the motion of a system of particles
can in general be replaced by the motion of the center of mass of that system
under the net force which is the sum of all external forces, applied on that system.
7.3 Conservation of linear momentum
The principle of linear momentum conservation states that if the net effective
force on a single particle or system of particles be zero then the momentum of
that particle or the vector sum of momentum of all particles of the system will
be constant or will remain invariant throughout the motion of that particle or
that system.
This conservation can easily be understood from Newton's law and for single
particle motion,
_ dV d ( _)
F = m- = - mv
dt dt
So for F = 0, mv = constant or conserved.
Also, for system of particles,
"F-(e) _" d
2
r; _" dV
j
-!!...-" -
F = L.J j - L.J mj dt2 - L.J mj dt - dt L.J m, Vj
I I I I
so for,
F
:, ( mi Vi) ~ 0 => L mi
V
' ~ Constant
This gives the conservation of linear momentum. So for effective force to
be zero for the motion, the net momentum will become a conserved quantity
or constant of motion.
7.4 Conservation of Angular Momentum
For a single particle rotation or for rigid body rotation if net torque acting
on the particle or the system be equal to zero then the angular momentum will
be conserved. This is angular momentum conservation. We can now discuss
such conservation principle for single particle motion or for motion of the system
of particles, in the following way-
For single particle rotation, the angular momentum is the moment of linear
momentum which is given by
L = F x Ii = F x (mv) = m(F x v)
So the torque acting on it, is
and for
- F- - djJ d (- -)
t = r x = r x dt = dt r x p
-
t
dL
dt
0,
(
dF - fi dF -)
.:- x p= 0, or-lip
dt dt
dL
- = 0 L = Constant
dt '
This is angular momentum conservation for single particle rotation. Also
for system of particles, the torque on ith particle.
- .r.::. - d( -) - -
r.. x "p. = r. x - mv = r.. x F
I I I dt I I I I
But,
So,
But we have, for all constituent particles in the system
L ~ x Fij
I#i
~ ~ ~ x Fij +Fj x Fji)
J"#l
Here, ~ - F
j
= ~ j and this IFijl gives the separation between ith and jth
particle. Since, the internal force between ith and jth particle Fij acts along
:. - 0) x Fij = 0 and the total internal torque x Fij = 0

So the effective torque on ith particle is effectively the torque for external
force F(e) acting on that particle,
I
- -(e)
Ie. x F
j
= x Fi
and the net torque on the whole system
1: =
"i:. x dPj x P')= dL
I dt dt I I dt
I I
(
d"l; _ . d"i_)
for, - x P = 0, SInce, _I Ilpi
dt dt
dL L = I("l; x Pj) = Constant
Now, for :r = 0, - = 0 ,
dt
This is conservation of angular momentum for system of particles by which
if net torque for all external forces acting on the system be zero, the total angular
momentum for the whole system will become constant.
7.5 Conservation of Energy
For a single particle motion or for motion of a system of particles, if
conservative force acts on that particle or on that system, then its total energy
is the sum of kinetic and potential energy will be constant.
Now a force F is said to be conservative if the close line integral of it
along a close path 'C' becomes zero. i.e., if F be conservative,
fF.dr = 0
c
and in that case we have from Stoke's theorem V x F = 0 the conservative
nature of force demands that this force can be expressed as a -ve gradient of
some scalar function, v(r), called scalar potential.
I.e. F = -Vv(r)
Under action of such conservative force, we have for a single particle system,
Conservation Principle and Constrained Motion
2 2 d-
Work done, W = f F.dr = f m .di
1 1
When particle goes from state 1 to state 2.
2 2 2
W f mv.av = f v) = f
1 1 1
1 2 1 2
- mv2 --mvl
2 2
143
But also, with respect to conservative force F , the work done in this case,
2 2
W f
F
.
dr
= - f VY.dr
I I
W
1 2 1 2
- mv2 --mvl
Y
I
- Y
2
and in this case, we finally have,
2 2
1 2 1 2
=
2 2
= (KE)I + (PE)I) (KE)2 + (PE)2
i.e. the sum of kinetic and potential energy will be constant in every state
of the particle. This is energy conservation.
Similarly, for N-particle system
the total potential energy
when,
Fint _Vjyint
I
Here, one thing is very clear that if the mutual interacting forces be
Newtonian, yIn! must be a function of the relative distance between two
interacting particles, i. e., for ith particle.

)
yint Iyr = + yr)
j
= -rjl)
i /F-i
1
Here clearly, the factor "2 comes because of the fact that while summing
the mutual potential energies, a pair of particle (i, j) appears twice, once for
ith due to Jth and other for jth due ith particle, where both are the same and
should appear only once.
So, for ith particle of the system,
But
p. = p(e) + "pi.nt
I I
fF-i
F
-(e) d- "F- d-
i . ri + ij' 1";
joF-i
-(e) d- "F- d-
F, . r
i
+ y' r
i

d(l
)
F
-(e)d- "F-d-
_ _ m .f..2 dt = i . ri + ij' 1";
dt 2 I I
Thus for whole system,
L = +
j dt 2 i i
But also, = +P
ji
.d0)
I rF-I i
= -0)
,
Conservation Principle and Constrained Motion
Also,
We then finally have,
d(""l .2] _ "V(V(e)) .. ( )
dt dt - - j j . d'i -2 'VI} v,; .dry
i ' , 1*'
" 1 .2 = ,,(e) 1"" '

. 2 ,2 , .,
I , , J*'
" ..!.m.f,2 + "v(e) +..!. ""v.1,nt
2 I' 2 y = Constant
, , , J*'
T+V
where, T
Constant
,,1 .2
total K.E = 2 miri
I
145
"V(e) l""V
int
and V = total potential energy = j + 2 ij
, , J
This is energy conservation for system of particles under conservative forces
acting on it.
7.6 Constrained Motion
Degree of Motion, Constraints: The motion of a particle or system of
particles under some conditions or restrictions is called constrained motion. For
much motion, we must have some knowledge about the degree of freedom and
constraints.
The minimum no. of independent variables which are required to specify
the motion of particle or system of particles as a whole, is called degree of
freedom for that particle or system. It can be shown frum mathematical version,
that if motion of a system ofN-particles are involve with 'm' no. of restrictions,
then degree of freedom of that system is d"" 3N - m. We will analyse the reason
behind it later on, but at this junction, what we want to say that if we take all
example of gas molecules, then for mono atomic molecule, N 1. m = 0,
d= 3.
For diatomic molecule, N = 2, m == I, d = 5 ... etc.
Now the mathematical restrictions which can be expressed through the
equations or inequations, are called constraints. So the constrained motion is
restricted by the 'so called' constraints. We can now consider some example of
constrained motion to clear these points.
Example 1 : Consider motion of a simple pendulum. If (x, y) be the cartesian
coordinates of the instantaneous position of bob and (r, 8) be the polar
coordinates of it, then the constraint is x
2
+ y2 = r2 = 12 where, r = I = effective
length.
So here the degree of freedom is 1 for independent variable x or y in
caltesian system or 8 in plane polar system.
Example 2 : Consider the motion of a particle which is confined on the
surface ofa sphere of radius 'r'.
So here, the constraint is x
2
+ y2 + z2 = r2
where, the degree of freedom is (3 -- 1) = 2 and independent variables are
(x, y) or (y, z) or (z, x) in cartesian system and (r, 8) in spherical polar system.
Example 3 : Consider the motion of a monatomic gas molecule within a
box of dimension (a x b x c). So here if (x, y, z) be the instantaneous position
of that molecule within box, then the constraints are,
o ::;; x ::;; a, 0 ::;; y ::;; b, 0 :5 z ::;; c.
and here, the degree of freedom = (3 - 0) = 3 where, the independent
coordinates are (x, y, z), in cartesian coordinate system, or (r, 8, <p) in spherical
polar system.
Now the constraints which are involve with the motion of particle or the
system of particles, can be classified into two types-One is called 'holonomic'
constraint and another is called 'non-holonomic' constraint.
Holonomic* constraints means those constraints which are independent of
velocity can be expressed in the form of equation, otherwise the constraint will
be non-holonomic. Also any constraint may be explicit function of time or not.
If it is not explicit function of time, then this constraint is called scleronomic
constraint, otherwise, the constraint is rheonomic constraint which is explicit
function of time as for example, in case of motion of a simple pendulum, the
constraint is scleronomic type holonomic constraint. But if the point of
suspension of the pendulum moves horizontally with some velocity v, then for
instantaneous position (x', y') of the hob, the constraint will be x,2 + y'2 = P
(x - vt)2 + y2 = P.
and this constraint is rheonomic type holonomic. Similarly, if we take
another example for rigid body motion, the constraint will be
I
-I 1- -I
r .. = r: - r'l = Constant
IJ I I J
This is the restriction for rigid body motion for which the separation between
any two constituent particles within rigid body will be constant during rigid
body motion. Clearly, such constraint is holonomic and scleronomous constraint.
7.7 Generalised Co-ordinates and other Generalised Parameters
The independent coordinates which are involve with the motion of a single
* Constraints are bilateral which can be expressed in equations and non holonomic
constramts are unilateral whIch can not be expressed in equations.
particle or system of particles are commonly known as generalised coordinates.
So it is a fact that for a given system, the no. of generalised cordinates gives
the degree of freedom (d). These generalized coordinates are in general
represented by q/s
where,j = 1,2, ... d
Basically, for motion ofN-particle system having no 'so called' restrictions
over the motion, d = 3N and for q; 's, j = 1, 2, 3 ... 3N.
As for examples, for pendulum motion,
ql = 8, degree of freedom = 1
for motion of mono atomic molecule within cuboidal box,
ql = X, q2 = y, q3 = z, degree of freedom = 3
for motion of particle confined on the surface of a sphere of radius r,
ql = 8, q2 = <p, degree of freedom = 2.
for motion of particle outside the surface of a sphere,
ql = X, q2 = y, z3 = z, degree of freedom = 3.
(with constraint: r> a, a = radius of sphere)
for motion of a N particle system,
ql = xi' q2 = Yi' q3 = zi' ... , q3N = zN
and degree of freedom = 3N.
It is a fact that for motion of any system, under any no. of constraints, the
generalised coordinate can be expressed in terms of general coordinates and the
converse is also true. As for example for motion of a simple pendulum, x2 + y2
=f2
.. x = ql = - i = ./i(xlyd
or, y ql _X2 =J;(x,y)
also,
ql = 8 = = j(XI y)
... etc.
Thus, we can now conclude that for generalised cordinates a/s we usually
have,
and,
where, for i = 1, 2, ... N
= qjh)
(qj)
j = 1, 2, ... d. (d = 3N)
for a N-particle system.
We will now discuss about some other relevent generalised parameters which
are significant to study with the motion of a single particle or the system of
particles. These parameters are given below.
(i) Generalised Velocity: The time rate of change of generalized coordinate
i.e. iI, is called generalised velocity.
Since, we have, F;
3N ",-
~ o .
oF; = ~ a q}
]=\ qj
Here, or, are arbitrary virtual displacement which is totally independent
of time and ot = o.
So here, oq/ s are called generalised displacements.
But for,
dF;
dt
F;(qj,t)
Here, the terms q j are generalised velocity and so we can conclude that
the ordinary velocity ~ are seen to be the linear functions of the generalised
velocity components, no matter how the generalised coordinates are defined.
(iiJ Generalised Acceleration
We have, F; = F;(qj,t)
a- a
2
- a
2
- a
2
-
'! = "-.!lq + "" ': qqk + 2" _---.!L
q

