Ordinary Differential Equations - BAMAT-201

ORDINARY DIFFERENTIAL
EQUATIONS
BAMAT-201
Self Learning Material
Directorate of Distance Education
SWAMI VIVEKANAND SUBHARTI UNIVERSITY

MEERUT-250005
UTTAR PRADESH
SIM Module Developed by :
Reviewed by :
Assessed by:
Study Material Assessment Committee, as per the SVSU ordinance No. VI (2).
Copyright © Laxmi Publications Pvt Ltd.
No part of this publication which is material protected by this copyright notice may be
reproduced or transmitted or utilized or stored in any form or by any means now known or
hereinafter invented, electronic, digital or mechanical, including photocopying, scanning,
recording or by any information storage or retrieval system, without prior permission from
the publisher.
Information contained in this book has been published by Laxmi Publications Pvt Ltd and has
been obtained by its authors from sources believed to be reliable and are correct to the best
of their knowledge. However, the publisher and its author shall in no event be liable for any
errors, omissions or damages arising out of use of this information and specially disclaim and
implied warranties or merchantability or fitness for any particular use.
Published by : Laxmi Publications Pvt Ltd., 113, Golden House, Daryaganj, New Delhi-110 002.
Tel: 43532500, E-mail: [email protected]
DEMC—
Typeset at: Printed at:
Edition: 2020
CONTENTS
UNIT I
1. DIFFERENTIAL EQUATIONS 1
 Differential Equation ......................................................................................................................................................... 1
 Formation of a Differential Equation Whose General Solution is Given ....................................................................... 4
 Solution of a Differential Equation ................................................................................................................................ 13
 Initial Value Problem ...................................................................................................................................................... 16
 Solution of a Differential Equation by The Method of Separation of Variables ........................................................... 17
 Homogeneous Differential Equations and their Solution .............................................................................................. 33
dy
 Solution of Linear Differential Equation + Py = Q, where P and Q are functions of x or constants ................... 50
dx
 Summary ......................................................................................................................................................................... 68
UNIT II
2. EXACT DIFFERENTIAL EQUATIONS 70

 Introduction ..................................................................................................................................................................... 70
 Theorem ........................................................................................................................................................................... 70
 Equations Reducible to Exact Equations ........................................................................................................................ 74
UNIT III
3. LINEAR DIFFERENTIAL EQUATIONS OF THE FIRST ORDER

 Definition ......................................................................................................................................................................... 81
dy
 To solve the equation + Py = Q, where P and Q are functions of x only (Leibnitz’s Equation) .......................... 81
dx
 Bernoulli’s Equation (Equations Reducible to the Linear Form) .................................................................................. 88
 Differential Equations of the First Order and Higher Degree ....................................................................................... 96
 Equations Solvable for p ................................................................................................................................................. 96
 Equations Solvable for y ................................................................................................................................................. 98
Self-Instructional Material i
 Equations Solvable for x ............................................................................................................................................... 100
 Clairaut’s Equation ........................................................................................................................................................ 102
4. LINEAR DIFFERENTIAL EQUATIONS OF SECOND AND HIGHER ORDER

 Definitions ..................................................................................................................................................................... 104
 The Operator d .............................................................................................................................................................. 105
 Theorems ....................................................................................................................................................................... 105
 Auxiliary Equation (A.E.) ............................................................................................................................................. 106
 Rules for Finding the Complementary Function ........................................................................................................ 1107
1
 The Inverse Operator
f ( D ) ....................................................................................................................................... 111
 Rules for Finding The Particular Integral ..................................................................................................................... 112
 Method of Variation of Parameters to Find P.I. ............................................................................................................ 127
 Homogeneous Linear Equations (Cauchy-Euler Equations) ....................................................................................... 131
 Legendre’s Linear Differential Equation ...................................................................................................................... 134
 Linear Differential Equations of Second Order ........................................................................................................... 137
 Complete Solution in terms of known Integral ............................................................................................................ 137
d2 y dy
 To find a Particular Integral of +P + Qy = 0 ............................................................................................. 138
dx 2 dx
 Removal of the First Derivative (Ruduction to Normal Form) ................................................................................... 146
 Transformation of the Equation by Changing the Independent Variable .................................................................... 153
 Method of Variation of Parameters ............................................................................................................................... 160
UNIT IV
5. POWER SERIES SOLUTIONS

 Introduction ................................................................................................................................................................... 167
 Definitions ..................................................................................................................................................................... 167
d2 y dy
 Power Series Solution, when x = 0 is an Ordinary Point of the Equation 2
+ P(x ) + Q(x)y = 0 ................. 169
dx dx
 Frobenius Method: Series Solution when x = 0 is a Regular Singular Ppoint of the Differential Equation
d2 y dy
2
+ P(x ) + + Q(x)y = 0 ................................................................................................................................... 177
dx dx
6. DIFFERENTIAL EQUATIONS
 Introduction ................................................................................................................................................................... 195
 Legendre’s Function of First kind Pn(x) ...................................................................................................................... 197
 Legendre’s Function of Second kind Qn (x) ................................................................................................................ 197
 Solution of Legendre’s Equation .................................................................................................................................. 198
 Generating Function for Pn(x) ...................................................................................................................................... 198
ii Self-Instructional Material
 Rodrigue’s Formula ....................................................................................................................................................... 202
 Recurrence Relations .................................................................................................................................................... 206
 Beltrami’s Result ........................................................................................................................................................... 208
 Orthogonality of Legendre Polynomials ...................................................................................................................... 208
 Laplace’s Integral of First Kind .................................................................................................................................... 210
 Laplace’s Integral of Second Kind ............................................................................................................................... 210
 Cristoffel’s Expansion Formula .................................................................................................................................... 211
 Cristoffel’s Summation Formula .................................................................................................................................. 212
 Expansion of a Function in a Series of Legendre Polynomials (Fourier-Legendre Series) ....................................... 213
7. BESSEL’S DIFFERENTIAL EQUATION

 Introduction ................................................................................................................................................................... 224
 Solution of Bessel’s Equation ....................................................................................................................................... 224
 Series Representation of Bessel Functions ................................................................................................................... 228
 Recurrence Relations for Jn(x) ..................................................................................................................................... 229
 Generating Function for Jn(x) ....................................................................................................................................... 237
 Integral form of Bessel Function .................................................................................................................................. 238
 Equations Reducible to Bessel’s Equation ................................................................................................................... 241
 Modified Bessel’s Equation .......................................................................................................................................... 243
 BER and BEI Functions ................................................................................................................................................ 244
 Orthogonality of Bessel Functions ............................................................................................................................... 245
 Fourier-Bessel Expansion of f(x) .................................................................................................................................. 246
Self-Instructional Material iii

SEMESTER II
Course I
Course Name: Ordinary Differential Equations (ODE) Course Code: BAMAT-201
Course The main objectives of this course are to introduce the students to
Objectives: the exciting world of Differential Equations and their applications.
Unit 1: Formation of differential equation, Degree, order and solution of
a D.E., Ordinary differential equations of first order: initial and
boundary conditions,Seperation of variables method, homogeneous
equations:equation reducible to Homogeneous Form, linear equations,
Equation reducible to homogeneous form
Unit 2: Exact differential Equation. Necessary and sufficient condition for exact
differential equation, First order higher degree equations solvable for
x, y, p. Singular solution and envelopes,Clairaut’s equation,Equation
Reducible to Clauriat,s form.
Unit 3: Linear differential equations with constant coefficients; Determinaton of
C.F. and the P.I., homogeneous linear differential equations, Determinaton
of C.F. and the P.I., linear differential equations of second order with
variable coefficients,
Unit 4: Series solutions of differential equations. Introduction Frobenious Method
Solution near an ordinary point and a regular singular point, Method of
differentiation, Bessel and Legendre equations. Solution of Legendre
equation, Defination of Legendre polynomials, Bessel and Legendre
functions.
Course Learning Outcomes: The course will enable the students to:
1. Formulate Differential Equations for various Mathematical models.

2. Solve first order non-linear differential equation and linear differential equations of higher
order using various techniques.
3. Apply these techniques to solve and analyze various mathematical models.
References:
1. Barnes, Belinda & Fulford, Glenn R. (2015). Mathematical Modelling with Case Studies,
Using Maple and MATLAB (3rd ed.). CRC Press, Taylor & Francis Group.
2. Edwards, C. Henry, Penney, David E., & Calvis, David T. (2015). Differential Equation and
Boundary Value Problems: Computing and Modeling (5th ed.). Pearson Education.
Ross, Shepley L. (2004). Differential Equations (3rd ed.). John Wiley & Sons. India
UNIT I Differential Equations
NOTES
1. DIFFERENTIAL EQUATIONS
STRUCTURE
Differential Equation
Formation of a Differential Equation Whose
General Solution is Given
Solution of a Differential Equation
Initial Value Problem
Solution of a Differential Equation by the Method of Separation of Variables
Homogeneous Differential Equations and their Solution
DIFFERENTIAL EQUATION
An equation involving independent variables, dependent variables and at least one

derivative/differential of these variables is called a differential equation.
The following are some of the examples of differential equations:
dy
1. = x log x 2. dy = cos x dx
dx
d2y dy dy
dy
2
3.
dx 2
4
dx
12y = x4 4. y = x
dx
+ a 1

dx
.
Order and Degree of a Differential Equation

The order of a differential equation is the order of the derivative of the highest order,
occurring in the differential equation.
Consider the differential equation
F [ F[
3 + + y = sin x.
FZ FZ
Self-Instructional Material 1
Ordinary Differential
Equations
F [
The highest order derivative occurring in this equation is and its order
FZ
is 3.
 Order of given differential equation is 3.
NOTES
The degree of a differential equation is defined if it can be written as a
polynomial equation in the derivatives and for such a differential equation its degree
is given by the highest power of the highest order derivative appearing in it, provided
the derivatives are made free from radicals and fractions.
F[ F[
y=2
FZ
+ 3 
FZ . ...(1)
This equation is not free from radicals.
F[ = 3  F[
(1)  y2
FZ FZ
[ F[ = 9  F[


FZ FZ
F[
y + 4 4y
F[ F[
9 18 = 0

 2
FZ FZ FZ
F[
14 + 4y
F[

FZ y +9=0 2
FZ
The highest order derivative in this equation is
F[ and its highest power is 2.
FZ
 Degree of given differential equation is 2.
Linear Differential Equation

A differential equation is said to be linear, if the dependent variable and its derivatives
occur only in the first degree and are not multiplied together.
In general, a linear differential equation of order n is of the form
FP [ FP [ F[
P0 + P +  + Pn 1 + Pn y = Q,
FZ P 1
FZ P FZ
where P0, P1, ..., Pn 1, Pn, Q are functions of x or constants.
In particular, a linear differential equation of order one is of the form
P0
F[ + P y = Q.
FZ 1
dy d3 y y
The differential equations: cos x + y sin x = 1 and 3 + = x2 log x are linear
dx dx x
differential equations.
A differential equation which is not linear is called non-linear.
The degree of a linear differential equation is always one. But, the converse is
dy
not true. For example, the degree of y  7  sin x is one and it is not a linear
dx
differential equation.
2 Self-Instructional Material
SOLVED EXAMPLES Differential Equations
Example 1. Determine the order and degree, if defined, of the following

differential equations. State also, if these are linear or non-linear:
NOTES
dy  [
 Z  Z

dy dy
1
3
(i) xy
dx
=
 Z
(ii) y =
dx
+
dx .
dy (1  y 2 )(1  x  x 2 )
Solution. (i) The given differential equation is xy = .
dx 1  x2
dy
Order of the highest order derivative is 1.
dx
dy
Highest power of the highest order derivative is 1.
dx
 Order and degree of the given differential equation are 1 each.
dy
The given differential equation is non-linear, because y and are multiplied
dx
together.
dy dy 3
(ii) The given differential equation is y =
dx
+ 1
dx . ...(1)
dy
Order of the highest order derivative is 1.
dx
dy dy 3
y  dy 2
dy 3
(1)  y
dx
= 1
dx 
dx =1+
dx
dy
This is expressible as a polynomial in .
dx
dy
Highest power of the highest order derivative is 3.
dx
 Order and degree of the given differential equation are 1 and 3 respectively.
dy
The given differential equation is non-linear because is multiplied by itself.
dx
EXERCISE A
Determine the order and degree, if defined, of the following differential equations. State also, if
these are linear or non-linear (Q No. 14):

F [  Z F[ = 0 F[  Z F [ = 0
1. (i) Z
FZ
FZ (ii)
FZ FZ
F[  [ F [ = 0 (iv)
F [  [ F[ + x = 0
FZ FZ FZ
(iii) 3
FZ
dy
2
d2 y d y dy
2
 x
3 d y 3
2
dx

2. (i) 5 x 
dx
dx 2
 6 y  log x (ii) y
dx 2 dx 3
2
dy dy
(iii) y2 2y y + 1 = 0 (iv) y = x +a 1
dx dx
Equations dy
(i) 1 
2 "# 3/ 2
d2 y
! dx
3. =5 (ii) 1  x2 dx + 1  y2 dy = 0
#$ dx2
4
NOTES ds d2 s dy
(iii) + 3s =0 (iv) y = px + a2 p2  b2 , where p =
dt dt2 dx
2
d4y d2 y dy
4. (i)  sin ( y  ) = 0 (ii)  cos =0
dx 4
dx 2 dx
(iii) ( y)2 + (y)3 + (y)4 + y5 = 0 (iv) y + (y)2 + 2y = 0.

5. Write the sum of the order and degree of the following differential equations:
F %K F[ (K 
(i) &K FZ )K (ii)
F [ F[
 
 Z 
FZ ' * FZ FZ
Answers
1. (i) Order = 2, degree = 2, non-linear (ii) Order = 2, degree = 1, non-linear
(iii) Order = 2, degree = 1, non-linear (iv) Order = 2, degree = 3, non-linear
4. (i) Order = 4, degree not defined, non-linear
(ii) Order = 2, degree not defined, non-linear
(iii) Order = 3, degree = 2, non-linear
(iv) Order = 2, degree = 1, non-linear
5. (i) 3 (ii) 5.
FORMATION OF A DIFFERENTIAL EQUATION WHOSE

GENERAL SOLUTION IS GIVEN
If we have an equation between two variables, involving arbitrary constants,

then these arbitrary constants can be eliminated by using derivatives and as a result,
a differential equation is formed whose solution is the given equation.
I. Method of Forming a Differential Equation

To form a differential equation from a given equation in x, y and containing
arbitrary constants, the given equation is differentiated w.r.t. x successively as many
times as there are arbitrary constants. These equations are used to eliminate the
arbitrary constants. The equation obtained by eliminating the arbitrary constants is
the required differential equation.
In general, if the equation between two variables contains n arbitrary constants,
then the differential equation, obtained by eliminating these arbitrary constants, will
be of order n.
Differential Equations
Remark. The following are some of the important results of coordinate geometry which
are used in this section.
1. Equation of non-vertical line is y = mx + c, where m and c are arbitrary constants.
2. Equation of a non-vertical line passing through the origin is y = mx, where m is
NOTES
arbitrary constant.
3. Equation of the circle having centre (h, k) and radius r is (x h)2 + (y k)2 = r2.
4. Equation of circle in the general form is x2 + y2 + 2gx + 2fy + c = 0. Its centre and
radius are ( g, f) and g 2  f 2  c respectively.
5. Equation of a circle passing through the origin is x2 + y2 + 2gx + 2fy = 0, where g and
f are arbitrary constants.
6. Equation of a circle passing through the origin and having centre on the
x-axis is (x a)2 + y2 = a2, where a is arbitrary constant.
7. Equation of a circle passing through the origin and having centre on the
y-axis is x2 + (y a)2 = a2, where a is arbitrary constant.
8. Equation of a parabola with axis parallel to the x-axis is (y k)2 = 4a(x h), where a,
h and k are arbitrary constants.
9. Equation of a parabola with axis parallel to the y-axis is (x h)2 = 4a(y k), where a,
h and k are arbitrary constants.
10. Equation of an ellipse having centre at the origin and axes along the coordinate axes
x2 y2
is  = 1, where a and b are arbitrary constants.
a2 b2
Working Steps for the Formation of Differential Equations

Step I. Write the given equation.
Step II. Count the number of distinct arbitrary constants present in the given
equation.
Step III. Differentiate the given equation successively as many times as the
number of arbitrary constants.
Step IV. Eliminate the arbitrary constants by using the given equation and
equations obtained in the step III. The equation so obtained is the
required differential equation.
SOLVED EXAMPLES
Example 2. Form the differential equation of the following families of curves:
(i) y = mx, where m is an arbitrary constant.
(ii) (x a)2 + 2y2 = a2, where a is an arbitrary constant.
Solution. (i) We have y = mx. ...(1)
dy
Differentiating (1) w.r.t. x, we get =m
dx
F[ x.
Elimination of m. Putting the value of m in (1), we get y =
FZ
This is the required differential equation.
(ii) We have (x a)2 + 2y2 = a2 i.e., x2 2ax + 2y2 = 0 ...(1)
Differentiating (1)  2x 2a + 4yy = 0  a = x + 2yy
Ordinary Differential Elimination of a. Putting a = x + 2yy in (1), we get
Equations x2 2(x + 2yy)x + 2y2 = 0
 x2 2x2 4xyy + 2y2 = 0  4xyy + x2 2y2 = 0
 4xy
F[ + x2 2y2 = 0. This is the required differential equation.
NOTES FZ
Remark. The differential equation obtained for each system in the above example is of
order one. This is so, because each system contained only one arbitrary constant.

B
(i) y = Ax + , where A, B are arbitrary constants.
x
(ii) y = Ae3x + Be5x, where A, B are arbitrary constants.
B
Solution (i) We have y = Ax + .
x
 xy = Ax2 + B ...(1)
dy
Differentiating (1) w.r.t. x, we get x + y 1 = A 2x + 0
dx
dy
 x + y = 2Ax ...(2)
dx
Z F [  F[ 

Differentiating (2) w.r.t. x, we get FZ FZ
+
dy
dx
= 2A 1.
d2y dy
 x +2 = 2A
dx 2 dx
Elimination of A and B. Putting the value of 2A in (2), we get
dy d2y dy
x
dx
+y= x 2 2
dx dx
x
 x2
F[ +x
F[ y = 0. This is the required differential equation.
FZ FZ
(ii) We have y = Ae3x + Be5x. ...(1)
dy
Differentiating (1) w.r.t. x, we get = 3Ae3x + 5Be5x ...(2)
dx
d2y
Differentiating again = 9Ae3x + 25Be5x ...(3)
dx 2
Elimination of A and B.
d2y dy 1 d 2y 5 dy
(3) 5(2)  2 5 = 6Ae3x  + = Ae3x
dx dx 6 dx 2 6 dx
d2y dy 1 d2y 3 dy
(3) 3(2)  3 = 10Be5x  = Be5x
dx 2 dx 10 dx 2 10 dx
F [ F[ F [ F[
 (1)  y= 
FZ FZ
+
FZ FZ
d2y dy d2y dy
 30y = 5 + 25 +3 9
dx 2 dx dx 2 dx
d 2y dy Differential Equations
 2 16 + 30y = 0
dx 2 dx

F [ 8
F[ + 15y = 0. This is the required differential equation.
FZ FZ NOTES
Alternative Method
We have y = Ae3x + Be5x. ...(1)
(1)  y1 = 3Ae3x + 5Be5x ...(2)
(2) 3(1)  y1 3y = 2Be5x ...(3)
(3)  y2 3y1 = 10Be5x  y2 3y1 = 5(y1 3y)  y2 8y1 + 15y = 0

F [

8
FZ FZ
Remark The differential equation obtained for each system in the above example is of
order two. This is so, because each system contained two arbitrary constants.

(i) y = aex + be2x + ce3x, where a, b, c are arbitrary constants.
(ii) x2 + y2 + 2ax + 2by + c = 0, where a, b, c are arbitrary constants.
Solution (i) We have y = aex + be2x + ce3x. ...(1)
(1)  y1 = aex + 2be2x 3ce3x ...(2)
(2) (1)  y1 y = be2x 4ce3x ...(3)
(3)  y2 y1 = 2be2x + 12ce3x ...(4)
(4) 2(3)  y2 y1 2(y1 y) = 20ce3x
 y2 3y1 + 2y = 20ce3x ...(5)
(5)  y3 3y2 + 2y1 = 60ce3x ...(6)
(6) + 3(5)  y3 3y2 + 2y1 + 3(y2 3y1 + 2y) = 0
 y3 7y1 + 6y = 0

d y
7
dx FZ
(ii) We have x2 + y2 + 2ax + 2by + c = 0. ...(1)
Differentiating (1) w.r.t. x, we get 2x + 2yy1 + 2a + 2by1 + 0 = 0.
 x + yy1 + a + by1 = 0 ...(2)
Differentiating (2) w.r.t. x, we get 1 + (yy2 + y1y1) + 0 + by2 = 0
 by2 = (1 + yy2 + y12) ...(3)
Differentiating (3) w.r.t. x, we get by3 = (0 + yy3 + y1y2 + 2y1y2)
 by3 = (yy3 + 3y1y2) ...(4)
by2 
 [[  [
Elimination of a, b and c. Dividing (3) by (4), we get =
by3 
[[  [ [
[  [[  [
 =  yy2y3 + 3y1y22 = y3 + yy2y3 + y12y3
[ [[  [ [
 3y1y22 = y3 + y12y3  (1 + y21)y3 3y1y22 = 0
 F[ "# d y F[ d y = 0.
Equations
dx
! FZ #$
 3
dx FZ
NOTES
Remark The differential equation obtained for each system in the above example is of
order three. This is so, because each system contained three arbitrary constants.
Example 5. Form the differential equation of all lines in a plane which are at a
constant distance p from the origin.
Solution. The distance of the lines of the Y
family from the origin is p. Let L be a line of this
L
family. Draw OK perpendicular to this line. Let OK
make angle  with the x-axis.
 The equation of the line L is K
x cos  + y sin  = p.
p
 x cos  + y sin  p = 0 ...(1) a
Differentiating (1) w.r.t. x, we get X¢ O X
cos  + y1 sin  0 = 0 ...(2)
Y¢
Solving (1) and (2), we get
cos  UKP  1
= =
0  py1 R xy1  y
R[ R
 EQU   CPF UKP  
Z[ [ Z[ [
We have cos2  + sin2  = 1.
R[ 

R


Z[ [ Z[
[
 p2 y12  p2  ( xy1 – y)2
 R [  R  Z [  [ Z[[

R Z [  Z[[  R [ 


R Z  F[   Z[
F[  R [ 
 FZ  FZ
Example 6. Form the differential equation of the system of circles touching the
x-axis at the origin. Y
Solution. The circles in the system will have their
centres on the y-axis. Let (0, a) be the centre of a circle
touching the x-axis at the origin. |a|
(0, a)
 The radius this circle must be |a|*, otherwise the
circle will not touch the x-axis.
 The equation of the system of circle is
X¢ X
(x 0)2 + (y a)2 = (|a|)2 Y¢
or x2 + y2 2ay = 0, where a is an arbitrary constant.
*Why this step If the centre of the circle is below the origin, then a is negative. For
such a circle, the radius of circle is a, which is equal to |a|.
Differentiating w.r.t. x, we get Differential Equations
2x + 2yy1 2ay1 = 0 ...(1)
Elimination of a.
x  yy1 NOTES
(1)  a=
y1
Putting the value of a in x2 + y2 2ay = 0, we get
x  yy y  0 .
x2  y2 – 2 y
1
1
 x2 y1  y2 y1 – 2 xy – 2 y2 y1  0  ( x2 – y2 ) y1  2 xy

Z [ F[  Z[. This is the required differential equation.
FZ
Remark. This differential equation also represent the system of circles passing through
the origin and having centre on the y-axis.
Example 7. (i) Form the differential equation of all parabolas with latus rectum
4a and whose axes are parallel to the x-axis.
(ii) Form the differential equation of all parabolas whose axes are parallel to the
y-axis.
Solution. (i) The equation of a parabola with latus rectum 4a and axis parallel
to the x-axis is
(y k)2 = 4a(x h), ...(1)
where h and k are arbitrary constants.
Differentiating (1) w.r.t. x, we get
2(y k)(y 0) = 4a(1 0) i.e., (y k)y = 2a ...(2)
y 2
(y k)y + (y 0)y = 0 i.e., yk= ...(3)
y
Elimination of h and k.
y 2
. y  2 a C
F [  F[ 
(2) and (3) 
y
  FZ  
FZ  
(ii) The equation of a parabola whose axis is parallel to the y-axis is given by

Z J  C
[ M , ...(1)
where h, k and a are arbitrary constants.
(1)  2( x – h)(1 – 0)  4 a( y – 0)  x – h  2ay
 1 – 0  2ay  0  2ay  y  0
(Q a  0)

F [  This is the required differential equation.
FZ
Ordinary Differential Example 8. Form the differential equation representing the family of ellipses
Equations having foci on the x-axis and centre at the origin.
Solution. Let the equation of the family of ellipses be
NOTES x2 y2
= 1,  ...(1)
a 2 b2
where a and b are parameters and a > b > 0.
Differentiating (1) w.r.t x, we get
2x 2 yy  yy b2
2
 2 = 0   2 ...(2)
a b x a
Z = [[  [ [ ?  [[ 
=0 i.e., xyy + xy2 yy = 0.
Z
F [ Z F[ [ F[
 xy
FZ
FZ FZ = 0.
EXERCISE B
1. Form the differential equation of the family of curves given by:
(i) y = kx + k2 + k3 (ii) y +  sin x = 0.
2. Form the differential equation of all straight lines passing through the origin.
3. Form the differential equation of the family of all non-vertical lines y = mx + c, in the
xy-plane.
4. (i) Form a differential equation of the family of curves y = a sin (bx + c) where a and c
being arbitrary constants.
(ii) Form a differential equation of the family of curves y = a sin (bx + c) where a, b and c
being arbitrary constants.
5. Obtain a differential equation that should be satisfied by the family of concentric circles
x2 + y2 = a2.
6. Form a differential equation of the family of circles given by x2 + y2 = 2ax.
(i) y = Ae2x + Be2x (ii) y = ax + bx2
(iii) xy = C cos x (iv) y =

#  $.
T
(i) y = ex(a cos x + b sin x), where a and b are arbitrary constants.
(ii) xy = Aex + Bex + x2, where A and B are arbitrary constants.
(iii) y2 = a(b x2), where a and b are arbitrary constants.
(iv) y = e2x(a + bx), where a and b are arbitrary constants.
9. (i) Form the differential equation of the system of circles touching the y-axis at the origin.
(ii) Form the differential equation of the system of circles which passes through the origin
and having centres on the x-axis.
10. (i) Form the differential equation of all circles in the first quadrant which touch the
coordinate axes.
(ii) Form the differential equation of all circles in the second quadrant and touching the Differential Equations
coordinate axes.
11. (i) Form the differential equation of the family of circles of radius 2 units and having
centre on the x-axis.
(ii) Form the differential equation of the family of circles having centre on the NOTES
y-axis and radius 3 units.
12. (i) Form the differential equation of the family of circles (x a)2 + (y b)2 = r2 by eliminating
a and b.
(ii) Form the differential equation of the family of circles having radii 3.
13. Form the differential equation of all circles in the xy-plane.
14. (i) Form the differential equation of the family of parabolas having vertex at the origin
and axis along the positive y-axis.
(ii) Form the differential equation of the family of parabolas having vertex at the origin
and axis along the positive x-axis.
15. Form the differential equation of the family of ellipses having foci on the y-axis and centre
at the origin.
16. Form the differential equation of the family of hyperbolas having foci on the x-axis and
centre at the origin.
17. Show that the differential equation of which x2 y2 = c(x2 + y2)2 is a solution is
(x3 3xy2)dx = (y3 3x2y) dy.
Answers
dy dy  dy
2 3
dy
1. (i) y = x
dx

dx dx (ii)
dx
= y cot x
dy d2 y
2. y= x 3. 0
dx dx2
d2 y d3 y d2 y dy
4. (i)  b2 y  0 (ii) y  0
dx2 dx3 dx2 dx
dy dy
5. xy 0 6. 2 xy  x2 – y 2  0
dx dx
F [ d2 y dy
7. (i)  [  (ii) x2 2
 2x  2y  0
FZ dx dx
F[ F [ F[
(iii) Z  [  Z[ VCP Z  (iv) T  FT 
FZ FT
d2 y dy d2 y dy
8. (i) 2
2  2y  0 (ii) x 2  xy  x 2  2
dx dx dx2 dx
d2 y dy 2
dy d2 y dy
(iii) xy
dx 2
x
dx  y
dx
0 (iv)
dx 2
4
dx
 4y  0
dy dy
9. (i) 2 xy  x 2  y2  0 (ii) 2 xy  x 2  y2  0
dx dx
dy  x  y dy 2 2 dy "#  x  y dy
2 2
dx dx
! dx #$ dx
10. (i) ( x  y) 2 1  (ii) ( x  y)2 1 
dy 2 2
(i) y 1   4
dy
dx

11. 2 (ii) ( x 2  9) 2
 x 0
dx
d y  1  dy  0
2
2 2
3
d y  1  dy
2
2 2
3
dx dx dx dx
12. (i) r 2 (ii) 9 =0
2 2
Ordinary Differential 2 2
Equations

dy d3y dy d 2 y
13. 1  3 0

dx
dx 3 dx dx 2
dy dy
14. (i) x  2y = 0 (ii) y2 2xy 0
NOTES dx dx
2 2
d2y dy dy d2 y dy dy
15. xy x y =0 16. xy  x  y = 0.
dx 2 dx dx dx 2 dx dx
Hints
dy dy
2. y = mx  m  y x .
dx dx
4. (i) y = a sin (bx + c)  y1 = ab cos (bx + c)  y2 = ab2 sin (bx + c) = b2y
 y2 + b2y = 0.
(ii) y = a sin (bx + c)  y1 = ab cos (bx + c)  y2 = ab2 sin (bx + c)
 y3 = ab3 cos (bx + c)
y2 y
 y2 = b2y and y3 = b2y1   3  yy3 = y1y2.
y y1
8. (i) y = ex(a cos x + b sin x)
 y1 = ex (a cos x + b sin x) + ex( a sin x + b cos x)
 y1 = y + ex( a sin x + b cos x)
 y2 = y1 + ex ( a sin x + b cos x) + ex( a cos x b sin x)
 y2 = y1 + (y1 y) + ( y)  y2 2y1 + 2y = 0.
(ii) xy = Aex + Bex + x2  xy1 + y = Aex Bex + 2x
 (xy2 + y1) + y1 = (Aex + Bex) + 2  xy2 + 2y1 = (xy x2) + 2.
(iii) y2 = a(b x2)  2yy1 = a(2x)  yy1 = ax
yy1
 yy2 + y1 . y1 = a  yy2 + y12 = .
x
10. (i) Let the equation of the circle be
(x a)2 + (y a)2 = a2, where a is an arbitrary constant. Y
 2(x a) + 2(y a)y1 = 0 a>0
x  yy1
 x a + yy1 ay1 = 0  a =
1  y1
Putting the value of a in the equation of circle, we get
2 2 2 a (a, a)
x  yy1 x  yy1 x  yy1
x   y   a

1  y1 1  y1 1  y1
O X
 (x + xy1 x yy1)2 + (y + yy1 x yy1)2 = (x + yy1)2
 y12 (x y)2 + (y x)2 = (x + yy1)2.
11. (ii) Let the equation of family of circles be
(x 0)2 + (y a)2 = 9. ...(1)
x 2
x
 2x + 2(y a)y = 0  y a =  (1)  x2 + = 9.
y y 2
12. (i) We have (x a)2 + (y b)2 = r2. (1)
(1)  2(x a) + 2(y b)y1 = 0 (2)
(2)  2 + 2(y b)y2 + 2(y1)y1 = 0 (3)
 [
(3)  yb=  and
[
 [ [  [  [
(2)  x a = (y b)y1 = [ [
Now put the values of x a and y b in (1).
13. The equation of a circle in xy-plane is Differential Equations
x2 + y2 + 2gx + 2fy + c = 0, where g, f, c are arbitrary constants. (1)
(1)  2x + 2yy1 + 2g + 2fy1 + 0 = 0  x + yy1 + g + fy1 = 0 (2)
(2)  1 + yy2 + y12 + 0 + fy2 = 0  2
(y + f )y2 + y1 + 1 = 0 ...(3)
NOTES
(3)  (y + f )y3 + y1y2 + 2y1y2 + 0 = 0  (y + f )y3 + 3y1y2 = 0 ...(4)
Multiply (3) by y3, (4) by y2 and subtract.
14. (i) Take x2 = 4ay as the equation of the family of parabolas.
x2 y2
15. Take 2
 = 1, a > b > 0 as the equation of the family of ellipse.
b a2
x2 y2
16. Take 2
 = 1 as the equation of the family of hyperbolas.
a b2
17. We have x2 y2 = c(x2 + y2)2. ...(1)
(1)  2x 2yy1 = c.2(x2 + y2)(2x + 2yy1)
 x yy1 = 2c(x2 + y2)(x + yy1) ...(2)
Divide (2) by (1) and simplify.
SOLUTION OF A DIFFERENTIAL EQUATION
A solution of a differential equation is a functional relation between the variables

involved which satisfies the given differential equation.
A solution of a differential equation is called the general solution (or complete
solution), if it contains as many arbitrary constants as the order of the differential
equation.
Illustration y = Cx 4 is the general solution of the differential equation
dy
x 4y = 0, because the general solution contains one arbitrary constant C and the
dx
dy
order of the differential equation x 4y = 0 is also one.
dx
A solution obtained by giving particular values to arbitrary constants in the
general solution of a differential equation is called a particular solution of the
differential equation, under consideration.
Illustration y = 7x 4 is a particular solution of the differential equation
dy
x 4y = 0, because this solution has been obtained by giving a particular value 7
dx
to the arbitrary constant C in the general solution.
SOLVED EXAMPLES
Example 9. Show that y = bex + ce2x is a solution of y2 3y1 + 2y = 0.
Solution. We have y = bex + ce2x.
 y1 = bex + 2ce2x and y2 = bex + 4ce2x
 y2 3y1 + 2y = (be + 4ce ) 3(bex + 2ce2x) + 2(bex + ce2x)
x 2x
= bex(1 3 + 2) + ce2x(4 6 + 2) = bex(0) + ce2x(0) = 0.

 y = bex + ce2x is a solution of y2 3y1 + 2y = 0.
Ordinary Differential B d2y dy
Equations Example 10. Show that y = Ax + is a solution of x2 2 + x y = 0.
x dx dx
B
Solution. We have y = Ax + .
x
NOTES B
dy
 = A + B(1)x2 = A 2
dx x
d 2y 3
2B
and = 0 + B(1)(2) x =
dx 2 x3
d 2y dy 2B "# B B "# "#
 x2
dx 2
+x
dx
y = x2
x 3 + x A  2 Ax 
! $ x ! x $ ! $
2B B B
= + xA Ax =0
x x x
B d2y dy
 y = Ax + is a solution of x2 +x y = 0.
x dx 2 dx
Example 11. Show that y = A 2  x 2 , x  (A, A) is a solution of

dy
x+y = 0, y  0.
dx
dy 1 Z
Solution. We have y = A 2  x2 .  = (A2 x2)1/2 (2x) =
dx 2 #  Z

 x+y
dy
=x+
#

 Z Z
= x + (x) = 0.
dx #

 Z
dy
 y= A 2  x 2 is a solution of x + y = 0.
dx
EXERCISE C
2
dy dy
1. Show that x2 + 4y = 0 is a solution of +x y = 0.
dx dx
xy
2. Show that y = 1  x2 is a solution of y = .
1  x2
1 d2 y
3. Show that y = + Ax + B is a solution of x 3  1.
2x dx2
4. Show that y = a cos x + b sin x is a solution of y + y = 0.
d2 y dy
5. Show that y = 3 cos(log x) + 4 sin(log x) is a solution of x2 2
+x + y = 0.
dx dx
6. Show that y = ae2x
+ bex
is a solution of y2 y1 2y = 0.
7. Show that y = e3x (A + Bx) is a solution of y2 6y1 + 9y = 0.
8. Show that y = c1eax cos bx + c2eax sin bx is a solution of y2 2ay1 + (a2 + b2)y = 0.
F [ F[
9. Show that y = cos (cos x) is a solution of EQV Z  [ UKP Z = 0.
FZ FZ
10. Show that x + y = tan1 y is a solution of y2y + y2 + 1 = 0.
11. Show that y = x sin x is a solution of xy = y + x x2  y2 (x  0 and x > y or x < y).
12. Show that y cos y = x is a solution of (y sin y + cos y + x)y = y. Differential Equations
dy
13. Show that x2 = 2y2 log y is a solution of (x2 + y2) xy = 0.
dx
d2 y
14. Show that y = c1ex + c2ex is the general solution of  y = 0.
dx2 NOTES
15. x
Show that y = e + 1 is a solution of y y = 0.
16. Show that y = x2 + 2x + C is a solution of y 2x 2 = 0.
17. Show that y = cos x + C is a solution of y + sin x = 0.
18. Show that y = Ax is a solution of xy = y, x  0.
d2 y
19. Show that y = aex + bex + x2 is a solution of  y  x 2  2 = 0.
dx2
d2 y
dy
20. Show that y = ex (a cos x + b sin x) is solution of 2  2 y = 0.
dx2 dx
21. Show that x2 y2 = c(x2 + y2)2 is a solution of (x3 3xy2)dx = (y3 3x2y)dy.
D
22. Can y = ax + be a solution of the following differential equation
C
F[ D
y= Z  ?
FZ F[
FZ
If no, find the solution of the given differential equation. (CBSE 2018 SP)
Answer
22. Yes.
Hints
1 1 1
3. y= + Ax + B  y1 =  +A  y2 =  ( 2) x3  x3y2 = 1.
2x 2 x2 2
 UKP
NQI Z EQU
NQI Z
5. y = 3 cos (log x) + 4 sin (log x)  y1 = 
Z Z
 xy1 = 3 sin (log x) + 4 cos (log x)
 EQU
NQI Z UKP
NQI Z [
 xy2 + 1 y1 =   .
Z Z Z
9. y = cos (cos x)  y1 = sin (cos x)  sin x = sin xsin (cos x)
 y2 = cos x  sin (cos x) + sin x  cos (cos x) sin x
y
 y2 = (cos x) 1  (sin 2 x) y .
sin x
1
10. x + y = tan1 y  1 + y= . y  (1 + y2)(1 + y) = y
1  y2
 1 + y2 + y2y = 0.
11. y = x sin x  y = sin x + x cos x
y2
 xy = x sin x + x2 cos x = y + x2 1  sin 2 x = y + x2 1 = y + x x2  y2 .
x2
12. y cos y = x  y + (sin y)y = 1 ...(1)
Also, (y sin y + cos y + x) y = (y sin y + y x + x) y
= (y sin y + y) y = y(sin y + 1) y = y(1) = y.(By using (1))
2 y2
13. x2 = 2y2 log y  2x = 4yy log y + y  x = (2y log y + y) y
y
 xy = (2y2 log y + y2) y  xy = (x2 + y2) y
Equations INITIAL VALUE PROBLEM
A differential equation with given initial conditions is called an initial value

NOTES problem.
F[
= ysec x, y(0) = 1 is an initial value problem, because the solution of the
FZ
differential equation
F[ = ysec x is also to satisfy the initial condition y(0) = 1.
FZ
SOLVED EXAMPLES
Z
Example 12. Show that y = 2 is a solution of the initial value problem:
Z 
F[
F[
yx
FZ
= 2  Z

FZ
, y(1) = 1.