rj L: aqj J 7"-: aqj.aqk J L: oqj.Dt J ot
2
Here, the term iij is called generalised acceleration and here, we see that
the ordinary acceleration is not linear functions of components of generalised
acceleration alone but depends quadratically and linearly on the generalised
velocity iI j as well.
(iii) Generalised Momentum
We originally have the kinetic energy ofN-free particles, as,
N
" 1 .2 ,,1 (-'--'-)
T = = rj.ri
i=1 i
N {3N 3N } N (-) ( -)
1 " "uri. "uri. uri 1 " ari ari
+- + .- +- - . -
2 i=1 j=1 aqj k=1 aqk at 2 i=1 at at
" " ( aF; ) (aF; ) 1 " (aF;) (aF;) --. - -.-
" aqj' at 2, at at
I J I
Now if F; be not function of time explicitly then
= 0 and T = LLajkiJAk
at j k
where, a
jk
= Imi( aF; ).( aF; )
2 i aqj aqk
and in that case, T is homogeneous function of generalised velocities in degree
2. Now the generalised momentum Pj corresponding to generalised velocity iI
j
is defined as
(
oF. )
So, for stationary generalised system a:- = 0 ,this generalised momentum
is given by
p. = L L a,i! k which is again linear function of generalised velocities.
} I k
(iv) Generalised Force
The amount of workdone by the force L Fi on the system during an
i
arbitrary small displacement oy; of the system, is
,,- oF.
Q.= L..Fi-
'
J i oqj
where,
This parameter Q. is called generalised force associated with the co-ordinates
qj' Here we should follow that whatever dimension a generalised coordinate
has, the product of the generalised force and generalised displacement must have
the dimension of work. For that reason the generalised force need not have
always the dimension of force.
(v) Generalised Potential
If V be the position dependent potential energy of the system then
dw = oV = - " oV oq. = "Q .oq .
L..aq. } L.. } }
j } j
Since,
,,- oF.
= - L..Fi- ' = -Qj
i oqj
So, the definition of as generalised force is a natural one.
But when the system is not conservative for which the potential energy V
is also function of generalised velocities qj, we can define the generalised force
Q]
as,
oV d (avo 1
Q
j
= - oqj + dt oqj
How, V 0 may be called a 'velocity dependent potential' or simply
'generalised potential', since it gives rise to generalised force Qj"
7.8 Limitation of Newton's Law
The limitations for applicability of Newton's law are
(i) These laws are only valid in inertial frame. For non inertial frame,
the transformed version of Newton's Laws are made because of the
presence of pseudo force in that frame.
(ii) The velocity of particle on which Newton's law is required to apply,
must be much less than that of light in vacuum. Because, of the
particle velocity be close to that of light in vacuum (v ~ c), then by
Einstein's relativity, mass will change with velocity.
But Newton's law only is applicable for stationary mass.
SUMMARY
1. Conservation principle gives the constantness of parameter.
2. Conservation law gives the informations about motion even if forces
affecting motion are not known.
3. Force for a single particle motion, F = ~
4. Effective force on ith particle for the system of particles is
- -(e) ,,-int :.;
Fi = Fi + L.'pij = mj'f
ft-i
5. Effective force on system of particles
where
6. For momentum conservation, LmiVj = Constant for effective force,
F= LFi(e) = 0
i
7. For angular momentum conservation,
L)::'i == x vi) Constant
i
,,- - ,,-
for effective torque L rj x Fi :::: L. ri x Fi = 0
i i
8. For conservative force, f
F
.
dr
== 0 and F == - v( v)
c
9. For energy conservation,
1 2 I 2
-mvi + vI = - mv2 + V
2
2 2 /
(Single particle)
m/? + vi(e) + Vb
nt
= Constant (System of N particles)
I I I )
10. For constrained motion:
No. of constraints = No of restrictions involbe = m
No. of particles in the system = N
No. of degree of freedom = d = 3N - m
No. of generalised coordinates = d
i.e. for q/s,j = 1,2,3, ... d
Constraints: Holonornic Velocity dependent
Non holonomic Velocity independent
Bilateral Can be expressed in equations
Unilateral Can not be expressed in equations
Scleronomic Not explicit function of time
Rheonomic Depends on time explicitly
Generalised parameters:
Generalised coordinates: qj ; qj = qj(r;)
I
fJi
Generalised displacement: oq).; or.. = _I oqj
1 oq.
J J
Generalised velocity :
Generalised acceleration: ii
j
Generalised momentum: Pj
(for 01 = 0)
Generalised potential :
Worked Out Examples
1. Find the generalised coordinates for the following system, fly-wheel, a
particle moving inside the surface of a cone, Bead in abacus, Hydrogen mole<.:Ule.
Ans : (i) Fly wheel: 8, the angle between a finite radius of the fly wheel
and fixed line perpendicular to the axis.
(ii) A particle moving inside the surface of a cone:
r, 8; r = The radius vector drawn from the vertex as the origin to the position
of the particle 8 = the angle of the radius vector with a fixed slant edge of the
cone.
(iii) Beads in abacus: x, the cartesian co-ordinate along the horizontal wire.
(v) Hydrogen molecule: x, y, Z, <p, \jI;
(x, y, z) are the cartesian co-ordinates of the center of molecule,
<p and \jI are the angle of rotation about two mutually perpendicular axis
through center.
Example 2 : Find the nature of constraints in the following cases rigid
body, deformable bodies, simple pendulum with rigid support, pendulum with
variable length, rolling of disc without sliding, gas fitted hollow sphere.
ADS: Rigid body: Scleronomic, holonomic, bilateral
Deformable bodies: Rheonomic, holonomic, bilateral simple pendulum with
Rigid suffer : Scleronomic, holonomic, bilateral.
Pendulum with variable light: Rheonomic, holonomic, bilateral.
Rolling of disc without sliding: Non holonomic.
Gas filled hollow sphere : Scleronomic, holonomic, unilateral.
Example 3 : For conservative system, express kinetic energy of N-particle
system in terms of generalised momenta.
ADS: We have for conservative system,
~ 1 .2 "" ..
L '2
m
/ i = L..JL..Jajkqjqk
i j k
Kinitic energy, T =
where,
I ~ m i a ~ ).( ~ )
i 2 aqj aqk
so, this kinetic energy is a homogeneous function of generalised velocity
in degree 2.
Hence, from Euler's theorem
II .
T = - p.q.
2 J J
j
Example 4 :Write down Jacobian (1) for the transformation from a set of
coordinates q} (j = 1, 2, ... , d) to the set ri(i = 1, 2, ... N). (d = 3N)
Ans : J(ql'q2 ... ,q3N} == (O(QbQ2 ... ,q3N ))
xby),zN O(XbYbZN)
aQ) aQ2 aQ3N
ax) ax) ax)
aQ)
aQ2 aQ3N
aX2 aX2 aX2
aQ) aQ2 aQ3N
aZ
N
aZ
N
aZ
N
EXERCISES
1. Prove the laws of conservation of linear momentum, angular momentum
and energy for a system of interacting particles.
2. What are the advantages of using generalised coordinates.
3. Discuss the statement ~ 8Qj must have the dimension of work".
4. What are constrains? How do they affect motion of a mechanical system.
S. Show that of the equations of transformations do not involve time, the
kinetic energy can be expressed as a homogeneous quadratic function of
velocities.
-:0:-
Chapter-8
Variational Principle and
Lagrangian Mechanics
8.1 Introduction
In our present discussion, we will try to develope a new field of mechanics,
popularly known as Lagrangian Mechanics which is some thing different from
Newtonian mechanics. In this new field, we can deal the several mechanical
problems by considering the energy of the system, both kinetic and potential
rather than the force or the momentum. The development of Lagrangian
formalism can be made through the zone of Newtonian mechanics (by the
consideration of D'Alambert's principle) or from the basis of calculus of
variation. It is now important and also significant to realize the calculus of
variation at first and then its role to develop the new field, the Lagrangian field
in classical mechanics.
In our provinces chapter we have discussed about the limitations of
applicability of Newtonian mechanics. Keeping such restriction in mind, we
will discuss how to develop the range if the applicability of that Newtonian
mechanics in a new version through energetic consideration. This approach is
at first made by Lagrange and in that sense, Lagrangian mechanics was mostly
acceptable to several mechanical systems. This mechanics, or more clearly, the
laws of such mechanics, were at first developed on the basis of Hamilton's
integral principle which was actually an outcome of the calculus of variation.
After that, this laws are shown to be equivalent with Newtonian mechanics
through the developement ofD' Alambert's principle. We will try to cover more
or less all such discussions in our present chapter and also we will try to make
interest on such discussions as far as possible.
8.2 Forces of Constraint
The mathematical restrictions which is the constraints of the constrained
motion, not only interfere with the solution of the problems, but they are always
associated with the forces by virtue of which they restrict the motion of the
155
system. Such forces are known as "forces of constraints".
We generally formulate the laws of mechanics in such a way that the work
done by the forces of constraints is zero when the system is in motion. We
basically require our formulation to side track the effect of forces of constraints
without violating them.
So for the force of constraint F , the workdone for small time independent
displacement Or , is
OW= F.or = 0
By definition, the forces of constraint are workless forces. As for example,
if we take the motion of simple pendulum the tension of the string will be the
force of constraint which is perpendicular to every instantaneous displacement
of the bob of pendulum. Similarly, if we take the sliding of a body down a
friction less inclined plane, the normal reaction given by the plane on the body
will be the force of constraint which is also workless since it is perpendicular
to every instantaneous displacement of the body along the plane. So we see
that the role of the forces of constraints are very significant for constrained
motion although the forces are themselves workless.
8.3 Virtual Displacement
A virtual displacement is an arbitrary, instantaneous, infinitesimal
displacement of a dynamical system independent of time and consistent with
the constraints of the system. Since we have for N particle system
we have, virtual displacement,
d or.
= I-J oqj (Since, it is independent of time)
j=l
oq
j
So, the virtual displacement can be expressed as the linear superposition of
infinitesimal generalised displacements, all taken as independent of time.
8.4 Principle of Virtual Work
The work due to virtual displacement is known as virtual work. By the
principle of virtual work,
A system with workless constraints is in equilibrium under applied forces
if and only if zero virtual work is done by the applied force in an arbitrary
infinitesimal displacement satisfying the constraints.
Let us now consider a dynamical system which is in equilibrium. Then from
the total force one each particle of the system vanishes, i.e., Fi = 0 .
Now let us give a virtual displacement to the system, then the virtual work
of the ith particle is, Fi where is the virtual displacement if i-th particle.
So the total virtual work for all the particle is
Variational Principle and Lagrangian Mechanics 157

0
But
F
j
pta) + p(c)
, ,
where,
p.(a)
I
externally applied force
Ftc)
force of constraint
,
But Fj(c),s are all workless forces
"F- (e)
L..J' .ur, = 0
This is principle of virtual work.
8.S D' Alembert's Principle
We have from Newton's 2
nd
law, if Pi be the force on with particle of a
dynamical system,
o
Now for infinitesimal, time independent virtual displacement 8F; we have
from the principle of virtual work.
I[Pi 0
i
But
where, externally applied force,
p(c) force of constraints
,
"{- (a) - (e) ( :;)}-
L..J F; + F; + -m,r; .8r; = 0
But,
o
because, F (e) is workless.
I
:. =0
j
This is D' Alembert's law. Here, the force is called reverse effective
force and so by this law, the total workdone for all particles of the system by
the resultant of externally applied force and reverse effective force for
infinitesimal, time independent displacement of all particles of the system will
become zero. Here, we should note that we can equate the coefficients of
to zero since oF;' s are not independent to each other. There it is necessary to
transform oF; into the changes of generalised co-ordinates, dqp which are
independent to each other. The coefficient of every dqj will then be equated to
zero.
We should also note that, as this principle does not involve forces of
constraints in any way, it is sufficient to specify all the applied forces only.
Also, this principle is valid for all rheonomic and scleronomic systems that are
either holonomic or nonholonomic.
8.6 Lagrange's equations for a holonomic System
For a holonomic dynamical system, we have, the instantaneous position
vector of ith particle,
F; = F;(qj,t) i = 1,2, ... N
j = 1,2, ... d (d = 3N)
we get,
... (1)
Also the kinetic energy for the whole system,
T
II .2
2
mjrj
= I
. 2
! I
or
I
mr..--
... (2)
Oqk
11
0
'
j qk
and
or
I -
mj'i'--
... (3)
oqk
j oqk
" " d ( )
L..Jnljrj.-. + L..Jnljrj.- -.
. aqk . dt aqk
I I
But, (From basic rule of calculus)
So from equation (2),
[from equation (3)]
where,
So, we finally get,
,,- a;.
Q
k
= L..JFj .-
1
= Generalised force.
j aqk
a: = 0
k dt aqk aqk
Since, the system is holonomic, all 8q k 's are independent and arbitrary. So
it follow from the above equation,
d(aT) aT
dt aq k - aq k - Q
k
= 0 for all k.
This equation is called Lagranges equation of 1 st kind for the holonomic
system.
Now if the system be conservative, then there exist a potential energy
V = V(q) (j = 1, ... , d)
and
(k = 1, ... , d)
160
:. We have from above equation,
d(OT) oT OV
dt oq k - oq k = - oq k
oV
Since, V is independent of qj, oq. = 0
_ J
We now get,
_ oV V) = 0
dt oqk oqk oqj
d (OL) oL
=> dt oqk - oqk = 0
This is Lagrange's equation of 2nd kind for a conservative holonomic system
and here the term
Here,
L
L
T - V is called 'Lagrangian' of the system
L(qk' qk' t) (in general).
8.7 Lagrange's equation for a conservative, non-bolonomic system
Basically, for nonholonomic system, dqj's are n()t independent and they are
related by 'm' nonintegrable relations. (m < d), such that
+ aikdt) = 0,
k
where, a
fk
= afk(qk' t).
(l = 1,2, .... m)
Now these above equations are the equations of the constraints for a non
holonomic system. Hence, for time independent virtual displacements C5q k' we
have from above equation.
. .. (1)
Now from Hamilton's integral principle (which will be developed latter
on) we have for a conservative system
... (2)
We now have, by multiplying equation (1) by some undetermined multipliers
(Lagrange's) 1...[ (l = 1, 2, ... , m) and then integrating from tl to t2 and finally
adding with equation (2), we get,
... (3)
Since, all the 8q k 's are not independent for non holonomic system, we now
assume that some of them, 8q i' 8q 2' ... , 8q d-m are independent and the last m
are fixed.
We now choose A;U = 1,2, ... , m) such that
oL d (OL)
--- -. + 4...Alalk = 0
oqk dt oqk 1=1
... (4)
(k = d - m + I, ... , d)
With these A/ 's determined by equation (4), we can write equation (3) as,
tr{d-m[ oL d ( OL) m ]}
J L --- -. + LAlalk 8qk
dt
= 0
t\ k=1 oqk dt oqk 1=1
... (5) .
Therefore, 8a\, 8a
2
... 8a
d
_
m
are independent which follows
oL d(OL)
--- -. + 4...Alqlk = 0
oqk dt oqk 1=1
... (6)
(k=I,2, .. d-m)
Combining equation (4) with equation (6), we have fmally the complete
set of Lagrange's equations for non holonomic system,
d (OL) oL
- -- --- = 4...Alalk
dt oqk oqk 1=1
(k = 1, 2, ... , d)
S.S Introduction to Calculus of
variations
For the solution of a dynamical Y
problem, we usually are interested about the
position or location of the system or the
particle at a particular instant of time. We
are also interested about the path adoped by
the system and its information, wheather it
is maximum or minimum or extremum.
This whole study can be made easily by the
,
,
,
,-
,-
,
,
,
,
,
,
,
,,'
... (7)
,
technique of the 'calculus of variation'. ol----------x--
Here, consider the motion of the particle from the point-
I (xi' Y2) to the point-II (x
2
, Y2) along the path' A'. This
path may be the shortest path 'Ao' which is a straight line
path connecting this two said positions of the particle.
We can find this shortest straight line path for motion
of the particle by the method of calculus of variation in the
following way -
ds
I
,
I
I
I
: dy
,
I
I
I
I
_____________ I
dx
If ds be the length of infinitesimal portion of the path of particle then total
length of the path
S
2 2 x2
fds= = fV1+li) dx
I I Xl
X2
S
J f(y,y', x) dx
... (1)
where, fly, y', x) = + y,2
( where, y' = :)
The equation (1) is now called the action integral in the calculus of variation
and here, since, 'S' is minimum or shortest, we must have, in general,
oS 0
X2
or,
oJ f(y,y', x)dx = 0 ... (2)
Xl
This is the formulation of the problem of calculus of variations. Here we
should note that the o-variation is defined as the variation in the quantity to
which it is applied at the fixed value of the independent parameter i.e.
Oy = Y2 - y\ = (y)X=X2 -(y)X=Xl'
Here since we consider the path for motion of particle, the shortest path is
taken as the extremum path and the other path with which we compare it (in
our present discussion of finding the extremum path) are called comparison path.
Now,
o [f(y y' x)] = of oy+ of oy' + of ox
. " By By' ax
But 0 (o:)oa, where, a is some parameter.
Since x is not a function of a
ax
Ox = -oa=O
oa
(for, := = 0)
X2
o Jf(y,y',x)dx
XJ2[Of oy + Oy,ldx
oy OY'
Xl ...
X2[of 7.f d ]
J
dx
oy oy'dx
Xl
XJ2[8f d(Of) d(Of )]
-oy+- --oy -- --oy dx
oy dx By' dx oy'
Xl
XJ2[Of d (Of)] XJ2 d (Of )
By - dx By' oydx + dx By,oy dx
Xl Xl
X2[Of d (Of)] [Of ]X2
= ! By - dx By' oydx+ By,oy Xl
But there is no o-variation of y at x = x\ and also at x = x
2
.
i.e., (OY)X=XI = (OY)X=X2 = 0
X2
So we get, 0 J f(y, y', x) dx ... (3)
Xl
So fmally, we have from equation (2)
XJ2[of Of)]OYdx=O.
By dx oy'
Xl
But oy represents some arbitrary variation of y(x) with respect to the
arbitrary parameter a. about its extremum value (a. = 0). Since oy is arbitrary,
of = 0 of = 0
oy By
We get, '" (4)
These are the conditions that should be satisfied by the functionj(y, y1 if
the integral'S' of equation (1) is to be extremum.*
* Here we should note that if this calculus of variation be applied in formulating
Lagrange's equation, the path of a particle in configuration space of Lagrangian
formulation is uncertain and consequently a discussion of extremum path is an
important consideration for the system in motion. Also, Readers would note that
Lagrange's equation for conservative system (holonomic) bear a great similarity
with equation (4) which will be derived from this.
Now in our case of extremum path,j{y, y', x) = + y,2 .
of =0 of = y'
oy 'oy'
d ( y' J So from equation (4), - r:--:? = 0
dx V
1
+ y,2
y'
Const. = A (Say).
y' = c.
y = cx+ b. (b = another const.)
This is an equation representing a straight line.
So the shortest or extremum path must be a straight line path which is
usually obtained from the technique of this calculus of variation.
In another example, we can also apply this
method of the calculus of variation to the
'brachistocrone problem'. In this problem, we find
a curve joining two points along which a particle
falling from rest under the influence of gravity
travels from higher to the lower point in the
minimum time.
Here, the time spent in traversing ds portion
of the curve with speed v along the curve is x
2 ds
112 = f-;
I

2
But for vertical distance to fall x upto the point-2 we have from energy
conservation.
mgx :. v =
Variational Principle and Lagrangian Mechanics
2
f f(y,y' x)dx
R
Where, in this case,j(y, y' x) =
aj
dx oy' ay
of aj y'
But ay = 0, ay' =- + y,2
So, we get, - y = d [ , J
dx + y,2
x( 1 + y' 2) = Constant = C (Say).
-x) =x
2
.
y' =
x
1/2 1
On integration, y = a cos -I (1 - xl a) - ( 2ax - x
2
) + ci for 2a =
Here, c 1 is the new constant of integration.
Obviously, if c 1 be zero, then y will be zero for x to be zero.
1
In such case, Y = a cos -I (1 - ) -(2ax - x
2
)"z
165
'" (6)
Which represents an inverted cycloid with its base along y-axis and cusp at
the origin. This is an well known result for brachistochrone problem which is
not obtained from the calculus of variation.
8.9 Variational Technique for many independent variables: Euler-
Lagrange's differential equation:
Let us consider that the function 'f is a function of many independent
variables y/x) i.e. f= j(yj' yj', x). (j = 1, 2, ... , d)
In that case from the calculus of variation.
166
2
8 ff(Yj,Yj,x)dx = 0
)
Now taking, 8 4 for an arbitrary parameter u, we get,
OU
fI{-Of au + of }8U dx = 0
, 00. Oy" 00. oY}
1 } ,
JI{ of oYj + OYj} 8u dx = 0
) j oYj OU oYj OU
(
for 8x = ox 8u = 0 as ox = 0 because, x is independent of U)
OU 00.
=> I{J of oYj 80. dx +[ OYj]2 - J 1 OYj} 80. dx = 0
, oY)' OU oY)' 00. dt oY)' OU
) ) ) )
, (Oyj) (Oyj) =0
But at end pomts, OU 2 = OU )
:. I{J[ Of, _.!!..-If = 0
j ) Oy) dx oY) 00.
But 8u.dx is arbitrary and independent.
So, we must have each coefficient of the previous equation zero. i.e.,
of _.!!..-( Of 1 = 0 for all}.
Oyj dx Oyj
This mathematical equation is known, as 'Euler-Lagrange's equation' for
all independent variables Yj 'so
8.10 Hamilton's Variational Principle:
Following the method of the calculus of variation, Hamilton stated that the
motion of a dynamical system from time t) to time t2 will be such that

I = f L dt will be extremum for the path of the motion. i. e., of L dt = 0 .
This is Hamilton's variational principle.
where, L=L(qj,qj) = T(qj,qj)-V(qj)
is Lagrangian for that conservative system. To study with this principle,
we should at first have some knowledge about the motion of the system. For
motion of a single particle we should require 3-independent co-ordinate.
Similarly, for two body system 6 independent coordinates are required and in
that case, that two body system can be replaced by a single particle in 6
dimensional space. Similarly, any system ofN particles can be replaced by the
motion of a single particle in 3N dimensional space and that particle (single) is
called 'system point' and the required 3N dimensional space is called
'configuration space'. So the motion ofa system in realistic space can be replaced
by the motion of a system point in configuration space.
8.11 Derivation of Hamilton's principle from Lagrange's equation
d (aL) aL
From Lagrange's equation, we have dt aqj = aqj
where, L = L(qj' qj) = Lagrangian of the system which is holonomic,
conservative.
t2
Now, ofLdt
t1
where ( oq j ) = (oq j ) = 0
, t2 t1
for no o-variation of qj at the end points.
t2
of Ldt = o
This is Hamilton's principle which follows Lagrange's equation for
conservative, holonomic system.
8.12 Derivation of Lagrange's equations from Hamilton's principle:
t2
From Hamilton's principle, of Ldt = o.
tl
(
Since, 0 ~ ~ o and = 0)
aa aa
",[t
f
2
aL aqj 1 aL aqj 1
L..J --oa dt + -. -oa dt = o.
. aqj aa aqj aa
} ~ ~
",[t
f
2
aL aqj aL 1 d (a
q
j
) t
J
d ( aL) aqj ]
~ L..J --oadt+-. - - oadt- - -. -Badt
. aqj aa aqj dt aa dt aqj aa
} tl tl tl
[
a
q
. ]
} oa
But, aa
tl
So we finally get,
But since, &i
j
is arbitrary, each [for, j = ';:: ]
equation must be zero. So in that case,
coefficient of above
aL d [aL) = 0
8qj dt 8qj
or, !!.-( aL ) _ aL = 0
dt 8qj aqj
This is Lagrange's equation of 2nd kind:
Also for conservative system. L = L{qj,qj) = T(qj,qj) - v(qj)
aL aT h k
so, - = - were, T = me tIc energy
8qj 8qj
:. _ - V) = o.
dt 8qj aqj
-aT = - av = Q. = This is Lagrange's equation of 1st
dt aqj aqj aqj J
kind.
8.13 Derivation of Lagrange's equation from D' Alambert's principle:
We have from D'Alambert's principle, l:[F, = 0
i
But for,
... (1)
where,
,,- aF
Q. = F
i
'-
a
I Generalised force.
J I qj
But
"1 d ( . af::) . a (.)}
dt -
i q} q}
[
. af; )
smce -=-
, aqj aqj
m.F';) - m.F })]
7 dt aq j 2 I I I aq j 2 I I I

dt aq j i 2 aq j i 2
But for whole system, total kinetic energy
,,1 .2
T =
i
draT) aT
----
dt aqj aqj
We thus have from equation (1) and (2) as,
=> L[l!!...( ) -aT} -Qj]8
q
j
= 0
. dt aqj aqj
}
Since, all 8qj's are arbitrary, we have,
av av
Q. = --, -. =0
J aqj aqj
.. -V) -V) = 0
dt aqj aqj
or, !!...( ) - aL = O. (for L = T -V)
dt aqj aqj
... (2)
This is Lagrange's equation of 2nd kind for the conservative holonomic
system.
8.14 Derivation of Hamilton's Principle from D' Alambert's Principle
From D' Alambert's principle, we have,
L(Fi-Pi).Or;
i i
when, Pi =
LPi.Or;
... (l)
But
d( dr; s:-) dr; s:(dr;)
- m- uf, -m- u -
dt 1 dt 1 1 dt dt
!!"(m. dr; .0;:;) _ o["!'m.(dr; ).(dr;)]
dt 1 dt I 2 I dt dt
d ( d;:; -) (1 2)
- m-'.od -0 -mv
dt 1 dt 1 2 I I
forv. =_1
(
d;:;)
1 dt
So from equation (1)
:,[
But Li
miv
/
i
and,
!!..["(m. dr; .0;:;)]-0(" ..!.m.v.
2
)
dt I dt I 2 I 1
I I
= T = Total kinetic energy of the system,
:. :, [ m, .or, )] o(T - V) oL
(where, L = T - V)
... (2)
172
[ ]
12
dr; -
"m.-.or,.
'dt '
, II
o (for or;' s zero at the end points)
where, o(L dt) = (0 L)dt + U(dt)
= (oL)dt for o(dt) = 0
This is Hamilton's principle
8.15 Cyclic or Ignorable Co-ordinates
Among all generalised co-ordinates qj (j = I, 2, ... , d), if q k 's are some
generalised coordinates (k = I, 2, ... p, p < d) such that the Lagrangian of the
system (L) does not depend on these co-ordinates, then all these generalised
co-ordinates q k 's are called cyclic or ignorable co-ordinates.
So, for qk's are to be cyclic, L:;:. L (qk)'
oL
:. -;-- = 0 for all such k = 1,2, ,,, , p.
uqk
Now we have from Lagrange's equation.
d(OL) oL
dt oq k - oq k = O.
oL d(OL)
So for oqk = 0, dt oqk = 0
oL OT
or -;:- = -;:- = constant (since for conservative system, V :;:.V(qk)'
uqk uqk
Pk = Constant
So we see that generalised momentum corresponding to cyclic co-ordinate
will be conserved.
8.16 Conservation Theorems:
(a) Conservation of Linear momentum in Lagrangian formulation :
For linear motion (uniform) of any dynamic system, K.E = T = L m, Vi
2
i
is independent of some generalised co-ordinates, say qj
OT
.. -=0.
oqj
Also, we have the generalised force,
8V ,,- 8i}
Q. = -- = ~ F i -
J 8qj i 8qj
where, . ~ represents the component of the total force along the direction of
translatIOn of q ..
Thus, if th6 component of the total applied force along a given direction
vanishes, then ~ = O.
8V
and
d( 8LJ
dt 8qj
o and qj will be also cyclic in V.
8
-(T-V) =0
8qj
d(8TJ d
= -d -8' = -(Pj) = O.
t qj dt
.. Pi = Constant.
i.e., in that case, the corresponding component of the linear momentum
will be conserved. This is linear momentum conservation.
(b) Conservation of Angular momentum in Lagrangian formulation:
N ow consider that dq. represents a rotation of the system along the
generalised co-ordinate q. J 1\
J n
then Idi}1 = r
i
sin e dqj
dr;
dqj
r
i
sin e
and
8i}
n xri
8qj
Also, since q. cannot appear in total kinetic energy
T (because trans(ational velocities are not affected by
rotation)
Also the generalised force (which has not always
the dimension of force)
"n .. (r. x F.) A (L- F-)
'= n r,x
; I I
;
where i total torque = L i j
Also, the generalised momentum '
P
.= ..!..mv2)
} L2 j j
uq} VLfJ j
Because -' = -' =_,
(
ot. ot. or. )
'oqj oqj oqj
Where, L = I L; = Total angular momentum.
If now the effective torque along ii for rotation be zero then
oV
Q. = 0---
) - oqj
So qj is also cyclic in V.
oL
:,[ :,(Pj)=O ..
oqj
..
Pj
= const.
=>
ii.L
= const.
This gives conservation of angular momentum, by which, when component
of torque along ii vanishes, the component of angular momentum along ii
will be conserved.
(C) Conservation of Energy in Lagrangian Formulation
From Lagrange's equation, we have, ! ( ) = .
Now if we take that the constraints are all time independent then the
Lagrangian will not be the function of time explicity.
and
L L(qj,qj)
dL
dt
dL
dt
"'d[.8L)
7 dt qj 8qj
Iqj =0
dt j 8qj
But in that case, the total kinetic energy T will be a homogeneous quadratic
function of q j
Thus 2T
I' 8T
qj-;;-
j qj
(By Euler's theorem)
.. 2T
I' 8L
qj-;;-
j qj
(for V = V(q) only)
d
..
-(L-2T)
0
dt
d