Z
Solution. We have y = 2 . ...(1)
Z 

x=1  y=2 = 2 1 = 1  y(1) = 1


F[ = 0
Z  Z
=
(1) 
FZ
Z 

Z 
F[ = 2 Z x = 2 Z + Z
 yx
FZ Z 
Z

 Z 
Z 

Z  Z
Z   Z Z  Z 
= 

Z 
Z 
F[

  Z
and

 Z
FZ
Z 

Z  Z  Z  Z 

Z 
Z 
=
F[
=  Z
F[

FZ

 yx
FZ
Z
 y=2
Z 
is a solution of the given initial value problem.
Example 13. Show that y = x sin 3x is a solution of the initial value problem:
F [
+ 9y 6 cos 3x = 0, y(0) = 0.
FZ
Solution. We have y = x sin 3x. ...(1)
x=0  y = 0  sin 3(0) = 0  y(0) = 0
(1) 
F[ = x 3 cos 3x + 1  sin 3x = 3x cos 3x + sin 3x
FZ
F [ Differential Equations
 = (3x (3 sin 3x) + 3.1. cos 3x) + 3 cos 3x
FZ
= 9x sin 3x + 6 cos 3x

F [ NOTES
 + 9y 6 cos 3x = 9x sin 3x + 6 cos 3x + 9(x sin 3x) 6 cos 3x = 0
FZ
 y = x sin 3x is a solution of the given initial value problem.
EXERCISE D
1. Show that y = ex is a solution of the initial value problem:
F[
= y, y(0) = 1.
FZ
2. Show that y = sin x + cos x is a solution of the initial value problem:
F [ + y = 0, y(0) = 1, y(0) = 1.
FZ
3. Show that y = xex + ex is a solution of the initial value problem:
F [ F[
2 + y = 0, y(0) = 1, y(0) = 2.
FZ FZ
4. Show that 3x2y = 2x + y is a solution of the initial value problem:
x2 dy + (xy + y2) dx = 0, y(1) = 1.
5. Show that 2y = x(x + y) is a solution of the initial value problem:
x2
F[ = y2 + 2xy, y(1) = 1.
FZ
SOLUTION OF A DIFFERENTIAL EQUATION BY THE

METHOD OF SEPARATION OF VARIABLES
dy
Let us consider the differential equation = f(x) g(y), ...(1)
dx
where f(x) and g(y) are some functions of x and y respectively.
We know that
F[
dy = dx,
FZ
where dx and dy are respectively the differentials of the variables x and y.
 (1)  dy = f (x) g(y) dx
dy
 = f(x) dx, provided g(y)  0 ...(2)
g ( y)
In equation (2), the expressions involving y are on one side and the expressions
involving x are on the other side.
Such a differential equation is said to be with variables separable.
Integrating equation (3), we get
I F[
I
[
= I f(x) dx + C, where C is an arbitrary constant.
This represents the general solution of the differential equation (1).
Equations
G\
Working Steps for Solving = f(x) g(y)
G[
Step I. Identify the functions f(x) and g(y).
NOTES Step II. Bring expressions involving x on one side and expressions involving y
on the other side. Always keep dx and dy in the numerators.
Step III. Integrate both sides and add arbitrary constant C only on one side.
This gives the required general solution.
Step IV. If some initial condition is given, then find the value of the arbitrary
constant C, so that the initial condition is satisfied. Put the value of
C in the general solution to get the required particular solution.
dy
Tpye I. Solution of = f(x)
dx
If g(y) = 1, then
F[ = f(x) g(y) reduces to F[ = f(x).
FZ FZ
SOLVED EXAMPLES
Example 14. Solve the differential equations:
dy 1  cos x
(i) (ex + ex)dy (ex ex)dx = 0 (ii)  .
dx 1  cos x
Solution. (i) We have (ex + ex) dy = (ex ex) dx.
e x  ex
 dy = dx (Variables are separate)
e x  ex
Integrating, we get I I
1dy =
e x  ex
e x  ex
dx + C.

y = log|ex + ex| + C
y = log (ex + ex) + C.

'
I f  ( x)
f ( x)

dx  log| f ( x)|

F[  EQU Z 1 – cos x
(ii) We have  i.e., dy = dx. (Variables are separate)
FZ  EQU Z 1  cos x
Integrating, we get I I
1dy =
1  cos x
1  cos x
dx + C.
I
x
2 sin2
 y= 2 dx + C
x
2 cos2
2
 I
y= sec 2 x
2

 1 dx + C  y = 2 tan
Z x + C.

Example 15. Solve the differential equations: Differential Equations
F[ F[
(i) (1 + x2) x = 2 tan1 x (ii) cos x cos 2x = cos 3x.
FZ FZ
F[ NOTES
Solution. (i) We have (1 + x2)
x = 2 tan1 x.
FZ
F[ Z  VCP Z
 
FZ  Z
Z VCP Z
 F[ 
 Z

 Z
dx (Variables are separate)
Z
dy = 
dx

 Z
  VCP
Z
 Z
Integrating, we get
I F[ 

I Z
 Z
FZ  I VCP
Z

 Z
FZ + C.

VCP Z
 y= log |1 + x2| + 2 +C

 y= log (1 + x2) + (tan1 x)2 + C.

F[
(ii) We have cos x cos 2x = cos 3x.
FZ
F[ EQU Z  EQU Z
 
FZ EQU Z
EQU Z  EQU Z EQU Z
EQU Z

EQU Z
dy = EQU Z  EQU Z

dx.
 EQU Z UGE Z dx

dy = EQU Z 


 dy = (2 cos x + 2 + 2 cos 2x 3 sec x)dx
 dy = (2 cos x + 2 cos 2x 1 sec x)dx
Integrating, we get
UKP Z
y = 2 sin x + 2. x log sec x + tan x + C.

 y = 2 sin x + sin 2x x log |sec x + tan x| + C.
Example 16. Solve the differential equation:

dy
(x3 + x2 + x + 1) = 2x2 + x, y = 1 when x = 0.
dx
dy
Solution. We have (x3 + x2 + x + 1) = 2x2 + x.
dx
2x 2  x
x3  x 2  x  1
Ordinary Differential Integrating, we get
I I I
Equations
2 x2  x 2x 2  x
dy  dx + C  y= dx + C
x3  x2  x  1 ( x  1)( x 2  1)

I
1 3 1
NOTES
 y= 2  2
x
2 dx + C
x1 x2  1
(By resolving into partial fractions)
1 3 1
 y= log |x + 1| + log (x2 + 1) tan1 x + C
2 4 2
Also, y = 1 when x = 0 ...(2)
1 3 1
 (1)  1= log |1| + log (1) (0) + C  C=1
2 4 2
 The required solution is
1 3 1
y= log |x + 1| + log (x2 + 1) tan1 x + 1.
2 4 2
EXERCISE E
Find the general solution of the following differential equations (Q. No. 13):
dy dy x
1. (i) x2 =2 (ii) 
dx dx x2  1
dy dy 1  cos 4 x
(iii) = x2 + sin 3x (iv) 
dx dx 1  cos 4 x
dy
2. (i) (x + 2) = x2 + 4x 9 (ii) 1  x6 dy = x2 dx
dx
dy F[
(iii) = log (x + 1) (iv) 
dx FZ UKP Z  EQU Z
F[ = x5 tan1 x3 F[
3. (i) (ii) = sin3 x cos2 x + x ex
FZ FZ
F[ F[ = sin 1 x
(iii) = tan 1 x (iv)
Z FZ FZ
4. Solve the following initial value problems:
dy
(i) x + 1 = 0, y(1) = 0 (ii) edy/dx = x + 1, y(0) = 5
dx
dy dy
(iii) x(x2 1) = 1, y(2) = 0 (iv) sin = k, y(0) = 1.
dx dx
Answers
2 1
1. (i) y + =C (ii) y = log (x2 + 1) + C
x 2
x3 cos 3 x 1
(iii) y =  +C (iv) y = tan 2x x + C
3 3 2
x2 1
2. (i) y = + 2x 13 log | x + 2 | + C (ii) y = sin 1 x3 + C
2 3
1 tan x  cot x Differential Equations
(iii) y = (x + 1) log (x + 1) x + C (iv) y = tan 1 +C
2 2

3. (i) 6y = (x6 + 1) tan1 x3 x3 + C (ii) y = cos5 x cos3 x + (x 1) ex + C

2 NOTES
(iii) y = (x + 1) tan1 x x + C (iv) y = x sin1 x + Z + C

4. (i) y + log |x| = 0 (ii) y = (x + 1) log (x + 1) x + 5

Z
(iii) y = log (iv) y = x sin1 k + 1.
Z
dy
TYPE II. Solution of = g(y)
dx
If f(x) = 1, then
F[ = f(x) g(y) reduces to F[ = g(y)
FZ FZ
SOLVED EXAMPLES

dy 1  y 2 dy 1
(i)  =0 (ii)  .
dx y dx log y
dy 1  y 2
Solution. (i) We have  = 0.
dx y
dy 1  y2 y
   dy = dx
dx y 1  y2
I I
(Variables are separate)
y
Integrating, we get dy   dx + C.
1  y2

1
2 I 2y
1 y 2
dy = x + C 
1
2
log (1 + y2) + x = C.
dy 1
(ii) We have  .
dx log y
I I
 log y dy = dx (Variables are separate)
Integrating, we get log y dy = dx + C.
 I log y  1 dy = x + C  (log y) y
I 1
y
 y dy = x + C
 y log y y = x + C.

F[ F[  EQU [

(i) + cos2 y = 0 (ii) = 0.
FZ FZ  EQU [
dy
Solution. (i) We have + cos2 y = 0.
dx
dy
 = cos2 y  sec2 y dy = dx (Variables are separate)
dx
I I
Equations
sec2 y dy = dx + C  tan y = x + C.
dy 1  cos y
NOTES (ii) We have  = 0.
dx 1  cos y
dy 1  cos y 1  cos y
   dy = dx
dx 1  cos y 1  cos y
Integrating, we get
I 1  cos y
1  cos y
dy   I dx + C.
I I
y
2 sin 2
2 dy = x + C  y

2 cos 2 y
sec 2
2
 1 dy = x + C
2
y [
 2 tan y=x+C  x y + 2 tan = C.
2
EXERCISE F
dy dy 1
1. (i) + y = 1, y  1 (ii)  2
dx dx y  sin y
dy 1  y2 dy
(iii)  (iv) = [
dx y3 dx
FZ
2. (i) y dy =  (ii) (sin y cos y)dy = (sin y + cos y)dx
VCP [
(iii) (y5 tan 1 y3)dy = dx (iv) y log y dy = dx
F[ F[
3. (i) = sec y (ii) = sin2 y
FZ FZ
(iii)
F[ =
EQU [
(iv) (2y2 + y)
F[ = y3 + y2 + y + 1
FZ  EQU [ FZ
dy dy 
(i) + 2y2 = 0, y(1) = 1 (ii) + cos2 y = 0, y(0) = .
dx dx
Answers
y3
1. (i) log  1 y  + x + C = 0 (ii) x = cos y + C
3
y2 1
(iii) x =  log (y2 + 1) + C (iv) y = 2 sin (x + C)
2 2
1 2 [
2. (i) x = (y + 1) tan 1 y +C (ii) x + log  sin y + cos y  = C
2
(iii) 6x = (y6 + 1) tan 1 y3 y3 + C (iv) 4x = 2y2 log y y2 + C
3. (i) x = sin y + C (ii) x + cot y = C Differential Equations
(iii) x + cot y + y = C

(iv) x = log  y + 1  + log (y2 + 1) tan1 y + C

NOTES
4. (i) y = (ii) x + tan y = 1.
Z
dy
TYPE III. Solution of = f(x) g(y)
dx
SOLVED EXAMPLES

(i) (y + xy)dx + (x xy2)dy = 0
(ii) (y3 + 1)(ex + xex) dx xexy2 dy = 0.
Solution. (i) We have (y + xy)dx + (x xy2)dy = 0.
 y(1 + x)dx + x(1 y2)dy = 0  y(1 + x)dx = x(1 y2)dy
1 x y2  1
 dx = dy (Variables are separate)
x y
Integrating, we get
I 1+ x
x
dx = I y2  1
y
dy + C.
 I 1
x

 1 dx =
I y
1
y
dy + C

y2
 log |x| + x = log |y| + C
2
[
 log |xy| + x = + C.

(ii) We have (y3 + 1)(ex + xex)dx xexy2 dy = 0.
 (y3 + 1)ex (1 + x)dx = xexy2 dy
2
1+ x y
 dx = 3 dy (Variables are separate)
x y 1
Integrating, we get I 1  1 dx = 1

x 3 I 3y2
y3  1
dy + C.
1
 log |x| + x = log |y3 + 1| + C.
3
(i) sec2 x tan y dx + sec2 y tan x dy = 0
(ii) ex tan y dx + (1 ex) sec2 y dy = 0.
Solution. (i) We have sec2 x tan y dx + sec2 y tan x dy = 0.
 sec2 x tan y dx = sec2 y tan x dy
UGE Z UGE [
Equations  dx = dy (Variables are separate)
VCP Z VCP [
NOTES
Integrating, we get I sec2 x
tan x
dx = I sec2 y
tan y
dy + C.
 log|tan x| = log |tan y| + C  log |tan x tan y| = C

 |tan x tan y| = eC  tan x tan y = ± eC
 tan x tan y = C1. (Taking C1 = ± eC)
(ii) We have ex tan y dx + (1 ex) sec2 y dy = 0.
 ex tan y dx = (ex 1) sec2 y dy
ex sec 2 y
 dx = dy (Variables are separate)
ex  1 tan y
Integrating, we get I ex
ex  1
dx = I UGE [
VCP [
dy + C.
 log |ex 1| = log |tan y| + C

| e x  1| ex  1 ex  1
 log =C  = eC  =  eC
|tan y| tan y tan y
 ex 1 = C1 tan y. (Taking C1 = ± eC)
Example 21. Solve:
dy dy = 3x + 4y.
(i) 1 + x 2  y 2  x 2 y 2 + xy
dx
=0 (ii) log
dx
dy
Solution. (i) We have 1 + x 2  y 2  x 2 y 2 + xy = 0.
dx
dy
 (1  x 2 )(1  y 2 ) + xy =0
dx
1  x2 y
 dx + dy = 0
x 1  y2
Integrating, we get I 1  x2
x
dx + I y
1  y2
dy = C. ...(1)
1  x2 = z  1 + x2 = z2  2x dx = 2z dz
 I 1  x2
x
dx = I 1  x2
x2
x dx =
I z
z2  1
. z dz
=
I z2
z2  1
dz = I ( z 2  1)  1
z2  1
dz
= I! 1
1
( z  1)( z  1)
"#
$
dz = z + I !

"#
$

\ 


\ 
dz
1 1 z1
=z+ (log |z 1| log |z + 1|) = z + log
2 2 z1
1 1  x2  1 1 1  x2  1
= 1  x2 + log = 1  x2 + log
2 1  x2  1 2 1  x2  1
Also, 1  y2 = z  1 + y2 = z2  2y dy = 2z dz
I I I
NOTES
y \ F\
 dy = = 1 dz = z = 1  y 2
1  y2 \
1  Z 
 (1)   Z + 2 log +  [ = C.
 Z 
dy = 3x + 4y.
(ii) We have log dx
dy dy
 = e3x+4y  = e3x e4y
dx dx
I I
 e4y dy = e3x dx (Variables are separate)
Integrating, we get e4y dy = e3x dx + C.
Z
 G [ = G + C.

Example 22. Show that the general solution of the differential equation
dy y 2  y  1
 = 0 is given by (x + y + 1) = A(1 x y 2xy), where A is a parameter.
dx x 2  x  1
dy y 2  y  1 dy dx
Solution We have  = 0, i.e., + = 0.
dx x 2  x  1 y2  y  1 2
x x1
Integrating, we get I 2
y  y1
dy
+ I 2
x x1
dx
= C. ...(1)

I 1
dy
 3
2 2
+
I 1
dx
 3
2 2
=C
y
2 2 x
2 2
   
1 [  1 Z  
 tan1   + tan1   = C
3 /2   3 /2  
2    [     Z   
  VCP    VCP   = C
3     
 Z   1  [   = 3
 tan1   + tan   C = C1, say
    2
 Z  [  
 
 tan1   = C 
3 (2 x  1  2 y  1)
= tan C1
  Z  [  
1
3  (2 x  1)(2 y  1)
 
 
2 3 ( x  y  1) 3 ( x  y  1) = tan C
Equations  = tan C1  1
2  4 xy  2 x  2 y 1  x  y  2 xy
 x + y + 1 = tan C1 (1 x y 2xy)
NOTES 3
 x + y + 1 = A(1 x y 2xy),
tan C1
where A = is a parameter.
3
Example 23. Solve the following differential equations:
F[ Z
NQI Z 
(i) yex/y dx = (xex/y + y2)dy, y  0 (ii)  .
FZ UKP [  [ EQU [
dx x x/y
Solution. (i) We have yex/y dx = (xex/y + y2)dy i.e., ex/y = e + y. ...(1)
dy y
x dx
Put z=  x = zy and = z + y dz
y dy dy
dz dz dz
 (1)

 ez z  y
dy
= zez + y  ezy
dy
= y  ez
dy
=1
 ezdz = dy (Variables are separate)
Integrating, we get

I e z dz = I
ez = y + C
dy  C .
 ex/y = y + C.
F[ Z
NQI Z 
(ii) We have =
FZ UKP [  [ EQU [
 (sin y + y cos y) dy = 2x (log x + 1) dx
 
UKP [  [ EQU [F[ =  Z NQI Z FZ   Z FZ  %
  UKP [ F[ + [ EQU [ F[ =  NQI Z Z FZ  Z  %
+ ++ + ++
 Z Z 
 cos y +  [ UKP [   UKP [ F[ = 2 
NQI Z  FZ  + x2 + C
 Z 

Z
 cos y + y sin y + cos y = x2 log x + x2 + C

Z
y sin y = x2 log x + + C.

Example 24. Solve the following initial value problems:
(i) y = y tan 2x, y(0) = 2 (ii) 2xy = 3y, y(1) = 4.
Solution. (i) We have y = y tan 2x.
dy dy
 = y tan 2x  = tan 2x dx
dx y

dy
y I I
= tan 2x dx + C  log  y  =
log|sec 0|
log|sec 2x |
2
+ C.
Now, y(0) = 2 implies log  2  = +C

2
log 1
i.e., log 2 = + C or C = log 2
2
log|sec 2 x | Differential Equations
 The required particular solution is log  y  = + log 2.
2
 log  y  = log  sec 2x 1/2 + log 2
2
 log  y  = log (Assuming sec 2x > 0)
cos 2x NOTES

 y= . (Assuming y > 0)
EQU Z
dy dy 3 dx
(ii) We have 2xy = 3y i.e., 2x  3y or = .
dx y 2 x
 Idx
x
dy 3
I 3
+ C  log  y  = log  x  + C.
3
y
=
2 2
Now, y(1) = 4 implies log |4| = log |1|+ C or C = log 4

2
3
 log  y  = log  x  + log 4 = log 4 x 3/2
2
 y = 4x3/2. (Assuming x > 0, y > 0)
This is the required particular solution.
(i) cos y dy + cos x sin y dx = 0, y(/2) = /2
F[
(ii) = ey cos x, y(0) = 0.
FZ
Solution. (i) We have cos y dy + cos x sin y dx = 0.
cos y dy
 + cos x dx = 0  cot y dy + cos x dx = 0
sin y
 I I
cot y dy + cos x dx = C  log |sin y| + sin x = C
Now, y 2 = 2 implies log sin


2
+ sin

2
= C i.e., 0 + 1 = C or C = 1
 The required particular solution is log |sin y| + sin x = 1.

dy
(ii) We have = ey cos x.
dx
dy cos x
 =  ey dy = cos x dx
I I
dx ey
 ey dy = cos x dx + C  e y = sin x + C.
Now y(0) = 0 implies e0 = sin 0 + C or 1 = 0 + C or C = 1
 The required particular solution is ey = sin x + 1.

F[
(i) = 1 + x2 + y2 + x2y2 given that y = 1 when x = 0.
FZ
(ii) (x y)(dx + dy) = dx dy given that y = 1 when x = 0.
Solution. (i) We have

F[ = 1 + x2 + y2 + x2y2.
FZ
F[ F[ = (1 + x2) dx
 = (1 + x2)(1 + y2) 
FZ  [
Equations
Z
tan1 y = x + + C. ...(1)

Also, y = 1 when x = 0
NOTES
 (1)  tan 1 (1) = 0 + 0 + C  C = /4

Z
 (1)  tan 1 y = x + + .

(ii) We have (x y)(dx + dy) = dx dy.
dx  dy d ( x  y)
 dx + dy =  d(x + y) =
x y x y
Integrating, we get
I
d( x  y) 
Also, y = 1 when x = 0
I
d( x  y)
x y
 C  x + y = log |x y| + C ...(1)
 (1)  0 1 = log |0 (1)| + C  1 = 0 + C  C = 1

 (1)  x + y = log |x y| 1.
EXERCISE G
dy 4y dy 1  y2
1. (i) = (ii) + =0
dx x( y  2) dx 1  x2
dy F[  [
(iii) = (ex + 1)y (iv) Z G
dx FZ
dy dy x  1
2. (i) x5 = y5 (ii) 
dx dx 2  y
dy 1  y2
(iii)  (iv) y log y dx x dy = 0.
dx 1  x2
2 dy
3. (i) y(1  x )  x (1  y 2 ) (ii) (1 + x)y dx + (1 y)x dy = 0
dx
dy dy F[ .
(iii) y a = ay2 + x (iv) cosec x log y Z [ 
dx dx FZ
4. (i) (1 + x)(1 + y2)dx + (1 + y)(1 + x2)dy = 0
(ii) (x2 yx2) dy + (y2 + x2y2)dx = 0
dy dy
(iii) y x = a y2  (iv) x 1  y 2 dx + y 1  x 2 dy = 0
dx dx
dy
5. (i) = x + y + xy + 1 (ii) (1 ex) sec2 y dy 2ex tan y dx = 0
dx
dy
(iii) (1 + y2)(1 + log x)dx + x dy = 0 (iv) cos x cos y + sin x sin y = 0
dx
6. (i) (ex + 1)y dy = (y + 1)ex dx 2
(ii) tan y dx + sec y tan x dy = 0
dy
(iii) 1  x2 dy + 1  y 2 dx = 0 (iv) sin3 x = sin y
dx
dy Differential Equations
7. (i)  e x  y  e x  y (ii) (xy2 + 2x)dx + (x2y + 2y)dy = 0
dx
dy dy
(iii) = (cos2 x sin2 x)cos2 y (iv) xy = 1 + x + y + xy
dx dx
dy NOTES
8. (i) y(1 + ex)dy = (y + 1)ex dx (ii) = y2 tan 2x
dx
y
(iii) ex 1  y 2 dx + dy = 0 (iv) (1 + e2x)dy + (1 + y2)exdx = 0
x
Solve the following initial value problems (Q. No. 912):
9. (i) y = 2exy3, y(0) = 1/2 (ii) y = 4xy2, y(0) = 1
(iii) xyy = y + 2, y(2) = 0 (iv) y = y cot 2x, y(/4) = 2
10. (i) 2 2
x(1 + y )dx y(1 + x )dy = 0, y(0) = 1
(ii) (1 + y2)(1 + log x)dx + x dy = 0, y(1) = 1
(iii) (1 y2)(1 + log x)dx + 2xy dy = 0, y(1) = 0
(iv) sec2 y (1 + x2)dy + 2x tan y dx = 0, y(1) = /4
 UKP Z F[ dy
11. (i)   EQU Z [
 (ii) (x + 1) = 2ey 1, y(0) = 0.
 [ FZ dx
F[
(iii) (1 + e2x)dy + (1 + y2)ex dx = 0, y(0) = 1 (iv) Z
Z  = 1, y(2) = 0
FZ
dy dy
12. (i) = y tan x, y(0) = 1 (ii) log = 3x + 4y, y(0) = 0
dx dx
(iii) xy
F[ = (x + 2)(y + 2), y(1) = 1 (iv) (x2 yx2)dy + (y2 + x2 y2) dx = 0, y(1) = 1
FZ
Answers
1. (i) y = log (x4y2) + C (ii)sin1 x + sin1 y = C
(iii) log |y| = ex + x + C (iv) 2e2y = x4 + C
2. (i) x4 + y4 = C (ii)x2 + y2 + 2x 4y + C = 0
(iii) tan1 y = tan1 x + C (iv) y = eCx
3. (i) (1 x2)(1 + y2) = C (ii)log |xy| + x y = C
 NQI [
(iii) (a + x)(1 ay) = Cy (iv) + (2 x2) cos x + 2x sin x = C
[
1
4. (i) tan1 x + tan1 y + log (1 + x2)(1 + y2) = C
2
1 1
(ii) x  log| y|–  =C (iii) (x + a)(1 ay) = Cy
x y
(iv) 1  x2 + 1  y2 = C
1 2
5. (i) log |y + 1| = x +x+C (ii) tan y = C(1 ex)2
2
1
(iii) (1 + log x)2 + tan1 y = C (iv) sin y = C cos x
2
6. (i) y log |y + 1| = log (ex + 1) + C (ii) sin x tan y = C
1
(iii) (x + 1  x 2 )(y + 1  y2 ) = C (iv) cos y + cos3 x cos x = C
3
7. (i) ex
= +Cey ex (ii) (x2 + 2)(y2 + 2) = C
1
(iii) tan y = sin 2x + C (iv) y = x + log |x(1 + y)| + C
2
1 1
8. (i) y log |1 + y| = log C(1 + ex) (ii) = log |sec 2x| + C
y 2
(iii) (x 1)ex 1  y2 = C (iv) y + ex = C(1 yex)
Ordinary Differential 1
Equations 9. (i) y2(8 4ex) = 1 (ii) y =
2 x2  1
| x|
(iii) y = 2 log |y + 2| + log (iv) y2 = 4 sin 2x
8
NOTES 10. (i) y2 2x2 = 1 (ii) 2(1 + log x)2 + 4 tan1 y = 2 + 
(iii) (1 + log x)2 = 2 log |1 y2| + 1 (iv) (1 + x2) tan y = 2
11. (i) (2 + sin x)(1 + y) = 4 (ii) (x + 1)(2 ey) = 1
1 4| x 2  1|
(iii) tan1 y + tan1 ex = /2 (iv) y = log
2 3x2
12. (i) y = sec x (ii) 4e3x + 3e4y = 7
(iii) y x + 2 = 2 log x(y + 2)  (iv) x = x1 + y1 + log  y  1.
dy
Solution of = f(ax + by + c) by the Method of Separation of Variables
dx
dy
Consider the differential equation = f (ax + by + c), ...(1)
dx
where f (ax + by + c) is some function of ax + by + c.
dz
dz dy dy a
Let z = ax + by + c.  =a+b or = dx
dx dx dx b
dz
a
dx dz
 (1)  = f(z)  = bf(z) + a
b dx
dz
 = dx ...(2)
bf ( z )  a
In the differential equation (2), the variables x and z are separated.
Integrating (2), we get
I dz
bf ( z )  a
=
I 1dx + C.
 I F\
DH
\  C
= x + C, where z = ax + by + c.
G\
Working Steps for Solving = f(ax + by + c)
G[
Step I. Identify the function f(ax + by + c).
Step II. Put z = ax + by + c and differentiate it w.r.t. x. Solve this to find the
dy
value of .
dx
dy
Step III. Put the values of and ax + by + c in the given differential equation.
dx
Separate the variables z and x and integrate both sides.
Step IV. Replace the value of z. This gives the general solution of the given
SOLVED EXAMPLES

F[ F[ NOTES
(i) cos (x + y) = 1 (ii) cos2 (x 2y) = 1 2 .
FZ FZ
F[ F[ = sec (x + y).
Solution. (i) We have  cos (x + y) = 1 or ...(1)
FZ FZ
RHS of (1) is a function of x + y.
F\ F[ F[ = F\ 1
z=x+y  =1+ 
FZ FZ FZ FZ
F\ F\ = 1 + sec z
 (1)  1 = sec z 
FZ FZ

F\ = dx (Variables are separate)
 UGE \
EQU \ dz = dx

 EQU \
dz = dx 
 EQU \
Integrating, we get
I
dz =
 EQU \ I  FZ + C.
 z
I
EQU
\

dz = x + C
 z

I UGE
\

dz = x + C
\
VCP Z [
 z = x + C  x + y tan =x+C

Z [
 y = tan + C.

F[
(ii) We have cos2 (x 2y) = 1 2 .
FZ

F[ = EQU
Z [ = UKP
Z [
FZ

EQU
Z [
=

 (1 cos (2x 4y))

F[ = (1 cos (2x 4y)) ...(1)
FZ
RHS of (1) is a function of 2x 4y.
F\ F[ F[ F\
z = 2x 4y 
FZ
=24
FZ

FZ
=
FZ
F\

 (1) 

FZ
=

(1 cos z)
Ordinary Differential F\
F\
Equations  2 = 1 cos z  = 1 + cos z
FZ FZ

F\ = dx (Variables are separate)
NOTES  EQU \
Integrating, we get
I F\
 EQU \
= I  FZ + C.

I F\
EQU \

=x+C 

I UGE
\

dz = x + C
\
VCP Z [
 =x+C  tan =x+C

 tan (x 2y) = x + C.