-(2T- L)
0
dt
.. 2T- L Constant.
.. 2T - (T - V) Constant
.. T+V E = Total energy = Constant.
This is conservation of energy.
8.17 Gauge Function for Lagrangian
From the calculus of variation, we know that Euler's Lagrangian
is given by
176

This equation does not change its form if we add the time derivative of
any arbitrary function F(qj' t) with the Lagrangian L{qj,qj' t) of the system.
This function F( q j' t) is then called Gauge function.
Because, let the Lagrangian will now become,
aq; aq;
But, aqj = 0, aqj = By = I for i = j only
aF
and aqj = 0 for F * F(q;)
Thus taking i = j, we get
:'[a:):)]- a:):) :,(:;)- a:):)
= a
2
F q. a
2
F q. = 0
aq; ) aqj at aq; } aqj at)
So we get from equation (1)
d [aLI) au d (aL) aL
dt aqj aqj = dt aqj aqj
dF
177
Thus Lagrange's equation will remain invarient even if dt for F(q" t) be
added with Lagrangian of the system. This arbitary function F(q" t) is called
Gauge function for the Lagrangian L( q}, q}, t)
8.18 Invarience of Lagrange's equations under Generalised Co-ordinate
transformations
For generalised co-ordinate transformation, let qj and Q
j
are two sets of
co-ordinates and the corresponding Lagrangian as
where
where
and
L = L(q}, qj,t), L' = L/(Q},Oj' t)
qj = t), Q
j
= t)
aq}. aq}
aQ
j
Qj+Tt
Basically, such transformations are called point transformation for which
aq}
we usually consider aQ} = 1
Now, to show invarience of the form of Lagrange's equation
we have,
d (au) aL'
dt ao} - aQ} =
178
[
d ( 8L) 8L]( 8qj )
= dt 8qj - 8q} 8Q
j
d (8L 1 8L 8qi
= - - - - for --- 1 for point transformation.
dt 8qj 8qj 8Q
j
-
Thus the form of Lagrange's equation remains unchanged under generalised
co-ordinate transformation.
8.19 Concept of Symmetry: Homogeneity and Isotropy
If a function does not change its property under some operation, then that
function is said to be symmetric about that operation. As for example, if a
cylinder rotates along the surface about its axis, the cylinder's apparent shape
will not change and in that case, the cylinder is said to have rotational symmetry
about its axis. In our case, the Lagragian of the system may have similar kind
of symmetry under some operations, which are,
(i) Homogeneity of time : For such symmetry operation, the Lagrangian
of the close system will not be function of time explicity and in that case,
8L
at = 0 and there is no external force.
From our earlier discussion, this leads to the conservation of energy i.e.
E = T + V = Constant.
(ii) Homogeneity of Space: For this symmetry, Lagrangian of a closed
system, should not change due to any arbitrary small uniform translation of all
particles.
As for closed system, external force is zero, we have generalised force
= O. Thus for such symmetry 8L = 0 and we have from our previous discussion
oqj
the total linear momentum of that closed system will be conserved due to this
homogeneity of space.
(iii) Isotropy of Space: If for any arbitrary rotation about the origin of
any reference frame, the physical property of any closed system remains
unaffected, we say that space is isotropic.
For such rotation about iz, we have from our earlier discussion,
8F; 8F; A_
--=-=nxr:
8qj 89 I
and for such isotropy, we come to conclusion that if rotation co-ordinate is
cyclic, the angular momentum is conserved.
8.20 Invarience of Lagrange's equation under Galilean Transformation
For two inertial frames Sand S', when S' is moving with velocity Vo w.r.t.
S frame, we have from Galilean transformation, F' = r - vot
v' = v- Vo
Now in S frame, the Lagrangian of the system
and in S' frame,
1 2'
L = -mv -V
2
1 ,2 V'
L' = -mv -
2
But since, the potential energy V only depends on interparticle separation
IF; - F21 ' we must have,
V(li) -F
2
1) = V'(Ii)'-Fil)
i.e., the potential energy V will remain unchanged under Galilean
transformation.
L
But L'
L'
=>
L'
1 (_ _) (_ _)
- m v - Vo . v - Vo - V
2
1 2 1 2 -- V
-mv +-mvo --mv.vo-
2 2
1 2 (1 2 __ I
-mv - V + -mvo - mv.vo I
2 2 )
d( - - 1 2)
L+- -mvo.r +-mvot
dt 2
L+ dF (F,t)
dt
h
F(
- ) - - 1 2 h G c. . d h h
were, F = r,t = - mvo.r + -mvot 1S t e auge lunctlon an we ave sown
2
dF
in our earlier discussion that for additional term with the original
dt
Lagrangian, the Lagrange's equation will remain also unchanged in S' frame.
So Lagrange's equation will remain unchanged under Galilean transformation.
8.21 Application's of Lagrange's equation of motion in several mechanical
systems
(i) Linear Harmonic Oscillator :
For linear harmonic oscillator in one
dimension,
the kinetic energy of the system is T = .!.'nX
2
2
and the potential energy
x
V = -fFdx = -f(-/a;)dx=-l/a;2
o
Hence, the Lagrangian of the system
L = T - V = .!.lni
2
- .!./a;2
2 2
aL _ Ini aL - _ /a;
So, here, ax - 'ax-
we now have from Lagrange's equation of motion,
d (aL) aL
dt ax ax

x + w5x = 0 for Wo = g = Angular frequency
m
This is equation of motion for one dimensional linear harmonic oscillator.
(ii) Simple Pendulum 0
At angular position of the pendulum e(L4) the kinetic
energy
T
= .!.mv
2
='!'m(/8)2 =.!.m1
2
8
2
2 2 2 .
Also, taking horizontal reference plane through the
equilibrium position of the bob, the potential energy V = mg(1
- I cos e)
= mg/(1 - cos e)
Thus, Lagrangian of this pendulum is given by,
L = T - V = .!.mP8
2
- mg/(1- cose)
2
m
aL
so that,
aL
ae
ml2e, as = -mgl sin S
Putting these in Lagrange's equation,
d (aL) aL
dt ae - as
o
ml29 + mglsinS 0
But for small S (L4), sin 8 ::: 8
9+(7)S = 9+w58 =0
181
This is equation of motion of a simple pendulum where, Wo = Jf is the
circular or angular frequency of the pendulum with time period of oscillation
T = 21t = 21t fI.
Wo Vg
(iii) Spherical Pendulum :
For such pendulum, bob is constrained to move on a sphere rather than in
a circle. The position of bob is then located by independent spherical co-ordinates
(8, cp), such that for length of pendulum r,
the kinetic energy of the bob is