F[ Z  [  F[ Z  [ 
(i)  (ii)  .
FZ Z  [  FZ [  Z 
Solution. (i) We have

F[ =
Z  [  . ...(1)
FZ
Z  [ 
RHS of (1) is a function of x + 2y.
F\ F[ F[  F\ 
Let z = x + 2y.  =1+2 or = 
FZ FZ FZ  FZ 
 F\  \ F\ \ 
 (1)    =  1=
 FZ  \  FZ \
F\ \  F\ \ 
 = +1  =
FZ \ FZ \
\
 dz = dx (Variables are separate)
\ 
\
Integrating, we get  \  dz =  FZ  %
\    

 \ 
dz = x + C 
   \   dz = x + C
 
 
  \  NQI  \    = x + C  3z + 4 log  3z 1  = 9x + 9C
 
 3(x + 2y) + 4 log  3(x + 2y) 1  = 9x + C1, where C1 = 9C
 6(y x) + 4 log  3x + 6y 1 = C1.
F[ Z  [  F[
Z  [ 
(ii) We have = i.e.,  . ...(1)
FZ [  Z  FZ 
Z  [ 
RHS of (1) is a function of 2x y.
F\ F[ F[ F\ Differential Equations
Let z = 2x y.=2 or =2
FZ FZ FZ FZ
F\ \ F\ \  \
(1)  2 =  =2 
FZ  \  FZ  \   \  NOTES
F\ \ \ 
 =  dz = dx (Variables are separate)
FZ \  \
\ 
Integrating, we get  \
dz =  FZ + C.
 
   
 \ 
 dz = x + C 

z

log  z  = x + C
 2z log  z  = 5x + 5C
 2(2x y) log  2x y  = 5x + C1, where C1 = 5C
 x + 2y + log  2x y  + C1 = 0.
EXERCISE H
dy 2 dy
1. (i) = (ii) (x + y + 1) =1
dx x + 2y – 3 dx
dy dy
2. (i) = (3x + 2y + 1)2 (ii) (x + y)2 = k2
dx dx
F[ Z  [  F[ Z  [ 
3. (i) = (ii) 
FZ Z [ FZ Z  [ 
F[ = Z  [  F[ Z  [ 
4. (i) (ii)  .
FZ Z  [  FZ Z  [ 
(i) (x + y + 1)2 dy = dx, y(1) = 0 (ii) cos (x + y)dy = dx, y(0) = 0.
Answers
1. (i) 2y = 4 log  x + 2y + 1  + C (ii) y = log  x + y + 2  + C
2. (i) 2 (3x + 2y + 1) = 3 tan [ 6 (x + C)]

xy
(ii) y = k tan1 +C
k
3. (i) 2(y x) = log  2x + 2y + 1 + C (ii) x2 4xy + 4y2 + 6x 10y = C
4. (i) 3(2y x) + log  3x + 3y + 4  = C (ii) 4(2y x) + log  4x + 8y + 5  = C
Z [
5. (i) tan y = x + y + 1 (ii) y = tan .

HOMOGENEOUS DIFFERENTIAL EQUATIONS AND

THEIR SOLUTION
Homogeneous Function
A function f (x, y) of x and y is called a homogeneous function if
f (x, y) = n f (x, y). The number n is called the degree of the homogeneous function
f (x, y).
Ordinary Differential Illustration. Let f (x, y) = x3 + 2xy2 3y3.
Equations
 f (x, y) = (x)3 + 2(x)(y)2 3(y)3 = 3[x3 + 2xy2 3y3] = 3 f (x, y)
 x3 + 2xy2 3y3 is a homogeneous function of degree 3.
NOTES If f (x, y) is a homogeneous function of degree n, then f (x, y) can be expressed as

[ , where  [ is some function of [ .
xn 
Z Z Z
Illustration f (x, y) = x2 + 7xy 3y2 is a homogeneous function of degree 2 and
[ [

we have f (x, y) = x2 + 7xy 3y2 = x2  Z Z
[ [
[
and 1 +
Z Z is a function of
Z
.
Homogeneous Differential Equation

If f (x, y) and g(x, y) are homogeneous functions of same degree then the differential
equation
F[ H
Z [

FZ I
Z [
is called a homogeneous differential equation.
Let f (x, y) and g(x, y) be homogeneous functions of degree n each.
 f (x, y) = xnF(y/x) and g(x, y) = xnG(y/x) for some functions F(y/x) and
G(y/x) of y/x.

F[ H
Z [

F[ Z P (
[ Z = (
[ Z = (y/x), say
becomes 
FZ I
Z [ FZ Z P )
[ Z )
[ Z
 A homogeneous differential equation can also be expressed as
F[ = (y/x).
FZ
Illustration. Let
F[ Z [ ...(1)

FZ Z[  [
(1) is a homogeneous differential equation, because x3 2y3 and xy2 + 7y3 are
homogeneous functions of degree 3 each.
(1) can also be expressed as
F[
[ Z
 .
FZ
[ Z 
[ Z
Solution of Homogeneous Differential Equation
F[ H
Z [
Let  ...(1)
FZ I
Z [
be a homogeneous differential equation.
 f (x, y) and g(x, y) are homogeneous functions of same degree, say, n.
Let f (x, y) = xn F(y/x) and g(x, y) = xn G(y/x)
for some functions F(y/x) and G(y/x) of y/x.
 
H
Z [
P
Z (
[ Z (
[ Z = (y/x), say
 P =
I
Z [ Z )
[ Z )
[ Z
 (1) 
F[ = (y/x) ...(2)
FZ
F[ FX = v + x FX
Let y = vx.  = v(1) + x
FZ FZ FZ
FX FX FZ NOTES
 (2)  v + x = (v)   (Variables are separate)
X X Z
I I
FZ 

FX FZ + C.
Integrating both sides, we get 

X X Z
FX [

 X
X
= log | x | + C, where v =
Z
.
This equation is solved and v is replaced by y/x.
G\ \
Working Steps for Solving
G[
= 
[
Step I. Make sure that R.H.S. is either a function of y/x or the quotient of two
homogeneous functions of same degree.
dy dv
Step II. Put y = vx and differentiate it w.r.t. x to get =v+x .
dx dx
dy
Step III. Put the values of and y in the given differential equation. Separate
dx
the variables v and x and integrate both sides.
Step IV. Replace the value of v. This gives the general solution of the given
SOLVED EXAMPLES
x y
Example 29. Solve: y = .
x
dy xy
Solution. We have = . ...(1)
dx x
dy dv
This is a homogeneous differential equation. y = vx  =v+x
dx dx
dv x  vx
 (1)  v x   1 v
dx x
dv dx
 x  1  dv  (Variables are separate)
dx x
Integrating, we get I I 1dv =

dx
x
+ C.
[
 v = log |x| + C  = log |x| + C.
Z
dy
Example 30. Solve: x2 = x2 + xy + y2.
dx
dy dy x 2  xy  y 2
Solution. We have x2 = x2 + xy + y2.  = ...(1)
dx dx x2
dy dv
dx dx
Equations dv x 2  x (vx )  (vx )2
 (1)  v+x = = 1 + v + v2
dx x2
dv dv dx
NOTES  x = 1 + v2  = (Variables are separate)
dx 1  v2 x
Integrating, we get I dv
1  v2
= I dx
x
+C
[
 tan1 v = log |x| + C  tan1 = log |x| + C.
Z
F[
Example 31. Solve: 2xy = x2 + y2.
FZ
F[ Z  [
Solution. We have  . ...(1)
FZ Z[
F[ FX
FZ FZ
 (1)  v+x
FX = Z 
XZ =  X
FZ Z
XZ X
FX  X  X
 x = v=
FZ X X
X FZ
 dv = Z (Variables are separate)
X
X FZ
 dv + =0
X  Z
 log v2 1 + log  x  = log C
 log  x(v2 1)  = log C
 [ 
 x(v2 1) = ± C  x    = C, where C = C
Z 
 y2 x2 = Cx.
Example 32. Solve: x dy y dx = x 2  y 2 dx .
dy dy y y 2
Solution. We have x
dx
y= x 2  y2 . 
dx
=
x
+ 1
x ...(1)
dy dv
dx dx
dv dv dx
 (1)  v+x = v + 1  v2  =
dx 1v 2 x
 I dv
1v 2
= I dx
x
+ log C
 log |v + 1  v2 | = log |x| + log C
y y2 y x2  y2
 log  1 2 = log (C|x|)  = C|x|
x x x
 |y+ x 2  y 2 | = Cx2  y+ x 2  y 2 = ±Cx2

NOTES

 y+ Z [ = C1x2. (Putting C1 = C)
dy
Example 33. Solve: x2 = 2xy + y2.
dx
dy dy 2xy  y 2
Solution. We have x2 = 2xy + y2.
=  ...(1)
dx dx x2
dy dv
dx dx
dv 2 x(vx)  (vx)2
 (1)  v+x = = 2v + v2
dx x2
dv dv dx
 x = (2v + v2) v = v + v2  =
dx v(1  v ) x
v(1  v)
= I dx
x
+ log C.
I


X X

dv = log |x| + log C
 log |v| log |1 + v| = log C|x|

v y/ x
 log = log C|x|  = C|x|
1 v 1  y/ x
y
 = C|x|  |y| = C|x (x + y)|
x y
 y = Cx(x + y)
 y = C1x(x + y), where C1 = C.
dy y y
Example 34. Solve: = + sin .
dx x x
dy y y
Solution. We have = + sin . ...(1)
dx x x
dy dv
dx dx
dv
 (1)  v+x
= v + sin v
dx
dv dx
 x = sin v  cosec v dv = (Variables are separate)
I I
dx x
dx
Integrating, we get cosec v dv = + log C.
x
v
 log tan = log |x| + log C
2
y
 log tan = log C|x|
2x
Ordinary Differential y y
Equations  tan = C|x|  tan = Cx
2x 2x
[
 tan = C1x. (Putting C1 = C)
NOTES Z
dy y
Example 35. Solve: x = y x tan .
dx x
dy y
Solution. We have x = y x tan .
dx x
dy y y
 = tan ...(1)
dx x x
dy dv
dx dx
dv
 (1)  v+x = v tan v
dx
dv dx
 x = tan v  cot v dv =
dx x
Integrating, we get I cot v dv = I dx

x
+ C.
C
 log |sin v| = log |x| + log C  log |sin v| = log
|x |
C y
 |sin v| =  |x| sin =C
|x | x
y y
 x sin =C  x sin =C
x x
[
 x sin = C1. (Putting C1 = ± C)
Z
Example 36. Solve: (x3 3xy2)dx = (y3 3x2y)dy.
dy x 3  3xy 2
Solution. We have (x3 3xy2)dx = (y3 3x2y)dy i.e., = 3 . ...(1)
dx y  3x 2 y
dy dv
dx dx
dv x 3  3x (vx )2 x 3  3v2 x3 1  3v2
 (1)  v+x = 3 2 = 3 3 3 = 3
dx (vx )  3x (vx ) v x  3vx v  3v
dv 1  3v2 1  3v2  v4  3v2 1  v4

 x = 3 v= 3 = 3
dx v  3v v  3v v  3v
v3  3v dx
 dv = (Variables are separate)
1  v4 x
Integrating, we get I v3  3v
1  v4
dv = I dx
x
+ log C.
 I v3
1  v4
dv 3 I v
1  v4
dv = log |x| + log C

1
4 I X
 X
3
dv 2 I 2v
1 – v4
dv = log C|x|

1
4
log |1 v4|
3
2 I dt
1  t2
= log C|x|, where t = v2
NOTES
1 3 1 1 t
 log |1 v4|  log = log C|x|
4 2 2(1) 1 t
1 3 1  v2
 log |1 v4| log = log C  x 
4 4 1  v2
(1  v2 )3 (1  v2 )4
 log (1  v4 ) . = log (C|x|) 4  =
(1  v ) 2 3 2 2
(1  v ) %
Z
1  y
2
4
y 2
2
 C4x4 (1 + v2)4 = (1 v2)2  C4x4 x 2
= 1 
x 2
( x 2  y 2 )4 ( x 2  y2 ) 2
 C4x4 . =
x 8 x4
 C1(x2 + y2)4 = (x2 y2)2, where C1 = C4.
y y dx = x y sin y  x cos y dy.

Example 37. Solve:

y x cos
x
 y sin
x x x
y
y x cos  y sin
y
dy x x
Solution. The given equation is
dx
=

x y sin  x cos
y y
. ...(1)
x x
dy dv
dx dx
dv vx ( x cos v  vx sin v ) v (cos v  v sin v )
 (1)  v+x = =
dx x (vx sin v  x cos v ) v sin v  cos v
dv v (cos v  v sin v) 2v cos v
 x = v=
dx v sin v  cos v v sin v  cos v
v sin v  cos v dx
 =2 (Variables are separate)
I
v cos v x
Integrating, we get
I v sin v  cos v
dv = 2
dx
x
+ log C.
I
v cos v
cos v  v sin v
 dv = 2 log |x| + log C
v cos v
 log |v cos v| = log Cx2
1 x
 =  Cx2  = Cx2
v cos v y cos ( y/ x)
y =  1 [ = C , where C 1
 xy cos
x C  xy cos
Z 1 1 =
C
.
Note. The above question can also be given as follows:

y y
Solve: y(x dy y dx) sin = x(y dx + x dy) cos .
x x
Ordinary Differential Example 38. Check whether the following differential equation is homogeneous
Equations
F[  [
or not: x2 xy = 1 + cos   , x  0?
FZ  Z
NOTES Find the general solution of the differential equation using substitution y = vx.
Solution. We have
F[  [
x2 xy = 1 + cos   , x  0.
FZ  Z
 [
F[ Z[   EQU  Z 
 =
FZ Z
F[ [  [ 
   EQU   
 = ...(1)
FZ Z Z   Z 
This is not a homogeneous differential equation, because RHS is not a function
of y/x.
F[ FX
Let y = vx  =v+x
FZ FZ
FX
 (1)  v+x = v + (1 + cos v)
FZ Z
FX  EQU X FX FZ
 
FZ = Z  EQU X =
Z
Integrating, we get
FX FZ
  EQU X =  Z + C.
 EQU X Z 
  UKP X
dv =

+C

 
EQUGE X  EQV X EQUGE X dv =
Z
+C

 cot v + cosec v = +C
Z
[ [
 cosec cot + = C.
Z Z Z

y  y dx + x dy = 0 given that y =  when x = 1.
Example 39. Solve: x sin 2
x
4
y
Solution. We have x sin  y dx + x dy = 0.
x
2

y  x sin
y
x i.e., dy = y sin y
2
dy

x
= 2 ...(1)
dx x dx x
dy dv Differential Equations
dx dx
dv vx vx
 (1)  v+x
dx
=
x
 sin 2
x NOTES
dv dx
 x = sin2 v  cosec2 v dv =
dx x
Integrating, we get I dx
cosec 2 v dv =
x
C I
 cot v = log |x| + C
 log |x| cot (y/x) = C ...(2)
Now, y = /4 when x = 1.
 (2)  1
log |1| cot ( / 4) 1 = C 6 C = 0 1 = 1
 (2)  log |x| cot (y/x) = 1. This is the required solution.
Example 40. Solve: x2 dy + (xy + y2)dx = 0, given that y = 1 when x = 1.
dy y( x  y )
Solution. We have x2 dy + y(x + y) dx = 0.  = ...(1)
dx x2
dy dv
dx dx
dv vx ( x  vx )
 (1)  v+x = = v(1 + v)
dx x2
dv
 x = v v2 v = v(2 + v)
dx
dv dx
 = (Variables are separate)
v( 2  v) x
v( 2  v)
=
I dx
x
+ log C.
 I 

 X





 X

 dv = log |x| + log C

1 1 C
 log |v| log |2 + v| = log
2 2 |x |
1 v C v C
 log = log  log = 2 log
2 2v |x | 2v |x |
v C2 y/ x C2
 = 2  =
2v x 2  y/ x x2
y C2
  2
2x  y x
 x2y = k(2x + y), where k = ±C2
Now, y = 1 when x = 1.  (1)2 (1) = k(2(1) + 1) i.e., k = 1/3
1
 The required solution is x2y = (2x + y) or 3x2y = 2x + y.
3
dy x(2y  x)
Equations Example 41. Solve:  , y(1) = 1.
dx x(2y  x)
dy x(2 y  x) dy 2 y  x
Solution. We have  i.e.,  . ...(1)
NOTES dx x(2 y  x) dx 2 y  x
dy dv
dx dx
dv 2vx  x
 (1)  v+x = = 2v  1
dx 2vx  x 2v  1
dv 2v  1 2v  1  2v2  v 2v2  v  1
 x = v= =
dx 2v  1 2v  1 2v  1
2v  1 dx
2v2  v  1 x
Integrating, we get I 2v  1
2v2  v  1
dv =
I dx
x
 C.
 I 2v  1
2v2  v  1
dv = log |x| + C ...(1)
Now
I 2v  1
2
2v  v  1
dv =
1
2 I (4v  1)  3
2v2  v  1
dv
=
1
2 I 4v  1
2
2v  v  1
dv 
3
2 I 2
dv
2v  v  1
=
1
2
log |2v2 v + 1| +
3
4 I v
dv
1

2

7
4 16

X
1 3
= log|2v 2  v  1| +  tan1
2 4
1 3 4v  1
= log |2v2 v + 1| + tan1
2 7 7
1 3 4v  1
 (1)  log|2v2  v  1| + tan1 = log |x| + C
2 7 7
1 2 y2 y 3 4y  x
 log   1 + tan 1 = log |x| + C
2 x2 x 7 7x
1 3 4y  x
 log |2y2 xy + x2| log |x| + tan1 = log |x| + C
2 7 7x
1 3 4y  x
 log |2y2 xy + x2| + tan1 =C ...(2)
2 7 7x
We have y = 1 when x = 1.
 3 41
 (2)  log |2 1 + 1| + tan1 =C
2 7 7
1 3 3 Differential Equations
 C= log 2 + tan1
2 7 7
 Using (2), the required solution is
[  Z 1 3 3 NOTES
NQI^ [  Z[  Z ^ + VCP  = log 2 + tan 1 .
Z 2 7 7
dx
Solution of Homogeneous Differential Equation = y(x/y)
dy
We have
FZ = (x/y). ...(1)
F[
FZ FX FX
Let x = vy.  = v(1) + y =v+y
F[ F[ F[
FX FX F[
 (1)  v+y = (v)   (Variables are separate)
F[ 
X X [
Integrating both sides, we get I FX

X X

F[
[ I
+ C.
 I FX
\
X X
= log |y| + C, where v =
Z
[
.
Z
This equation is solved and v is replaced by
[
Remarks 1. The equation

FZ = (x/y) can also be solved after interchanging x and y in
F[
the equation and again interchanging x and y in the solution of the given equation.
2. Sometimes, a given homogeneous differential equation is conveniently solved by
expressing it in the form

FZ = (x/y).
F[
Example 42. Solve:

FZ = Z + sin Z .
F[ [ [
Solution. We have
FZ = Z + sin Z . ...(1)
F[ [ [
FZ
This is a homogeneous differential equation of the form = (x/y).
F[
Let x = vy 
FZ =v+y
FX
F[ F[
FX
 (1)  v+y = v + sin v
F[
 y
FX = sin v  cosec v dv =
F[ (Variables are separate)
F[ [
Integrating, we get I EQUGE X dv =
X
I F[
[
+ log C.
 log VCP = log |y| + log C

Equations Z
 log VCP = log C |y|
[
Z Z
 VCP = C y   VCP =  Cy
NOTES [ [
Z
 VCP = C1y. (Putting C1 = ±C)
[
Example 43. Solve: 2yex/y dx + (y 2xex/y)dy = 0.
Solution. We have 2yex/y dx + (y 2xex/y)dy = 0.
FZ ZG Z [
[ FZ
Z [ G Z [

 =  = ...(1)
F[ [G Z [
F[ G Z [
FZ
This is a homogeneous differential equation of the form = (x/y).
F[
FZ FX
Let x = vy.  =v+y
F[ F[
 (1)  v+y
FX =
XG X
F[ GX
FX XG X
 y = v
F[ GX
FX
 y = X  2evdv = dy
F[ G [
Integrating, we get

I GX FX =
[

I
dy + C.
2ev = log  y  + C  2ex/y + log | y | = C.

Z

Example 44. Solve: (1 + ex/y)dx + ex/y
[
dy = 0.

Solution. We have (1 + e )dx + e
Z
dy = 0.
[
x/y x/y
Z
G Z [

FZ

[ ...(1)
F[  GZ [
FZ
This is a homogeneous differential equation of the form  
Z [
F[
FZ FX
Let x = vy  =v+y
F[ F[
 (1)  v+y
FX =
G
X

X
F[  G
X
 y
FX =
X
G  XG
X
X
X
G  XG
X
X XG
X
F[  GX  GX
FX G
X
X
 y =
F[  G X
 GX F[ Differential Equations
X G
X [


I  GX
X G
log  v +
X
dv =
ev
I F[
[
+ log C
= log  y  + log C

NOTES
 log  v + ev = log

%  y(v + ev)  = C
^ [^
Z  G = C
Z [
 y
[  x + yex/y = C1. (Putting C1 = C)
FZ Z + y x sin Z = 0, y(/2) = 1.
Example 45. Solve: y
F[
sin
[ [
Solution. We have
FZ Z + y x sin Z = 0.
y
F[
sin
[ [

FZ =
Z UKP
Z [ [

FZ =

Z [ UKP
Z [
...(1)
F[ [ UKP
Z [ F[ UKP
Z [
This is a homogeneous differential equation of the form

FZ = (x/y).
F[
FZ FX
Let x = vy.  =v+y
F[ F[
FX X UKP X FX X UKP X
 (1)  v+y =  y = v
F[ UKP X F[ UKP X
FX F[
 y =  sin v dv =
F[ UKP X [
Integrating, we get I UKP v dv =
I F[
[
+ C.
 cos v = log y + C

 log  y  = cos (x/y)+ C ...(2)
We have y = 1 when x = /2.
 + C

 (2)  log |1| = cos

 0=0+C  C=0
 (2)  log | y | = cos (x/y). This is the required solution.
EXERCISE I
Solve the following differential equations (Q. No. 125):

1. (3xy + y2)dx + (x2 + xy)dy = 0 2. 2xyy = x2 + 3y2
3. (x2 + xy)dy = (x2 + y2)dx 4. (x2 y2)dx + 2xy dy = 0
dy y2
5. x + =y 6. x2y dx (x3 + y3)dy = 0
dx x
dy dy dy
7. x 2  y( x  y) 8. y x =x+y
dx dx dx
Ordinary Differential dy y( x  y ) dy
Equations 9. x2 = 10. x = y(log y log x + 1)
dx 2 dx
dy y y dy
11. = + tan 12. (x y) = x + 2y
dx x x dx
NOTES 13. x2y1 = x2 2y2 + xy 14. (x2 y2)dx + xy dy = 0
y
15. x2 dy + y(x + y)dx = 0 16. y dx + x log dy 2x dy = 0
x
 [  [ [ [ Z [ [
17. Z[  NQI  FZ   [  Z F[  18. EQU FZ   UKP  EQU  F[ 
Z 
NQI
 Z   Z Z  [ Z Z 
19. (y2 2xy)dx = (x2 2xy)dy 20. y2 dx + (x2 xy + y2)dy = 0
21. 2xy dx + (x2 + 2y2)dy = 0 22. (y2 x2)dy = 3xy dx
y dy y
23. x cos = y cos  x 24. (x y)dy (x + y)dx = 0
x dx x
dy y
25. x  y  x sin = 0.
dx x
Solve the following initial value problems (Q. No. 2645):
2 2 dy dy
26. y  x  xy , y(1)  1 27. x(x2 + 3y2)dx + y(y2 + 3x2)dy = 0, y(1) = 1
dx dx
y/ x dy
28. (y4 2x3y)dx + (x4 2xy3)dy = 0, y(1) = 1 29. xe  yx  0, y( e)  0
dx
30. (xey/x + y)dx = x dy, y(1) = 1 31. (x + y)dy + (x y)dx = 0, y(1) = 1
dy dy
32. 2xy + y2 2x2 = 0, y(1) = 2 33. 2x2 2xy + y2 = 0, y(e) = e
dx dx
34. 2yex/y dx + (y 2xex/y)dy = 0, y(0) = 1 35. (x2 y2)dx + 2xy dy = 0, y(1) = 1
dy
36. x2 = y2 + 2xy, y(1) = 1 37. dy  y  cosec y  0 , y(1) = 0
dx dx x x
y y
38. (x dy y dx)y sin = (y dx + x dy)x cos , y(3) = 
x x
F[ [ F[ [ [ 
39. x y + x sin = 0, y(2) = . 40. x sin + x y sin = 0, y(1) =
FZ Z FZ Z Z
[ F[ [ 
41. (3xy + y2) dx + (x2 + xy)dy = 0, y(1) = 1 42. x cos = x + y cos , y(1) =
Z FZ Z
43. (x2 + xy)dy = (x2 + y2)dx, y(1) = 0 44. (x y)
F[ = x + 2y, y(1) = 0
FZ
45. (x2 + y2)dy xy dx = 0, y(0) = 1.
Answers
1. x2y(2x + y) = C 2. x2 + y2 = Cx3
y
3. log |x| = log (x y)2 + +C 4. x = C(x2 + y2)
x
x x3
5. log |x| = +C 6. = log |y| + C
y 3 y3
1 y
7. y log |x| + x + Cy = 0 8. log (x2 + y2) + tan1 =C
2 x
y
9. xy2 = C(y x)2 10. log = Cx
x
y Z  [ Differential Equations
11. x = C sin 12. log |x2 + xy + y2| VCP  =C
x Z
1 x  2y
13. log = log |x| + C 14. y2 = x2(C 2 log|x|)
2 2 x  2y
NOTES
y
15. x2y = C(y + 2x) 16. Cy = log  1
x
Z [ y
17. log y2 + NQI  =C 18. y sin =C
[ Z x
1
19. x2y xy2 = C 20. y = Ce tan ( y / x)
21. 3x2y + 2y3 = C 22. y2(4x2 y2)3 = C
[ y 1
23. sin = log |x| + C 24. tan 1  log ( x2  y2 )  C
Z x 2
y C y y
25. 1  cos  sin 26. log |y| = 1
x x x x
27. x4 + 6x2y2 + y4 = 8 28. x3 + y3 = 2xy
1 1
29. y = x log log |x| 30. log |x| = 
e e y/ x
y 
31. log (x2 + y2) + 2 tan1 = + log 2
x 2
2x
32. + log |x| = 1 33. y log ex = 2x
y
34. 2ex/y + log |y| = 2 35. x2 + y2 = 2x
[
36. 2y = x(x + y) 37. EQU   NQI^ Z^
Z
[ [ [
38. 2xy cos = 3 39. Z EQUGE  EQV = 2
Z Z Z
[
40. log x  = cos 41. x2y2 + 2x3y = 3
Z
[ [
42. sin = log  x  + 43. + 2log  x y  log  x  = 0
Z Z
 Z  [
44. log  x2 + y2 + xy  + = tan1
Z
45. x2 = 2y2 log y.
dy a x  b1y  c1 a1 b
Solution of  1 , where  1 by Reducing it to a
dx a2 x  b2 y  c2 a2 b2
Homogeneous Equation
dy a x  b1 y  c1
a1 b
= 1  1., where ...(1)
dx a2 x  b2 y  c2
a2 b2
We substitute x = X + h and y = Y + k, where h and k are constants to be
properly chosen.
dy dy dY dX dY dY
 = × × =1× ×1=
dx dY dX dx dX dX
Ordinary Differential dY a (X  h)  b1 (Y  k)  c1
Equations  (1)  = 1
dX a2 (X  h)  b2 (Y  k)  c2
dY a X  b1Y  (a1h  b1k  c1)
 = 1 ...(2)
dX a2 X  b2 Y  (a2 h  b2 k  c2 )
NOTES The constants h and k are chosen so that a1h + b1k + c1 = 0 and
a2h + b2k + c2 = 0.
dY a X  b1 Y
 (2)  = 1 ...(3)
dX a2X  b2 Y
This is a homogeneous differential equation and can be solved by putting
Y = VX.
dY dV
Y = VX  =V+X
dX dX
dV a X  b1VX a b V
 (3)  V+X = 1 = 1 1
dX a2 X  b2 VX a2  b2 V
dV a b V a  b V  a2 V  b2 V 2
 X = 1 1 V = 1 1
dX a2  b2 V a2  b2 V
C  D 8 dX
 dV = ...(4)
C 
D  C 8  D 8 X
In the differential equation (4), the variables X and V are separated.
Integrating (4), we get I a2  b2 V

a1  (b1  a2 ) V  b2 V 2
dV 
I dX
X
 C.
 I C  D 8
C 
D  C 8  D 8
dV = log |X| + C,
where V = Y/X, X = x h and Y = y k.

C D F[ C Z  D [  E
Remark. If  in the differential equation  then it can be easily
C D FZ C Z  D [  E
solved by putting z = a1x + b1y or z = a2x + b2y.
G\ D [  E \  F D E
Working Steps for Solving = , where 
G[ D [  E \  F D E
Step I. Put x = X + h and y = Y + k. Given differential equation reduces to
dY a X  b1Y  ( a1h  b1k  c1 )
= 1 .
dX a2 X  b2Y  ( a2h  b2 k  c2 )
Step II. Solve a1h + b1k + c1 = 0 and a2h + b2k + c2 = 0 to get the values of h
dY a X  b1Y
and k. The resultant equation = 1 is a homogeneous
dX a2 X  b2Y
Step III. Put Y = VX. This gives a differential equation in X and V with variables
separated.
Step IV. Solve this differential equation and put V = Y/X, X = x h and
Y = y k to get the answer in original variables x and y.
dy x  2y  5
Example 46. Solve: = .
dx 2x + y – 4
dy x  2y  5 NOTES
Solution. We have = . ...(1)
dx 2 x+y–4
a1 1 b1 2 a1 b
Here = and = =2   1
a2 2 b2 1 a2 b2
Let x = X + h and y = Y + k.
dy dy dY dX dY dY
 = × × =1× ×1=
dx dY dX dx dX dX
dY (X  h )  2(Y  k )  5
 (1)  =
dX 2(X  h )  (Y  k )  4
dY X  2Y  (h  2k  5)
 = ...(2)
dX 2 X  Y  (2h  k  4)
Let h and k be such that h + 2k 5 = 0 and 2h + k 4 = 0.
 h = 1, k = 2 (On simplification)
dY X  2Y
 (2)  = ...(3)
dX 2X  Y
This is a homogeneous differential equation.
dY dV
Let Y = VX.  =V+X
dX dX
dV X  2(VX) 1  2V
 (3)  V+X = =
dX 2 X  VX 2V
dV 1  2V 1  2V  2 V  V2 1  V2
 X = V= =
dX 2V 2V 2V

2V
1 V 2 dV =
dX
X
 I 2V
1  V2
dV = I dX
X
+ log C
 I 2 V
(1  V)(1  V)
dV = log |X| + log C
 I !


 8

 8
"#
$
dV = log C|X|
1 3 log|1  V|
 log |1 + V| +  = log C|X|
2 2 1
1 V
 log = log C2X2
(1  V)3
1  Y/X XY
 = C2X2  = ±C2
(1  Y/X)3 (X  Y)3
 X + Y = C1(X Y)3, where C1 = ±C2
 (x 1) + (y 2) = C1((x 1) (y 2))3
 x + y 3 = C1(x y + 1)3.
Ordinary Differential EXERCISE J
Equations
Solve the following differential equations:
dy x  2y  3 dy y x1
1. = 2. =
NOTES dx 2x  y  3 dx yx5
dy x y1 dy 2x  y  1
3. = 4. = .
dx x y2 dx x  2y  3
Answers
y2 1
1. x + y = C(x y + 6)3 2. tan1 + log (x2 + y2 6x 4y + 13) = C
x3 2
3. y2 + 2xy x2 2x 4y = C 4. x2 y2 xy + x + 3y = C.
SOLUTION OF LINEAR DIFFERENTIAL EQUATION

G\
+ Py = Q, WHERE P AND Q ARE FUNCTIONS OF x
G[
OR CONSTANTS
dy
Let + Py = Q ...(1)
dx
be a linear differential equation, where P and Q are functions of x or constants.
P dx
Multiplying both sides of (1) by e I
, we get
dy
e I P dx + e I P dx Py = Q e
I P dx
.
dx
F
 e I P dx
dy
+
d I P dx
(e ). y = Q e I P dx '


 FZ
2FZ  2 
 
dx dx
d
 (y e I P dx ) = Q e I P dx
dx
 I ! d
dx
"#
( ye I P dx ) dx =
$ I Qe I P dx
dx + C
 yG I2 FZ
= I QG I2 FZ
dx + C.
P dx
This is the general solution of linear differential equation (1). The function e I
is called the integrating factor (I.F.) of (1).