Also the potential energy of the bob due to gravity relative to the horizontal
plane, = mgr cos 8.
The Lagrangian for this system is thus given by
So here,
:. cp is cyclic
L T - V = + sin
2
Scj>2) - mgrcosS.
aL
ae
aL
as
aL
acj>
mr2 sinS cosScj>2 + mgrsinS
aL
2 .2S - 0
mr sm cp, acp =
Here, Lagrange's equations can be written as,
d (aL) aL
dt a8 - as
: t ~ ~ )
From equation (2),
o
o
~ (mr2 sin
2
S<j 0
or, mr2 :t (sin
2
S<j 0
Also from equation (l)
... (1)
'" (2)
'" (3)
mr
2
e-mr
2
sin9cosS<j>2 -mgrsine = 0 '" (4)
These equations (3) and (4) are the equations of motion of a spherical
pendulum.
(iv) Compound Pendulum:
When a rigid body is capable for
oscillating in a vertical plane about a fixed
horizontal axis then that body is called a
compound pendulum.
For such pendulum of mass m and
moment of inertia I about the axis of
rotation, if I be the distance of its center of
mass from the point of suspension, then for
angle of deflection e of that body
the kinetic energy T = -.!. 18
2
.
2
Also the potential energy relative to a horizontal plane through the point
of suspension, is
v = - mgt cos S
So the Lagrangian of the system,
L
aL
a8
1 . 2
T - V = - Ie + mgl cos e
2
. aL
1 e; as = - mgl sin S
so from Lagrange's equation,
d (aL) aL
dt a8 - as
Ie + mglsinS
o
o
8+( mt)e
e +W2 S 0
This is equation of motion for compound pendulum with time period
T = 27t = 2 7 t ~ I
w mgl'
(v) Isotropic Oscillator in three dimension:
For such oscillator, the vibration is acted upon by a force directed always
towards or away from the position of equilibrium and the magnitude of such
force varies linearly with the distance from the position of equilibrium. This
force is given by
F = - kr when k is stiffness constant.
r 1 2
Now the potential energy, V = - J F dr = "2 kr .
o
Taking polar co-ordinates, the kinetic energy is then given by
T = k m(f2 + r
2
e
2
+ r2 sin
2
Sej>2)
we have, Lagrangian of the system
L = T - V = -km(r2 +r
2
e
2
+r2 sin
2
eej>2) - kkr
2
.
aL
Here, we have,
af
mr
aL
m,{)2 + mrsin
2
S<j>2 - kr
ar
aL
2 .
ae
mr e
cL
mr2 sin
2
S cosSej>2
-
as
aL
aej>
mr2 sin
2
Sej>
aL
a2 + kr = 0
(mr
2
S) - mr2 sinScosS2 = 0
( mr2 sin S<i = 0
All these equations are the required equations of motion in polar co-
ordinates, for an isotropic oscillator.
(vi) Particle's motion under central force:
Central force is that force which is always directed away or towards a fixed
center.
Here for motion of particle under such force, the kinetic energy at
instantaneous position (r, S) is
T = + r
2
(2)
r r k k
and potential energy, V = - f F dr = - f -2"dr = --
r r
co 00
for the attractive nature of the force.
So Lagrangian is
L
1 ("2 2' 2) k
T-v=-mr+rS +-
2 r
aL aL . aL '2
But
mf; as = mr
2
S; a; = mre
af
and also, S is cyclic.
The equations of motion for such case will be
d (aL) aL
dt af ar = 0
aL = 0
dt as as
We now have from these equations.
k
r2
or, o
.. ..62 k
Solving further, we get, mr - mru +"2 = 0
r
2mrr9 + mr2fj = 0
which are the well-known equations of
motion for the particle moving under the
influence of an attractive central force.
(vii) At wood's Machine:
This is a holonomic conservative system
with one degree of freedom and if the pulley
is frictionless, constraints are sc1eronomous.
Here for two masses m\ and m
2
attached
by inextensible string of length I, the
configuration can easily be specified by the two
coordinates x\ and x
2
We now
----T
I
have, x\ + x
2
+ 1t& = I = Constant.
(a = radius of pulley)
:. X2 = -XI; X2 = -XI
The kinetic energy is therefore
m,
1 .2 1 .2 1 ( ).2
T Zmjx\ +Zm2X2 = Z m) + m2 Xl
and potential energy,
V m\gx\ - m
2
gx
2
- (m\ - m
2
) gx) - m
2
g (/- 1ta)
= - (m\ - m
2
)gx) + C.
where, C -m
2
g (l-1ta) = const.
N ow Lagrangian of this system is
L
and
T - V = .!..(ml + m2)Xr + (ml - m2)gxl - C
2
d (OL) oL
Since, Lagrange's equation is dt OXI - oXl = 0
185
.. m2 -ml
and x2 -xI = g
ml +m2
These are the required accelerations of two masses.
SUMMARY
1. All forces of constraints are workless, i.e., :F/
e
) = 0
2. Virtual displacement (time independent) = I oqj
j 8qj
when F = F(q .)
,I I }
3. Principle of virtual work: =0
4. D' Alambert's principle:
I[ F}a) + = 0i
i .
5. Lagrange's equation of 1st kind (for conservative system)
=Qj
6. Lagrange's equation of 2nd kind for holonomic, conservative system:
!!.-( 8L ) _ 8L _ 0 [ . ]
dt 8qj 8qj - for L=L qj,qj,t =T-V
7. Lagrange's equation for conservative, nonholonomic system:
:qL. = i)l.j aij (j = 1,2, ... , d)
} } 1=1
.. XJ2, , dy(x)
8. For calculus of vanatIon, 0 f(y, y ,x) dx = 0 for y = dx
9. For calculus of variation, we have the condition,
d (8
f
) 8f dy
dx 8y' - 8y = 0 for f = f(y, y', x), y' = dx
10. For many variable function f(Yj,Yj, x), Euler-Lagrange's equation is
af -!!:..-( af J = o.
aYj dx aYj
11. Hamilton's variational principle:
Ir2
8 f L(qj,qj,t)dt = 0 for L(qj,qj) = T(qj' qj)- V(qj)
1\
12. Cyclic or ignorable co-ordinates :
qk's are cyclic when L =1= L(qk)
and then Pk = constant.
13. For conservative system, if constraints are time independent
aT
2L.J I a'. 2L.J I I
j ql j
dF
14. Gauge function in Lagrangian; F = F(qj' t) and L' = L + dt will have
same Lagranges equation and it is called Gauge invarience.
15. Lagrange's equation will remain invarience under generalised coordinate
transformation of point transformation.
16. Symmetry: Homogenity in Time: = 0 Energy is conserved
aL 0 L' . d
Homogenity in space: - = mear momentum IS conserve
aqj
Isotropy of space: Angular momentum is conserved.
17. For Galilean transformation r' = r - rot, v' = v - vo, the changed
Lagrangian in S' frame is
dF(r,t) -(_) _ _ 1 2
L' = L+ dt when, F r, t = - mVor + "2
mv
ot
But Lagrange's equation will remain unchanged under Galilean
transformations.
Worked out Examples
Example 1 : Obtain equation of a projectile path by using Lagrangian
equation for that projectile.
Ans : Now consider that a projectile is thrown with initial velocity v and
P(x, y) is its instantaneous position during the motion.
The kinetic energy is T = m( x
2
+ y2) (m = mass) and the potential energy,
taking x-axis as the reference plane
V =mgy.
The Lagrangian is then given by
1 (.2 .2)
L = "2 m x + y - mgy
aL aL
where we have, ax = mX, ay = my
ax
ax = - mg and x is cyclic.
Now from Lagrange's equations
d(OL) oL = 0
dt ox ox
~ O L ) _ oL
dt OJ By
0
We now get,
x
= 0
y+g = 0
x= At + B.
1 2
Y = --gt +ct+D
2
where, all A, B, C, D are constraints of integration.
But at t = 0 x = 0, x = v cosa (a = angle of projection)
y= 0, y= vsina
x
y
v cos at
. 1 2
vsmClt--gt
2
1 x
2
and with this two equations, we fmally have y = x tan a - -2 g 2 2
V cos a
This is the equation of the path of projectile which is a parabola
Example 2 : Obtain equation of motion and frequency for oscillation of a
conical pendulum.
Ans : In this case, the bob of the pendulum moves in a horizontal circle as
shown in the figure.
So far the angular position 8 of the pendulum we have,
T
1
2
.
2
1
/
2.2
8
.2
Kinetic:, e ~ = "2 mr <p ="2 m sm <P. Potential energy V = -mgl
cos 9.
So Lagrangian of the system
L = T - V = ~ m P sin
2
9cp2 + mgl cos9
2
Since in this case, 9 is constant, only cp is the variable
oL 12.29. oL 0
here and -. = m sm cp - =
'ocp , ocp
we then have from Lagrange's equation for cp
mP sin
2
9q> = 0 or, q> = 0 as 9 = const.
and cp = constant = (0.
I
I
a
189
",---i----..........
, r: '\
----\:;.) '.
I ,'<P ,
\ , I
' , ~ "
............. _------, ... '
d (OL) oL
Ifwe now write down the Lagrange's equation for 9 the dt oe - De = 0
~ 0 - ( mP sin 9 cos 9cp
2
- mgl sin 9) = 0
2
=
g
..
cp
Icos9
(0 =
. ~
cp = Icos9
Example 3 : Starting from Lagrange's equation of motion, obtain the
av
equation of motion mf = - a;
for a particle of mass m moving with acceleration 'f in a potential V.
Am : Here, we have, :, ( :; ) - :; 0
So for motion of particle, in x-direction
~ ( ~ ~ ) - ~ ~ = o.
I .2
But L = T - V, T ="2
mx
,V = V(x)
aL aT
ax = ax = mX, T = T(x)
d (OT) oV
we get, dt ax + ox = 0
190
av
mf= --
ax
This is equation of motion.
Example 4 : Use Lagrange's equations of motion to detennine the motion
of a mass m, sliding without friction down an inclined plane with angle of
inclination a..
Ans : Suppose, at any instant the mass slides a distance x along the plane
so that the vertical distance fallen is h = x sin a..
1 .2
:. The kinetic energy, T(K. E) = 2
mx
The potential energy V = -mgh = -mg x sin a.
1 .2 .
:. Lagrangian is L = 2mx + mgxsmo.
so the equation of motion will be
mX - mgsino. = 0
x = g sin a..
This is equation of motion.
Example 5 : Use Hamilton's principle to find the equations of motion of a
particle of mass m moving on a plane in a conservative field.
Ans : Let P(x, y) be the position of a particle moving on the xy plane under
the action of the forces Fx and Fy where,
av av
Fx -a; , Fy = -ay
1 (.2 .2) V
L = T - V = 2m x + y -
By Hamilton's principle, we have,
12 t2
8 J Ldt = 0 J 8 Ldt = o.
Therefore, we have,
12
J(x8x+y8y-8v)dt = 0
1}
... (1)
12
Now,
f XOX
I}
12 12
f
x!!.-(Ox)dt = XOX y2 - fxoxdt
dt I}
I} I}
12
- f xoxdt
I}
for Ox = 0 at t = t 1 and t = t2
12 12
Similarly, f yoydt = - f jioydt
So we finally get from equation (1)
Since, dx and dy are arbitrary and independent
we have,
x =
_ OV =F
ox x
OV
--=F
Oy y
This gives the equation of motion.
191
Example 6 : Find Lagrange's equation of motion for an electrical circuit
having inductance L and capacitance C in series. The charge in capacitor is q
and current of the circuit is i.
Ans : For such a circuit,
. 1 2 1 .2
The magnetic energy T M = 2 Lj = 2 L q
. dq .
l = - = q
dt
where,
1 q2
and the electrical energy V E = 2C
2
S h
L . LIT ",2 1 q
o t e agranglan ="2 LoAf - 2C
Thus from Lagrange's equation,
d (OL) oL
dt oq - oq = 0
L
q
q+-
C
1
when W=--
.JLC
This is the equation of motion of charge in this circuit.
Example 7 : A spring of mass M and spring constant k is r
hung vertically. Another mass m is suspended from it. Write
down the Lagrangian of the system and show that the system
will execute SHM of period. I

27t
k
Ans : Here, at lower end, where m is connected, velocity
is given by y. It is maximum at lower end (z = l) and zero at
fixed end z = o.
1
m
mg
So at any distance, z, from the fixed end the velocity is given by (f Y) .
So the kinetic energy associated with dz length of the spring is
dT = r = r
where, P is the mass/length of the spring.
So for whole spring, kinetic energy is
But M = pL
T
L 2 ( 3)1 1 z . 1.2 1 z 1
-fPdZ(-Y) = -ply - - = -Ipi
2 I 2 P 3 6
o 0
1 M. 2
T = - y
6
1
The potential energy is V = 2" ky2 and so the
(
1 .2 1 M.) 1 ky2
Lagrangian L = 2"m
y
+"6 y -2"
Then from Lagrange's equation
Time period, To
EXERCISES
1. What is virtual work? Write the principle of virtual work.
2. Obtain the Euler-Lagrange differential equation for a single variable function
from variational procedure.
3. Discuss energy conservation for a conservative system with time independent
constraints.
4. Show that the shortest distance between two points in a plane is along the
straight line joining them.
5. Derive equation of motion for a particle whose potential energy is given by
V(r) = - J f(r)dr in polar co-ordinates.
6. Obtain Lagrangian for a charged particle moving in an electromagnetic
field.
7. Obtain equation of motion of a dumb-bell in Lagrangian procedure.
S. Obtain equation of motion of a double pendulum in Lagrangian procedure.
9. Discuss that the configuration space is a virtual device introduced for
displaying the motion of the system employing the Lagrangian approach.
10. A particle of mass m moves under the influence of gravity on the inner
surface of the paraboloid of revolution x
2
+ jl = a z, which is frictionless.
Obtain the equations of motion.
-:0:-
Chapter-9
Hamiltonian Formulation in Mechanics
9.1 Introduction
Up till now we have an idea about Lagrangian mechanics. As we have
constructed, the Lagrangian of a system is basically the function of independent
generalised co-ordinates qp and time t. Although, it is a function of generalised
velocity qj also, but this generalised velocities are only the time derivatives of
generalised coordinates and so generalised velocities are basically dependent
variables. In construction of Hamiltonian of the system, we remove the presence
of these dependent variables q j and we introduce a new independent variables,
called generalised momentum Pj which is defined as
aL aT
Pj = aqj = aqj for T = T(qj' qj' t), v = V(q). (for conservative system)
Similar to Lagrangian of the system, this new function Hamiltonian (H) is
defined in such new formation of mechanics and H = H (q., p., t).
So, the fact is that from Lagrangian formulation to Hamilto'nian formulation,
there is a change of basis from (q j' q j' t) to (Pj' qp t) set. The mechanical state
of the system can thus be described completely, provided qj and Pj are given as
a function of time. Also in this new formation, since, we provide equation status
to 'co-ordinates' and 'momenta' configuration space no longer remains adequate
to specify the system as a whole.
9.2 Hamiltonian of the System
For holonomic conservative system, all the constraints are independent of
time, and then we have Lagrangian of the system
L = L(q},qj)
194
dL
dt
So, we get,
dL L
d
[8L) L
d
( .)
- -' - p.q.
- = dt 8' q, = . dt } }
dt j q, }
~ dL ~ [ P.q.) = 0
dt dt ~ } }
j
~ ~ [ p.q . - L) = 0
dt ~ } }
}
195
Here we take, H = LPjqj - L as a new function of the system and it is
j
called 'Hamiltonian' of the system. In functional dependence,
H = H(qj,Pj,t) = LPjqj -L(qj,qj,t)
j
dH
and for conservative system, dt = 0
or, H = constant and this 'H' represent the total energy of the system.
If 'H' does not involve time, it is said to be a constant of motion. In that
case, H represent the total energy of the system. But we should remember that
in another way it is possible that H may be a constant of motion but not the
total energy.
9.3 Concept of Phase Space
In construction of Hamiltonian of the system, we basically provide the
generalised co-ordinates and generalised momentum (both independent) in the
same footing and in that case, configuration space no longer remain adequate
to represent the overall specification of the system. The configuration space
provides the path of motion of the system in 6N dimensional space instead of
3N dimensional space. So in such configuration for all, each particle contributes
one dimension for each position and one for each momentum component. This
new space is called 'phase space'.
Regarding such space, we should always keep in mind that since the
equations of motion of the system which describe the motion of the system in
this new formulation will involve the basis of (qj' Pj' t), specification of initial
values of (qJ' p) at any instant on that path will fix the whole path of the moving
system. It can, therefore, be stated as, "there is only one possible path in phase
space". It shows that there is less arbitrariness about the path in phase space as
compared to the path in configuration space in Lagragian formulation. Basically,
the path in phase space almost refers to actual dynamical path. This is the
importance of phase space which is required to construct in Hamiltonian
formalism.
H = H(qj' Pj' t) = LPjqj -L(qj,qj,t)
J
Now dH = I(Pjdqj + qjdpj) - dL
j
But,
z:(
oL oL. J oL
dL= -dq+-dq +-
oqj J oqj J ot
dH = "(p, + "q .dp. - " oL dq. _ oL
j oqj' j j j oqj' j ot
j j j
Also,
H = H(qj' PI t)
,,( oR oH J oH
dH= L..J -dqj +-dpj +-
j oqj OPj ot
But since, Pj = ,we have from equation (1) and equation (2).
oqj
"[OH dq. + oH d
P
']+ oH = ,,[q.d
P
' _ oL dq .]- aL
L..J oq. j ap . j ot L..J j ) aq . } at
}}} j j
oL d(OLJ d() .
But also, aqj = dt oqj = dt Pj = Pj
... +PJr
J
+ -4}j+(: +
Since, dq. dp. all are arbitrary and independent,
J. }
we must have,
oH .
-+p'
oqj j
oH .
0; OPj -qj = 0
... (1)
... (2)
and
aH aL
-+-
at at
o
aH aH aH
So finally, we have, iJj = apj ; Pj = - aqj and at
197
aL
at
All these equations are called "Hamilton's canonical equations" and there
equations has important role to describe the motion of the system completely.
9.5 Hamilton's Canonical Equations in different Co-ordinate System:
Here, we will find the forms of Hamilton's canonical equations in several
co-ordinate system.
(i) In cartesian co-ordinate system.
The point has co-ordinates (x, y, z).
So the kinetic energy,
1 (.2 .2.2)
T = 2
m
x + y + z and potential energy V = Vex, y, z)
1 (. 2 . 2 2 ) v( )
L = T - V = 2m x + y + z - x,y,z
aT . aT. aT.
P
x
= ax = mx ,P
y
= ay = my , P
z
= ai = mz
:. Hamiltonian H = LPjiJj - L
= (xPx + YPy +ipz) - L
. . . 1 (.2 2 .2) v( )
H= xPx+YPy+zPz-2m x +y +z + x,Y,z
= mi
2
+ mi + mi
2
- ~ m x 2 + i +i
2
) + V(x,y,z)
1 ( . 2 . 2 .2) V( )
= 2m x + y + z + x,Y,z
1 (2 2 2)
= - P
x
+ P
y
+ P
z
+ V
2m
where,
aH P
x
. _ aH _ Py . aH P
z
x - --=-. y-----. Z=-=-
- apx m' apy m' apz m
All these equations are Hamilton's canonical equations in cartesian co-
ordinates.
(v) In cylinderical polar co-ordinates, the point has instantaneous co-ordinates
(r, <p, z)
198
where,
oT .
P
r
= or = mr
oT 2.
Pr.p = ocj> = mr cp
and P
z
= mi
:. Hamiltonian of the system,
1 (.2 22 .2) V( )
H=T+V= "2mr +r cp +z + r,cp,z
1 1 (2 2 2 2) ( )
H = -- P
r
+ r Pcp + P
z
+ V r, cp, Z
2m
oH Pr' oH oL
. = ---' p =--=--
r ,r or or
uP
r
m
. oH r2 Pr.p . _ oH _ oV
cp = oPr.p =---;;;-; Pr.p - - ocp - - ocp
oH Pz. oH OV
i = opz =-;;; Pj = -a;=-a;
All these equations are the equations of Hamiltonian canonical equations
in cylinderical polar co-ordinates.
(vi) In spherical polar co-ordinates, the point has co-ordinates Cr, e, cp). In
that case, the kinetic energy of the system is
T = +r
2
e
2
+r2 sin
2
8cj>2)
oT. oT 2'
P = -=mr Pa =-. =mr 8
r or ' 08
OT 2.2e'
Pr.p = ocj> = mr sm cp
Now the Hamiltonian of the system is
H = T + V = +r
2
e
2
+r2 sin
2
8cj>2) + V(r,8, cp)
_ 1 (2 ( )
- - Pr + 2""" + 2 . 2 + V r, e, cp
2m r r sm 8
oH
r = oPr
and
and
and also,
= Pr
m
. _ aH 1 ( 2 2 2 2) ay
Pr - -a;= 2m -,?Pe - r3sin2e Pcp -a;
Pr = __ + .p; )_ ay
mr
3
sm
2
e ar
2
= Pcp (sin
3
ecose)- ay
acp ae
. aH. ay
P = --=--
cp acp acp .
199
All these equations are Hamilton's canonical equations in spherical polar
co-ordinates.
9.6 Hamilton's Canonical equations from Hamilton's Intergral Principle:
12
From Hamilton's integral principle, oJ Ldt = 0
or,
But
11
L = -H .
j
oL = I(Pjoqj + qjoPj) - oH
j
8H
Here, taking 'H' which is not an explicit function of time t, at = 0 .
12
Since, f 8L dt = 0
12 12
But fp HJ dt = fp ~ 8 q .)dt
j j j dt j
II II
12
= [p .8q .]/
2
- f ~ p .)8q dt
j j II dt j j
II
12
= 0 - f P j 8q j dt
II
12 12
f Pj 8qj dt = - f P
j
8qj dt
So we finally get from equation (1)
Since, 8qp 8pj all are arbitrary,
8H . 8H
___ po =o=> Pj=--
8qj ) 8qj
8H . 8H
q.-- =o=> qj=-
} 8pj 8pj
,
Hamiltonian Formulation in Mechanics 201
These are Hamilton's canonical equations. Here, we should note that these
equations are only derivable from integral principle if and only if H be not a
function of time t explicitly, where as Lagrangian is also not function of t
explicitly.
9.7 Physical Significance of Hamiltonian of the System
The basic significance of Hamiltonian, H of the system are
(i) Like Lagrangian, H has also dimension of energy but in all circumstances,
it is not equal to total energy.
(ii) For conservative system, L P/lj = 2T
}
and in that case, only
H = LP/i} - L = 2T - (T - V)
= T + V = E = Total energy.
(iii) If qj be cyclic in Lagrangian, it is also cyclic in it Hamiltonian, because,
JL
for - = 0 , p. = const.
Jqj }
. 0 JH JH 0 . 1" H
p. = = - - - = q. IS cyc IC III
} Jqj .. Jq; }
(iv) The method of Hamiltonian formulation can determine energy of the
system but it does not require the individual indentification of force components.
9.8 Advantage of Hamiltonian Approach:
Hamiltonian approach of solving some mechanical problems has some
advantages over Lagrangian approach. These are,
(i) In Lagrangian approach, two variables qj and qj are not given equal
status because. q; 's are independent but q j' s are not independent
variables. But in Hamiltonian approach, co-ordinates (q) and momenta
(p) are placed at equal footing. This provides frequent freedom of
choosing co-ordinates. This is very important in solving the problem.
(ii) Hamiltonian approach providing 'equality of status' of co-ordinates
and momenta gives a convenient basis for the development of quantum
mechanics and the statistical mechanices.
(iii) Knowledge of Hamiltonian of a system is very important in quantising
a dynamical system.
9.9 Principle of Least Action:
For a dynamical system, the action 'A' or the action integral is defined as
12
A = f 2T dt in interval (ti' t
2
)
I)
Where T is the kinetic energy of the system.
Now the principle of least action states that the variation of A with time
i.e. IlA will become zero on the actual path as compared with some neighbouring
paths provided the Hamiltonian, H, is constant throughout that actual path.
To establish this principle of least action we will use an identity
/'if = of + jilt for any functionj(q), t)
So, the action is now given by