Thus, the solution of (1) can also be written as
y(I.F.) =
3
+( FZ %
log f ( x )
Remark. In evaluating integrating factor (I.F.), the results e = f (x) is frequently
used.
G\
Working Steps for Solving + Py = Q
G[
F[
Step I. If the coefficient of is not unity, it must be made unity by dividing NOTES
FZ
F[
the equation by the coefficient of .
FZ
Step II. Identify P and Q and make sure that these are functions of x or
constants.
Step III. Evaluate  P dx.
Step IV. Find e P dx . This is the integrating factor (I.F.).
I
Step V. Put the value of I.F. in the general solution y(I.F.) =  Q(I.F.) dx + C and
simplify it. This gives the general solution of the given differential
equation.
SOLVED EXAMPLES
dy
Example 47. Solve: + 2y = ex.
dx
dy
Solution. We have + 2y = ex. ...(1)
dx
This is a linear differential equation. Here P = 2 and Q = ex.
 2 FZ = I 2 dx = 2x  I.F. = e
I P dx
= e2x
The solution of (1) is y(I.F.) = I Q(I.F.) dx + C.


ye2x =
G
ye2x = ex + C
Z
G Z FZ  % 

ye2x = I ex dx + C
y = ex + Ce2x.
Example 48. Solve: y 2y = cos 3x.
dy
Solution. We have 2y = cos 3x. ...(1)
dx
This is a linear differential equation. Here P = 2 and Q = cos 3x.
P dx
I
 P dx =  2 dx = 2x  I.F. = e I
= e2x
The solution of (1) is y(I.F.) = Q(I.F.) dx + C.
 EQU Z . e G
 Z
 ye2x = 2x dx + C  ye2x = cos 3x dx + C
 y = e2x I e2x cos 3x dx + Ce2x ...(2)
G
 Z
Let I= cos 3x dx
sin 3x sin 3x e2 x sin 3x 2

 I = e2x
3
  e2x
3
dx =
3
+
3 G
 Z
sin 3x dx
e2 x sin 3x 2   Z   EQU Z    EQU Z  
  G
 Z
Equations = +
3 
G     FZ 
3     
NOTES
=
e2 x sin 3x
3
2
e2x cos 3x
9
4
9 I e2x cos 3x dx
e 2x 4
 I= (3 sin 3x 2 cos 3x) I
9 9
1  4 I = e 2x
e 2 x

9 9 (3 sin 3x 2 cos 3x)  I=
13
(3 sin 3x 2 cos 3x)
 (2)  y = e2x
e 2 x "#
(3 sin 3 x  2 cos 3 x ) + Ce2x
! 13 #$
1
 (3 sin 3x 2 cos 3x) + Ce2x.
y=
13
dy
Example 49. Solve: + ay = emx.
dx
dy
Solution. We have + ay = emx. ...(1)
dx
This is a linear differential equation. Here P = a and Q = emx.
P dx
 P dx =  a dx = ax  I.F. = e = eax
I
The solution of (1) is y(I.F.) =

 3
+( dx + C.
OZ
G G
 yeax = eax dx + C  yeax =
C  O Z dx + C
e ( a m ) x G  CZ  G
C O Z
 yeax = +C  y= + C  eax
am CO
(Provided a + m  0)
G OZ
 y= + Ceax.
CO
This is the required solution of the given differential equation.
dy
If a + m = 0, then the given differential equation (1) becomes my = emx.
dx
Here P = m and Q = emx.
I 2 dx = I m dx = mx  I.F. = e I P dx
= emx
 The solution is yemx = I emx emx dx + C.
 y = emx
! I "#
 FZ  %
$  y = emx(x + C).
This is the required solution of the given differential equation provided

a + m = 0.
dy y
Example 50. Solve: = 2x2.
dx x
dy y
Solution. We have = 2x2. ...(1)
dx x
1 Differential Equations
This is a linear differential equation. Here P = and Q = 2x2.
I I
x
1 1
P dx = dx = log |x| = log (Assuming x > 0)
x x
1 NOTES
P dx
log
x 1
 I.F. = e =e =
I
x
The solution of (1) is
y(I.F.) = I Q(I.F.) dx + C.  y
1 
x I 2 x2
1 dx  C
x

y
x
2
Idy
x dx  C 
y
x
=2
x2
2
+C  y = x3 + Cx.
Example 51. Solve: cos2 x + y = tan x.

dx
dy dy
Solution. We have cos2 x + y = tan x .  + y sec2 x = tan x sec2 x
dx dx
...(1)
This is a linear differential equation. Here P = sec2 x and Q = tan x sec2 x.
P dx
I
 P dx =  sec2 x dx = tan x  I.F. = e I
= etan x
 yetan x = I tan x sec2 x etan x dx + C ...(2)
Let I= I tan x sec2 x etan x dx. z = tan x  dz = sec2 x dx
 I= I zez dz = zez I
= zez ez = (z 1) ez = (tan x 1) etan x
1.ez dz
 (2)  y etan x= (tan x 1)etan x + C

 y = tan x 1 + Cetan x.
dy
Example 52. Solve: (x2 1)
+ 2(x + 2)y = 2(x + 1).
dx
dy
Solution. We have (x2 1) + 2(x + 2)y = 2(x + 1).
dx
dy 2( x  2) 2
 + 2 y= ...(1)
dx x 1 x 1
2( x  2) 2
This is a linear differential equation. Here P = and Q= .
x2  1 x 1
I P dx = I 2( x  2)
x2  1
dx = I 3

1
x 1 x 1
dx

( x  1)3
= 3 log (x 1) log (x + 1) = log
x 1
(Assuming x 1, x + 1 > 0)
3
( x  1)
I P dx
log
x 1 ( x  1)3
I.F. = e =e =
I
x 1
Equations  y
( x  1)3
x 1
= I 2
x 1
.
( x  1)3
x 1
dx + C = 2 I x 2  2x  1
x 1
dx + C
NOTES
=2 I x  3  4 dx + C = 2 x
x  1 !2
2 "#
 3 x  4 log ( x  1) + C
#$

Z
 y = x2 6x + 8 log (x + 1) + C.
Z 
dy
Example 53. Solve: sin x + 3y = cos x.
dx
dy dy
Solution. We have sin x + 3y = cos x.  + 3y cosec x = cot x
dx dx
...(1)
I I
This is a linear differential equation. Here P = 3cosec x and Q = cot x.
x x
P dx = 3 cosec x dx = 3 log tan = log tan3
2 2
x
log tan 3 x
P dx 2
 I.F. = e I
=e = tan3
I
2
I I
x
x x 1  tan 2
 y tan3 = cot x tan3 dx + C = 2 . tan3 x dx + C
2 2 x 2
2 tan
2
=
1
2 I tan 2
x
2
dx + C
 tan 4

x
2
=
1
2 I !tan
x
2
2
 tan
x
2
sec
x
2
2 "
 1 # dx + C
$
2
=
1
2 I! 2 tan
x
2
2
 tan
x
2
sec
x"
2 #$
2
dx + C
2
=
1
2 I !
2 sec
x
2

 1  tan

2 x
2
sec
x "
2 $
# dx + C
2 2
1 4 tan x  2x – 2 tan x "# + C = 2 tan x x 1 tan

3 x
! 2 2$
= 3 +C
2 3 2 3 2
x x 1 x
 y tan3 = 2 tan x tan3 +C
2 2 3 2
[  Z Z

tan3

= 2 tan

x + C.
e 2 x dx = 1
Example 54. Solve: x


x dy
y
(x  0).
e 2 x
y dx = 1 dy e 2 x y
Solution. We have
x

x dy
i.e.,
dx

x

x
dy 1 2 x
 e
dx x
 y= ...(1)
x
1 e 2 x Differential Equations
This is a linear differential equation. Here P = and Q = .
I I
x x
P dx 
1
dx = 2 x  I.F. = e I P dx
=e
2 x
x
NOTES
The solution of (1) is y(I.F.) =
 3
+( FZ  % .
 ye 2 x
 I e 2 x
x
. e2 x
dx  C = I x 1/2 dx  C = 2 x + C
 [G Z = 2 Z + C.
dy
Example 55. Solve: x + y x + xy cot x = 0 (x  0).
dx
x dy + y x + xy cot x = 0 i.e.,
dy 1
Solution. We have
dx dx
+
x
 cot x y = 1.

...(1)
1
This is a linear differential equation. Here P = + cot x and Q = 1.
I I
x
P dx = 1
x
 cot x dx = log |x| + log |sin x|
= log |x sin x| = log (x sin x) (Assuming x sin x > 0)

I P dx
I
 I.F. = e = elog (x sin x) = x sin x
The solution of (1) is y(I.F.) = Q(I.F.) dx  C .
 y(x sin x) =
 
Z UKP Z FZ  %
= x( cos x)
I 
 EQU Z FZ  % = x cos x + sin x + C
 y = cot x +

+
% .
Z Z UKP Z
dy 
Example 56. Solve:  3y cot x = sin 2x, y = 2 when x = .
dx 2
Solution. We have dy (3 cot x)y = sin 2x. ...(1)
dx
This is a linear differential equation. Here P = 3 cot x and Q = sin 2x.
I P dx = I EQV Z dx = 3 log |sin x| = 3 log sin x = log (sin x)3
(Assuming sin x > 0)
 I.F. = e I P dx
= e log (sin x)
3
= (sin x)3
y(I.F.) =
 3
+( FZ  %
 y(sin x)3 =
I UKP Z
UKP Z  FZ  %
=2 I
UKP Z  EQU Z FZ  % = 2

UKP Z 

%
Ordinary Differential  y = 2 sin2 x + C sin3 x ...(2)
Equations
Now, y = 2 when x = /2.
 
 (2)  2 = 2 sin2 + C sin3  2 = 2 + C(1)  C=4
2 2
NOTES  (2)  y = 2 sin2 x + 4 sin3 x.
F[ 
Example 57. Solve: tan x = 2x tan x + x2 y, y = 0 when x = .
FZ
Solution. We have
F[
tan x = 2x tan x + x2 y. ...(1)
FZ
F[ Z [
 = 2x + 
FZ VCP Z VCP Z
F[
 + (cot x) y = 2x + x2 cot x.
FZ
This is a linear differential equation. Here P = cot x, and Q = 2x + x2 cot x.
 2FZ =  EQV Z FZ = log  sin x  = log sin x (Assuming sin

x > 0)
2FZ
 I.F. = G  = GNQI UKP Z = sin x
y(I.F.) =  3
+( FZ  %

 y sin x = 
Z  Z EQV Z sin x dx + C

 y sin x =  Z UKP Z FZ   Z EQU Z FZ  %

=  Z UKP Z FZ +  Z UKP Z   Z UKP Z FZ  + C
= x2 sin x + C
 y = x2 + C cosec x ...(2)

Now, y = 0 when x = .

 
 0 =   + C cosec

 
 0= + C . (1)  C=


 (2)  y = x2 cosec x.

EXERCISE K
dy dy
1. (i) + 2y = e4x (ii) 2y = e3x
dx dx
dy dy Differential Equations
(iii) + 2y = 6ex (iv) 4 + 8y = 5e3x
dx dx
dy dy
2. (i) +y=1 (ii) + y = ex
dx dx
dy dy NOTES
(iii) + y = e3x (iv) 4y = ex .
dx dx
dy FZ
3. (i) +y=2x (ii) (y + 3x2) =x
dx F[
(iii) x dy + (y x3)dx = 0 (iv) x dy (y + 2x2)dx = 0
dy dy
4. (i) + 2y = sin x (ii) y = cos x
dx dx
dy dy
(iii) + 2y = xe4x (iv) + y = cos 2x
dx dx
dy 1  x log x dy 1  sin x
5. (i) +y= (ii) +y=
dx x dx 1  cos x
dy dy
(iii) + y = cos x sin x (iv) x + 2y = x cos x
dx dx
dy y dy
6. (i) = 2x2 (ii) 2x + y = 6x3
dx x dx
dy y dy y
(iii)   x2 (iv) + = 3x2
dx x dx 2x
dy dy y
7. (i) sec x y = sin x (ii) + = ex
dx dx x
dy dy y sin x
(iii) x + 2y = x2 log x (iv) + = cos x +
dx dx x x
dy dy
8. (i) = y tan x 2 sin x (ii) + y sec x = tan x
dx dx
dy dy
(iii) y tan x = ex (iv) cos3 x + y cos x = sin x
dx dx
dy
9. (i) x y=x+1 (ii) (1 + x2)dy + 2xy dx = cot x dx
dx
dy dy
(iii) x + 2y = x2 (iv) (1 + x2) + y = tan1 x
dx dx
dy F[
10. (i) x log x + y = 2 log x (ii) Z  [  G Z
dx FZ
dy 3 x2 sin 2 x dy
(iii) + y = (iv) (1 + x2) + y = etan 1 x
dx 1  x3 1  x3 dx
dy 4x 1 dy 4x 1
11. (i) + y= (ii) + 2 y=
dx x2  1 ( x 2  1)3 dx x 1 ( x 2  1) 2
dy [ 1 dy
(iii) + = (iv) + y tan x = 2x + x2 tan x
dx Z NQI Z x dx
dy dy
12. (i) x ay = x + 1 (ii) y tan x = 2 sin x
dx dx
dy dy
(iii) + 2y tan x = sin x (iv) + y cot x = 2 cos x
dx dx
Ordinary Differential dy 2 dy 2
Equations 13. (i) x log x +y= log x (ii) (x2 1) + 2xy =
dx x dx x2  1
dy dy
14. (i) (x2 + 1) + 2xy = x2  4 (ii) x log x + y = log x
dx dx
NOTES F[ F[ Z  [ EQU Z
(iii) + y tan x = x2 cos2 x (iv) + =0
FZ FZ  UKP Z
dy dy
15. (i) sin x + y cos x = cos x sin2 x (ii) + y cot x = x2 cot x + 2x
dx dx
dy
(iii) (1 + x2) 2xy = (x2 + 2)(x2 + 1)
dx
dy
(iv) (1 x2) + xy = ax.
dx
Solve the following initial value problems (Q.No. 1618):
dy dy
16. (i) x  y = (x + 1)ex, y(1) = 0 (ii)  y cot x = 4x cosec x, y(/2) = 0
dx dx
dy
(iii) x  y = x cos x + sin x, y(/2) = 1
dx
dy
(iv) (x2 + 1) 2xy = ( x 4  2 x 2  1) cos x, y(0)  0
dx
dy dy
17. (i)  2 y tan x  sin x , y( /3 )  0 (ii) + y sec2 x = tan x sec2 x, y(0) = 1
dx dx
(iii)
F[ y = cos x, y(0) = 1 (iv) x
F[ + 2y = x2, y(2) = 1
FZ FZ
dy 1 dy
18. (i) (1 + x2) + 2xy = 2 , y(1) = 0 (ii)  y cot x = 2x + x2 cot x, x  0, y(/2) = 0
dx 1 x dx
(iii) (1 + y + x2y)dx + (x + x3)dy = 0, y(1) = 0
(iv) cos x dy = sin x(cos x 2y)dx, y(/3) = 0.
Answers
e4 x
1. (i) y = + Ce2x (ii) y = e3x + Ce2x
6
5 3x
(iii) y = 2ex + Ce2x (iv) y = e + Ce2x
4
1 x
2. (i) y = 1 + Cex (ii) y = e + Cex
2
1 3x 1 x
(iii) y = e + Cex (iv) y = e + Ce4x
2 3
3. (i) y = 3 x + Cex (ii) y = 3x2 + Cx
Z %
(iii) y =  (iv) y = 2x2 + Cx
Z
1 1
4. (i) y = (2 sin x cos x) + Ce 2x (ii) y = (sin x cos x) + Cex
5 2
1 4x 1 4x 1
(iii) y = xe  e  Ce2 x (iv) y = (2 sin 2x + cos 2x) + Cex
6 36 5
x
5. (i) y = log x + Cex (ii) y = tan + Cex
2
(iii) y = cos x + Cex (iv) yx2 = (x2 2) sin x + 2x cos x + C
6 3 C Differential Equations
6. (i) y = x3 + Cx (ii) y = x 
7 x
x3 C 6 3 C
(iii) y = + (iv) y = x +
4 x 7 x
x–1 x C
NOTES
7. (i) y + 1 + sin x = C esin x (ii) y = e +
x x
1 2 1 2 C C
(iii) y = x log |x| x + (iv) y = sin x +
4 16 x 2 x
cos 2x xC
8. (i) y cos x = +C (ii) y = 1
2 sec x  tan x
ex
(iii) y cos x = (sin x + cos x) + C (iv) y = tan x 1 + Cetan x
2
9. (i) y = x log x  1 + Cx (ii) (1 + x2)y = log |sin x| + C
x2 C
(iv) y = tan1 x 1 + C e tan x
1
(iii) y   2
4 x
10. (i) y log x = (log x)2 + C (ii) [G Z

 Z %
1 1
(x sin x cos x) + C (iv) yetan 1 x = e2 tan x + C
1
(iii) y(1 + x3) =
2 2
11. (i) y(x2 + 1)2 = tan1 x + C (ii) y(x2 + 1)2 = x + C
1 C
(iii) y = log x + (iv) y = x2 + C cos x
2 log x
x 1 1
12. (i) y =  + Cxa (ii) y = cos 2x sec x + C sec x
1 a a 2
1
(iii) y = cos x + C cos2 x (iv) y = cos 2x cosec x + C cosec x
2
2 log x 2 x1
13. (i) y log x = +C (ii) y(x2 1) = log +C
x x x1
x x2  4
14. (i) y(x2 + 1) = + 2 log |x + x2  4 | + C
2
1 C
(ii) y = log x  (iii) y sec x = (x2 2) sin x + 2x cos x + C
2 log x
Z
(iv) (1 + sin x) y + =C

1
15. (i) y = sin2 x + C cosec x (ii) y = x2 + C cosec x
3
(iii) y = (1 + x2)(x + tan 1 x + C) (iv) y = a + C 1  x2
16. (i) y = xe1 ex (ii) y sin x = 2x2 2/2
(iii) y = sin x (iv) y = (x2 + 1) sin x
17. (i) y = cos x 2cos2 x (ii) y = tan x 1 + 2etan x

(iii) y = (sin x cos x) + ex (iv) 4y = x2

2
18. (i) (1 + x2)y = tan1 x /4 (ii) y = x2
4 sin x

(iii) xy + tan1 x = (iv) y = cos x 2 cos2 x.
4
Equations
Hints
I
18. (ii) Here, I.F. = sin x.
 Solution is y sin x = (2 x  x 2 cot x) sin x dx  C .
I I I
NOTES
(2x + x2 cot x) sin x dx = 2x sin x dx + x2 cos x dx
= I (sin x)2x dx + I x2 cos x dx

= (sin x) x2 
I (cos x) x2 dx 
(iii) We have (1 + y + x2y)dx + (x + x3)dy = 0.

I x2 cos x dx = x2 sin x.
 (1 + y(1 + x2))dx + x(1 + x2)dy = 0
dy 1  y (1  x 2 ) 1 1

dx

x (1  x 2 )
= 
x(1  x 2 )

x
y.

dy
Solution of + Py = Qyn, where P and Q are Functions of x or
dx
Constants, by Reducing it to a Linear Differential Equation
dy
Consider the differential equation + Py = Qyn, ...(1)
dx
where P and Q are functions of x or constants and n  0, 1.
Equation (1) is known as Bernoullis equation.
dy
Dividing (1) by yn, we get yn + Pyn+1 = Q. ...(2)
dx
Let z= yn+1.
dz dy dy dz
 = (n + 1)yn or yn = ·
dx dx dx  P dx
dz dz
 (2)  · + Pz = Q  + P(1 n)z = Q(1 n). ...(3)
 P dx dx
(3) is a linear differential equation with z as the dependent variable.
G\ n
Working Steps for Solving + Py = Qy
G[
dy
Step I. Divide the equation by yn and get yn + Pyn+1 = Q ...(1)
dx
dz dy dy 1 dz
Step II. Put z = yn+1.  = ( n + 1)yn or yn = .
dx dx dx 1  n dx
dy
Putting the values of yn+1 and yn in (1), we get
dx
dz
+ P(1 n)z = Q(1 n). ...(2)
dx
Step III. (2) is a linear differential equation with dependent variable z.
dy 2 x Differential Equations
Example 58. Solve: + y= .
dx 3 y
dy 2 x
Solution. We have + y= . ...(1)
dx 3 y NOTES
This is a Bernoullis equation.
dy 2
Multiplying (1) by y , we get y
+ y3/2 = x. ...(2)
dx 3
dz 3 1/2 dy dy 2 dz
Let z = y3/2.  = y or y =
dx 2 dx dx 3 dx
2 dz 2 dz 3
 (2)  + z=x  +z= x ...(3)
3 dx 3 dx 2
3
Here P = 1 and Q= x.
I I
2
P dx
 P dx = 1  dx = x and we have I.F. = e = ex
I
The solution of (3) is z(I.F.) = I Q (I.F.) dx + C.
 zex = I
3
3
2
x ex dx + C  y3/2 ex =
3
2
3
ZG
!
Z
 I  GZ FZ "# + C
$
 y3/2 ex = (x 1)ex + C  y3/2 = (x 1) + Cex.
2 2
Example 59. Solve: y(x2y + ex)dx ex dy = 0.
Solution. We have y(x2y + ex)dx ex dy = 0
dy [
Z [  G Z dy x2 y2
 =  = +y
dx GZ dx ex
dy x y
2

dx
+ (1)y = e
x
2 ...(1)
This is a Bernoullis equation.

1 dy 1 x2
Dividing (1) by y2, we get + (1) = . ...(2)
y 2 dx y ex
1 dz dy 1 dy dz
Let z = .  = (1)y2 or =
y dx dx y 2 dx dx
dz x2 dz x2
 (2)  + (1)z = x  +1z= x ...(3)
dx e dx e
(3) is a linear differential equation with dependent variable z.
x2
Here P = 1 and Q= .
ex
 I P dx = I 1 dx = x and we have I.F. = e I P dx

= ex
Equations The solution of (3) is z(I.F.) = I Q(I.F.) dx + C.
 zex = I –
x2
e x
e
x dx + C
I
NOTES
1 Z
 ex = x2 dx + C  ex = + C.
y [
EXERCISE L
dy y dy y y2 dy x
1. + = y2 2. + = 3. + y=x y
dx x dx x x2 dx 1  x2
1
dy 2 y3 dy x2 dy
4. + y= 5. + xy = y2 e 2 sin x 6. + xy = xy5.
dx x x3 dx dx
Answers
1
1. + log x = C 2. 2x y = Cx2y
xy
1 1
3. 3 y x2 + 1 = C(1 x2)1/4 4. = +C
y2 x 4 3 x6
1 2
 x 1
= 1 + C e2 x .
2
5. e 2 = y(cos x + C) 6.
y4
dy
Solution of f (y) + Pf(y) = Q, where P and Q are Functions of x or
dx
Constants and f(y) is Some Function of y, by Reducing it to a Linear
Differential Equation
dy
Consider the differential equation f (y) + Pf (y) = Q, ...(1)
dx
where P and Q are functions of x or constants and f(y) is some function of y.
dz dy
Let z = f(y).  = f (y)
dx dx
dz
 (1)  + Pz = Q. ...(2)
dx
G\
Working Steps for Solving f ¢( y) + Pf (y) = Q
G[
dz dy
Step I Put z = f(y).  = f (y)
dx dx
dy
Step II Put the values of f(y) and f (y) in the given differential equation
dx
dz
and get + Pz = Q.
dx
This is a linear differential equation with dependent variable z.
dy 1 ey
Example 60. Solve: + = 2 .
dx x x
dy 1 ey NOTES
Solution. We have + = 2. ...(1)
dx x x
dy 1 1
Dividing (1) by ey, we get ey +  ey = 2 .
dx x x
dy 1
1
 ey
dx
+ 
x
ey = 2
x
...(2)
dy
(2) is a differential equation of the form f (y) + Pf(y) = Q, where f(y) = ey.
dx
dz dy
Let z = f(y) = ey.  = ey
dx dx
dz
1 1
 (2) 
dx
+ 

x
z= 2.
x
...(3)

1 1
Here P = and Q= .
I I
x x2
1 1
 P dx = dx = log x = log (Assuming x > 0)
x x
1
P dx
log 1
 I.F. = e I
= e x =
I
x
The solution of (3) is z(I.F.) = Q(I.F.) dx + C.
 ey 
1
x

I –
1
x 2
.
1
x
dx  C 
e y
x
=
Z 

+C
e y 1
 = +C  2xey = 1 + 2Cx2.
x 2x 2
dy
Example 61. Solve: sin y = cos y(1 x cos y).
dx
dy
Solution. We have sin y = cos y(1 x cos y).
dx
dy
 sin y cos y = x cos2 y ...(1)
dx
sin y dy cos y
Dividing (1) by cos2 y, we get = x.
cos2 y dx cos2 y
dy
 sec y tan y + (1) sec y = x ...(2)
dx
dy
(2) is a differential equation of the form f (y) + Pf (y) = Q, where f(y) = sec y.
dx
dz dy
Let z = f (y) = sec y.  = sec y tan y
dx dx
dz
 (2)  + (1)z = x ...(3)
dx
Here P = 1 and Q = x.
 I P dx = I 1 dx = x and we have I.F. = e P dx
= ex.
I
I
Equations
The solution of (3) is z(I.F.) = Q(I.F.) dx + C.
NOTES  (sec y)ex = I x  ex dx + C
 ex sec y = Z 
!
G

Z
 I 
G

Z "#
#$
FZ + C
GZ
 ex sec y = xex +C  sec y = x + 1 + Cex.

Remark. Please note carefully the placing of cos y on the LHS in equation (1). The
placing of x cos2 y on the LHS of (1) will not reduce the given differential equation to the
desired form.
EXERCISE M
dy dy 1
1. (1  x)  1  ex – y 2. + y = y3
dx dx x
dy x 2  y2  1 dy
3. = 4. 2 tan y + x sin2 y = x3 cos2 y
dx 2xy dx
dy 1 1 dy y y
5. + tan y = 2 tan y sin y 6. log y = 2 .
dx x x dx x x (log y )2
Answers
1. ey(1 + x) = ex + C 2. 2xy2 + Cx2y2 = 1
1 2
 x
3. y2 + 1 = x2 + Cx 4. tan2 y = x2 2 + C e 2
3
5. 2x cosec y = 1 + 2Cx2 6. (log y)3 = + Cx3.
4x
Hints
dy x 2  y2  1
3. We have = .
dx 2xy
dy x 2  y2  1 dy y2 1 dy 1
 2y =  2y =x+ +  2y +  (y2 + 1) = x.
dx x dx x x dx x
dy
Solution of Linear differential equation + Px = Q, where P and Q are
dx
functions of y or constants
FZ
Let + Px = Q ...(1)
F[
be a linear differential equation, where P and Q are functions of y or constants.
Multiplying both sides of (1) by e  P dy, we get
FZ
G I 2 F[ G I 2 F[ 2Z  3 G I 2 F[
.
F[
FZ F F Differential Equations
 G I 2 F[ 
F[ F[

G I

2 F[  Z  3G I 2 F[ '

 F[  2F[  2 
 
F 2 F[  3 G 2 F[

Z G
I I
F[ NOTES
F

 F[

Z G
 2 F[
F[ 

3 G 2 F[ F[  %
 Z G 2 F[  
3 G 2 F[ F[  %

This is the general solution of linear differential equation (1).

e  P dy is called the integrating factor (I.F.) of (1).
Thus, the solution of (1) can also be written as x(I.F.) =  3

+( dy + C.
G[
Working Steps for Solving + Px = Q
G\
Step I. Identify P and Q and make sure that these are functions of y or
constants.
Step II. Evaluate  P dy.
Step III. Find eP dy. This is the integrating factor (I.F.).
Step IV. Put the value of I.F. in the general solution x(I.F.) =  Q(I.F.) dy + C and
simplify it. This gives the general solution of the given differential
equation.
SOLVED EXAMPLES
Example 62. Solve: y dx + (x y3) dy = 0.

Solution. We have y dx + (x y3)dy = 0.
FZ FZ x = y
 y
F[
+ x y3 = 0 
F[
+
[ 2 ...(1)
FZ
(1) is a linear differential equation of the form + Px = Q.
F[

Here, P= and Q = y2.
[
I I
2 F[ 
[
dy = log y  I.F. = e P dy = elog y = y
(Assuming y > 0)
The solution of (1) is x(I.F.) = I 3

+( F[ + C.
 xy = I [ [ F[ + C  xy =
[

+ C.
dy
Example 63. Solve: (x + 3y2) = y, y > 0.
dx
dy
Solution. We have (x + 3y2) = y.
dx
Equations dx dx x dx 1
 y
dy
= x + 3y2 
dy
=
y
+ 3y 
dy
+ 
y
x = 3y ...(1)
FZ
NOTES This is a linear differential equation of the form + Px = Q.
F[
1
Here, P= and Q = 3y.
y
I P dy = I 1
y
dy = log y = log y1 = log

1
1
y
(y > 0  |y| = y)
 I.F. = e P dy = elog 1/y =

I
I
y
The solution of (1) is x(I.F.) = Q(I.F.) dy + C.
 x
1 =
y I 3y
1 dy + C
y 
Z
[
= 3y + C.
FZ
Example 64. Solve: y2 Z = 0.
F[ [
FZ
Solution. We have y2 Z = 0.
F[ [
FZ FZ

 y2
F[
Z
[

F[

[
Z
[
...(1)
FZ
F[

Here, P= and Q= .
[ [
I 2 F[  I [

F[ 

[
 I.F. = e P dy = e1/y
The solution of (1) is x (I.F.) = I 3

+( F[ + C.
 xe1/y = I
[
G [ F[ + C ...(2)
Let I= I
[
G [ F[

Let z=  y= and dy = dz
[ \ \
 I= I \ G \

F\  I \G F\
\
I
\
\G G F\"# = ze + e = e (1 z) = e 
! $ [
\ \ z z z 1/y
=
 + C  x = 1 + + Ce .
  xe 1/y = e 1/y
[
(2) 1/y
[
F[
Example 65. Solve: (1 + y2) + (x G VCP = 0.
[
)
FZ

[ F[
Solution. We have (1 + y2) + (x G VCP ) = 0.
FZ
FZ FZ G VCP [
 (1 + y2)  Z G VCP [
=0   Z  ...(1)
F[ F[  [  [
FZ
Equation (1) is a linear differential equation of the form + Px = Q. NOTES
F[

[
G
VCP
Here P= and Q = .
 [  [
I 2 F[  I  [

F[  VCP
[  I.F. = e  P dy = G VCP

[
The solution of (1) is x(I.F.) = I 3

+( F[ + C.
I

VCP [

[ G
[
 x G VCP = G VCP F[  %
 [
 x G VCP

[
= I G
\
F\ + C, where z = tan1 y

\ VCP [
G G
%

 xG VCP [
= +C  2x G VCP [
= G VCP [
+ 2C.

Example 66. Solve: y dx (x + 2y2)dy = 0, y > 0 given that y = 1 when x = 2.
Solution. We have y dx (x + 2y2)dy = 0.
dx dx x dx 1 x = 2y
 y
dy
= x + 2y2 
dy
=
y
+ 2y 
dy
+ 
y ...(1)
FZ
F[
1
Here P= and Q = 2y.
y
I P dy = I
1
y
dy = log y = log y1 = log
1
y
(y > 0   y  = y)
1
log 1
P dy y
 I.F. = e =e =
I
The solution of (1) is x(I.F.) = I Q(I.F.) dy + C.
 x
1 =
y I 2y
1 dy + C
y 
x
y
= 2y + C ...(2)
2
Now y = 1 when x = 2.  (2)  = 2(1) + C or C = 0
1
x
 Using (2), the required solution is = 2y + 0 i.e., x = 2y2.
y
Ordinary Differential EXERCISE N
Equations
F[
1. (i) y dx (x + 2y2)dy = 0 (ii) (x + 2y3) = y, y > 0
NOTES FZ

F[
(iii) (3y2 x)dy = y dx, y > 0 (iv) y2 + Z

[ FZ
=0
F[
2. (i) (1 + y2)dx = (tan1 y x)dy (ii) (1 + y2) + (2xy cot y) 
FZ
F[ F[
(iii) (2x 10y3) +y=0 (iv) (x + y) = 1.
FZ FZ

VCP [
3. (i) (1 + y2)dx + (x G )dy = 0, y(0) = 0
(ii) (1 + y2)dx = (tan 1 y x)dy, y(1) = 0
(iii) (x sin y)dy + (tan y)dx = 0, y(0) = 0.
FZ 
(iv) + x cot y = 2y + y2 cot y, (y  0), y(0) = .
F[
Answers
1. (i) x = 2y2
+ Cy (ii) x = y3 + Cy
(iii) xy = y3 + C (iv) x = 1 + y 1 + Ce1/y

VCP [
2. (i) x = tan1 y 1 + C G (ii) x(1 + y2) = log  sin y  + C
(iii) x = 2y3 + Cy 2 (iv) x + y 1 = Cey

3. (i) x GVCP [ = tan 1 y (ii) (x tan1 y + 1) GVCP [ = 2
(iii) 2x = sin y (iv) 4(x y2) sin y + 2 = 0.
SUMMARY
1. An equation involving independent and dependent variables and at least one derivative/
differential of these variables called a differential equation.
2. The order of a differential equation is the order of the derivative of the highest order,
occurring in the differential equation.
3. The degree of a differential equation is defined if it can be written as a polynomial equation
in the derivatives and for such a differential equation its degree is given by the highest
power of the highest order derivative appearing in it, provided the derivatives are made
free from radicals and fractions.
4. A differential equation is said to be linear, if the dependent variable and its derivatives
occur only in the first degree and are not multiplied together.
5. (i) A solution of a differential equation is a functional relation between the variables
involved which satisfies the given differential equation.
(ii) A solution of a differential equation is called the general solution (or complete
solution), if it contains as many arbitrary constants as the order of the differential
equation.
(iii) A solution obtained by giving particular values to arbitrary constants in the general
solution of a differential equation is called a particular solution of the differential
equation, under consideration.
6. (i) If
F[
FZ
= f (x), then dy = f (x) dx. 
I I F[  H
Z FZ + C.
This represents the general solution of the given differential equation.
(ii) If
F[
FZ
= g(y), then
F[
I
[
= f (x). 
I I F[
I
[
 H
Z  % .
NOTES
(iii) If
dy
dx
= f(x) g(y), then
dy
g ( y)
= f(x) dx.  I 1
g( y)
dy 

I f ( x) dx  C .
dy
7. If = f(ax + by + c), then z = ax + by + c reduces the given differentiable equation to
dx
variable separable form.
dy y
8. If
dx
= f
x is a homogeneous equation, then y = vx reduces the given differential
equation to variable separable form.
F[ C Z  D [  E C D
9. If  CPF  then put x = X + h and y = Y + k where h and k are
FZ C Z  D [  E C D
constant such that a1h + b1k + c1 = 0, a2h + b2 k + c2 = 0. The substitution Y = VX
reduces the resultant equation to variable separable form.
dy
10. If + Py = Q is a linear differential equation, where P and Q are functions of x or
I
dx
constants, then [G I 2 FZ =
3G
I
2 FZ FZ  %
is the general solution of the given differential equation.
dy
11. A differential equation of the form + Py = Qyn, where n  0, 1 and P and Q are functions
dx
of x or constants, is solved by putting z = yn+1. This substitution reduces the given
differential equation to a linear differential equation.
dy
12. A differential equation of the form f (y) + f(y)P = Q, where P and Q are functions of x
dx
or constants, is solved by putting z = f(y). This substitution reduces the given differential
equation to a linear differential equation.
FZ
13. If + Px = Q is a linear differential equation, where P and Q are functions of y or
F[
constants, then Z GI 2 F[ 
equation.
I
3 G 2 F[ F[  % is the general solution of the given differential
I
Equations UNIT II
NOTES
2. EXACT DIFFERENTIAL
EQUATIONS
STRUCTURE
Introduction
Theorem
Equations Reducible to Exact Equations
INTRODUCTION
A differential equation obtained from its primitive directly by differentiation,

without any operation of multiplication, elimination or reduction etc. is said to be an
exact differential equation.
Thus a differential equation of the form M (x, y) dx + N (x, y) dy = 0 is an exact
differential equation if it can be obtained directly by differentiating the equation u
(x, y) = c, which is its primitive.
i.e., if du = Mdx + Ndy.
For example, the equation xdx + ydy = 0 is an exact differential equation, as it
can be obtained from its primitive x2 + y2 = c2 directly by differentiation.
THEOREM
The necessary and sufficient condition for the differential equation Mdx + Ndy =
0 to be exact is
M N
 .
y x
The condition is necessary
The equation Mdx + Ndy = 0 will be exact, if du = Mdx + Ndy
u u
But du = dx  dy
x y
u u
 Mdx + Ndy = dx  dy
x y
Equating co-efficients of dx and dy, we get
u u
M= and N =
x y
Exact Differential
M  2 u N  2 u Equations
  and 
y yx x xy
 2u  2u
But  NOTES
yx xy
M N
 
y x
which is the necessary condition of exactness.
The condition is sufficient.
Let u= I
y constant
Mdx
u  2 u M
 M and 
x yx y
 2u  2u M N
But  and 
yx xy y x
N  2 u  u
  
x xy x y
u
Integrating both sides w.r.t. x treating y as constant, we have N =  f ( y)
y
u u%& () ' M
u
,N=
u
 f ( y)
"#
 Mdx + Ndy =
x
dx 
y
 f ( y) dy
' * ! x y $
u dx  u dy + f(y) dy = du + f(y) dy = d[u +  f(y) dy]
=
x y
which shows that Mdx + Ndy is an exact differential and hence Mdx + Ndy = 0 is an
exact differential equation.
Note. Since Mdx + Ndy = d[u +  f(y) dy]
 Mdx + Ndy = 0  d[u + f(y) dy] = 0
I
Integrating, u +  f(y) dy = c
But u= Mdx and f(y) = terms of N not containing x

y constant
I I
Hence the solution of Mdx + Ndy = 0 is
Mdx  (terms of N not containing x) dy = c.
y constant
SOLVED EXAMPLES
Example 1. Solve (5x4 + 3x2y2 2xy3) dx + (2x3y 3x2y2 5y4) dy = 0.