A = f 2T dt = f (H + L) dt = H(t2 - t
l
) + f L dt
where, H = constant along actual path.
. . A = H(t
2
- t
l
) + l(t
2
) - I(t
l
)
for 1 = f Ldt
M = H(llt
2
- Ill
I
) + IlI(t
2
) - Ill(t
l
)
H(llt2 - Ilt
l
) + 01(t
2
) + i(t2)llt2 - ol(td - i(tdlltl
[H Ilt];: + 0[I(t
2
) - I(t,)] + [i 1llJ;:
where, i
d (8L) 8L
But from Lagranges equation, dt 8qj = 8qj
1
Since,
12
of Ldl
II
So we finally get,
But H
(H + L)
M = [(H + L)Mt -[(H + L)lltt = 0
M = 0
This is the principle of least action.
203
9.10 Difference between Hamilton's Principle and the principle of least
action
Since both the principles are variational principles with fixed end points,
but in case of Hamiltons principle, the time interval (/
2
- (1) is prescribed in a
configuration, while in the principle of least action, there is no such restriction
on the interval (t2 - 11)' but the total energy between the end points is prescribed.
This is only the difference between Hamilton's principle and the principle of
least action.
9.11 Application of Hamilton's Canonical Equations:
(I) Simple Pendulum:
In this case, T = K.E = mPe
2
V P.E = mg/(1 - cos e).
Po
a ~ = mPe => e = Pe
as mP
So the Hamiltonian of the system,
..
..
and
2
H = Pee - L = Pe - (T - V)
mP
H
8
Pe
2
Pe 1 2' 2 ( )
~ - - ml e + mgl 1- cos e
ml"" 2
2 2
Pe _ ---.EL + mgl( 1 - cos e)
mP 2mP
2
Pe 2 + mg/(1- cose)
2ml
aH
Pe
-
= mP
ape
aH I . S
--=-mg sm
ae
For small e, sine = e
..
Pe
ml28 = - mgle
i.e.,
8+(02
e o for (0 = If .
This is the equation of motion of simple pendulum.
(2) Compound Pendulum :
Here, T = K.E ~ 18
2
(I = moment of inertia)
2
But since,
V P.E = -mgl cos e.
H
H
Pee - L = 8(:)-(T-V)
.(.) (1 '2 ) 1'2
e I8 - "2 Ie + mgl cos e ="2 Ie - mgl cos S
p2
-..!L - mgl cos e
21
Hamiltonian Formulation in Mechanics 205
and
for
Pe
oR Pe .
De = -I => Pe = Ie
- oR = _ mgl sin e = - mgl e
oe
Ie + mgle 0
o
co =
~ t
[for small e]
This is equation of motion of one dimensional linear harmonic oscillator.
(4) Motion under Attractive Central force:
1 (.2 22)
In this case, T = K.E = 2m r +r e
But
Also,
v
k
P.E= --
r
(k = constant)
R (fpr+8Pe)-L
aT. aT 2
P r = or = mr; Pe = oe = mr e
R
.2 22 (1 .2 1 2.2) k
mr +mr e - -mr +-mr e --
2 2 r
oR _ Pr 8 _ oR _ Pe
aPr - -;;; ; - aPe - mr2
oR p ~ k
--=---+-
or mr
3
r2
d
2
r _r(d8)2 =
dt
2
dt
This is radial equation for particle's motion under attractive central force.
8H
Again,
Po = -00 = 0
(for 8 to be cyclic)
m :t (r
2
e) = 0
i.e., r
2
e = Const.
This is another welknown result for such motion under central force.
SUMMARY
(1) Hamiltonian of the System:
H = H(p" q" t) = LPlij - L(qj,qj' t)
j
(2) For conservative holonomic System:
d
-(H) = 0, H = T + V = E = Constant
dt
(3) Hamilton's Canonical Equation :
q). = 8H p)' =_ 8H 8H =_ 8L
8p j , 8q j' 8t 8t .
(4) It qk be cyclic in Lagrangian, it will also become cyclic in Hamiltonian.
(5) For dynamical system, the action integral
A = = J2T dt
tl ) tl
(6) Principle of Least Action:
t2
I1A = 11 f 2T dt = 0 along actual path of the system.
tl
Worked Out Examples
Example 1 : Write down Hamiltonian for a charged particle in an
electromagnetic field.
Ans : Lagrangian for such particle is
(
L-)
L = T-q
where,
Now
T
1 2
K.E =-mvj
2
<!> Scalar potential
Vector potential
A
or
Pj =-;-:-
uqj
aT q
-=mv.+-A
avo J c J
J
:. Hamiltonian of the particle
H = - L= mVj + ! A
j
)- L
J J
" + " !I v .A . - L
J J J
J J
H
1
2'mv2 +q<p .
207
Example 2 : Find equation of motion of charged particle in an
electromagnetic field.
Ans : For such case,
P
and H
where, L
-q<p +;(V.A)
H
2 q-- 1 2 q--
mv +-v.A--mv +q<p --v.A
c 2 c
( )
2
1 q -
- p--A +q<p
2m c
So from canonical equations of motion,
r =
v= aH
ap m c
-p
V(H) = qV(<p) +
qV<p -!iV(v.A)
c
p
Example 3 : Find equations of motion of particle moving near the surface
of earth.
Ans : Let us consider z axis along upward vertical direction, the kinetic
energy is
T = m( x
2
+ i + ;i2)
Further the applied force on the body is its weight acting in negative z
direction, i.e.,
av
F = F
z
- mg = - az
This gives V = mgZ, on setting additive constant to zero
Now Lagrangian is
So that
giving
1 (.2 2 .2) Z
L T - V = "2 m x + y + z - mg
aL
ax
Px
m
Also, Py = my, pz = mi
. pz
z=-
m m
Hamiltonian for such a system is conserved, i.e.,
H T+V
1 (2 2 2)
- Px + P
y
+ P
z
+ mgZ
2m
giving equations of motion
so we finally get,
_ oH = 0 p = _ oH = 0
ox y Oy
oH
--=-mg
OZ
oH P
x
. oH P
y
. oH Pz
--=-- Y=--=- Z=-=-
oP
x
m' OPy m' oZ m
dH oH
Example 4 : Show that dt = at
Ans : Since H H(qj' Pj' t)
dH
dt
dH
dt
dH
dt
oH
ot .
EXERCISES
209
1. Define Hamiltonian of the system and show that it is constant of motion for
the conservative system.
2. Obtain Hamilton's canonical equations from Hamiltons principle.
3. Obtain Hamilton's canonical equation from Lagranges equations of motion.
4. Show that if qj be cyclic in Lagrangian, it will also be cyclic in Hamiltonian.
5. What are the advantages of Hamilton's formulation over Lagrangian
formulation.
6. State and explain principle of least action.
7. The Hamiltonian of a dynamical system is given as H = qp2 - qp + bp
where, b is constant.
Solve the problem and obtain equations of motion of the system.
8. A particle of mass m exists in a force field of potential V. Write the Hamilton's
equations of motion in spherical polar co-ordinates.
9. A particle moves under the influence of gravity on the frictionless inner
surface of a cone x2 + y2 = C2Z2 (C = constant)
Obtain the equations of motion.
-:0:-
Chapter-10
Canonical Transformations
10.1 Introduction to Canonical Transformations
Although in Hamiltonian fonnulation, the way of solving any given problem
is straight forward and is also very similar to the way of Lagrangian fonnulation,
but in Hamiltonian formulation, equal status accorded to co-ordinates and
momenta, as independent variables assures a greater freedom in selecting the
physical quantities to be designed as 'co-ordinates'. But there exist one type of
problem in Hamiltonian formulation. If we take system for which the
Hamiltonian of the system is constant and all the co-ordinates qj'S are cyclic
for which all momenta Pj'S are all constants, then that problem can easily be
solved. But it is a rare case. Here the canonical transformation or contact
transfonnation is that types of transfonnation by which we can find out one set
of co-ordinates from another set of co-ordinates such that all the co-ordinates
in the new set are cyclic.
Such transfonnation from one set of co-ordinates qj to a new set Q
j
can be
expressed as Q = t)
Also the transfonnation of the independent co-ordinates and
momenta (qJ' p) to a new set P) can be represented in the fonn
= O;(qJ' Pj' t)
P
j
= P/qi Pi t)
This transfonnations in which P
j
are canonical, are called contact or
canonical transfonnations of phase space. They are characterised by the property
that they leave the fonn of Hamilton's equations of motion invarient.
. oK oK
Now if there exist a function Pj' t) such that Qj = oP. ' Pj = - oQ.
} }
then and are called 'canonical co-ordinates' and the transfonnations
qj P
j
where,
= Pi t) Piqi Pi t) are known as canonical transfonnations.
210
Canonical Transformations 211
Here 'K' plays the role of 'Hamiltonian' and the new function K(Q., P, t)
is called 'Kamiltonian' of the system. :J J
Thus for such new set of co-ordinates, P) must satisfy Hamilton's
variational principle. Therefore,
- K(Q
j
, Pj' t)] dt = 0
11 J
... (1)
But also we have, for old set of co-ordinates.
. .. (2)
Let us now take
... (3)
where
for of(t
2
) = of(t
l
) = 0 at the end points.
So for such relation (3), both the equations (1) and (2) will remain valid
simultaneously.
Here the function F is called the generating function and there are four
different possibilities for F as,
F = F\(qj' t);F = F
2
(qj' Pj' t)
F = F
3
(PJ' t); F = F4(Pj' Pj' t)
Case-I:
Now for F = t), we have from equation (3), LP/lj - H(qj,Pj,t)
j
L
( ) L of\. L of\ OFl
= PQ.-KQP.t+ -q.+ -+-
J J J' J' J oQ. ot
j j U J J
... (4)
Comparing the coefficients of qj' OJ from both side of equation (4), we
get
OFI
p. =
}
oqj
... (5)
oFl
P. =
}
oQ
j
... (6)
and K=
H+ oF
1
... (7)
at
On solving equation (5), we fmd, Q. = Q.(q., PI t) ... (8)
Which when substituted in equation
J
(6), ~ i j s
P. = P.(q.,p., t) ... (9)
Equations (8) and (9) are ili/ de;ired transformation equations and also the
relation (7) gives connection between old and new Hamiltonian.
Case-2 :
For F = F
2
(qj' P
j
, t), we should now apply Legendre transformation. Let
us now at first discuss this transformation in brief.
Let, f = fix, y)
of of
df = ax dx + Oy dy = u dx + v dy
Let us now change the basis of description from (x, y) to independent
variables (u, y), that for any new function g(u, y) defind by the equation
g(u, y) = fix, y) - ux
We have, dy = dF - udx - x du
= udx - vdy - udx - xdu
= vdy-xdu
Which is exactly in the form desired so that we can now write
og og
dg = Oy dy + AU du = udy - xdu
og og
v= - x=--
Oy' au
where,
of .
Thus if u = ax ' then the relatIOn
g(u, y) = fix, y) - ux
would be appropriate to effect a change from the basis (x, y) to (x, y).
Now we apply this to the present case. Here since
_ oF
1
P. =
} OQ
j
get
Putting u = -Pj' x = 0.,., y = qj' g = F
2
,j= Fl with one more variable t, we
F1(qj, Qj, t) + IPjQj
j
I
( ) I aF
2
I aF
2

= PQ-K Q. p. t + -q.+ -po
} } }' }' aq.} ap}
j j} j}
aF
2
I' I' +-- p.Q.- p.Q.
at } } } }
j j
I
K( ) aF
2
I
aF
2 I
aF
2 .
= - p.Q.- Q. p. t +-+ -q.+ -p.
. }} }' }' at . aq}.} . ap}. }
} }}
or, I(aF2 - pj)qj + I(aF2 -Qi)P
j
+ H+ aF
2
-K = 0
j aqj j aP
j
at
Since qj' P
j
both are independent, we have from equation (11),
aF
2
p. =
)
aqj
aF
2
Q. =
J
aP
j
and K=
H+ aF
2
at
and also, equation (12) can be solved to give
P. = P. (qp p., t)
Which when substinited ii equadon (13) gives on solving,
0.,. = 0.,.( qp Pj' t)
which are desired transformations.
Case-3 :
'" (10)
'" (11)
... (12)
'" (13)
'" (14)
'" (15)
... 16)
Here ~ r F = F 3(Pj Qj' t), we can connect F 3 with F 1 by Legendre
transformatlOns as .
Then LP/Jj-H
j
But
elF
3
dt
F)(qj,Qj' t) - LPjqj
j
"p.Q. -K+.!!..-+(F3 +" P.q.)
~ )) dt ~ ) )
j j
" p.Q . _ K + elF
3
+ " P .q . + "p .q' .
~ ) ) dt ~ ) } ~ ) )
} } }
L
L oF
3
. L oF
3
oF3 L . L .
= P.Q.-K+ -p'+ -Q.+-+ p.q.+ p.q.
. ) ) . op),) OQ).) at .)) .))
)) ))
So comparing the coefficients of Q
j
and P
j
, we get
... (17)
q. =
_ oF3
)
OPj
... (18)
and K=
H+ oF3
at
... (19)
This equation (17) also states that ~ = ~ ( q i Pi' t) and knowing ~ from
equation (17) we can determine Pj from equation (18) in terms of (qi' Pi' t).
Case-4 :
In this case, we choose F = F ipi' Pi' t)
we have p.= of) . P.= _ of)
') :::. ') ;)
uqj V'<-j
So from Legendre transformation, we get
F) (qj' Qj' t) + LPjQj - LPjqj
j
L
elF)
PQ.-K+-
) } dt
j
... (20)
= ,",p.Q. - '"'P.Q. + '"' P.q.J
)) dt I L.J ) ) L.J))
) \ I )
= -K+ - LP)Q) - LP)Q) + LP/J) + LP)q)
) ))))
dF
4
L' L'
:. -H = -K+-- p.Q. + p.q.
dt .)) .))
) )
L
oF4
. L oF4
OF4 L' L .
=>-H= -K+ -p.+ -p.+-- p.Q.+ p.q.
. op).) . oP).) at .)) .))
)) ))
So comparing the co-efficients of p) and P J , we get
qj
_ OF4
op)
... (21)

OF4
oP)
... (22)
H =
K- oF4
at
... (23)
Equation (21) states that P. can be determined in terms of (q., p., t).
. ) J J
Knowing Pp we can determme from equation (22) in terms (qp Pj' t)
10.2 Hamilton-Jocobi Method:
As we have discussed earlier that canonical transformations provide a general
procedure for easy solution of mechanical problems, but there are two ways of
affecting such transformations.
One way to obtain the solution of mechanical problem is to transform old
set of co-ordinates into new set of co-ordinates that are all cyclic and consequently
all momenta are constant. In this way, the new equations of motion can be
integrated to give a solution and is adopted when Hamiltonian H is conserved.
Another way to obtain the solution is to seek canonical transformation from
co-ordinates and momenta (q, p) at time t to a new set of constant quantities
which may be zd (d = degree of freedom) initial value (qo' Po) at t = O.
The transformation equation will be therefore
q = q(qo' Po' t)
p = p (qo' Po' t)
The greatest advantage of such a transformation is that we are here
benefitted in two ways. That is in obtaining transformation equation we arrive
at that solution as well. This way is due to Jacobi. Thus Jacobi's way is a
transformation as well as a method in it self. That is, in it we do not first
transform and then apply Hamilton's equations of motion but while performing
transfommtion, we are arriving at the result as well.
This procedure is more general because it can be applied, in principle, to
the cases for which Hamiltonian involves the time.
Under this kind of transformation, as we k.tl.OW, the new set is of constant
co-ordinates (initial values qo' po). If we require that the transformed
Hamiltonian or Kamiltonian K is zero, then new equations of motion (involving
co-ordinates of transformed ste Pi' Q) are
OJ = oK =0
oP
j
_ oK =0
oQj
which ensures that (Q,;, P.) are constant in time, is both the co-ordinates
are rendered cyclic in the
of
But we have K = H +-
, at
more, F is generating function,
we have, for K = 0
of
H+- = 0
at
H(qj' Pj,t) + (qj,Pj,t) = 0
But
Pj = (qj,Pj,t)
We get, H(qj' :F ,tJ + of (qj' Pj ' t) = 0
uqj at
for j = 1, 2, ..... a.. (1)
This equation (1) is a partial differential equation in (d + 1) variables ('a.'
for q/s and one for t. This partial differential equation (1) is called 'Hamilton-
Jacobi's equation', the solution of which is called 'Hamilton's principal function'
(S).
To solve Hamilton-Jacobi equation (1) we should integrate this equation
(1) to obtain its solution S.
Since this equation (1) is a partial differential equation in (a. + 1) variables,
a complete variable must involve (a. + 1) independent constants of integration 0.
1
,
0.
2
... a.
u
+!. We only use derivatives of S with respect to qj and t will appear in the
equation. This means (S + a) will also be the solution of equation (1), where a
any additive constant. Out of (a + 1) constants of integration one should therefore
be an additive constant to S but since this constant will have no effect in
tranformation, we can take S to be involving only constants (a
p
a
2
. aa)'
Hence a complete solution of equation (1) can be written as
S = S(q), a), t) (j = 1, 2, .,. d) .. , (2)
Where none of d (constants) is solely additive. We can take these no. of
constants as the new momenta Pi' i.e., a) = P
j
.. , (3 J
Therefore, d transformation equations can be written as
and under specific initial conditions, we get
a) = aiqj' Pj' t)
P. = P. (q., p., t)
Wh
. . J J))
lch IS one of the transformation equations.
.. , (4)
.. , (5)
.. , (6)
The 2nd transformation equation which provide the new constant coordinates
appear as
oS ( oS )
Q.= ~ . = - qj,aj,t) = -(qj,aj,t
J ) oP
j
oa j
.. , (7)
and the equation (7) can then be 'turned inside out' to furnish qj in terms
of aj' ~ j t as qj = qj ( ~ j a
p
t) .. , (8)
Which solves the problem by giving the co-ordinates as function of time.
10.3 Application of Hamilton-Jacobi method to the particle falling freely
We can now apply Hamilton-Jacobi method to study the motion of a particle
falling freely under gravity.
Here the Hamiltonian of the system is given by
1 2 p2
H=T+V= -mY +mgy=-+mgy=E ... (1)
2 2m
where, E is total energy.
Here we take y-axis in vertical direction.
oS
Now let p= oy' we get
1 (OS)2
H = H(y,p,t)=- - +mgy
2m oy
Now for new Hamiltonian K = 0, we also have
oS
H+- =0
ot
... (2)
:. _1_( as)2 + mgy + as = 0
2m ay at
Writing the solution of above equation in the form
S(y, a, t) = W(y, a) - at.
as aw
we find that
ay ay
as
and
at = -a.
Then equation (3) a'ssumes the form
_1_(aw)2 +mgy-a = 0
2m ay
aw
BY =
W =&f dy+C
where C is the constant of integration.
Therefore, the (4) is
S = &f .dy+C-aJ
= J f a mgy t
_1 mgy)
lm
1 2 a
y = --g(p+t) +-
2 mg
Let the initial conditions be such that at t = O.
Y = Yo,P = 0
so that from equation (5), we get
aw
p = ay = =0
giving a = mgyo for y = Yo
... (3)
. .. (4)
... (5)
... (6)
which when substituted in equation (6) gives
1 g( \2
Y = l3+
t
) +Yo
L.
Further at t = 0, Y = Yo so that from equation (7).
219
... (7)
We infer that f3 = 0 and hence finally the equation of path of a particle
falling martially under the action of gravity comes out to be
1 2
Y= -2
gt
+ Yo.
10.4 Hamilton's Characteristics function:
Hamilton-Jacobi equation for Hamilton's principal function Seq, P, t)
becomes

Let us assume the solution of the form
S(qp ai' t) = W(qj' a) - alt
from which it follows that
as oW as
oqj = oqj and at =-al
So from equation (1), we get
H(qj' OW) = a
oqj I
which is time independent
Now let W W(qj' a)
then
dW
dt
oW.
= L-qj
. aqj
S, J h
mce a
j
s are constants, we now ave
oW as
aqj
-=p.
oqj J
dW
LP/lj
so that
dt j
W
s( action A.
... (1)
... (2)
Thus W is identified as action A. Here this function W is called "Hamilton's
charact<!ristics function".
10.5 Action and Angle Variables:
We know that in classical mechanics the action or the phase integral is
defined as
A= fIp/iidt= f2Tdt
}
Now we will try to extend Hamilton-Jacobi method so as to apply it for
the solution of the problems in periodic motions. Here we will not choose a
j
as the new momenta, but we will define constant J
j
which forms a set of 'd'
no. of independent functions of the a/s and these J/s are called action variables.
Now we have from Hamilton-Jacobi differential equation
Pj = (qi,aj)
... (1)
which gives Pj = p;(q;, a)J = 1,2, ... , d.
We shall now define the phase integral or action variable conjugate to the
co-ordinate qj by the integral
J
j
= f Pjdqj ... (2)
where the integration is to be carried over a complete period of oscillation
or rotation cycle of q)" I case qj is cyclic, the conjugate Pj is the case
21t
and the J
j
= Pj f dqi = 2rtpj ... (3)
o
Compairing this equation (2) with the action given above, it becomes quite
obvious why J) is designated as action variables.
So we also have,
f
aW(qj,a j)
J = dq
j aqj }
... (4)
and then after integrating for qj' we get
J
j
= J A a j ) J = 1, 2, ... , d.
and also, a
j
= a A J j )
So, ultimately, we can define Hamiltonian characteristics function W as
W = W(qj' J) ... (5)
The generalised co-ordinates conjugate to J
j
are called angle variables Wj
and are given by
0). = aw (qj,a j)
} aJ j
... (6)
Since J
j
has dimension of angular momentum, co-ordinate conjugate of it
should be an angle and so its name is 'angle variables'.
10.6 Application of Action Angle Variables to Harmonic Oscillator Problem
For such a problem if we consider the single variable oscillator (medium),
there is only one action variable
J = f pdq = f (q, a) dq
... (1)
But we previously have
aW(q,a) = .f,;;kJ2a-l
aq k
aw aw &[ 1 2
[From equation (5) of section 10.3, aq = ay = 2m V a - 2. kq
Because, for medium oscillator problem, the potential energy term mghy
1 2 1 2 aw
should be replaced by 2.
xy
=2.
kq
.:. aq =vmkfK"-q ]
We have from equation (1)
J =
-l}q
i.e., q
p' a
T
sm
we have, J
fi2n
2a ; f cos
2
ada
0