Sol. Here M = 5x4 + 3x2y2 2xy3 and N = 2x3y 3x2y2 5y4
M N
  6 x 2 y  6 xy 2 
y x
I I
Thus the given equation is exact and its solution is
Equations
Mdx  (terms of N not containing x) dy = c
y constant
NOTES
i.e.,
I
y constant
(5x4 + 3x2y2 2xy3) dx + I 5y4dy = c
or x5 + x3y2 x2y3 y5 = c.
Example 2. Solve [cos x tan y + cos (x + y)] dx + [sin x sec2 y + cos (x + y)] dy = 0.
Sol. Here, M = cos x tan y + cos (x + y)
and N = sin x sec2 y + cos (x + y)
M N
 = cos x sec2 y sin (x + y) =
y x
I Mdx  I (terms of N not containing x) dy = c
I
y constant
i.e., [cos x tan y + cos (x + y)] dx = 0

y constant
or sin x tan y + sin (x + y) = c.

dy y cos x  sin y  y
Example 3. Solve   0.
dx sin x  x cos y  x
Sol. The given equation can be written as
(y cos x + sin y + y) dx + (sin x + x cos y + x) dy = 0
Here M = y cos x + sin y + y and N = sin x + x cos y + x
M N
 = cos x + cos y + 1 =
y x
I I
Mdx  (terms of N not containing x) dy = c
I
y constant
i.e., (y cos x + sin y + y) dx = c

y constant
or y sin x + (sin y + y) x = c.
EXERCISE A
Solve the following differential equations (1 to 22):
1. (1 + 4xy + 2y2)dx + (1 + 4xy + 2x2) dy = 0 2. (3x2 + 6xy2)dx + (6x2y + 4y3) dy = 0
3. y(y2 3x2)dy + x(x2 3y2) dx = 0, y(0) = 1 4. (2x3 xy2 2y + 3) dx (x2y + 2x)dy = 0
dy ax  hy  g  y2 1 1 x
2 
5.  0 6.  2
  dx    2  dy  0
dx hx  by  f  ( y  x) x  y ( x  y) 
2 Exact Differential
xdy  ydx a ( xdy  ydx )
7. xdy + ydx + 2 2
0 8. xdx  ydy  2 2 Equations
x  y x y
9. dx 
y
1  x 2 y2
dx 
x
1  x 2 y2
dy 
10. 2x 1 
2
x  y dx  2
x  y dy
NOTES
11. ( y cos x + 1) dx + sin x dy = 0
12. (i)  y  1  1   cos y  dx  ( x  log x  x sin y) dy  0

  x 
  1 
(ii)  y  1   cos y  dx  (x  log x )(cos y  y sin y) dy  0
  x 
13. 2xy  y  tan y  dx  ( x 2  x tan2 y  sec2 y)dy  0

x/y  x  x/y
14. (1  e ) dx   1   e dy  0 15. eydx + (xey + 2y) dy = 0
 y
2 2
16. yexy dx + (xexy + 2y) dy = 0 17. ( y2e xy  4x 3 ) dx  2xye xy  3 y2 ) dy  0
18. (sec x tan x tan y ex)dx + sec x sec2 y dy = 0
19. (sin x cos y + e2x) dx + (cos x sin y + tan y) dy = 0
2 2 2
20. (2xy cos x  2xy  1) dx  (sin x  x ) dy  0
x
21. e (cos y dx  sin y dy)  0, y(0)  0
 1 sin x 9
22.  cos x log(2 y  8)  x  dx  y  4 dy  0, y (1) 
2
23. Find the value of  for which the differential equation ( xy 2   x 2 y)dx  ( x  y)x 2dy  0 ,
is exact. Hence solve it.
Answers
1. (x + y) (1 + 2xy) = c 2. x3 + 3x2y2+ y4 = c 3. x4 6x2y2 + y4 = 1
4. x4 x2y2 4xy + 6x = c 5. ax2 + 2hxy + by2 + 2gx + 2fy + c = 0
y2 y 1  x 
6.  log  c 7. xy  tan    c
yx x  y
x 1  xy
8. x 2  y2  2a 2 tan 1    c 9. log  2x  c
2
10. 3x  2(x  y)
2 3/2
c
y 1  xy
11. y sin x + x = c 12. (i) y(x + log x) + x cos y = c (ii) y cos y (x + log x) = c
13. x2y + xy x tan y + tan y = c 14. x + yex/y = c 15. xey + y2 = c
2
16. exy + y2 = c 17. e xy  x 4  y3  c 18. sec x tan y ex = c
1 2x
19. cos x cos y + e  log sec y  c 20. y sin x2 x2y + x = c
2
21. ex cos y = 1 22. sin x log (2y 8) + log x = 0
1 2 2 3
23.  = 3; x y x y  c.
2
Equations EQUATIONS REDUCIBLE TO EXACT EQUATIONS
Differential equations which are not exact can sometimes be made exact after
NOTES multiplying by a suitable factor (a function of x and/or y) called the integrating factor.
For example, consider the equation y dx x dy = 0 ...(1)
Here, M = y and N=x
M N , therefore the equation is not exact.

y x
1
(i) Multiplying the equation by 2 , it becomes
y
ydx  xdy x
y2
= 0 or d
y
=0

which is exact.
1
(ii) Multiplying the equation by , it becomes
x2
ydx  xdy y = 0
x2
=0 or d
x which is exact.
1
(iii) Multiplying the equation by , it becomes
xy
dx dy
 = 0 or d(log x log y) = 0 which is exact.
x y
1 1 1
 , and are integrating factors of (1).
y2 x 2 xy
If a differential equation has one integrating factor, it has an infinite number of
integrating factors.
I.F. Found by Inspection

In a number of problems, a little analysis helps to find the integrating factor.
The following differentials are useful in selecting a suitable integrating factor.
xdy  ydx y
(i) ydx + xdy = d(xy) (ii)
x 2 x
d
ydx  xdy x xdy  ydx

 d tan
y

x
1
(iii) 2
d (iv)
y y x y 2 2
xdy  ydx
y "# ydx  xdy
(v)
xy
 d log
!
x $ (vi)
xy
= d [log (xy)]
xdx  ydy 1
log ( x 2  y 2 )
"# xdy  ydx 1 x y
(vii)
x 2  y2
d
2 ! $ (viii)
x 2  y2
d
2
log
x y
.

SOLVED EXAMPLES
3
Example 4. Solve ydx xdy + 3x2y2 e x dx = 0.
3 3 3
Sol. Since 3 x 2 e x = d( e x ), the term 3 x 2 y2 e x dx should not involve y2.
1
This suggests that may be an I.F.
y2
1 ydx  xdy 3 Exact Differential
Multiplying throughout by 2
, we have 2
+ 3x2 e x dx = 0 Equations
y y
x  d (e x3
or d
y ) = 0, which is exact.
NOTES
x 3
Integrating, we get  e x  c , which is the required solution.
y
Example 5. Solve xdy ydx = x x 2  y 2 dx.
xdy  ydx
y
1
2
x2
Sol. The given equation is xdy ydx = x2
x dx or

y
2
 dx
1

x
y

1
or d sin = dx, which is exact.
x
y
Integrating, we get sin1 =x+c or y = x sin (x + c), which is the required
x
solution.
a 2 (xdy  ydx)
Example 6. Solve: xdx + ydy = .
x2  y2
y

1
Sol. The given equation is xdx + ydy a2d tan =0
x
x2 y2 y
Integrating, we get   a 2 tan 1  c
2 2 x
y
or x2 + y2 2a2 tan1 = C, where C = 2c.
x
EXERCISE B
1. xdy ydx = (x2 + y2) dx 2. xdy ydx = (x2 + y2) (dx + dy)
3. y(2xy + ex) dx = ex dy 4. (y log y 2xy) dx + (x + y) dy = 0
5. xdy ydx = xy2 dx 6. xdy = (x2y2 y) dx
2 dy   dy 
7. ( x  y)  x  y   xy  1   8. xdy  ydx  (4x 2  y2 )dy
 dx   dx 
9. (y + y2 cos x) dx (x y3) dy = 0.
Answers
1 y 1 y
1. tan  xc 2. tan  x yc
x x
ex
3. x2  c 4. x log y x2 + y = c
y
x x2 1
5.  c 6.   xc
y 2 xy
1 1 1  y 
7. log (xy)   c 8. tan    yc
x y 2  2x 
x y2
9.  sin x   c.
y 2
Ordinary Differential Hints
Equations
xdy  ydx  y 1 ye x dx  e x dy  ex 
1. 2 2
 dx  d  tan 1   dx 3. I.F.  2
and 2
 d 
x y  x y y  y
NOTES 1 x xdy  ydx d( xy )
4. I.F.  and dy  log ydx  d ( x log y) 6.  dx   dx
y y x 2 y2 ( xy )2
xdy  ydx dx  dy d( xy ) d( x  y )
7.   
xy ( x  y )2 xy ( x  y )2
 y
d 
xdy – ydx ( xdy  ydx )/x 2  x   dy
8.  dy   dy  .
4 x 2  y2 4  ( y/x )2 y
2
4 
x 
I.F. for a Homogeneous Equation

1
If Mdx + Ndy = 0 is a homogeneous equation in x and y, then is an I.F.
Mx  Ny
provided Mx + Ny  0.
Note. If Mx + Ny consists of only one term, use the above method of I.F. otherwise,
proceed by putting y = vx.
Example 7. Solve: (x2y 2xy2) dx (x3 3x2y) dy = 0.

Sol. The given equation is homogeneous in x and y with
M = x2y 2xy2 and N = x3 + 3x2y
Now, Mx + Ny = x3y 2x2y2 x3y + 3x2y2 = x2y2  0
1 1
 I.F. =  2 2.
M x  Ny x y
1
Multiplying throughout by , the given equation becomes
x y2 2
1  2 dx  x  3 dy = 0, which is exact.
y x y y 2
The solution is I 1  2 dx  3 dy  c

y x
y constant
y I
x
or 2 log x + 3 log y = c.
y
EXERCISE C
1. (xy 2y 2) dx (x2 3xy) dy = 0 2. x2y dx (x3 + y3) dy = 0
3. (3xy2 y3) dx (2x2y xy2) dy = 0 4. (x2 3xy + 2y2) dx + x(3x 2y)dy = 0.
Answers
3
x x
1.  2 log x  3 log y  c 2. log y  3
c
y 3y
3. y 2
4. x log x  3xy  y  cx
2 2
3 log x  2 log y  c
x
I.F. for an Equation of the Form f1(xy) ydx + f2(xy) xdy = 0. Exact Differential
Equations
1
If Mdx + Ndy = 0 is of the form f1(xy) ydx + f2(xy) xdy = 0, then is an
Mx – Ny
I.F. provided Mx Ny  0. NOTES
Example 8. Solve: y (xy + 2x2y2) dx + x(xy x2y2) dy = 0.

Sol. The given equation is of the form f1(xy) y dx + f2(xy) x dy = 0.
Here, M = xy2 + 2x2y3 and N = x2y x3y2
Now, Mx Ny = x2y2 + 2x3y3 x2y2 + x3y3 = 3x3y3  0
1 1
 I.F. =  3 3
Mx  Ny 3 x y
1
Multiplying throughout by , the given equation becomes
3 x3 y3
1  2 dx  1  1 dy  0
3 x y 3 x 3xy 3 y
2 2
which is exact. The solution is

I
1  2 dx   1 dy  c
3x y 3x 3y
y constant
2 I
1 2 1
or   log x  log y = c
3 xy 3 3
1
or  + 2 log x log y = C, where C = 3c.
xy
EXERCISE D
1. (1 + xy) ydx + (1 xy)xdy = 0. 2. (x2 y2 + xy + 1) ydx + (x2y2 xy + 1)xdy = 0.
3. y(2xy + 1)dx + x(1 + 2xy x3y3)dy = 0. 4. (xy2 + 2x2y3)dx + (x2y x3y2) dy = 0.
5. (y xy2)dx (x + x2y) dy = 0.
6. (xy sin xy + cos xy) ydx + (xy sin xy cos xy) xdy = 0.
Answers
1 x x 1
1.   log    c 2. xy  log    c
xy  y  y  xy
1 1 1
3. 2 2
 3 3
 log y  c 4.   2 log x  log y  c
x y 3x y xy
x
5. log    xy  c 6. y cos xy = cx.
 y
For the equation Mdx + Ndy = 0

M N

y x
(i) If = f(x), a function of x only, then ef(x)dx is an I.F.
N
Ordinary Differential N M
Equations 
x y
(ii) If = g(y), a function of y only, then eg(y)dy is an I.F.
M
NOTES
SOLVED EXAMPLES
1
3
Example 9. Solve: (xy2 ex ) dx x2y dy = 0.
1
3
Sol. Here, M= xy2 ex and N = x2y
M N

y x 2 xy  ( 2 xy) 4
 2
  , which is a function of x only.
N x y x
 I.F. = e
I 
4
x
dx
 e 4 log x 
1
x4
y 1
, we have
2 1
x  x e dx  x
1 y x3
Multiplying throughout by dy = 0
x4 3 4 2
I y 1 2 1
which is exact. The solution is x  x e dx  c x3
y constant 3
4
I I
1
y2 1 3 3 y2 1 1
or    e x dx  c or   e t dt = c, where t =
2x 2
3 x 4
2x 2
3 x3
1
y2 1 3 y2 3
or  2
 et  c or  2
 2e x = C, where C = 6c.
2x 3 x
Example 10. Solve: (xy3 + y) dx + 2(x2y2 + x + y4) dy = 0.

Sol. Here, M = xy3 + y and N = 2x2y2 + 2x + 2y4
N M

x y 4 xy 2  2  3 xy 2  1 xy2  1 1
 3
 2

M xy  y y( xy  1) y
which is a function of y only.
 I.F. = e
I 1
y
dy
= elog y = y
Multiplying throughout by y, we have (xy4 + y2) dx + 2 (x2y3 + xy + y5) dy = 0

I
y constant
(xy4 + y2) dx + I 2y5 dy = c
x2 y4 y6
or  xy2   c.
2 3
EXERCISE E Exact Differential
Equations
1. (x2 + y2 + x) dx + xydy = 0 2. (x2 + y2 +1) dx 2xydy = 0.
3. (x2 + y2 + 2x)dx + 2ydy = 0. 4. (y4 + 2y)dx + (xy3 + 2y4 4x) dy = 0. NOTES
 y
3
x 
2
1 2
5. y    dx  (x  xy )dy  0. 6. (x sec2 y x2 cos y)dy = (tan y 3x4)dx.
 3 2  4
7. (xyex/y + y2) dx x2ex/y dy = 0. 8. (3xy 2ay2)dx + (x2 2axy)dy = 0.
9. (x4ex 2mxy2 )dx + 2mx2ydy = 0. 10. y(2x2y + ex)dx = (ex + y3)dy.
11. ydx xdy + log x dx = 0. 12. (2x log x xy) dy + 2ydx = 0.
13. (3x2y4 + 2xy)dx + (2x3y3 x2) dy = 0. 14. y log y dx + (x log y) dy = 0.
15. (xy3 + y)dx + 2 (x2y2 + x + y4)dy = 0.
Answers
2
y 1 x 2 2
1. 3x4 + 6x2y2 + 4x3 = c 2. x   c 3. e (x  y )  c
x x
 2  2 tan y
4. y  2 x  y  c 5. 3x2y + x4y3 + x6 = c 6.  x 3  sin y  c
 y  x
2
x  y
7. ex/y + log x = c 8. x2y(x ay) = c 9. e  m    c
x
2x 3 e x y2 1 2
10.   c 11. 1 + y + log x = cx 12. 2 y log x  y c
3 y 2 2
2 2 4 6
3 2 x 1 2 x y 2 y
13. x y  c 14. x log y  (log y)  c 15.  xy   c.
y 2 2 3
I.F. for an equation of the form

xayb (my dx + nx dy) + xcyd (py dx + qx dy) = 0
where a, b, c, d, m, n, p, q are all constant is xhyk, where h, k are so chosen that after
multiplication by xhyk the equation becomes exact.
Example 11. Solve (2x2y2 + y) dx + (3x x3y) dy = 0.

Sol. The equation can be written as 2(x2y2 dx x3ydy) + (y dx + 3xdy) = 0
or x2y (2ydx xdy) + x0y0 (y dx + 3xdy) = 0
which is of the form mentioned above. Therefore, it has an I.F. of the form xhyk.
Multiplying the given equation by xhyk, we have
(2xh+2yk+2 + xh yk+1) dx + (3xh+1yk xh+3yk+1) dy = 0
M N
For this equation to be exact, we must have 
y x
i.e., 2(k + 2) xh+2yk+1 + (k + 1) xhyk = 3(h + 1) xhyk (h + 3) xh+2yk+1
which holds when 2(k + 2) = (h + 3) and k + 1 = 3(h + 1)
i.e., when h + 2k + 7 = 0 and 3h k + 2 = 0
11 19
Solving these equation, we have h = ,k
7 7
 
Equations 7 y 7
 I.F. = x
11 19
 
Multiplying the given equation by x 7 y 7 , we have
NOTES
3
5 11 12 4 19 10 12
dy = 0
2x
y x 7 y

dx + 3 x y
7

7

7

7

7  x7 y

7

2 x y I
y constant
3
7

5
7 x

11

7 y 7
12 dx  c

10 5 4 12 10 5 4 12
7  7 7     20
or x 7 y 7  x y 7 c or 4x 7 y 7  5x 7 y 7  C , where C = c.
5 4 7
Note. The values of h and k can also be determined from the relations
ah1 bk1 ch1 dk1
 and  .
m n p q
Comparing the given equation
x2y(2y dx x dy) + x0y0(y dx + 3x dy) = 0
with xayb (my dx + nx dy) + xcyd (py dx + qx dy) = 0
we have a = 2, b = 1, c = 0, d = 0
m = 2, n = 1, p = 1, q = 3
a h1 b k1 2 h1 1 k1
   
m n 2 1
or 3 + h = 4 2k or h + 2k + 7 = 0 ...(1)
ch1 dk1 0h1 0k1
Also,   
p q 1 3
or 3h k + 2 = 0 ...(2)
11 19
Solving (1) and (2), we have h= ,k .
7 7
EXERCISE E
1. (x2y + y4)dx + (2x3 + 4xy3)dy = 0 2. (y2 + 2x2y)dx + (2x3 xy)dy = 0
3. (2x2y 3y4)dx + (3x3 + 2xy3)dy =0 4. (y3 2x2y)dx + (2xy2 x3)dy = 0
5. (2ydx + 3xdy) + 2xy(3ydx + 4xdy) = 0 6. x(3ydx + 2xdy) + 8y4 (ydx + 3xdy) = 0
7. (2y2 4x2y) dx + (4xy + 3x3) dy = 0.
Answers
3/2
 y
1. 7x11/2y11 + 11x7/2y1/4 = c (I.F. = x5/2y10) 2. 6 xy    c (I.F. = x5/2y1/2)
x
3. 5x36/13 y24/13 12x10/13 y15/13 = c (I.F. = x49/13y28/13)
4. x2y4 y2x4 = c (I.F. = xy) 5. x2y3 (1 + 2xy) = c (I.F. = xy2)
2 4 20 5
  
11 y 11  x 11 y 11  c (I.F. = x13/11y26/11)
6. x3y2 + 4x2y6 = c (I.F. = xy) 7. 5x
UNIT III linear Differential
Equations of the
First Order
3. LINEAR DIFFERENTIAL NOTES
EQUATIONS OF THE FIRST ORDER
STRUCTURE
Definition
To Solve the Equation + Py = Q, where P and Q are Functions of x only
(Leibnitzs Equation)
Bernoullis Equation (Equations Reducible to the Linear Form)
Differential equations of the first order and higher degree
Equations Solvable For p
Equations Solvable For y
Equations Solvable For x
Clairauts Equation
DEFINITION
A differential equation is said to be linear if the dependent variable and its

derivative occur only in the first degree and are not multiplied together.
Thus, the standard form of a linear differential equation of the first order is
dy
+ Py = Q, where P and Q are functions of x or constants (i.e., independent of y).
dx
TO SOLVE THE EQUATION

G\ + Py = Q, WHERE P AND
G[
Q ARE FUNCTIONS OF x ONLY (Leibnitzs Equation)
dy
The given equation is + Py = Q
dx
P dx
Multiplying throughout by e I
, we get
dy P dx P dx P dx
.e I
+ Py . e I
=Q. e I
...(1)
dx
Ordinary Differential dy d
d P dx P dx P dx
Equations Now, [y e I
]= .e I
+y. [e I
]
dx dx dx
dy P dx P dx d
= .e I
+y. e I
. [ P dx]
NOTES dx dx
' d
> C
e f ( x)  e f ( x) .
d
; @"#$
! dx dx
f ( x)
dy P dx P dx dy P dx P dx
= .e I
+y. e I
.P= .e I
+ Py . e I
dx dx
d
 From (1), [y . e P dx ] = Q . e I I P dx
dx
Integrating both sides w.r.t. x, we have
y . e P dx =  Q . e P dx dx + c
I I
which is the required solution of the given linear differential equation.
Note 1. The factor e I P dx , on multiplying by which the LHS of the differential equation
becomes the differential co-efficient of some function of x and y, is called an integrating factor of
the differential equation and is shortly written as I.F.
dy
Note 2. The solution of the linear equation + Py = Q, where P and Q are
dx
functions of x only, is
y(I.F.) =  Q(I.F.) dx + c
Note 3. Sometimes a differential equation becomes linear if we take y as the independent
variable and x as dependent variable. In that case, the equation can be put in the form
dx
+ Px = Q, where P and Q are functions of y (and not of x) or constants.
dy
I.F. (in this case) = e  P dy , and the solution is

x(I.F.) =  Q. (I.F.) dy + c.
Note 4. While evaluating the I.F., it is very useful to remember that
elog f ( x ) = f(x).
2
Thus, elog x = x2.
dy
Note 5. The co-efficient of , if not unity, must be made unity by dividing
dx
throughout by it.
SOLVED EXAMPLES
Example 1. Solve the following :

dy dy
(i) (1 + x2) + 2xy = 4x2 (ii) = y tan x 2 sin x.
dx dx
dy
Sol. (i) Given equation is (1 + x2) + 2xy = 4x2
dx
dy

Dividing throughout by 1 + x2, to make the co-efficient of
dx
unity.

dy 2x 4 x2 linear Differential
+ . y  ...(i) Equations of the
dx 1  x2 1  x2 First Order
dy
It is of the form + Py = Q
dx NOTES
2x 4 x2
Here, P= ,Q
1  x2 1  x2
 I.F. = e
I P dx
=e
I 2x
1  x2
dx
 e log (1  x
2
)
= 1 + x2
Hence the solution is
y . (I.F.) = I Q . (I.F.) dx + c
or y(1 + x2) = I 4 x2
1  x2
. (1 + x2) dx + c
or y(1 + x2) = I 4x2dx + c
4 x3
or y(1 + x2) = + c.
3
dy
(ii) Given equation is (tan x) . y = 2 sin x
dx
dy
dx
Here P = tan x, Q = 2 sin x
 I.F. = e P dx = e  tan x dx = e( log cos x)
I I
= elog cos x = cos x

y (I.F.) = I Q . (I.F.) dx + c
or y cos x = I 2 sin x cos x dx + c
=
1
I sin 2x dx + c =
 cos 2 x
2
+c
or y cos x = 2
cos 2x + c.
Example 2. Solve the following:

dy dy
(i) sec x = y + sin x (ii) x log x + y = 2 log x
dx dx
dy
Sol. (i) Given equation is sec x . y = sin x
dx
dy
Dividing throughout by sec x, to make the co-efficient of unity,
dx
dy
(cos x) . y = sin x cos x
dx
dy
dx
Ordinary Differential Here, P = cos x, Q = sin x cos x
Equations
P dx  cos x dx
 I.F. = e I
=e I
= e sin x
I
NOTES y . (I.F.) = Q . (I.F.) dx + c
or y . e sin x = I sin x cos x . e sin x dx + c = I tet dt + c, where t = sin x
=t.
et
1
 1.
et
1 I
dt + c = tet et + c
= et(t + 1) + c = e sin x (sin x + 1) + c

or y = (sin x + 1) + c esin x.
dy
(ii) Given equation is x log x + y = 2 log x
dx
dy
Dividing throughout by x log x to make the co-efficient of unity,
dx
dy 1 2
+ .y= .
dx x log x x
dy
dx
1 2
Here, P= , Q=
x log x x
 I.F. = e
I P dx
=e
I 1
x log x
dx
=e
I 1/x
log x
dx
= elog log x = log x
y . (I.F.) = I Q. (I.F.) dx + c
or y log x =
2
I
log x dx + c
I
x
1
or y log x = 2 . log x dx + c
x
(log x) 2
' I[ f ( x ) n
f ( x ) dx
n 1
=2. +c [ f ( x )]
2  ,n  1
n 1
or y log x = (log x)2 + c.
dy
Example 3. Solve: x(x 1) (x 2)y = x3(2x 1).
dx
Sol. Given equation is
dy
x(x 1) (x 2)y = x3(2x 1)
dx
dy x2 x 2 (2 x  1)
or y=
dx x( x  1) x1
dy
dx
linear Differential
x2 x 2 (2 x  1) Equations of the
Here, P= ,Q=
x( x  1) x1 First Order
 I.F. = e I P dx
=e
I 
x2
x ( x  1)
dx
e
I 

2

1
dx
x x  1 NOTES
2
= e [2 log x log (x 1)] = e  [log x  log ( x  1)]
x
1
x2
x2 1
log 2
 log x1

x  1
x1 x  1
= e =e 
x2
 The solution is
y.
x1
x2
= I x 2 (2 x  1) x  1
x1
.
x2
dx + c = I (2x 1) dx + c = x2 x + c
or y(x 1) = x2(x2 x + c).

dy
Example 4. Solve: x (1 x2) + (2x2 1)y = x3.
dx
dy
Sol. Dividing by x(1 x2) to make the co-efficient of unity, the given equation
dx
becomes
2
dy 2 x2  1 x
+ y=
dx 2
x (1  x ) 1  x2
dy
dx
2 x2  1 x2
Here P= , Q=
x (1  x 2 ) 1  x2
2x 2  1 1 1 1
Now P= =   [Partial fractions]
x(1  x)(1  x) x 2(1  x) 2(1  x)
 I Pdx = log x
1
2
1
log (1 x) log (1 + x)
2
= log [x (1 x)1/2 (1 + x)1/2]
= log [x 1  x 2 ] = log (x 1  x 2 )1
P dx log ( x 1  x 2 ) 1 1
 I.F. = e =e =
I
x 1  x2
The solution is
y.
1
x 1  x2
= I x2
1 x 2
.
1
x 1  x2
dx + c
= I (1  x )
x
2 3/ 2
dx + c =
1
2 I (1 x2)3/2 . ( 2x) dx + c
Equations 1 (1  x 2 ) 1/ 2
= . +c
2 1

2
NOTES y 1
 = +c  y = x + cx 1  x2
2
x 1 x 1  x2
which is the required solution.
e 2 x dx = 1.
Example 5. Solve: x

y

x dy
Sol. The given equation is

e 2 x
x

y

x
dx
dy
=1
dy e 2 x y dy 1 e 2 x
or =  or + y
dx x x dx x x
dy
dx
1 e 2 x
Here P= , Q=
x x
 I.F. = e
I 1
x
dx
=e I x 1/ 2 dx
= e
2 x
 Hence the solution is
y. e
2 x
= I e 2
x
x
.e
2 x
dx + c
or y. e
2 x
= I 1
x
dx + c
or ye2 x
= 2 x +c or y = e 2 x
(2 x + c).
dx
Equations of the Form + Px = Q where P and Q are functions of y
dy
only.
tandy 1
y dy
(i) (1 + y2) + ( x  e =0 ) (ii) (2x 10y3) + y = 0.
dx dx
Sol. (i) The given equation is
1 dy
(1 + y2) + ( x  e tan y
) =0
dx
dx tan  1 y
or (1 + y2) +x e =0
dy
1
dx 1 e tan y
or + . x =
dy 1  y2 1  y2
dx linear Differential
It is of the form + Px = Q Equations of the
dy First Order
I.F. = e
I 1
1  y2
dy
=e
tan 1
y
NOTES
 The solution is
I I
1
tan 1
y e tan y
tan 1
y
x. e = 2 .e dy + c = et . et dt + c where t = tan1 y
1 y
tan 1
y
= I e2t dt + c =
1
1
2
e2t + c
or x. e = 1
2 e 2 tan y
+ c.
dy
(ii) The given equation is (2x 10y3) +y=0
dx
dx dx 2
or y. + 2x 10y3 = 0 or + . x = 10y2
dy dy y
dx
It is of the form + Px = Q
dy
2
 dy
e  P dx
2
log y
I.F. = = e
y
= e2 log y = e  y2
 The solution is xy2 = I

10 y 5
10y2 . y2dy + c = 10 I y4 dy + c
or xy2 = + c = 2y5 + c.
5
EXERCISE A
dy y dy
1.   x2 2.  y sec x  tan x
dx x dx
dy dy
3.  y tan x  sec x 4. (1 + x2) + 2xy = cos x
dx dx
dy x  y 1 dy
5. = 6. (x + 1) ny = ex (x + 1)n + 1
dx x 1 dx
dy dy
7. cos2 x + y = tan x 8. (1 + x2) + y = tan1 x
dx dx
dy 2x 1
9.  .y 2 given that y = 0 when x = 1
dx x 2  1 ( x  1)2
dy 
10. + 2y tan x = sin x given that y = 0 when x =
dx 3
dy dy
11. x  2 y  x 2 log x 12.  y cos x  sin 2 x
dx dx
dy dy
13. = x(x2 2y) 14. sin x + y cos x = 2 sin2 x cos x
dx dx
Ordinary Differential dy dy
Equations 15. (1 x2) + xy = ax 16. x(x 1) y = x2(x 1)2
dx dx
dy
17. y dx x dy + log x dx = 0 18. + 2y cot x = 3x2 cosec2 x
dx
NOTES dy dy
19. sin 2x = y + tan x 20. (x + 2y3) =y
dx dx
21. (1 + y2) dx = (tan1 y x) dy 22. ey dx + (1 + xey) dy = 0
dx
23. + 2x = 6ey 24. 2y + 4y = x2 x
dy
dy
25. y 2y = cos 3x 26. + y cot x = 2x + x2 cot x
dx
1  x log x
27. y + y = 28. xy y = (x 1) ex
x
Answers
1
1. xy = x4 + c 2. y(sec x + tan x) = sec x + tan x x + c
4
3. y = sin x + c cos x 4. y(1 + x2) = sin x + c
y
5. = log (x + 1) + c 6. y = (x + 1)n (ex + c)
x 1
7. y = tan x 1 + cetan x 8. y = tan1 x 1 + cetan x

9. y(x2 + 1) = tan1 x 10. y = cos x 2 cos2 x
4
x4 x4
11. x2y = log x  c 12. y = 2(sin x 1) + cesin x
4 16
1 2 2 2
13. y = (x 1) + ce  x 14. y sin x = sin3 x + c
2 3
 1   x3 
15. y = a + c 1  x 2 16. y = 1     c
 x 3 
17. y + 1 + log x = cx 18. y sin2 x = x3 + c
19. y = tan x + c tan x 20. x = y3 + cy

1
21. x = tan1 y 1 + ce  tan y
22. xey + y = c
1
23. x = 2ey + ce2y 24. y = (x 1)2 + ce2x
4
1
25. y = (3 sin 3x 2 cos 3x) + ce2x 26. y sin x = x2 sin x + c
13
27. y = log x + cex 28. y = ex + cx
BERNOULLIS EQ
EQUUATION (EQUATIONS REDUCIBLE TO
(EQU
THE LINEAR FORM)
dy
To solve the equation + Py = Qyn, where P and Q are functions of x
dx
only
dy
The given equation is + Py = Qyn ...(i)
dx
Dividing both sides of (i) by yn, to make the RHS a function of x only. linear Differential
Equations of the
dy First Order
yn + Py1n = Q ...(ii)
dx
Put y1n = z, then NOTES
dy dz dy 1 dz
(1 n) . yn = or yn = .
dx dx dx 1  n dx
1 dz
 (ii) becomes . + Pz = Q
1  n dx
dz
or + (1 n) . Pz = (1 n) Q.
dx
which is a linear equation in z and can be solved.
In the solution, putting z = y1n, we get the required solution.
SOLVED EXAMPLES
dy y y2
Example 7. Solve: 2 = + 2.
dx x x
dy y y2
Sol. The given equation is 2. = 2 .
dx x x
Dividing throughout by y2
dy 1 1
2y2 . y1 = 2 ...(i)
dx x x
dy dz
Put y1 = z, then y2 =
dx dx
 (i) becomes
dz 1 1 dz 1 1
2 z= 2 or + z=
dx x x dx 2x 2x2
which is linear in z.
1 1
P= ,Q=
2x 2x2
I.F. = e
I 1
2x
dx
1
 e2
log x
 e log x
= x
 The solution is z . x = I
2x2
1
x dx + c
or y1 x =
1
2 I x3/2 dx + c or
y
x

1
x
+c
or x = y(1 + c x ).
Example 8. Solve the following :
dy 1 ey dy x
(i) +  2 (ii) + y  x y.
dx x x dx 1  x2
Ordinary Differential dy 1 ey
Equations Sol. (i) The given equation is +  2
dx x x
Dividing throughout by ey
NOTES dy dy 1 1
ey + ey + = 2 ...(i)
dx dx x x
dy dz
Put ey = z, then ey =
dx dx
 (i) becomes
dz 1 1 dz 1 1
+z. = 2 or .z= 2
dx x x dx x x
1 1
which is linear in z. P= ,Q= 2
x x
I.F. = e
I 
1
x
dx
 e  log x  e
log
1
x
=
1
x
 The solution is z.
1
x
=
x
1
2 I
.
1
x

dx + c
or
or
1
ey . =
x
1
x3 I
dx + c
2x = ey + 2cx2ey.
or ey .
1
=
1
x 2x2
+c
dy x
(ii) The given equation is + y=x y
dx 1  x2
Dividing throughout by y,
dy x
y1/2 . + y1/2 = x ...(i)
dx 1  x2
1 1/2 dy dz
Put y1/2 = z ; then y . 
2 dx dx
 (i) becomes
dz x dz x x
2. + .z=x or
+ .z
dx 1  x2 dx 2
2(1  x ) 2
x x
which is linear in z. P= ,Q 
2(1  x 2 ) 2
 I.F. = e
I x
2( 1  x 2 )
dx
e