a =
2n m
Since, the system is conservative, H is constant. If we choose this constant
as a,
{K
2n V-;;;
:. The frequency of Oscillation is therefore
ill = : = 2 ~ ~
Which is the angular frequency of harmonic oscillator.
10.7 Poisson's Bracket:
For any two dynamical variables x and y, if x = x(q., P.), y = y(q, p.) the
. ] ] J ]
POlsson Bracket of x and y IS defined as
)'( ox oy ox By J
[x, y] = 7 oqj OPj - OPj oqj
Now we can use this new rotation of Poisson's Bracket to obtain the time
derivative of any dynamical va:iable F(qj' Pj' t) as
dF dF L
oF
. L
oF
. of
(
t) -;::--q. + -p. +-
-d qj'Pj' = oq ] on. ] ot
dt t j ] k YJ
But
The basic properties of such Poisson Bracket are now given below-
(i) [x, y] = -[y, x]
(ii) [x, x] = 0
(iii) [x, y + z] = [x, y] + [x, z]
(iv) [x, yz] = y[x, z] + [x, y]z.
Although, we apply such Poisson Bracket notation to the dynamical variables
x(qj' p), y(qr p), but we can also apply such notion to the variables qj' Pj also.
That Poisson Brackets are known as fundamental or basic Poisson Bracket.
These fundamental Poisson Brackets are now given below-
(i) [qJ' q) = [Pl' p) = 0
(ii) [qj' p) = Ojj = 0li = 1 = [qi' Pi]
(iii) [qi' p.] = 0,; = 0 for i -:t:. j
= 1 for i = j
(iv) [qi' q) = [P" p) = 0
,,( oqi oqj oqi oqj I
Because, [q" q.] = ~ ~
] x ~ oqk oPk oPk oQk J
and
But
The advantage of this Poisson Bracket notation is that the equation of motion
of the particle can be expressed in terms of Poisson Bracket form.
As we have already mentioned earlier that the total time derivative of any
dynamical variable F(qp Pj t) can be expressed as
. dF aF
F = - = [p, H] +-
dt at
aF . [ ]
Then for F not a function of time explicity at = 0 and F = F, H .
Now from earlier discussion of fundamental Poisson Bracket
Similarly,
aH
-a (for i = j)
'Pi
[qp H]
fpp H]
These are the equation of motion in 'Poisson bracket form'.
10.8 Poisson's Theorem:
We have for any dynamical variable F(qp Pp t)
dF = [F H]+ aF
dt ' at .
So if F is a constant of motion and F does not contain time explicity, then
dF=O aF=O
dt 'at
[F, H] = 0
This is Poisson's theorem.
So this theorem states that "all functions whose Poisson bracket with
Hamiltonian vanish, will be constants of motion and conversely Poisson brackets
of all constants of motion with Hamiltonian must vanish."
10.9 Jacobi's Identity
For any three dynamical variables (x, y, z) we have
[x, [x, z]] + y, [z, x]] + [z, [x, y]] = 0
This identity is known as "Jacobi's identity".
We can show such identity in the following manner.
let P = [x, [y, z]] + [y, [z, x]]
= [x, [y, z]] - [y, [x, z]]
=
j aqj apj aPj aqj j aqj apj apj aqj
=
j aqj apj j aPj aqj j aqj apj j aPj aqj
But we have an identity
[x, yz] = [x, y]z + [x, z]y
...
=0
-:j [Y' :j]+ :;; [y, :j]}
But we have also another identity
= [aA 'B]+[A' aB]
ax ax ax
So, using it we have
P = [x, [y, z]] + [y, [z, x]]

j aqj aPj apj aqj
= - [z, [x, y]]
:. [x, [y, z]] + [y, x ]] + [z, [x, y]] = 0
This Jacobi's ideptity is established.
Let us now consider that z = H = Hamiltonian.
:. [x, [y, H]] + [y, [H, x]] + [H, [x, y]] = 0
If we now consider that x and yare both constants of motion then [x, H]
[y, H] = 0
:. [H, [x, y]] = O.
which means that [x, y] is also constant of motion. Hence the Poissons
bracket of two constant of motion is itself a constant of motion this is an outcome
of Jacobi's identity.
10.10 Lagrange's Brackets:
Lagrange's Bracket of (a, b) with respect to the basis (qj' p) is defined as
I[a
q
j
aPj apj a
q
;)
(a, = j aa' - aa
The basic characteristics of such brackets are given below
(i) Under canonical transformation, Lagrangi's bracket will remain
invarient for the change of basis from (q, p) to (Q, P).
i.e., (a, p) = (a, P)
(ii) Lagrange's brackets do not obey the commutative law,
i.e. (a, = a)
(iii) (qi' q) = 0 = (Pi' p)
(qi' p) = oij
Which are fundamental Lagrange's brackets.
Because, (qi' p)
l
'since, 8qu = 8pu = 0)
8pj 8qi
Here we should note that for any two dynamical variables (u, v)
211
~ u / , Ui)[U/,Uj] = Olj
1=1
This is the relation between Lagrange and Poisson brackets. This relation
between the two types of brackets holds even if the coordinates are not canonical
and is true for any arbitrary transformation from (qj'Pj) to (qj, pj).
10.11 Liouville's Theorem
We all know that the dynamical state of a system at some instant of time
can be represented by a point in the phase space. This point will not be stationary
but will move along a definite path or trajectory which can be determined from
the equations of motion,
. 8H . 8H
qj = 8pj and Pj = - 8qj .
Where, H is Hamiltonian of the system.
As a result of this motion, the phase space density of changes with time.
To find ~ at a given point in phase space, Liouville's theorem states that,
(i) The conservation of density in phase space will occur is ~ = 0 .
(ii) The volume at a disposal of a particular number of phase points is
conserved through out the phase space, is the principle of conservation
d
of extension in phase space occurs for which dt (oV) = 0 .
This two statements are consistent to each other since, phase space no. oN
= poV = const.
d
-(pov) = 0
dt
dp d
==> -oV + p-(oV) = 0
de dt
Since dr = 0 by the conservation of phase space density
= 0
dt
Which is the principle of conservation of extension in phase space.
SUMMARY
227
1. Canonical transformations from old set (qj' p) to new set (Qj' P) are given
by
Q
j
= Q/qj' Pj' t)
P
j
= P/qj' Pj' t)
. oK p. =_ oK
where, Q
j
= oP.; J oQ.
] J
and K = K(Qj' Pj' t).
2. For generating function 'F' for canonical transformation
LP/lj - H(qj,Pj,t) = - K(Qj,Pj,t)+ dF
. . &
] ]
where, for four possibilities,
F = F
l
(%, t), F = Fiqj' Pj' t)
F = F
3
(Pj' Q
j
, t), F = F4(Pj' Pj' t).
3. For Fl = Fiqj' Q
j
, t),
. = OF2 . p. = oFl . K = H + oFI
P J oq j' J oQ j , ot
4. For F2 = Fl(qj' Pi' t)
F2 = Fl(qj,Pj,t) + LPjQ
j
and
j
K = H+.?F3
ot
(By Lagendre transformation)
6. For F4 = F
4
(P/ P
J
, t)
F4 = F
1
(qj, Qj, t) + LPjQj - LPjqj
j j
_ aF
4
aF
4
aF
4
and qj--- Q.=- K=H+-
ap j' ) ap 4 ' at
7. HamiIton-Jacobi's Equation:
H[q), aF ,t)+ aF (qj,Pj,t)=O
aqj at
for j = 1, 2, 3, .... d.
8. Hamilton's principal function, S = S(qp up t).
as
where, Pj = aqj , P
j
= u
j
as as
Q
j
= ~ j =-(qj,Uj,t)=-(aj,Uj,t)
aP
j
aUj
9. Hamilton's characteristics function, W = W(qp u)
where, w: f( Pj,q j )dt : action (A)
as aw
andp=-=-
) aqj aqj
10. Action variable, J j = f Pjdqj
11. Poisson's Bracket:
[X,Y]= L [ ~ ~ - ~ ~
j aqj apj apj aqj
12. Equation of motion through Poisson's Bracket:
dF = [F, H] + aF
dt at
cjj=[qj,H], Pj=[pj,H]
13. Jacobi's Identity:
[x, [y, z]] + [y, [z, x]] + [z, [x, y]] = 0
14. Lagrange's Bracket:
Worked Out Examples
1. Show that the transformation
1 (2 2) Q _ -l(q). .
P = 2" p + q , - tan p IS canoncial.
229
Ans : Here, K = K(Q, P) and the generating function does not involve
time.
8H oH
..
P =
-- q=-
... (1)
oq , op
But p
=
_ op . + op Q. . = oq p + Q
op P oQ ,q 8P oQ
... (2)
oH oK OP oK oQ oH oK OP oK oQ
oq = oP oq + oQ aq' op = 8P op + oQ op ... (3)
From the given transformation equations, we have
oP oP oQ q oQ p
- = p - = q - = - = ---,-"'--..,...
8p 'oq 'op p2 +l oq p2 +q2
Also, different transformation equations w.r.t. P and Q resp. we find
1
= op + oq I 0 _ ( oq _ op) ~ 2 + 2 )
P oP q oP , - P oP q oP / I P q
O
= ~ + op 1 _ (p oq _ op) ~ 2 + 2)
P8Q qoQ' - 8Q qoQ t
P
q
Solving simultaneously, we find
op
oP
p . '. q . .
p = 2 P - qQ q = 2 2 P + pQ
p2+q 'p +q
.. , (4)
oH oK p oK oH oK q oK
= q-+ -=p----=---
oq oP p2 + q2 oQ' op 8P p2 + q2 oQ
230
Thus from equations (1), (4), (5) we have,
p . .
(
2 2)P-QQ
p +Q
aK q aK
p ap - (p2 +q2) aQ
Solving these simultaneously, we find
oK . aK
P = -aQ Q=ap
So P and Q are canonical and the given transformation is canonical
3. Show that the generating function for the transformation
I q
p = Q' q = PQ2 is F = Q .
_ of p __ of
ADS: We have P - oq' - oq - Q - Q2
Both these equations on integration and taking for a particular case, the
q
constants of rntegration equal to zero, give F = Q .
3. Solve the Hamilton-Jacobi equation for the system whose Hamiltonian
is given by
p2 I.l.
H=---
2 q
ADS: We have from HJ equation,
_1:+ oS = 0
2 aq q at
Let the solution is S = Jet) + <p(q)
. of a<p r
I.l.
0
.. at 2 oq
of
a<p r

at q 2 oq
Where a is a constant
aF
I.l.
Now for
at a
I.l.
a
Canonical TransJormations
J=
1::(
a
and
o<p r =

a 2 oq
a
and on integration we get,
+[ -q) r
4. Show that [u + v, w] = [u, w] + [u, w]
_ v) oq _ o(u+ v) ow]
Ans : [u + v, w] - oq. Op. op. oq.
j } } } }
= oW ow]+ Ow Ow]
j oqj OPj OPj oqj j oqj OPj OPj oqj
= [u, w] + [v, w].
EXERCISES
231
1. What is canonical transformations? Show that a canonical transformation is
given by
1
q = (+)2 cos(21tQ)
21t mv
P = -(2vmp)1I2 sin (21t Q)
reduces the Hamiltonian for a linear oscillator
+n2mN]
to a new Hamiltonian which is a function of generalised momenta P only.
2. If the transformation equations between two sets of co-ordinates are
P 2(1 + ql/2 cos p) ql/2 sin P
Q = log (1 + q1l2 cos p)
then show that
(0 the transformation is canonical, and (iO the generating function of this
transformation is F3 = -(e
Q
- 1)2 tan p.
3. Give an account of the Hamilton-Jacobi theory and illustrate it by applying
it to the motion of one dimensional harmonic oscillator.
4. What is Poisson bracket of two dynamical variables. Show Jacobi identity
regarding it.
5. If [<p, \jI] be the Poission bracket of <p and \jI then prove that
; [<p, \jI) = [ : ~ , \jI ] + [ <p, : ~ ]
6. What is Lagrange's bracket of two variables.
7. How can you obtain equations of motion in Poisson bracket from.
8. What do you understand by Hamilton's characteristics function? Use it to
solve the problem of projectile motion.
9. What do you mean by Hamilton's principal function? Use it to solve the
problem of a particle falling freely under the action of gravity.
10. Show that the transformations
(a) P q cotp
Q
10' s i n p
(b) q
..fiP sinQ
p
..fiP cosQ
are canonical.
11. Using the fundamental Poisson brackets, find out the following Poisson
brackets
(i) [Q, F(P)], (ij) [P, g(Q)].
-:0:-

Chinmoy Taraphdar - The Classical Mechanics (2007)

Enviado por

Dados do documento

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Chinmoy Taraphdar - The Classical Mechanics (2007)

Enviado por

Direitos autorais:

Formatos disponíveis

The Classical Mechanics

Você também pode gostar