1
4 I  2x
1  x2
dx
| Note
1
 log ( 1  x 2 ) 2 1 / 4
= e 4  e log ( 1  x )
= (1 x2)1/4
 The solution is z . (1 x2)1/4 =

2
x
I
(1 x2)1/4 dx + c
or y . (1 x2)1/4 =
1
4 I
2x(1 x2)1/4 dx + c
1 (1  x 2 ) 3 / 4
or y . (1 x2)1/4 = . +c
4 3
4
1
or y = (1 x2) + c(1 x2)1/4.
3
Example 9. Solve: (x2y3 + xy) dy = dx. linear Differential
Equations of the
Sol. The given equation is (x2y3 + xy)dy = dx First Order
dx
or = x2y3 + xy
dy NOTES
dx dx
or xy = x2y3 Form + Px = Qxn
dy dy
Dividing throughout by x2
dx
x2 x1 y = y3 ...(i)
dy
dx dz
Put x1 = z, then x2 =
dy dy
 (i) becomes
dz dz
zy = y3 or + y.z = y3
dy dy
which is linear in z. P = y, Q = y3
2
y dy y /2
I.F. = e =e
I
I
2 2
 The solution is z . e 1/2 y = y3 . e 1/2 y dy + c
or
2
x1 . e 1/2 y = I 2
y 2 . y . e 1/2 y dy  c
or
2
= I 2t et dt + c,
x1 . e 1/2 y = 2et (t 1) + c

where t = 1
2
y2
x1 . e 1/2 y = 2 e 1/2 y ( 21 y2 1) + c

2 2
or
2
or x1 = y2 + 2 + c e  1/2 y .
dy dy
(i) (x + 1) + 1 = 2ey (ii) = exy (ex ey)
dx dx
dy
Sol. (i) The given equation is (x + 1) + 1 = 2ey
dx
dy 1 2e  y
or + =
dx x1 x1
dy 1 2
or ey . + . ey = ...(i)
dx x1 x1
dy dz
Put ey = z, then ey . =
dx dx
dz 1 2
 From (i), + .z=
dx x1 x1
1 2
which is linear in z. P= ,Q=
x1 x 1
I.F. = e
I 1
x1
dx
 e log ( x  1) = x + 1
or
 The solution is z (x + 1) =
2
x1
cy . (x + 1) = 2x + c.
I. (x + 1) dx + c
Ordinary Differential (ii) The given equation is
Equations dy dy
= exy (ex ey) or = e2x . ey ex
dx dx
dy dy
or + ex = e2x . ey or ey . + ex . ey = e2x ...(i)
NOTES dx dx
dy dz
Put ey = z, then ey . =
dx dx
dz
 (i) becomes + ex . z = e2x
dx
which is linear in z. P = ex, Q = e2x
e x dx x
I.F. = e  ee
I
I
x x
 The solution is z . ee = e2x . e e dx + c
or
x
ey . e e = I ex . ex . e e dx + c
x
= I t et dt + c,
= et (t 1) + c
where t = ex
x x x
or ey . e e = e e (ex 1) + c or ey = ex 1 + c e  e .
dy
Example 11. Solve: + (2x tan1 y x3)(1 + y2) = 0.
dx
dy
+ (2x tan1 y x3)(1 + y2) = 0
dx
1 dy
or . + 2x tan1 y x3 = 0
1 y 2 dx
1 dy
or . + 2x tan1 y = x3 ...(i)
1 y 2 dx
1 dy dz
Put tan1 y = z, then 2
. =
1 y dx dx
dz
 From (i), + 2xz = x3
dx
which is linear in z. P = 2x, Q = x3
2
2 x dx x
=e
I
I.F. = e I
2 2
x
 The solution is z . ex = x3 . e dx + c
or
x
tan1 y . e =
2
1
I x
2x . x2 e dx + c
2
I
2
1
= 2
t et dt + c, where t = x2
1
= 2 et (t 1) + c
2 2
x
or tan1 y . e = 1
2
e x (x2 1) + c
2
or tan1 y = 1
2
(x2 1) + c e  x .
dy y y
(i) (x3y2 + xy) dx = dy (ii) + log y  2 (log y)2
dx x x
Sol. (i) The given equation is (x3y2 + xy) dx = dy linear Differential
dy dy Equations of the
or = x3y2 + xy or xy = x3y2 First Order
dx dx
dy
Dividing both sides by y2, y2 xy1 = x3 ...(i)
dx NOTES
dy dz
Put y1 = z, then y2 =
dx dx
dz dz
 (i) becomes xz = x3 or + xz = x3
dx dx
which is linear in z. P = x, Q = x3
x2
x dx
I.F. = e
I
 e2
I I
2
x x2 x2
 The solution is z. e2 = x3 . e2 dx + c = x2 . xe 2 dx + c
= I 2tet dt + c, where t =
x2
I
2
= 2tetdt + c = 2et (t 1) + c
x2 x2 x  1
2
y1 . e2 = 2e 2 2 + c
x2

or = + 2 + ce y1 . x2 2
dy y y
(ii) The given equation is + log y = 2 (log y)2
dx x x
Dividing both sides by y (log y)2, we get
1 dy 1 1 1
. + .  2 ...(i)
y(log y) 2 dx log y x x
1 1 dy dz
Put = (log y)1 = z, then (log y)2 . =
log y y dx dx
1 dy dz
or . =
y(log y) 2 dx dx
dz 1 1 dz 1 1
 From (i), +z. = 2 or z= 2
dx x x dx x x
1 1
which is linear in z. P= ,Q= 2 .
x x
I.F. = e
I 
1
x
dx
 e  log x  e log x
1
= x1 =
1
x
 The solution is z.
1
x
= I
1 1
. dx + c
x2 x
or z.
1
x
= I x3 dx + c =
x 2
2
+c
1 1 1 1 1
or . = +c or = + cx.
log y x 2 x2 log y 2x
Ordinary Differential Example 13. Show how to solve an equation of the form
Equations
dy
f (y)
+ Pf(y) = Q where P, Q are functions of x only.
dx
Sol. (a) The given equation is
NOTES
dy
f (y)
+ Pf(y) = Q ...(i)
dx
where P, Q are functions of x only.
dy dz
Put f(y) = z, then f (y) =
dx dx
dz
 (i) becomes + Pz = Q
dx
which is linear in z and can be solved.
I.F. = e P dx and the solution is

I
z(I.F.) =  Q . (I.F.) dx + c
or f(y) . (I.F.) =  Q . (I.F.) dx + c
dy dy
(i) (x + 1) + 1 = exy (ii) = y tan x y2 sec x.
dx dx
Sol. (i) The given equation is
dy ex dy ey ex
(x + 1) +1= y or ey + = ...(i)
dx e dx x1 x1
dy dz
Putting ey = z so that ey =
dx dx
dz z ex
 (i) becomes + =
dx x1 x1
1 ex
which is linear in z with P= ,Q=
x1 x1
I.F. = e
I P dx
e
I 1
x1
dx
 e log ( x  1) = x + 1
 The solution is z (x + 1) = I ex
x1
. (x + 1) dx + c or ey (x + 1) = ex + c.
(ii) The given equation is

dy
y tan x = y2 sec x
dx
1 dy 1
or 2
. + tan x = sec x ...(1)
y dx y
1 1 dy dz
Putting = z so that 2 =
y y dx dx
dz linear Differential
 Equation (1) becomes + z tan x = sec x Equations of the
dx First Order
which is linear in z with P = tan x, Q = sec x
P dx tan x dx
I.F. = e
I
e I
 e log sec x = sec x NOTES
 The solution is z . sec x = I sec x . sec x dx + c
1 1
or sec x = tan x + c or = sin x + c cos x.
y y
EXERCISE B
dy y
1.  = y2 2. y + y = y2
dx x
dy dy 2 x3
3. = x3y3 xy 4. 3  y 2
dx dx x  1 y
dy y dy
5.  = x2y6 6. x + y = x3y4
dx x dx
dy
7. 2y tan x = y2 tan2 x 8. (y log x 1)y dx = x dy
dx
dy dy tan y
9. + x sin 2y = x3 cos2 y 10.  = (1 + x) ex sec y
dx dx 1  x
11. (x y2) dx + 2xy dy = 0 12. cos x dy = y(sin x y) dx
dy 2
 y3 e  x
3
13. xy 14. ( xy2  e1/ x ) dx x2y dy = 0
dx
 dy 
15. ey   1   ex 16. (xy 2x log x) dy = 2y dx
 dx 
dy y
17.  = y2 log x 18. y (2xy + ex) dx = exy
dx x
dy dy
19. x + y = y2x3 cos x 20. sin y = cos y (1 x cos y)
dx dx
Answers
1 1
1. + log x = c 2. y =
xy 1  ce x
x 6 2x5 x 4
3. y2 = x2 + 1 + ce x
2
4. y2(x + 1)2 =   c
6 5 4
1 5 3 1
5.  x + cx5 6. = 3x3 log x + cx3
y5 2 y3
1 tan3 x 1
7. sec2 x   c 8. = log x + 1 + cx
y 3 y
1 2 3
9. tan y = ( x  1)  ce  x 10. sin y = (1 + x)(ex + c)
2
1
11. y2 = x (c log x) 12. = sin x + c cos x
y
Equations 2
13. y2 . e x = 2x + c 14. 3y2 = 2x 2e x  cx 2
3
1 2x y2
15. ex y  e c 16. y log x = +c
2 4
NOTES 1 1
17.   (log x )2  cx 18. ex = y(c x2)
y 2
1
19. = x sin x cos x + c 20. sec y = x + 1 + cex
xy
DIFFERENTIAL EQUATIONS OF THE FIRST ORDER AND

HIGHER DEGREE
So far, we have discussed differential equations of the first order and first degree.
Now we shall study differential equations of the first order and degree higher than the
dy
first. For convenience, we denote by p.
dx
A differential equation of the first order and nth degree is of the form
pn + P1pn1 + P2pn2 + ...... + Pn = 0 ...(1)
where P1, P2, ......, Pn are functions of x and y.
Since it is a differential equation of the first order, its general solution will
contain only one arbitrary constant.
In the various cases which follow, the problem is reduced to that of solving one
or more equations of the first order and first degree.
EQ
EQUUATIONS SOLVABLE FOR p
SOLV
Resolving the left hand side of (1) into n linear factors, we have
[p f1(x, y)] [p f2(x, y)], ......, [p fn(x, y)] = 0
which is equivalent to p f1(x, y) = 0, p f2(x, y) = 0, ......, p fn(x, y) = 0
Each of these equations is of the first order and first degree and can be solved
by the methods already discussed.
If the solutions of the above n component equations are
F1 (x, y, c) = 0, F2 (x, y, c) = 0, ......, Fn (x, y, c) = 0
then the general solution of (1) is given by
F1 (x, y, c)  F2 (x, y, c) ...... Fn (x, y, c) = 0.
SOLVED EXAMPLES
dy 2
dy
Example 15. Solve: x2
dx  xy
dx
6y2 = 0.
dy linear Differential
Sol. The given equation is x2p2 + xyp 6y2 = 0 where p = Equations of the
dx
First Order
Factorising (xp + 3y) (xp 2y) = 0
 xp + 3y = 0 or xp 2y = 0
NOTES
Now, xp + 3y = 0
dy dy dx
 x + 3y = 0 or 3 =0
dx y x
Integrating, log y + 3 log x = log c or x3y = c
Also, xp 2y = 0
dy dy dx
 x 2y = 0 or 2 =0
dx y x
y
Integrating, log y 2 log x = log c or = c or y = cx2
x2
 The general solution of the given equation is (x3y c) (y cx2) = 0.
Example 16. Solve xyp2 + p(3x2 2y2) 6xy = 0.
Sol. Solving the given equation for p, we have
 (3 x 2  2 y 2 )  (3 x 2  2 y 2 ) 2  24 x 2 y 2
p=
2 xy
(2 y 2  3 x 2 )  (3 x 2  2 y2 ) 2 y 3x
=  or 
2 xy x y
2y dy 2 y dy 2dx
Now, p=   or  =0
x dx x y x
y
Integrating, log y 2 log x = log c or = c or y = cx2
x2
3x dy 3x
Also, p=   or ydy + 3xdx = 0
y dx y
y2 3 x 2
Integrating,  = C or y2 + 3x2 = c
2 2
 The general solution of the given equation is (y cx2) (y2 + 3x2 c) = 0.
Example 17. Solve p2 + 2py cot x = y2.
Sol. The given equation can be written as (p + y cot x)2 = y2 (1 + cot2 x)
or p + y cot x =  y cosec x
 The component equations are
p= y ( cot x + cosec x) ...(1)
and p= y ( cot x cosec x) ...(2)
dy
From (1), = y ( cot x + cosec x)
dx
dy
or = ( cot x + cosec x) dx
y
x
x c tan
Integrating, log y = log sin x + log tan + log c = log 2
2 sin x
Ordinary Differential x
Equations c tan
2 c c
or y=  
x x x 2 1  cos x
2 sin cos 2 cos
2 2 2
NOTES or y(1 + cos x) = c
dy
From (2), = y( cot x cosec x)
dx
dy
or = ( cot x cosec x) dx
y
x c
Integrating, log y = log sin x log tan + log c = log
2 x
sin x tan
2
c c
or y=  or y(1 cos x) = c
x 1  cos x
2
2 sin
2
 The general solution of the given equation is
[y(1 + cos x) c] [y(1 cos x) c] = 0.
EXERCISE C
Solve the following equations:
dy 2
dy
1. p2 7p + 12 = 0 2. xy
dx (x2 + y2)
dx
+ xy = 0
dy 2
dy
3. yp2 + (x y)p x = 0 4. x2
dx + 3xy
dx
+ 2y2 = 0
dy dx x y
5.    6. p2 2p sinh x 1 = 0
dx dy y x
7. p (p + y) = x (x + y) 8. 4y2p2 + 2pxy (3x + 1) + 3x3 = 0.
Answers
1. (y 4x c) (y 3x c) = c 2. (y2 x2 c) (y cx) = 0 3. (y x c) (x2 + y2 c)
=0
4. (xy c)(x2y c) = 0 5. (xy c)(x2 y2 c) = 0 6. (y ex c) (y ex c)
=0
1 2 1 2
7. (y x + c) (y + x + cex 1) = 0 8. (y2 + x3 c)(y2 + x c) = 0.
2 2
EQ
EQUUATIONS SOLVABLE FOR y
SOLV
If the equation is solvable for y, we can express y explicitly in terms of x and p.

Thus, the equations of this type can be put as y = f (x, p) ...(1)
dy dp
dx
= p = F x, p,
dx ...(2)
Equation (2) is a differential equation of first order in p and x.

Suppose the solution of (2) is (x, p, c) = 0 ...(3)
Now elimination of p from (1) and (3) gives the required solution. linear Differential
Equations of the
If p cannot be easily eliminated, then we solve equations (1) and (3) for x and y
First Order
to get
x = 1 (p, c), y = 2 (p, c)
NOTES
These two relations together constitute the solution of the given equation with
p as parameter.
SOLVED EXAMPLES
Example 18. Solve y + px = x4p2.

Sol. Given equation is y = px + x4p2 ...(1)
Differentiating both sides w.r.t. x,
dy dp dp
=p=px + 4x3p2 + 2x4p
dx dx dx
dp dp
or 2p + x
dx
2px3 2 p  x
dx
=0

2 p  x dp (1 2px ) = 0
dx
or 3
dp
Discarding the factor (1 2px3), which does not involve , we have
dx
dp dp dx
2p + x = 0 or 2 =0
dx p x
Integrating, log p + 2 log x = log c
or log px2 = log c
c
or px2 = c  .p=
x2
c
Putting this value of p in (1), we have y = + c2.
x
Example 19. Solve y = 2px p2.
Sol. The given equation is y = 2px p2 ...(1)
dy dp dp
Differentiating both sides w.r.t. x, = p = 2p + 2x 2p
dx dx dx
dp
or p + (2x 2p) =0
dx
dx
or p + 2x 2p = 0
dp
dx 2
or  x=2 ...(2)
dp p
which is a linear equation.
I 2
p
dp
 e 2 log p
 p2
I
I.F. = e
 The solution of (2) is x I.F. = 2 I.F.dp + c
or xp2 = I 2 p2dp + c
Ordinary Differential 2 2 3
Equations or xp2 = p + cp2
p + c or x= ...(3)
3 3
2
Putting this value of x in (1), we have y = 2p
3
p  cp2 p2

NOTES 1
or y = p2 + 2cp1. ...(4)
3
Equations (3) and (4) together constitute the general solution of (1).
EXERCISE D
1. xp2 2yp + ax = 0 2. y 2px = tan1 (xp2) 3. 16x2 + 2p2y p3x = 0
4. y = x + 2 tan1 p 5. y = 3x + log p 6. x yp = ap2
dy 4
dy
7. x2
dx + 2x
dx
y = 0. 8. y = 2px xp2
Answers
a
1. 2y = cx2 + 2. y = 2 cx + tan1 c 3. 16 + 2c2y c3x2 = 0
c
p1 p 1 3
4. x = log tan1 p + c, y = log + tan1 p + c 5. y = 3x + log
p2  1 2
p 1 1  ce3 x
p 1
6. x= (c + a sin1 p), y = (c + a sin1 p) ap. 7. y = c2 + 2 cx .
2
1 p 1  p2
8. y = 2 cx  c.
EQ
EQUUATIONS SOLVABLE FOR x
SOLV
If the equation is solvable for x, we can express x explicitly in terms of y and p.

Thus, the equations of this type can be put as x = f(y, p)
...(1)
dx 1
dp
Differentiating (1) w.r.t. y, we get   F y, p,
dy p
dy ...(2)
Equation (2) is a differential equation of first order in p and y.

Suppose the solution of (2) is  (y, p, c) = 0 ...(3)
Now elimination of p from (1) and (3) gives the required solution.
If p cannot be easily eliminated, then we solve equations (1) and (3) for x and y
to get
x = 1 (p, c), y = 2 (p, c)
These two relations together constitute the solution of the given equation with
p as parameter.

SOLVED EXAMPLES linear Differential
Equations of the
First Order
Example 20. Solve y = 2px + y2p3.

1 y
 y2 p2
NOTES
Sol. Solving for x, we have x=

2 p
Differentiating both sides w.r.t. y
dx 1 1 1 y dp dp 
=    2  2 yp2  2 y 2 p 
dy p 2  p p dy dy 
dp dp
or 2p = p y  2 yp4  2 y2 p3
dy dy
dp dp dp
or p + 2yp4 + y + 2y2p3 = 0 or p(1 + 2yp3) + y (1 + 2yp3) = 0
dy dy dy
p  y dp (1 + 2yp ) = 0
dy
or 3
dp
Discarding the factor (1 + 2yp3) which does not involve , we have
dy
dp dy dp
p+y = 0 or + =0
dy y p
c
Integrating, log y + log p = log c or py = c or p=
y
Putting this value of p in the given equation, we have
2cx c 3
y=  or y2 = 2cx + c3
y y
which is the required solution.
p .
Example 21. Solve p = tan x  1 p 2
p
Sol. Solving for x, we have x = tan1 p + ...(1)
1  p2
Differentiating both sides w.r.t. y,
dx 1 1 dp (1  p2 )  2 p2 dp
    
dy p 1  p2 dy (1  p2 )2 dy
1 2(1  p2 )  2 p2 dp 2p
or  or dy = dp
p (1  p2 ) 2 dy (1  p2 ) 2
1
Integrating, y=c ...(2)
1  p2
Equations (1) and (2) together constitute the general solution.

Ordinary Differential EXERCISE E
Equations
1. y = 3px + 6p2y2 2. y = 2px + p2y
NOTES 3. p3 4xyp + 8y2 = 0 4. y2 log y = xyp + p2
5. x = y + a log p 6. x = y + p2
Answers
1. y3 = 3cx + 6c2 2. y2 = 2cy + c2 3. 64y = c(c 4x)2
p
4. log y = cx + c2 5. x = c + a log , y = c a log (p 1)
p 1
6. x = 2p 2 log (1 p) + c, y = p2 2p 2 log (1 p) + c.
CLAIRAUTS EQ
CLAIRAUTS EQUUATION
An equation of the form y = px + f(p) ...(1)

is known as Clairauts equation.
dp dp dp
p=p+x + f (p) or [x + f (p)] =0
dx dx dx
dp
Discarding the factor [x + f (p)], we have =0
dx
Integrating, p=c
Putting p = c in (1), the required solution is p = cx + f(c)
Thus, the solution of Clairauts equation is obtained by writing c for p.
SOLVED EXAMPLES
Example 22. Solve (y px) (p 1) = p.

p p
y px = or y = px +
p1 p1
c
This is of Clairauts form. Hence putting c for p, the solution is y = cx + .
c1
Note. Many differential equations can be reduced to Clairauts form by suitably changing
the variables.
Example 23. Solve e4x (p 1) + e2y p2 = 0.
Sol. [In problems involving elx and emy, put X = ekx and Y = eky, where k is the
H.C.F. of l and m].
Put X = e2x and Y = e2y
so that dX = 2e2x dx and dY = 2e2y dy
dy e 2 x d Y X dY
 p=  2y  P , where P =
dx e d X Y dX

linear Differential
X  X2
The given equation becomes X2  P  1   Y  2 P2 = 0. Equations of the
Y  Y First Order
or XP Y + P2 = 0 or Y = PX + P2 which is of Clairauts form.
 Its solution is Y = cX + c2 and hence e2y = ce2x + c2. NOTES
Example 24. Solve (px y) (py + x) = 2p.
Sol. Put X = x2 and Y = y2 so that dX = 2x dx and dY = 2ydy
dy x dY X dY
   p= P, where P =
dx y dX Y dX
The given equation becomes
 X  X  X
 P X  Y P Y  X2 P
Y  Y  Y
  
2P
or (PX Y) (P + 1) = 2P or PX Y =
P 1
2P
or Y = PX which is of Clairauts form.
P1
2c 2c
 Its solution is Y = cX and hence y2 = cx2 .
c1 c1
EXERCISE F
a
1. y = xp + 2. y = px + a2 p2  b2
p
3. sin px cos y = cos px sin y + p 4. xp2 yp + a = 0
5. (x a)p2 + (x y)p y = 0 6. p = log (px y)
7. p = sin (y px) 8. p2(x2 1) 2pxy + y2 1 = 0
9. e3x (p 1) + p3e2y = 0 10. x2 (y px) = yp2
11. (y + px)2 = x2p.
Answers
a
1. y = cx + 2. y = cx + a2c2  b2 3. y = cx sin1 c
c
a ac 2
4. y = cx + 5. y = cx 6. y = cx ec
c c 1
7. y = cx + sin1 c 8. (y cx)2 = 1 + c2 9. ey = cex + c2
10. y2 = cx2 + c2 [Hint. Put x2 = X, y2 = Y] 11. xy = cx c2. [Hint. Put xy = v]

Equations
NOTES 4. LINEAR DIFFERENTIAL

EQUATIONS OF SECOND AND
HIGHER ORDER
STRUCTURE
Definitions
The Operator D
Theorems
Auxiliary Equation (A.E.)
Rules for Finding The Complementary Function
The Inverse Operator
Rules for Finding The Particular Integral
Method of Variation of Parameters to Find P.I.
Homogeneous Linear Equations (Cauchy-Euler Equations)
Legendres Linear Differential Equation
Linear Differential Equations of Second Order
Complete Solution in Terms of Known Integral
To Find a Particular Integral of + P + Qy = 0
Removal of the First Derivative (Ruduction to Normal Form)
Transformation of theEquation by Changing the Independent Variable
Method of Variation of Parameters
DEFINITIONS
A linear differential equation is that in which the dependent variable and

its derivatives occur only in the first degree and are not multiplied together. Thus, the
general linear differential equation of the nth order is of the form
dn y d n 1 y d n 2 y dy
n
 P1 n 1
 P2 n 2
+ ...... + Pn1 + Pny = X, where P1, P2, ......, Pn1, Pn and
dx dx dx dx
X are functions of x only.

A linear differential equation with constant co-efficients is of the form Linear Differential
Equations of Second and
dn y d n1 y dn  2 y dy Higher Order
n
 a1 n 1
 a2 n 2
 ......  an1  an y  X
...(1)
dx dx dx dx
where a1, a2, ......, an1, an are constants and X is either a constant or a function of x NOTES
only.
THE OPERATOR D
d dy
The part of the symbol may be regarded as an operator such that when
dx dx
it operates on y, the result is the derivative of y.
d2 d3 dn
Similarly, , , ...... , may be regarded as operators.
dx 2 dx 3 dx n
d d2 2 dn
For brevity, we write  D,  D , ......,  Dn
dx dx 2 dx n
Thus, the symbol D is a differential operator or simply an operator.
Written in symbolic form, equation (1) becomes
(Dn + a1Dn1 + a2Dn2 + ...... + an1D + an)y = X
or f(D)y = X
where f (D) = Dn + a1 Dn1 + a2Dn2 + ...... + an1D + an
i.e., f (D) is a polynomial in D.
The operator D can be treated as an algebraic quantity.
Thus D(u + v) = Du + Dv
D(u) =  Du
DpDq u = Dp+q u
DpDqu = DqDpu
The polynomial f(D) can be factorised by ordinary rules of algebra and the factors
may be written in any order.
THEOREMS
Theorem 1
If y = y1, y = y2,..., y = yn are n linearly independent solutions of the differential
equation
(Dn + a Dn1 + a Dn2 + ... + a )y = 0
1 2 n ...(i)
then u = c1y1 + c2y2 + ... + cnyn is also its solution, where c1, c2, ..., cn are arbitrary
constants.

Ordinary Differential Proof. Since y = y1, y = y2,..., y = yn are solution of equation (i).
Equations
 Dny + a Dn1y + a Dn2y + ... + a y = 0
1 1 1 2 1 n 1 (K
Dn y2 + a1Dn1y2
+ a2 2 +
... + a y = 0
Dn2y KK
)K
n 2
NOTES ... ... ... ... ... ...(ii)

... ... ... ... ...
KK
Dnyn + a1Dn1yn + a2Dn2yn + ... + anyn = 0 *
Now Dnu + a Dn1u + a Dn2u + ... + a u
1 2 n
= (c1y1 + c2y2 + ... + cnyn)
Dn
+ a1Dn1(c1y1 + c2y2 + ... + cnyn)
+ a Dn2(c y + c y + ... + c y )
2 1 1 2 2 n n
+ ... ... ... ...
...
+ an(c1y1 + c2y2 + + cnyn)
= c1(Dny1 + a1Dn1y1 + a2Dn2y1 + ... + any1)

+ c2(Dny2 + a1Dn1y2 + a2Dn2yn + ... + any2)
+ ... ... ... ...
+ c (Dny + a Dn1y + a Dn2y + ... + a y )
n n 1 n 2 2 n n
= c1(0) + c2(0) + ... + cn(0) '

[ of (ii)]
=0
which shows that u = c1y1 + c2y2 + ...... + cnyn is also the solution of equation (i).
Since this solution contains n arbitrary constants, it is the general or complete
solution of equation (i).
Theorem 2
If y = u is the complete solution of the equation f(D)y = 0 and y = v is a particular
solution (containing no arbitrary constants) of the equation f(D)y = X, then the complete
solution of the equation f(D)y = X is y = u + v.
Proof. Since y = u is the complete solution of the equation f(D)y = 0 ...(i)
 f(D)u = 0 ...(ii)
Also y = v is a particular solution of the equation f(D)y = X ...(iii)
 f(D)v = X ...(iv)
Adding (ii) and (iv), we have f(D)(u + v) = X
Thus y = u + v satisfies the equation (iii), hence it is the complete solution
(C.S.) because it contains n arbitrary constants.
The part y = u is called the complementary function (C.F.) and the part y = v
is called the particular integral (P.I.) of the equation (iii).
 The complete solution of equation (iii), is y = C.F. + P.I.
Thus in order to solve the equation (iii), we first find the C.F. i.e., the C.S. of
equation (i) and then the P.I. i.e., a particular solution of equation (iii).
AUXILIARY EQUATION (A.E.)

(Dn + a1Dn1 + a2Dn2 + ...... + an)y = 0 ...(i)

Let y = emx be a solution of (i), then Dy = memx, D2y = m2emx ,......, Dn2y = mn2emx Linear Differential
Dn1y = mn1emx, Dny = mnemx Higher Order
Substituting the values of y, Dy, D2y, ......, Dny in (i), we get
(mn + a mn1 + a mn2 + ...... + a ) emx = 0
1 2 n NOTES
or mn + a1 mn1 + a2mn2 + ...... + an = 0, since emx  0 ...(ii)
Thus y = emx will be a solution of equation (i) if m satisfies equation (ii).
Equation (ii) is called the auxiliary equation for the differential equation (i).
Replacing m by D in (ii), we get Dn + a Dn1 + a Dn2 + ...... + a = 0
1 ...(iii)
2 n
Equation (ii) gives the same values of m as equation (iii) gives of D. In practice,
we take equation (iii) as the auxiliary equation which is obtained by equating to zero
the symbolic co-efficient of y in equation (i).
Definition. The equation obtained by equating to zero the symbolic co-efficient
of y is called the auxiliary equation, briefly written as A.E.
RULES FOR FINDING THE COMPLEMENTARY FUNCTION
Consider the equation (Dn + a1Dn1 + a2Dn2 + ...... + an)y = 0 ...(i)

where all the ais are constant.
Its auxiliary equation is Dn + a1Dn1 + a2Dn2 + ...... + an = 0 ...(ii)
Let D = m1, m2, m3,......, mn be the roots of the A.E. The solution of equation (i)
depends upon the nature of roots of the A.E. The following cases arise:
Case I. If all the roots of the A.E. are real and distinct, then equation (ii) is
equivalent to
(D m1) (D m2) ...... (D mn) = 0 ...(iii)
Equation (iii) will be satisfied by the solutions of the equations
(D m1)y = 0, (D m2)y = 0,......, (D mn)y = 0
dy
Now, consider the equation (D m1)y = 0, i.e.,  m1 y  0
dx
It is a linear equation and I.F. = e

I  m1dx
 e  m1 x
 its solution is I
y . e  m1 x  0 . e  m1 x dx  c1 or y = c1e m1 x
m x
Similarly, the solution of (D m2)y = 0 is y = c2 e 2
................................................................
the solution of (D mn)y = 0 is y = cn e mn x
Hence the complete solution of equation (i) is
y = c1 em1x  c2em2 x  ......  cnemn x ...(iv)

Case II. If two roots of the A.E. are equal, let m1 = m2.
The solution obtained in equation (iv) becomes
y = (c1 + c2) em1x  c3em3 x  ......  cnemn x
= ce 1  c3e 3  ......  cne n
mx mx m x

Ordinary Differential It contains (n 1) arbitrary constants and is, therefore, not the complete solution
Equations of equation (i).
The part of the complete solution corresponding to the repeated root is the
complete solution of
NOTES
(D m1)(D m1)y = 0
dv
Putting (D m1)y = v, it becomes (D m1)v = 0 i.e., m1v = 0
dx
As in case I, its solution is v = c1 e m1 x
dy m x
 (D m1)y = c1 e m1 x or  m1 y  c1e 1
dx
m x
which is a linear equation and I.F. = e 1
or
 its solution is y . e  m1 x  I c1e m1 x . e – m1 x dx + c2 = c1x + c2
y = (c1x + c2) e m1 x
Thus, the complete solution of equation (i) is
y = (c1x + c2) em1x  c3em3 x  ......  cnemn x
If, however, three roots of the A.E. are equal, say m1 = m2 = m3, then proceeding
as above, the solution becomes
y = (c1x2 + c2x + c3) em1x  c4 em4 x  ......  cn emn x
Case III. If two roots of the A.E. are imaginary, let

m1 =  + i and m2 =  i
The solution obtained in equation (iv) becomes
y = c1 e(   i) x  c2e(   i) x  c3e 3  ......  cne n

mx m x
x i x
= e (c1e  c2 e
i x
)  c3em3x  ......  cnemn x
= ex[c1(cos x + i sin x) + c2 (cos x i sin x)]
+ c3e 3  ......  cne n

mx m x
[' By Eulers Theorem, ei = cos  + i sin ]
= ex[(c1+ c2) cos x + i (c1 c2) sin x] + c3e 3  ......  cn e n

mx m x
= ex(C1 cos x + C2 sin x) + c3 em3 x  ......  cnemn x

[Taking c1 + c2 = C1, i(c1 c2) = C2]
Case IV. If two pairs of imaginary roots be equal, let

m1 = m2 =  + i and m3 = m4 =  i
Then by case II, the complete solution is
y = ex [(c1x + c2) cos x + (c3x + c4) sin x] + c5e 5  ......  cn e n .
mx m x

SOLVED EXAMPLES Linear Differential
Higher Order
d3y dy
Example 1. Solve: 7 6y = 0.
3
dx dx
NOTES
Sol. Given equation in symbolic form is (D3 7D 6)y = 0
Its A.E. is D3 7D 6 = 0 or (D + 1) (D + 2) (D 3) = 0
whence D = 1, 2, 3
Hence the C.S. is y = c1ex + c2e2x + c3e3x.
Example 2. Solve: (D3 4D2 + 4D)y = 0.
Sol. The A.E. is D3 4D2 + 4D = 0 or D(D2 4D + 4) = 0
or D(D 2)2 = 0
whence D= 0, 2, 2
Hence, the C.S is y= c1e0x + (c2x + c3)e2x or y = c1 + (c2x + c3)e2x.
d4 y d2 y
Example 3. Solve:  13 + 36y = 0.
dx 4 dx 2
Sol. Given equation in symbolic form is (D4 + 13D2 + 36)y = 0
Its A.E. is D4 + 13D2 + 36 = 0
or (D2 + 4)(D2 + 9) = 0  D =  2i,  3i
Hence the C.S. is y= e0x (c1 cos 2x + c2 sin 2x) + e0x (c3 cos 3x + c4 sin 3x)
or y= c1 cos 2x + c2 sin 2x + c3 cos 3x + c4 sin 3x.
d4 x
Example 4. Solve: + 4x = 0.
dt 4
d
Sol. Given equation in symbolic form is (D4 + 4)x = 0, where D =
dt
Its A.E. is D4 + 4 = 0 or (D4 + 4D2 + 4) 4D2 = 0
or (D2 + 2)2 (2D)2 = 0 or (D2 + 2D + 2)(D2 2D + 2) = 0
2 4 2 4
whence D= and i.e., D = 1  i and 1  i
2 2
Hence the C.S. is x = et (c1 cos t + c2 sin t) + et (c3 cos t + c4 sin t).
Example 5. Solve: y 2y + 10y = 0, given y(0) = 4, y(0) = 1.
Sol. Given equation in symbolic form is
(D2 2D + 10)y = 0
Its A.E. is D2 2D + 10 = 0
2  4  40 2  6i
 D=  = 1 ± 3i
2 2
The C.S. is y = ex (c1 cos 3x + c2 sin 3x) ...(1)
Now y(0) = 4  y = 4, when x = 0
 4 = c1
Equation (1) becomes y = ex (4 cos 3x + c2 sin 3x) ...(2)
so that y = ex (4 cos 3x + c2 sin 3x) + ex ( 12 sin 3x + 3c2 cos 3x)
Since y(0) = 1 i.e., y = 1, when x = 0
 1 = 4 + 3c2  c2 = 1
Equation (2) becomes y = ex (4 cos 3x sin 3x), which is the required particular
solution.

Ordinary Differential EXERCISE A
Equations
d2 y dy d2 y dy
1. 4  5y  0. 2.  ( a  b)  aby  0 .
NOTES dx 2 dx dx 2 dx
d2 y dy d2 x dx
3. 2
4  y  0. 4. 8  16 x  0 .
dx dx dt 2 dt
3 2
d y d y dy d3 y d2 y dy
5. 3
3 2
3  y  0. 6. 3
6  11  6y  0 .
dx dx dx dx dx2 dx
d4 y d2 y d4 y d2 y
7. 4
5 2
 4y  0. 8. 6  9y  0 .
dx dx dx4 dx2
d3 y
9. (D2 + 1)3 (D2 + D + 1)2 y = 0. 10.  y  0.
dx3
d2 y  
11. + y = 0, given that y(0) = 2 and y   = 2.
dx 2 2
d2 x dx dx
12. 3 + 2x = 0, given that, when t = 0, x = 0 and  0.
2 dt dt
dt
d2 y dy dy
13. 4 + 29y = 0, given that, when x = 0, y = 0 and = 15.
dx2 dx dx
d4 x
14. If = m4x, show that x = c1 cos mt + c2 sin mt + c3 cosh mt + c4 sinh mt.
dt4
15. Solve the differential equation: 9y+ 3y 5y + y = 0.
d3 y d2 y dy
16. Solve the differential equation 6  12  8 y = 0 under the conditions y(0) =
3
dx dx 2 dx
0, y(0) = 0 and y(0) = 2.
d2i R di i
17. Solve the differential equation 2
  = 0, where R2C = 4L and R, C, L are
dt L dt LC
constants.
Answers
1. y = c1 e5x + c2ex 2. y = c1 eax + c2ebx
3. y = c1 e(2  3)x
 c2 e(2  3)x
4. x = (c1 + c2t) e4t
5. y = (c1 + c2x + c3x2) ex 6. y = c1ex + c2e2x + c3e3x
7. y = c1ex + c2ex + c3e2x + c4e2x 8. y = (c1 + c2x) cos 3 x + (c3 + c4x) sin 3 x
9. y = (c1 + c2x + c3x ) cos x + (c4 + c5x + c6x2) sin x
2
1
 x
2
(c  c8 x) cos
3
x  (c9  c10 x) sin
3
x
"#
+e
! 7
2 2 #$
3x 3x
10. y = c1ex + ex/2 c2 cos 2
 c3 sin
2 11. y = 2 (cos x sin x)
1
x
12. x=0 13. y = 3e2x sin 5x 15. y = c1ex + (c2 + c3x) e 3
Rt

16. y = x2e2x 17. i = (c1 + c2t) e 2L .

Linear Differential
THE INVERSE OPERATOR Higher Order
I '
NOTES
1
Definition. X is that function of x, free from arbitrary constants, which
f (D)
when operated upon by f(D) gives X.
%& 1 X()  X
Thus f(D)
' f (D) *
1
 f(D) and are inverse operators.
f (D)
1
Theorem 1. X is the particular integral of f(D)y = X.
f ( D)
Proof. The given equation is f(D)y = X ...(1)
1
Putting y = X in (1), we have
f (D)
%& 1 X () = X
f(D)
' f (D) * or X = X
which is true.
1
 y= X is a solution of (1).
f (D)
Since it contains no arbitrary constants, it is the particular integral of f(D) y = X.
Theorem 2.
1
D
X I X dx.
1
Proof. Let Xy
D
1 X = Dy dy
Operating both sides by D, we have D
D or X=
dx
Integrating both sides w.r.t. x
y= I X dx ,
1
no arbitrary constant being added since y = X contains no arbitrary constant.
D

1
D
X= I X dx.
Theorem 3.
1
Da
X  e ax I Xe  ax dx .
1
Proof. Let X y
D a

Ordinary Differential 1 X = (D a)y
Equations Operating on both sides by (D a), (D a)
D a
dy dy
or X=  ay i.e., ay = X
NOTES dx dx
which is a linear equation and I.F. = e I  adx
I
= e ax
 Its solution is yeax = X eax dx, no constant being added
or y = eax I X eax dx
Hence,
1
Da
X  e ax I e  ax X dx .
RULES FOR FINDING THE PARTICULAR INTEGRAL
Consider the differential equation,

(Dn + a1Dn1 + a2Dn2 + ...... + an1D + an)y = X
It can be written as f (D)y = X
1
 P.I. = X
f (D)
Case I. When X = eax
Since, D eax = a eax
D2 eax = a2 eax
...........................
...........................
Dn1 eax = an1 eax
Dneax = an eax
 (Dn + a1 Dn1 + a2 Dn2 + ...... + an1D + an) eax
= (an + a1 an1 + a2 an2 + ...... + an1 a + an) eax
or f(D) eax = f(a) eax
1
Operating on both sides by .
f (D)
1 1 1
( f (D) e ax )  ( f (a)e ax ) or eax = f(a) e ax
f (D) f (D) f (D)
1 ax 1 ax
Dividing both sides by f(a), e  e , provided f(a)  0
f (a) f (D)
1 1
Hence, eax  eax , provided f(a) ¹ 0.
f (D) f (a)
Case of failure. If f(a) = 0, the above method fails.

Since f(a) = 0, D = a is a root of A.E. f(D) = 0
 D a is a factor of f(D).
Let f(D) = (D a)  (D), where  (a)  0 ...(i)

1 ax 1 1 1 ax 1 1 ax Linear Differential
Then e  e ax  . e  . e Equations of Second and
f (D) (D  a)  (D) D  a  (D) D  a  (a)
I
Higher Order
1 1 1 ax
= . e ax  e e ax . e  ax dx [By
 (a) D  a  (a) NOTES
Theorem 3]
=
1 ax
 (a)
e I 1 dx  x .
1 ax
 (a)
e ...(ii)
Differentiating both sides of (i) w.r.t. D, we have f  (D) = (D a)  (D) +  (D)

 f (a) =  (a)
1 1
 From (ii), we have eax  x . eax , provided f ¢ (a) ¹ 0
f (D) f (a)
1 ax 1
If f (a) = 0, then e  x2 . e ax , provided f (a)  0
f (D) f  (a)
and so on.
SOLVED EXAMPLES
Example 6. Find the P.I. of (4D2 + 4D 3) y = e2x.

1 1
Sol. P.I. = 2
e2 x  2
e 2 x (replacing D by 2)
4D  4D  3 4(2)  4(2)  3
1 2x
e .
=
21
Example 7. Find the P.I. of (D2 + 3D + 2)y = 5.
1
Sol. P.I. = 2
D  3D  2
(5e 0 x ) '
[ e0x = 1]
1
=5. e0 x (replacing D by 0)
002
5
= .
2
Example 8. Find the P.I. of (D3 3D2 + 4)y = e2x.
1
Sol. P.I. = 3 2
e2 x .
D  3D  4
Here the denom. vanishes, when D is replaced by 2. It is a case of failure.
We multiply the numerator by x and differentiate the denominator w.r.t. D.
1
 P.I. = x . e2 x 2
3D  6D
It is again a case of failure. We multiply the numerator by x and differentiate
the denominator w.r.t. D.
1 1 x2 2x
 P.I. = x2 . e2 x  x 2 . e2x  e .
6D  6 6(2)  6 6

Ordinary Differential Case II. When X = sin (ax + b) or cos (ax + b)
Equations
D sin (ax + b) = a cos (ax + b)
D2 sin (ax + b) = ( a2) sin (ax + b)
NOTES D3 sin (ax + b) = a3 cos (ax + b)
D4 sin (ax + b) = a4 sin (ax + b)
or (D2)2 sin (ax + b) = ( a2)2 sin (ax + b)
..................................................................
..................................................................
In general, (D2)n sin (ax + b) = ( a2)n sin (ax + b)
 f(D2) sin (ax + b) = f ( a2) sin (ax + b)
1
Operating on both sides by ,
f (D 2 )
1 1
2 (f(D2) sin (ax + b)) = [f( a2) sin (ax + b)]
f (D ) f (D 2 )
1
or sin (ax + b) = f( a2) sin (ax + b).
f (D 2 )
Dividing both sides by f( a2),
1 1
sin (ax  b)  sin (ax + b), provided f( a2)  0.
f ( a2 ) f (D 2 )
1 1
Hence, sin (ax  b) = sin (ax + b), provided f( a2) ¹ 0
f (D2 ) f (  a2 )
1 1
Similarly, cos (ax  b) = cos (ax + b), provided f( a2) ¹ 0
f (D ) 2
f (  a2 )
Case of Failure. If f( a2) = 0, the above method fails.
Since cos (ax + b) + i sin (ax + b) = ei(ax + b) | Eulers Theorem
1 1
 [cos (ax + b) + i sin (ax + b)] = ei(ax + b)
f (D ) 2
f (D 2 )
[If we replace D by ia, f(D2) = f( a2) = 0, so that it is a case of failure]
1 1
=x. 2
e i( ax  b)  x . [cos (ax + b) + i sin (ax + b)]
f  (D ) f  (D 2 )
Equating real parts
1 1
2
cos (ax  b) = x . cos (ax + b), provided f ¢( a2) ¹ 0
f (D ) f (D2 )
Equating imaginary parts
1 1
2
sin (ax + b) = x . sin (ax + b), provided f ¢( a2) ¹ 0
f (D ) f (D2 )
If f( a2) = 0, then
1 1
sin (ax + b) = x2 . sin (ax + b), provided f ( a2)  0
2
f (D ) f  (D 2 )
1 1
cos (ax + b) = x2 . cos (ax + b), provided f ( a2)  0
2
f (D ) f  (D 2 )
and so on.

Example 9. Find the P.I. of (D3 + 1)y = sin (2x + 3). Linear Differential
1 1 Higher Order
Sol. P.I. = sin (2x + 3) = sin (2x + 3)
3
D 1 D( 2 2 )  1
[Putting D2 = 22]
NOTES
1
= sin (2x + 3)
1  4D
Multiplying and dividing by (1 + 4D)
1  4D 1  4D
= sin (2x + 3) = sin (2x + 3)
(1  4D)(1 + 4D) 1  16D 2
1  4D
= sin (2x + 3) [Putting D2 = 22]
1  16( 2 2 )
1
= [sin (2x + 3) + 4D sin (2x + 3)]
65
1 ' d "#
=
65
[sin (2x + 3) + 8 cos (2x + 3)]
! D=
dx $
Example 10. Find the P.I. of (D2 + 4)y = cos 2x.
1
Sol. P.I. = 2
cos 2 x
D 4
Here the denominator vanishes when D is replaced by 22 = 4. It is a case of
failure. We multiply the numerator by x and differentiate the denominator w.r.t. D.
 P.I. = x .
1
2D
cos 2 x 
x
2
cos 2 x dx I 1
D
'
f ( x)  f ( x) dx
! I "#
$
x
= sin 2x.
4
Case III. When X = xm, m being a positive integer.
1
Here, P.I. = xm
f (D)
Take out the lowest degree term from f(D) to make the first term unity (so that
Binomial Theorem for a negative index is applicable).
The remaining factor will be of the form 1 +  (D) or 1  (D)
Take this factor in the numerator. It takes the form
[1 +  (D)]1 or [1  (D)]1
Expand it in ascending powers of D as far as the term containing Dm, since
Dm+1 (xm) = 0, Dm+2 (xm) = 0 and so on.
Operate on xm term by term.
Example 11. Find the P.I. of (D2 + 5D + 4)y = x2 + 7x + 9.
1 1
( x 2  7 x  9) 
Sol. P.I. = 2
D  5D  4 5D D 2 (x2 + 7x + 9)
4 1
4

4
1 5D D 2 "# 1
! #$
= 1  (x2 + 7x + 9)
4 4 4

1 5D D 2
5D D 2  ......"# (x + 7x + 9)
2
Equations
= 1
   #$
2
4
! 4 4 4 4
1 5D D 2
25D 2
= 1 ...... (x + 7x + 9)
NOTES  
4
2
4 4 16
1 5D 21D 2
= 1  ...... (x + 7x + 9)
4
 2
4 16
1 5 21 "
2 2
D ( x  7 x  9)#
2 2
4!
( x  7 x  9)  D ( x + 7 x  9) +
=
4 16 $
1 21 " 1 23
(2)#  x  x 
$ 4 2 8 .
2 5 9 2
4!
= ( x  7 x  9)  (2 x  7) 
4 16
Case IV. When X = eax V, where V is a function of x.
Let u be a function of x, then by successive differentiation, we have
D(eax u) = eax Du + a eax u = eax (D + a)u
D2 (eax u) = D [eax (D + a) u] = eax (D2 + aD) u + aeax (D + a)u
= eax (D2 + 2aD + a2) u = eax (D + a)2 u
Similarly, D3 (eax u) = eax (D + a)3 u
In general, Dn (eax u) = eax (D + a)n u
 f(D) (eax u) = eax f (D + a) u
1
Operating on both sides by ,
f (D)
1 1
[f(D) (eax u)] = [eax f(D + a)u]
f (D) f (D)
1
 eax u = [eax f(D + a)u] ...(i)
f (D)
1
Now let f(D + a) u = V, i.e., u= V
f (D + a)
1 1
 From (i), we have eax V= (e ax V )
f (D + a) f (D)
1 1
or (eax V) = eax V.
f (D) f (D  a)
1
Thus eax which is on the right of may be taken out to the left
f (D)
provided D is replaced by D + a.
Example 12. Find the P.I. of (D2 4D + 3)y = ex cos 2x.
1 1
Sol. P.I. = 2
ex cos 2x = ex cos 2x
2
D  4D  3 (D  1)  4(D  1)  3
1 1
= ex cos 2x = ex 2
cos 2x [Putting D2 = 22]
2
D  2D  2  2D
1 x 1 1 2D
= e cos 2x = e x cos 2 x
2 2D 2 (2  D)(2  D)

1 x 2D 1 2D Linear Differential
= e 2
cos 2 x   e x cos 2 x Equations of Second and
2 4D 2 4  ( 2 2 ) Higher Order
1 x 1 x
= e (2 cos 2x D cos 2x) = e (2 cos 2x + 2 sin 2x)
16 16 NOTES
1 x
= e (cos 2x + sin 2x).
8
Case V. When X is any other function of x.
Resolve f(D) into linear factors.
Let f(D) = (D m1)(D m2) ...... (D mn)
1 1
Then P.I. = X= X
f (D) (D  m1 )(D  m2 ) ...... (D  mn )
 A1 A2 An 
=    ......  X (Partial Fractions)
 D  m1 D  m2 D  m2 
1 1 1
= A1 X  A2 X  ......  A n X
D  m1 D  m2 D  mn
I
= A 1 e m1 x X e m1 x dx  A 2 e m2 x I X e m2 x dx  ......  A n e mn x I X e mn x dx
Remark. We know that ei = cos  + i sin 

'
!
1
Dm
X = e mx I
(Eulers Theorem)
"#
Xe  mx dx .
$
 xn sin ax = Imaginary part of xn (cos ax + i sin ax)
= I.P. of xneiax
and xn cos ax = Real part of xn (cos ax + i sin ax)
= R.P. of xneiax.
Example 13. Solve (D3 6D2 + 11D 6)y = e2x + e3x.

Sol. A.E. is D3 6D2 + 11D 6 = 0 or (D 1)(D 2)(D 3) = 0
whence D = 1, 2, 3
 C.F. = c1ex + c2e2x + c3 e3x
1
P.I. = (e2x + e3x)
D 3  6D 2  11D  3
1 1
= 3 2
e 2 x  3 2
e 3 x
D  6D  11D  6 D  6D  11D  6
1 1
= e 2 x  e3x
( 2) 3  6( 2) 2  11( 2)  6 ( 3) 3  6( 3) 2  11( 3)  6
1 2x 1 3x 1
= e e = (2e2x + e3x)
60 120 120
Hence the C.S. is y = C.F. + P.I.
1
i.e., y = c1ex + c2e2x + c3e3x (2e2x + e3x).
120

Ordinary Differential Example 14. Solve (D 2)2y = 8(e2x + sin 2x + x2).
Equations
Sol. A.E. is (D 2)2 = 0 whence D = 2, 2
 C.F. = (c1 + c2x)e2x
NOTES 1
P.I. = [8(e2x + sin 2x + x2)]
(D  2) 2
=8
1 e2 x 
1
sin 2 x 
1
x2
"#
! (D  2) 2
(D  2) 2
(D  2) 2
$
1 1
Now, 2
e2x = x . e2x | Case of failure
(D  2) 2 (D  2)
1 2x
= x2 . e | Case of failure
2
x 2 2x
= e
2
1 1 1
2
sin 2 x  2
sin 2 x  2
sin 2x
(D  2) D  4D  4  2  4D  4
[Putting D2 = 22]
=
1
4D
sin 2 x  
1
4
sin 2 x dx  
1
4
Icos 2 x
2
1
 cos 2x
8

1 1 1 1 D 2
x2
(D  2) 2
x2 
(2  D) 2
x2 
D 2
x2 
4
1
2

4 1
2

1 D ( 2)( 3) D ......"# x 2
! #$
= 1 2   2
4 2 2 2
1 3 "
1  D  D  ......# x 2 2
4! $
=
4
1 3 "
x  D ( x )  D ( x )#
2 2 2 2
4! $
=
4
x e  1 cos 2 x  1 x  2 x  3 "#
2
4 2 $
2x 2
 P.I. = 8
!2 8
= 4x2 e2x + cos 2x + 2x2 + 4x + 3
Hence the C.S. is y = (c1 + c2x) e2x + 4x2 e2x + cos 2x + 2x2 + 4x + 3.
Example 15. Solve: (D + 2)(D 1)2 y = e2x + 2 sinh x.

Sol. A.E. is (D + 2)(D 1)2 = 0 so that D = 2, 1, 1
 C.F. = c1e2x + (c2 + c3x) ex
1
P.I. = (e2x + 2 sinh x)
(D  2)(D  1) 2
=
1
(e2x + ex ex)
' sinh x 
e x  e x "#
(D  2)(D  1) 2
! 2 $
Now
1
e 2 x 
1 1
e 2 x 
1 "# 1
e 2 x
"# Linear Differential
(D  2)(D  1) 2 !
D  2 (D  1) 2
$
D  2 (  2  1) 2 ! $ Higher Order
1 1
= . e2x | Case of failure
9 D2 NOTES
1 1 2 x x 2 x
= x. e  e
9 1 9
1
ex 
1 1
ex 
1 1 ex
"# "#
(D  2)(D  1) 2 2
(D  1) D  2 2
!
(D  1) 1  2 $ ! $
1 1
= . ex | Case of failure
3 (D  1) 2
1 1
= .x ex | Case of failure
3 2 (D  1)
1 2 1 x 1 2 x
= .x . e  x e
3 2 6
1 1 1
ex  e  x  ex
(D  2)(D  1) 2 (1  2)( 1  1) 2
4
x 2x x 2 x 1 x
 e + P.I. = e + e
9 6 4
x x 2 x 1 x
Hence the C.S. is y = c1e2x + (c2 + c3x)ex + e2x + e + e .
9 6 4
d2 y
Example 16. Solve 4y = x sinh x.
dx 2
Sol. Given equation in symbolic form is (D2 4)y = x sinh x
A.E. is D2 4 = 0 so that D=2
 C.F. = c1 e2x + c2e2x
1 1
e x  e– x
P.I. = 2
D 4
x sinh x =
D2  4
x
2
=
1 1
ex . x  2
1
ex . x
"#
!2 D2  4 D 4 $
1 "
x#
1x 1 x
= e xe
2 ! (D  1)  4 (D  1)  4 $
2 2
1 e 1
x
xe
1
x#
" x
2 ! D  2D  3 D  2D  3 $
= 2 2
1 e 1x 1 "
x #
x
=
 31 
2D D xe
2
2D D ## 2
3
 3 1 
3
2
! 3

3

$
%K 2D D (K %K1  2D  D (K x"#
2
1
2
1
e &1  ) &K 3 3 )K #
1 x x
! K' 3 3 K*
=  x  e
6 ' * $
1  x  2D   2D  
Equations = e 1 ......  x  e  x 1  ......  x 
6   3   3  
1  x 2 x  2 
NOTES =  e  x  3   e  x  3 
6     

x e x  ex 2 e x  e x x 2
=
3 2 
9 2 3 9
  sinh x cosh x
x 2
Hence the C.S. is y = c1e2x + c2e2x sinh x cosh x.
3 9
d4 y
Example 17. Solve y = cos x cosh x.
dx 4
Sol. Given equation in symbolic form is (D4 1)y = cos x cosh x
A.E. is D4 1 = 0 or (D2 1)(D2 + 1) = 0 so that D =  1,  i
 C.F. = c1 ex + c2 ex + e0x (c3 cos x + c4 sin x)
= c1 ex + c2 ex + c3 cos x + c4 sin x
e  e
cos x
1
x x
2
1
P.I. = 4
cos x cosh x = 4
D 1 D 1
1 1 "
e cos x #
1 x x
= e cos x 
2 !D  1 4
D 1 $ 4
1 "
cos x #
1
x 1 x
= e cos x  e
2 ! (D  1)  1 4
(D  1)  1 $ 4
1 x 1 x 1 "#
= e cos x  e cos x
2 ! D  4D  6D  4D 4 3
D  4D 2 4 3 2
 6D  4D $
1 x 1
= e cos x
2 ! ( 1 )  4D(  1 )  6 ( 1 )  4D
2 2 2 2
 ex
1
cos x
"#
2 2 2
( 1 ) – 4D(  1 )  6 (  1 ) – 4D 2
$

1 x 1 1 1 e x  e x "# cos x
=
2
e
!5
cos x  e x
5
cos x  
5 2 $
1
= cosh x cos x
5
1
Hence the C.S. is y = c1ex + c2ex + c3 cos x + c4 sin x cos x cosh x.
5
dy d2 y
Example 18. Solve + y = xex sin x.
2
2
dx dx
Sol. Given equation in symbolic form is (D2 2D + 1)y = xex sin x
A.E. is D2 2D + 1 = 0 or (D 1)2 = 0 so that D = 1, 1
 C.F. = (c1 + c2x)ex
1 1
P.I = ex . x sin x = ex . x sin x
(D  1) 2 (D + 1  1) 2

= ex
D2
1
x sin x = ex
1
D I x sin x dx Integrating by parts
Linear Differential
I
Higher Order
1 1 "#
= ex
D !
x ( cos x)  1( cos x) dx  e x
D
( x cos x + sin x)
$ NOTES
= ex
I (  x cos x  sin x) dx  e x
= ex[ x sin x cos x cos x] = ex(x sin x + 2 cos x)

 Jx sin x 
! I L
1 .sin x dx  cos x "#
$
Hence the C.S. is y = (c1 + c2x)ex ex(x sin x + 2 cos x).
d2 y
Example 19. Solve 4y = cosh (2x 1) + 3x.
dx 2
(D2 4)y = cosh (2x 1) + 3x
A.E. is D2 4=0  D=2
 C.F. = c1 e2x + c2e2x
1
P.I. = 2 [cosh (2x 1) + 3x]
D 4
=
1
e 2 x 1  e  (2 x  1)
 e log 3
x "# ' cosh t  e t
 e t
and u  elog u
"#
2
D 4 2 ! $! 2 $
=

1 1
e2 x 1  2
1
e ( 2 x  1)
"#  1 e x log 3
! 2
2 D 4 D 4 $ D 4
2
1 1 1
2 x 1 "#  1  ( 2 x  1)
 x. e x log 3
2 ! 2D $ (log 3)
= x. e e 2
2D 4
1 1 1
2 x 1 "#  1  ( 2 x  1)
e x log 3
2 ! 2D $ (log 3)
= x. e  x. e 2
2D 4
= e
x
4! I 2 x 1
dx  e dx" 
# I 3
$ (log 3)  4
 (2 x  1)
x
x e 2 x1
e "#  3 ( 2 x  1) x
= 
4! 2  2 $ (log 3)  4 2
x e e 2 x 1 "#  3  ( 2 x  1) x
=
4! 2 $ (log 3)  4 2
x 3x
= sinh (2x 1) +
4 (log 3) 2  4
x 3x
Hence the C.S. is y = c1e2x + c2e2x +sinh (2x 1) + .
4 (log 3) 2  4
Example 20. Solve (D2 + 1)y = x2 sin 2x.
Sol. A.E. is D2 + 1 = 0  D=i
 C.F. = c1 cos x + c2 sin x
1 1
P.I. = 2 x2 sin 2x = I.P. of 2 x2e2ix
D 1 D 1

Equations = I.P. of e2ix 2 x2 = I.P. of e2ix 2 x2
(D  2i)  1 D  4 iD  3
1
x2
= I.P. of e2ix
4 D 2
NOTES  3 1 3
iD 
3
e 2ix 4 iD  D 2 "# x 1
= I.P. of
3 !
1 3 #$ 2
1
4 iD  D 4 iD  D 2 "# 2 2
1 3
 
#
2ix ...... 2
= I.P. of e
3
! 3
$
x
1 1  4iD  1  16 D  ......"# x
! 3 3 9
2ix 2 2
= I.P. of e
3 $
1 x  4i (2x)  13 (2)"#
2
! 3 $
= I.P. of e 2ix
3 9
= I.P. of (cos 2x + i sin 2x) x 

1 26  8 x i"#
! 9 3 $
2
3
1 8x 26 sin 2x"#
cos 2 x  x 
9
2
3!3 $
=
1
= [24x cos 2x + (9x2 26) sin 2x]
27
1
Hence the C.S. is y = c1 cos x + c2 sin x [24x cos 2x + (9x2 26) sin 2x].
27
Example 21. Solve (D4 + 2D2 + 1)y = x2 cos x.
Sol. A.E. is (D2 + 1)2 = 0  D =  i,  i
 C.F. = (c1x + c2) cos x + (c3 x + c4) sin x
1 1
P.I. = 2 2
x2 cos x = R.P. of x2 (cos x + i sin x)
(D  1) (D  1) 2
2
1 1
= R.P. of x2eix = R.P. of eix x2
(D 2  1) 2 [(D  i) 2  1]2
1 1
= R.P. of eix x2 = R.P. of eix 2
(D  2iD) 2
2
2iD 1  D #" x 2
! 2i $
1
1  i2D
ix 2
1 e
4 D
2
= R.P. of eix x = R.P. of . x2
iD
 4D 1 –
2 2
2
2
1 iD iD "#
2
1 2  3
1
= R.P. of e .
4 D !
ix
2 2 2
 ...... x2
#$
1 1 3 "
x  i(2 x)  (2)# 2
D ! $
= R.P. of e . ix
2
4 4

1
= R.P. of eix .
1 x3 3
 ix 2  x
"# Linear Differential
4 D 3 2 ! $ Higher Order
= R.P. of
1 ix x
e
4
i
x3 3x 2

"#
4 !
12 3 4 $ NOTES
1
= R.P. of (cos x + i sin x) [(x4 9x2) + (4x3)i]
48
1
= [(x4 9x2) cos x 4x3 sin x]
48
Hence the C.S. is
1
y = (c1x + c2) cos x + (c3x + c4) sin x [(x4 9x2) cos x 4x3 sin x].
48
d2 y
Example 22. Solve + y = cosec x.
dx 2
Sol. Given equation in symbolic form is (D2 + 1)y = cosec x
A.E. is D2 + 1 = 0  D =  i
 C.F. = c1 cos x + c2 sin x
1 1
P.I. = 2 cosec x = cosec x
D 1 (D  i)(D  i)
1 1 1
=


2i D  i D  i
cosec x (Partial Fractions)
1 1
=
2i D  i
cosec x 
1
Di
cosec x

Now
1
Di
cosec x = e ix
I
cosec x e dx
' 1 X = e Xe dx"#
! Da
ix
$
ax
I – ax
= eix I
= eix (log sin x ix)
cosec x (cos x i sin x) dx = eix I (cot x i) dx
1
Changing i to i, we have cosec x = eix (log sin x + ix)
Di
1 ix
 P.I. = [e (log sin x ix) eix (log sin x + ix)]
2i
e ix
 e  ix eix  e  ix
= log sin x 2i
x
2
= log sin x . sin x x cos x
Hence the C.S. is y = c1 cos x + c2 sin x + sin x log sin x x cos x.
d2 y
Example 23. Solve + a2y = tan ax.
dx 2
Sol. Given equation in symbolic form is (D2 + a2)y = tan ax
A.E. is D2 + a 2 = 0  D =  ia
 C.F. = c1 cos ax + c2 sin ax
1 1
P.I. = 2 2 tan ax = tan ax
D a (D  ia)(D  ia)
Equations =

1 1

1 "#
tan ax (Partial Fractions)
!
2ia D  ia D  ia $
1 1 1 "
= tan ax  tan ax#
NOTES 2ia ! D  ia D  ia $
Now
1
D  ia
tan ax = eiax I tan ax . eiax dx
= eiax I tan ax (cos ax i sin ax) dx = eiax I sin ax  i sin ax dx

2
cos ax
= eiax I sin ax  i

cos ax

1  cos 2 ax
dx = eiax I [sin ax i(sec ax cos ax)] dx
= eiax 
cos ax i sin ax "#
! a
 log (sec ax  tan ax)  i
a a $
1 iax
= e [(cos ax i sin ax) + i log (sec ax + tan ax)]
a
1 iax iax
= e [e + i log (sec ax + tan ax)]
a
1
= [1 + ieiax log (sec ax + tan ax)]
a
1 1
Changing i to i, we have tan ax = [1 ieiax log (sec ax + tan ax)]
D  ia a
1  1
 P.I. = 
2ia   a

1  ieiax log sec ax  tan ax  
1 
 {1  ie iax log (sec ax  tan ax )  
a 
1  eiax  e iax 
= log (sec ax + tan ax)  
a2  2 
1
=  log (sec ax + tan ax) . cos ax
a2
1
Hence the C.S. is y = c1 cos ax + c2 sin ax cos ax log (sec ax + tan ax).
a2
EXERCISE B
d3 y d2 y
1. 3 + y = 3 + 5ex.. 2. 4y = (1 + ex)2.
dx dx2
d2 y dy d2 y dy
3. 2
4 + 5y = 2 cosh x. 4. 2
2 + 5y = sin 3x.
dx dx dx dx
d y 3 2
d y dy d2 y dy
5. (i)   + y = sin 2x. (ii)   cos 2 x
dx 3
dx 2 dx dx2 dx

d3 y x Linear Differential
6. (i) + y = sin 3x cos2 (ii) (D3 + 1)y = 2 cos2 x Equations of Second and
dx3 2
Higher Order
d2 y dy d3 y d2 y dy
(iii) 2
2 + y = e2x cos2 x (iv) 3
2  = ex + sin 2x
dx dx dx dx2 dx
d
NOTES
(v) (D3  D) z  2 y  1  4 cos y  2 e y , where D 
dy
(vi) (D2 + D + 1) y = (1 + sin x)2
7. (D2 4D + 3)y = sin 3x cos 2x.
d2 y
8. (D2 3D + 2)y = 6e3x + sin 2x. 9. + 4y = ex + sin 2x.
dx2
d3 y d2 y dy d2 y
10. 3
2 2
4 = e2x + sin 2x. 11. 4y = x2 + 2x.
dx dx dx dx2
d3 y d2 y dy d2 y dy
12.  6 = 1 + x2. 13.  = x2 + 2x + 4.
dx3 dx2 dx dx2 dx
d2 y x
14. + y = e2x + cosh 2x + x3. 15. (D2 3D + 2)y = 2ex cos .
dx2 2
d2 y dy d4 y
16. 2
3 + 2y = xe3x + sin 2x. 17. y = ex cos x.
dx dx dx4
18. (i) (D2 2D)y = ex sin x. (ii) y 2y + 2y = x + ex cos x
19. (D2 + 4D + 8)y = 12e2x sin x sin 3x.
d2 y
20. (i) + 2y = x2e3x + ex cos 2x. (ii) (D2 + 4D + 3)y = ex sin x + x e3x.
dx2
21. (D3 + 2D2 + D)y = x2e2x + sin2 x. 22. (D2 4D + 4)y = 8x2 e2x sin 2x.
x d2 y
23. (D 1)2(D + 1)2y = sin2 + ex + x. 24. + 4y = x sin x.
2 dx2
d2 y
25. (D2 1)y = x2 sin x. 26. 9y = x cos 2x.
dx2
27. (D2 1)y = x sin x + (1 + x2)ex. 28. (D2 1)y = x sin 3x + cos x.
d2 y d2 y
29. 2
+ a2y = sec ax. 30. + 4y = 4 tan 2x.
dx dx2
d2 y dy x
31. 2
+3 + 2y = e e .
dx dx
d2 y dy 
32. Solve 2  10 y  37 sin 3x = 0 and find the value of y when x = being given
dx 2 dx 2
dy
that y = 3, = 0 when x = 0.
dx
Answers
1
x c 3 3 5
1. y = c1 ex + e 2
2 cos
2
x  c3 sin
2
x  3  ex
2
1 2 x 1 2x
2. y = c1e2x + c2e2x  e  xe
4 3 4
1 x 1 x
3. y = e2x(c1 cos x + c2 sin x) e  e
10 2
1
4. y = ex(c1 cos 2x + c2 sin 2x) + (3 cos 3x 2 sin 3x)
26

Equations 5. (i) y = c1ex + c2 cos x + c3 sin x + (2 cos 2x sin 2x)
15
x 1
(ii) y  c1  c2e  (sin 2 x  2 cos 2 x )
10
NOTES 1
x c 3 3 1
6. (i) y = c1 ex + e 2
2 cos
2
x  c3 sin
2
x 
730
(sin 3x + 27 cos 3x)
1 1
 (cos x sin x)
2 4
x
c 3 3 1
(ii) y  c1e x
 e 2

2 cos
2
x  c3 sin
2
x  1
65
(cos 2 x  8 sin 2 x)
1 2x 1 1
(iii) y = (c1 + c2x)ex + e  (3 cos 2x 4 sin 2x)
9 2 50
2
x x 1
(iv) y = c1 + (c2 + c3x)ex e  (3 cos 2x 4 sin 2x)
2 50
y y 2 y
(v) z  c1  c2 e  c3e  y  y  2 sin y  ye
x
c 3 3 3 1 3


(vi) y  e 2
1 cos x  c2 sin x   2 cos x  sin 2 x  cos 2 x
2 2 2 13 26
1 1
7. y = c1ex + c2e3x + (10 cos 5x 11 sin 5x) + (sin x + 2 cos x)
884 20
3 3x 1
8. y = c1ex + c2e2x + e + (3 cos 2x sin 2x)
10 20
1 x
9. y = c1 cos 2x + c2 sin 2x + e x  cos 2x
5 4
1 2x
10. y = c1 + ex (c2 cos 3 x + c3 sin 3 x) + (e + sin 2x)
8
1 1
11. y = c1e2x + c2e2x x 2  2 x 
4 2
1 x2 25
12. y = c1 + c2e3x + c3e2x
18
x3 
2

6
x
x3
13. y = c1 + c2ex + + 4x
3
1 2x 1
14. y = c1 cos x + c2 sin x + e  cosh 2x + x3 6x
5 5
8 x x x
15. y = c1ex + c2e2x e 2 sin  cos
5 2 2
1 1
16. y = c1ex + c2e2x + e3x (2x 3) + (3 cos 2x sin 2x)
4 20
1 x
17. y = c1ex + c2ex + c3 cos x + c4 sin x e cos x
5
1 x x 1 x
18. (i) y = c1 + c2e2x e sin x (ii) y  e (c1 cos x  c2 sin x )  ( x  1  xe sin x )
2 2
1 2x
19. y = e2x (c1 cos 2x + c2 sin 2x) + e (3x sin 2x + cos 4x)
2
e3 x
x2 
12 50 ex
20. (i) y = c1 cos 2 x + c2 sin 2x+
11 11
x
121

17
(4 sin 2x cos 2x)
1 x 1 3x 5
(ii) y = c1ex + c2e3x e (sin x + 2 cos x) + e x 
5 24 12

1 2x 1 1 Linear Differential
21. y = c1 + (c2 + c3x)ex + e (6 x 2  14 x  11)  x  (3 sin 2x + 4 cos 2x) Equations of Second and
108 2 100
Higher Order
22. y = (c1 + c2x)e2x e2x[4x cos 2x + (2x2 3) sin 2x]
1 1 x2 x NOTES
23. y = (c1 + c2x)ex + (c3 + c4x)ex +  cos x + e +x
2 8 8
1
24. y = c1 cos 2x + c2 sin 2x + (3x sin x 2 cos x)
9
1 2
25. y = c1 ex + c2 ex x cos x (x 1) sin x
2
1
26. y = c1 e3x + c2 e3x (13x cos 2x 4 sin 2x)
169
1 1
27. y = c1ex + c2ex (x sin x + cos x) + xex (2x2 3x + 9)
2 12
1 3 1
28. y = c1ex + c2ex x sin 3 x  cos 3 x  cos x
10 5 2
1 log cos ax
29. y = c1 cos ax + c2 sin ax + x sin ax  cos ax
a a
30. y = c1 cos 2x + c2 sin 2x cos 2x log (sec 2x + tan 2x)

x
31. y = c1ex + c2e2x + e2x . e e
32. y = ex (c1 cos 3x + c2 sin 3x) + 6 cos 3x sin 3x ; y = 1.
METHOD OF VARIATION OF PARAMETERS TO FIND P.I.
Consider the linear equation of second order with constant co-efficients

d2 ydy
2
 a1
+ a2y = X ...(1)
dx dx
Let its C.F. be y = c1y1 + c2y2 so that y1 and y2 satisfy the equation
d2 y
dy
 a1 + a2y = 0 ...(2)
dx 2 dx
Now, replacing c1, c2 (regarded as parameters) by unknown functions u(x) and
v(x), let us assume that the P.I. of (1) is y = uy1 + vy2 ...(3)
Differentiating (3) w.r.t. x, we have y = uy1 + vy2 + uy1 + vy2 = uy1 + vy2
...(4)
assuming that u, v satisfy the equation uy1 + vy2 = 0 ...(5)
Differentiating (4) w.r.t. x, we have y = uy1 + uy1 + vy2 + vy2
Substituting the values of y, y and y in (1), we get
(uy1 + uy1 + vy2 + vy2) + a1(uy1 + vy2) + a2(uy1 + vy2) = X
or u(y1 + a1y1 + a2y1) + v(y2 + a1y2 + a2y2) + uy1 + vy2 = X
or uy1 + vy2 = X ...(6)
since y1 and y2 satisfy (2).

0 y2 y y2 yX
Equations Solving (5) and (6), we get u =  1  2
X y2 y1 y2 W
y1 0 y y2 yX
and v =  1  1
NOTES y1 X y1 y2 W
y1 y2
where W= is called the Wronskian of y1, y2.
y1 y2
Integrating, u= Iy1X

W
y2 X
dx
W
dx , v = I
Substituting in (3), the P.I. is known. Thus P.I. = y1
y2 X
W
dx  y2 I I y1X
W
dx .
Note 1. As the solution is obtained by varying the arbitrary constants c1, c2 of the C.F.,
the method is known as variation of parameters.
Note 2. Method of variation of parameters is to be used if instructed to do so.
SOLVED EXAMPLES
Example 24. Apply the method of variation of parameters to solve
d2 y
+ 4y = 4 sec2 2x.
dx 2
Sol. Given equation in symbolic form is (D2 + 4)y = 4 sec2 2x
Its A.E. is D2 + 4 = 0 so that D =  2i
 C.F. is y = c1 cos 2x + c2 sin 2x
Here, y1 = cos 2x, y2 = sin 2x and X = 4 sec2 2x
y1 y2 cos 2x sin 2x
 W=  =2
y1 y2  2 sin 2x 2 cos 2x
P.I. = y1 I y2 X
W
dx  y2 I y1X
W
dx
= cos 2x
I sin 2 x . 4 sec 2 2 x
2
dx  sin 2 x I cos 2 x . 4 sec 2 2 x
2
dx
= 2 cos 2x I sec 2x tan 2x dx + 2 sin 2x I sec 2x dx
sec 2 x 1
= 2 cos 2x . + 2 sin 2x . log (sec 2x + tan 2x)
2 2
= 1 + sin 2x log (sec 2x + tan 2x)
Hence the C.S. is y = c1 cos 2x + c2 sin 2x 1 + sin 2x log (sec 2x + tan 2x).
Example 25. Solve by the method of variation of parameters:
d2 y dy e 3x
2
6  9y  2 .
dx dx x
e3 x
(D2 6D + 9)y =
x2

Its A.E. is (D 3)2 = 0  D = 3, 3 Linear Differential
 C.F. is y = (c1 + c2x)e3x
Higher Order
e3 x
Here, y1 = e3x, y2 = xe3x and X =
x2 NOTES
y1 y2 e3x xe 3 x
 W=  = e6x
y1  y2  3e 3 x (3 x  1)e 3 x
P.I. = y1 I y2 X
W
dx  y2 I y1X
W
dx
e3x e3x
= e3x I xe 3 x .
e 6x
x 2 dx  xe 3 x
I e3 x .
e 6x
x 2 dx
= e3x I 1
x
dx  xe 3 x I 1
x2
dx
1 = (1 + log x) e
= e3x log x + xe3x 
x 3x
Hence, C.S. is y = (c1 + c2x) e3x (1 + log x) e3x

or y = [(c1 1) + c2x log x] e3x
or y = [(C1 + c2x log x) e3x , where C1 = c1 1.
Example 26. Solve by the method of variation of parameters
d2 y
y = ex sin (ex) + cos (ex).
dx 2
(D2 1)y = ex sin (ex) + cos (ex)
Its A.E. is D2 1 = 0  D = 1
 C.F. is y = c1 + c2ex
ex
Here, y1 = ex, y2 = ex and X = ex sin (ex) + cos (ex)
y1 y2 ex ex
 W=  x =2
y1  y2  e  ex
P.I. = y1 I y2 X
W
dx  y2 I y1X
W
dx
e  x [ e  x sin ( e  x )  cos ( e  x )]
= ex  2
dx
e x [ e  x sin ( e  x )  cos ( e x )]
 ex  2
dx
=
1 x
e I ex [ex sin (ex) + cos (ex)] dx
I
2
1 x
e ex [ex sin (ex) + cos (ex)] dx ...(1)
2

Now, I ex [ex sin (ex) + cos (ex)] dx
I
Equations
= (t sin t + cos t) dt, where t = ex
NOTES = [t( cos t) I 1 . ( cos t) dt + sin t]

= ( t cos t + 2 sin t) = ex cos (ex) 2 sin (ex)
Also, I ex [cos (ex) + ex sin (ex)] dx
= ex cos (ex)
| Form I ex [ f (x) + f (x)] dx = exf(x)
 From (1), we have

1 x x 1
e [e cos (ex) 2 sin (ex)] ex . ex cos (ex)
P.I. =
2 2
1 1
= cos (ex) ex sin (ex) cos (ex) = ex sin (ex)
2 2
Hence, C.S. is y = c1ex + c2ex ex sin (ex).
EXERCISE C
Solve by the method of variation of parameters:
d2 y
1. + y = cosec x.
dx2
d2 y d2 y
2. (i) 2 + 16 y = 32 sec 2x (ii) + a2y = sec ax
dx dx2
(iii) y + y = sec2 x (iv) y + 3y + 2y = sin (ex)
d2 y d2 y
3. + y = tan x. 4. + 4y = tan 2x.
dx2 dx2
d2 y
5. (i) + y = x sin x. (ii) (D2 + 1)y = cosec x cot x
dx2
d2 y dy
6. (i) y 2y + 2y = ex tan x. (ii) 2 = ex sin x.
dx2 dx
d2 y dy 1 d2 y dy 12e4 x
7. 6  9 y  3 e3x. 8. 8  16 y  .
dx2 dx x dx2 dx x4
d2 y dy
9. 4 + 4y = e2x sec2 x. 10. y 2y + y = ex log x.
dx2 dx
d2 y 2 d2 y 1
11.  y . 12.  y .
dx 2
1 e x
dx2 1  sin x
Answers
1. y = c1 cos x + c2 sin x x cos x + sin x log sin x
2. (i) y = c1 cos 4x + c2 sin 4x + 8 cos 2x 4 sin 4x log (sec 2x + tan 2x)
1 1
(ii) y = c1 cos ax + c2 sin ax + 2 cos ax log (cos ax) + x sin ax
a a
(iii) y = c1 cos x + c2 sin x 1 + sin x log (sec x + tan x)
(iv) y = c1ex + c2e2x + e2x sin (ex)

3. y = c1 cos x + c2 sin x cos x log (sec x + tan x) Linear Differential
1 Higher Order
4. y = c1 cos 2x + c2 sin 2x cos 2x log (sec 2x + tan 2x)
4
x x2 NOTES
5. (i) y = c1 cos x + c2 sin x + sin x  cos x
4 4
(ii) y = c1 cos x + c2 sin x + cos x log sin x x sin x
6. (i) y = ex (c1 cos x + c2 sin x) ex cos x log (sec x + tan x)
1 x
(ii) y1 = c1 + c2 e2x e sin x
2
1 3x 2 e

8. y = c1  c2 x 

7. y = c1  c2 x  4x
e 2
2x x
1 2 x
9. y = (c1 + c2x log cos x) e2x 10. y = (c1 + c2x) ex + x e (2 log x 3)
4
11. y = c1ex + c2ex 1 xex + (ex ex) log (1 + ex)
12. y = c1 cos x + c2 sin x + sin x log (1 + sin x) x cos x 1.
HOMOGENEOUS LINEAR EQUATIONS (Cauchy-Euler

Equations)
An equation of the form
dn y d n 1 y dy d n 2 y
xn + a1 xn1 + ...... + an1 x
n 1
+ a2xn2 + any = X ...(i)
dx n
dx dx n 2 dx
where ais are constants and X is a function of x, is called Cauchys homogeneous
linear equation.
Such equations can be reduced to linear differential equations with constant
co-efficients by the substitution x = ez or z = log x
dy dy dz dy 1 dy dy d
so that  .  . or x  = Dy, where D =
dx dz dx dz x dx dz dz
d2 y d 1 dy
1 dy 1 d 2 y dz
dx 2
 .
dx x dz
 2
 . .
x dz x dz 2 dx
1 dy 1 d2 y ' dz 1
=  2
x dz x dz 2
2 
dx x
d2 y d2 y dy
or x2 2
 2
 = D2y Dy = D(D 1)y
dx dz dz
3
d y
Similarly, x3 = D(D 1)(D 2)y and so on.
dx 3
Substituting these values in equation (i), we get a linear differential equation
with constant co-efficients, which can be solved by the methods already discussed.

Ordinary Differential SOLVED EXAMPLES
Equations
d3y 2
2 d y 1
Example 27. Solve x3
dx 3
 2x
dx 2
+ 2y = 10 x 
x
.

NOTES
Sol. Given equation is a Cauchys homogeneous linear equation.
Put x = ez i.e., z = log x
dy d2 y
so that x = Dy, x2 = D(D 1)y
dx dx 2
d3 y d
x3 = D(D 1)(D 2)y, where D =
dx 3 dz
Substituting these values in the given equation, it reduces to
[D(D 1)(D 2) + 2D(D 1) + 2]y = 10(ez + ez)
or (D3 D2 + 2)y = 10(ez + ez)
which is a linear equation with constant co-efficients.
Its A.E. is D3 D2 + 2 = 0 or (D + 1)(D2 2D + 2) = 0
2 48
 D = 1, = 1, 1  i
2
c1
 C.F. = c1ez + ez (c2 cos z + c3 sin z) = + x[c2 cos (log x) + c3 sin (log x)]
x
1 1 1
P.I. = 10 3
D – D2  2
(ez + ez) = 10
D3  D2  2
ez  3
D  D2  2

e z
 1 1 
= 10  3 2
ez  z . 2
ez 
1  1  2 3D  2D 
1 z 1 
= 10  e  z . 2
ez 
 2 3(  1)  2(  1) 
2
= 5ez + 2zez = 5x + log x
x
c1 2
Hence the C.S. is y = + x[c2 cos (log x) + c3 sin (log x)] + 5x + log x.
x x
2
d y dy
Example 28. Solve x2 x 3y = x2 log x.
dx 2 dx
Put x = ez i.e., z = log x so that
dy d2 y d
x = Dy, x2 = D(D 1)y, where D = .
dx dx 2 dz
[D(D 1) D 3]y = ze2z or (D2 2D 3)y = ze2z
Its A.E. is D2 2D 3 = 0 or (D 3)(D + 1) = 0
 D = 3, 1
c2
C.F. = c1e3z + c2ez = c1x3 +
x

1 Linear Differential
P.I. = (e2z . z) Equations of Second and
D 2  2D  3 Higher Order
1 1
= e2z 2
z  e2 z 2
z
(D  2)  2(D  2)  3 D  2D  3 NOTES
1 1 2z
2D D 2 "# 1
! #$
= e2z z e 1  z
2D D 2 3 3 3
 3 1 3

3
1 2z   2D D2  
= e 1      ...... z
3   3 3  
1 2z  2 x2  2
=  e  z  3    3  log x  3 
3    
c2 x 2 2
Hence the C.S. is y = c1x3 +
x

3
log x 
3
.

d2 y dy
Example 29. Solve x2 2
x + y = log x sin (log x).
dx dx
dy d2 y
Put x = ez i.e., z = log x so that x = Dy, x2 = D(D 1)y
dx dx 2
d
where D= .
dz
[D(D 1) + D + 1]y = z sin z
or (D2 + 1)y = z sin z
Its A.E. is D2 + 1 = 0 so that D=±i
C.F. = c1 cos z + c2 sin z = c1 cos (log x) + c2 sin (log x)
1 1
P.I. = 2 z sin z = Imaginary part of 2
zeiz
D 1 D 1
1 1
= I.P. of eiz z = I.P. of eiz 2 z
(D  i) 2  1 D  2iD
1 1
= I.P. of eiz
D
z = I.P of eiz
iD z
2iD 1 +
2i 2iD 1 
2
1 iz 1 iD
1
1 iz 1
iD
= I.P. of
2i
e
D
1
2
z = I.P. of
2i
e
D
1
2
 ...... z
= I.P. of
1 iz 1
2i
e
D
z
i
2

= I.P. of
1 iz
2i

e
z

i

2
dz I
i
= I.P of e
z  i z = I.P. of e  i z  z
2
2 2 4 4
iz iz 2
2
i z z cos z  z sin z
= I.P. of (cos z + i sin z)  z   
2
4 4 4
2
4
1 1
= (log x)2 cos (log x) + log x sin (log x)
4 4
Ordinary Differential Hence the C.S. is
Equations y = c1 cos (log x) + c2 sin (log x) 1
(log x)2 cos (log x) + 1
log x sin (log x).
4 4
d2 y
dy
Example 30. Solve: x2  4x + 2y = ex.
NOTES dx 2 dx
Put x = ez i.e., z = log x so that
dy d2 y d
x = Dy, x2 = D(D 1)y, where D =
dx dx 2 dz
z z
[D(D 1) + 4D + 2]y = e e or (D2 + 3D + 2)y = e e
Its A.E. is D2 + 3D + 2 = 0 or (D + 1)(D + 2) = 0
 D = 1, 2
C.F. = c1ez + c2e2z = c1x1 + c2x2
1 z 1 z
P.I. = 2
ee  ee
D  3D  2 (D  1)(D  2)
1  1 e ez 1 z
ee 
1
ee
z
=
D  1 D  2 
D  ( 1) D  ( 2)
= ez I z
e e . e z dz  e 2 z I z
e e . e 2 z dz ' D 1 a X  e  X . e



ax  ax 
dx 

= ez I z
e e . e z dz  e 2 z I z
e e . ez . ez dz | Put ez = t
= ez I et dt e2z
= ez . et e2z (t 1)et

I tet dt
| Integrating by parts
z z
= ez . e e e2z(ez 1) e e
z z
= (ez ez + e2z) e e = e2z . e e
= x2ex
Hence the C.S. is y = c1x1 + c2x2 + x2ex or y = (c1x + c2 + ex )x2.
LEGENDRES LINEAR DIFFERENTIAL EQUATION
An equation of the form

n 1
dn y n1 d y ...... dy
(a + bx)n + a1 (a + bx) + + an1 (a + bx) + an y = X ...(i)
dx n
dx n 1 dx
where ais are constants and X is a function of x, is called Legendres linear equation.
Such equations can be reduced to linear differential equations with constant
co-efficients, by the substitution a + bx = ez i.e., z = log (a + bx) so that
dy dy dz b dy
 . 
dx dz dx a  bx dz
dy dy d
or (a + bx) b = b Dy, where D =
dx dz dz

d2 y d b
dy b2 dy
b d 2 y dz
Linear Differential


dx 2 dx a  bx dz
  
. .
(a  bx) 2 dz a  bx dz 2 dx Higher Order
b2 dy b d2 y b b2 d y  dy
2
=  . 
(a  bx) 2 dz a  bx dz 2 a  bx (a  bx) 2
dz dz
2
NOTES
d2 y
or (a + bx)2 = b2 (D2y Dy) = b2 D(D 1)y
dx 2
d3 y
Similarly, (a + bx)3 = b3 D(D 1)(D 2)y.
dx 3
Substituting these values in equation (i), we get a linear differential equation
with constant co-efficients, which can be solved by the methods already discussed.
SOLVED EXAMPLES
d2 y dy
Example 31. Solve (3x + 2)2 + 3(3x + 2) 36y = 3x2 + 4x + 1.
dx 2 dx
Sol. Given equation is a Legendres linear equation.
dy
Put 3x + 2 = ez i.e., z = log (3x + 2) so that (3x + 2) = 3Dy,
dx
d2 y d
(3x + 2)2 = 32 D(D 1)y, where D = .
dx 2 dz
e  2
z 2
e  2 + 1
z
[32 D(D 1) + 3.3D 36]y = 3 3 4 3
1 2z 1 1
or 9(D2 4)y =e  or (D2 4)y = (e2z 1)
3 3 27
Its A.E. is D2 4 = 0  D = ± 2
C.F. = c1e2z + c2e2z = c1 (3x + 2)2 + c2 (3x + 2)2
P.I. =
1
.
1
( e 2 z  1) 
1 1
e2 z  2
1
e0 z
"#
27 D 2  4 27 D 2 – 4 !
D 4 $
=

!
1
27
z.
1 2z
2D
e 
1
04
e0 z
"#  1 z
$ 27 ! 2 I e 2 z dz 
1
4
"#
$
1 z 2z 1"# 1 1
( ze 2 z  1) 
27 ! 4

= e
4

$ 108 108
[(3x + 2)2 log (3x + 2) + 1]
Hence the C.S. is

1
y = c1(3x + 2)2 + c2(3x + 2)2 + [(3x + 2)2 log (3x + 2) + 1].
108

Ordinary Differential EXERCISE D
Equations
Solve:
d2 y dy
1. (i) x2 y + 4xy + 2y = 0. (ii) x2  9x + 25y = 50.
NOTES dx2 dx
d2 y 1 dy d2 y
2. x2 2y = x2 +
. 3. x2  2x 20y = (x + 1)2.
dx2 x dx2 dx
d3 y d2 y dy
4. x2 3
 4x 6 = 4. [Hint. Multiply throughout by x]
dx dx2 dx
d3 y d2 y dy d2 y dy
5. (i) x4 3
 2 x3 2
 x2 + xy = 1. (ii) x2 2  4 x + 6y = x2.
dx dx dx dx dx
d2 y
dy 1
(iii) x2  3x  y .
dx2 dx (1  x) 2
6. The radial displacement u in a rotating disc at a distance r from the axis is given by
d 2u du
r2 r u + kr3 = 0, where k is a constant. Solve the equation under the conditions
dr 2 dr
u = 0 when r = 0, u = 0 when r = a.
d2 y dy d2 y dy
7. x2 2
x + y = log x. 8. x2 2
x + 2y = x log x.
dx dx dx dx
d2 y dy d2 y dy
9. x2  2x 12y = x3 log x. 10. x2  2x 4y = x2 + 2 log x.
dx 2 dx dx2 dx
d2 y dy d3 y d2 y dy
11. x2 2
 3x + 5y = sin (log x). 12. x3  3 x2 x + 8y = 65 cos (log x).
dx dx dx3 dx 2 dx
d2 y dy d2 y dy sin (log x)  1
13. x2  3x + 5y = x2 sin (log x). 14. x2 3x + y = log x .
dx2 dx dx2 dx x
d2 y 1 dy 12 log x
15. (i)  .  . (ii) x2y 4xy + 8y = 4x3 + 2 sin (log x)
dx2 x dx x2
d2 y dy
16. (i) (1 + x)2 2
 (1  x) + y = 4 cos log (1 + x).
dx dx
d2 y dy
(ii) (1 + x)2 2
+ (1 + x) + y = 2 sin [log (1 + x)]
dx dx
d2 y dy
(iii) (1 + x)2 2
+ (1 + x) + y = sin [ 2 log (1 + x)]
dx dx
d2 y dy
17. (x + 1)2 2
+ (x + 1) = (2x + 3) (2x + 4)
dx dx
d2 y dy
18. (1 + 2x)2 6(1 + 2x) + 16y = 8(1 + 2x)2
dx2 dx
d2 y dy
19. (2x + 3)2 2(2x + 3) 12y = 6x.
dx2 dx
Answers
1. (i) y = c1 x1 + c2 x2 (ii) y = x4 [c1 cos (3 log x) + c2 sin (3 log x)] +
2
c2 1 2 1 x2 x 1
2. y = c1x2 +  x  log x 3. y = c1x5 + c2x4  
x 3 x 14 9 20
2 1
4. y = c1 + c2x3 + c3x4 + x 5. (i) y = (c1 + c2 log x)x + c2x1 +
log x
3 4x
1 1 x
(ii) y = c1x2 + c2x3 x2 log x (iii) y = (c1  c2 log x)  log
x x 1 x

kr 2 Linear Differential
6. u= (a r2) 7. y = (c1 + c2 log x)x + log x + 2 Equations of Second and
8
Higher Order
x3
8. y = x[c1 cos (log x) + c2 sin (log x)] + x log x 9. y = c1x3 + c2x4 + log x (7 log x 2)
98
x2 1 3 NOTES
10. y = c1x1 + c2x4  log x 
6 2 8
1
11. y = x2[c1 cos (log x) + c2 sin (log x)] + [sin (log x) + cos (log x)]
8
12. y = c1x2 + x[c2 cos ( 3 log x) + c3 sin ( 3 log x)] + 8 cos (log x) sin (log x)
1 2
13. y = x2[c1 cos (log x) + c2 sin (log x)] x log x cos (log x)
2
1 
14. y = c1 x2 3 + c2 x2  3 +  log x {5 sin (log x )  6 cos (log x )}
61x 
2  1
 {27 sin (log x )  191 cos (log x )}   (1 + log x)
61  6x
15. (i) y = c1 + c2 log x + 2 (log x)3
7 log x  c 7 log x  2x "#

(ii) y  x
5 /2
c1 cos 2 sin 2 3

5
cos (log x) 
7
sin (log x)
#$
!
2
37 37
16. (i) y = c1 cos [log (1 + x)] + c2 sin [log (1 + x)] + 2 log (1 + x) sin [log (1 + x)]
(ii) y = c1 cos [log (1 + x)] + c2 sin [log (1 + x)] log (1 + x) cos [log (1 + x)]
1
(iii) y = c1 cos [log (1 + x )] + c2 sin [log (1 + x)] sin [2 log (1 + x)]
3
17. y = c1 + c2 log (x + 1) + [log (x + 1)]2 + x2 + 8x
18. y = (1 + 2x)2[c1 + c2 log (1 + 2x) + {log (1 + 2x)}2]
3 3
19. y = c1(2x + 3)1 + c2(2x + 3)3 (2x + 3) + .
16 4
LINEAR DIFFERENTIAL EQUATIONS OF SECOND ORDER
The general form of linear equation of second order may be written as

d2 y dy
+P + Qy = R
dx 2 dx
where P, Q and R are the functions of x only. There is no general method of solving this
type of equations, but we will consider some particular cases in which the integral can
be found.
COMPLETE SOLUTION IN TERMS OF KNOWN INTEGRAL
If an integral included in the complementary function of a linear equation of

second order be known then the complete solution can be found. Let y = u be an integral
in the complementary function of the equation
d2 y dy
2 + P + Qy = R ...(1)
dx dx

Ordinary Differential dy
Equations Then put, y = uv so that = u1v + uv1
dx
d2 y
and = u2v + 2u1v1 + uv2
NOTES dx 2
Putting in (1), we get
(u2v + 2u1v1 + uv2) + P(u1v + uv1) + Quv = R
or uv2 + (2u1 + Pu)v1 + (u2 + Pu1 + Qu)v = R
Since y = u is a solution of
d2 y dy
2 +P + Qy = 0  u2 + Pu1 + Qu = 0
dx dx
2 u R
So, we have uv2 + (2u1 + Pu)v1 = R or v2 +
u 1

 P v1 =
u
dp
Putting v1 = p, so that v2 = , we get
dx
dp 2
R
dx
+
u
. u1  P p =

u ...(2)
which is a linear equation in p.
 I.F. = e
I 2

u

. u1  P dx

e
I 2
u
du  I Pdx
2 log u   P dx
= e  u2 e  P dx
 We have p.u2 e I P dx
= I R 2
.u e I P dx dx + c

I
1
u
 p = u2 e  I P dx
4Rue 9 dx + c u e
I P dx 2  I P dx
I
1
or v1 =
dv
dx
= u2 e  I P dx
4Rue 9 dx + c u I P dx
1
2
e I P dx
Integrating again, we have
v= I u2 e  I P dx
 The complete solution of equation (1) is

. I Ru e I P dx
dx dx + c1
I4 u 2 e  I P dx
9 dx + c 2
y = uv = u
I u 2 e  I P dx
The above solution contains only two arbitrary constants.

. I Ru e I P dx
dx dx + c1u I4 u –2 e  I P dx
9 dx + c u 2
TO FIND A PARTICULAR INTEGRAL OF

G \ G\
+P + QY = 0
G[ G[
y = emx is a solution
If y = emx
dy d2 y
Then = memx and = m2emx
dx dx 2

 If y = emx is a solution of (1), then Linear Differential
(m2 + Pm + Q)emx = 0 or m2 + Pm + Q = 0.
Higher Order
Deduction. (i) y = ex is a solution of (1), if 1 + P + Q = 0.
(ii) y = ex is the solution of (1), if 1P+Q=0 NOTES
P Q
(iii) y = eax is the solution of (1), if a2 + Pa + Q = 0 or 1 +  = 0.
a a2
y = xm is a solution
If y = xm
dy d2 y
Then = mxm1 and = m(m 1)xm2
dx dx 2
 If y = xm is a solution of (1), then m(m 1)xm2 + Pmxm1 + Qxm = 0
or m(m 1) + Pmx + Qx2 = 0.
Deduction. (i) y = x is the solution of (1), if P + Qx = 0.
(ii) y = x2 is the solution of (1), if 2 + 2Px + Qx2 = 0.
Note. One integral belonging to the complementary function can be found by
inspection. For this following rules are observed :
(i) y = x is a part of C.R., if P + Qx = 0
(ii) y = ex is a part of C.F., if 1 + P + Q = 0 (i.e., sum of the co-efficients are zero)
(iii) y = ex is a part of C.F., if 1 P + Q = 0
P Q
(iv) y = eax is a part of C.F., if 1 +  =0
a a2
(v) y = x2 is part of C.F., if 2 + 2Px + Qx2 = 0.
SOLVED EXAMPLES
d2 y dy
Example 32. Solve: x2 2
2x(1 + x) + 2(1 + x)y = x3.
dx dx
d2 y 1  1 dy + 2 1  1 y = x
dx 2 2 x dx x x 2
1 1  1 = 0
P + Qx = 2  1 + 2x
x
where 2
x x
 y = x is a part of C.F.
Putting y = vx so that
dy dv
= x+v
dx dx
d2 y d 2v dv
and = +2 , we get
dx 2
dx 2 dx
d 2v dv dp dv
2 =1 or 2p = 1 where p=
dx 2
dx dx dx

Ordinary Differential which is a linear equation
Equations
2 dx
I.F. = e = e2x
I
I
1 2x
 pe2x = 1 . e2x dx + c1 = e + c1
NOTES 2
dv 1
 p= = + c1e2x
dx 2
1 c1 2 x
Integrating, we get v = x + e + c2
2 2
 The complete solution is
1 2 c1 2 x
y = vx = x + xe + c2 x.
2 2
d2 y dy
Example 33. Solve: x2 2
2 (x + 2x) + (x + 2)y = x3ex.
dx dx
d2 y 2
dy 1 2 y = xe
dx 2
1
x
dx
+  2
x x x
Here
2
P = 1  , Q=
1

2
and R = xex
x x x2
Since P + Qx = 0
 y = x is a part of the C.F.
Putting y = vx, so that
dy dv d2 y d2v dv
= .x+v and 2 = x+2
dx dx dx dx 2 dx
d2v dv dp dv
We get = ex or p = ex, where p =
dx 2 dx dx dx
which is a linear equation,
 dx
I.F. = e = ex
I
 pex = I ex . ex dx + c1 = x + c1
dv
 p= = xex + c1ex
dx
Integrating, we get v = xex ex + c1ex + c2
y = vx = x2ex xex + c1x ex + c2 x.
d2y
Example 34. Solve: sin2 x . = 2y given y = cot x is a solution.
dx 2
Sol. Putting y = v cot x, so that
dy dv
= cot x v cosec2 x
dx dx
d2 y d2v dv
and 2
= 2
cot x 2 cosec2 x + 2v cosec2 x cot x
dx dx dx

in the given equation, we get Linear Differential
d2v dv Higher Order
cot x sin2 x 2 =0
dx 2 dx
d2 y 2 dv NOTES
or =0
dx 2 sin x cos x dx
dp 2 dv
or = p, where p =
dx sin x cos x dx
dp 2 2 sec 2 x
or = dx = dx
p sin x cos x tan x
Integrating, we get
log p = 2 log tan x + log c  p = c1 tan2 x
dv
or = c1 tan2 x = c1(sec2 x 1)
dx
Integrating, v = c1 (tan x x) + c2
y = v cot x = c1(1 x cot x) + c2 cot x.
dy d2 y
Example 35. Solve : x
dx
y = (x 1)
dx 2
 x1 .
Sol. The given equation may be written as
d2 y x dy y
. + =x1
dx 2 x  1 dx x1
Here P + Qx = 0
dy dv d2 y d2v dv
 Putting y = vx, so that = x+v and 2 = x+2
dx dx dx dx 2 dx
d2v x 2 dv x1

We have,
dx 2
+ 
x  1

x dx
=
x
dp x 2
x1
dv
or
dx
+  
x1 x
p=
x
, where p =
dx
I.F. = e
I 
x
x1
dx  I 2
x
dx
e
I
1

1
dx 
x  1 I 2
x
dx
 x  log ( x  1)  2 log x x2 x
= e  e
x1
 p
x2 e x
x1
= I x  1 x2
x
.
x1
ex dx + c1 = I xex dx + c1 = xex ex + c1
dv x  1 ( x  1) c1 ( x  1) e x 1 1 1 e
 p=
dx
=
x

x 2

x 2 = 1 + 2  c1
x
 2
x x x

Equations Integrating, v=x + c1 ex + c2
x x
y = vx = x2 1 + c1ex + c2x = c1ex + c2x (1 + x2).
NOTES
Example 36. Solve:
d2 y dy
(x sin x + cos x)
2
x cos x + y cos x = sin x (x sin x + cos x)2.
dx dx
Sol. The given equation may be written as
d2 y x cos x dy cos x
.  y = sin x(x sin x + cos x)
dx 2 x sin x  cos x dx x sin x  cos x
Here P + Qx = 0  y = x is a part of C.F.
 Putting y = vx the equation reduces to
d 2v 2
x cos x dv sin x ( x sin x  cos x)
dx 2 +


x x sin x  cos x dx
=
x
dp 2
x cos x sin x
or
dx
+ 

x x sin x  cos x
p=
x
(x sin x + cos x)

 I.F. = e
I

2

x cos x
dx
x x sin x  cos x
=e
2 log x  log ( x sin x  cos x)

x2
( x sin x  cos x)
 p.
x2
( x sin x  cos x)
= I x sin x dx + c1 = x cos x + sin x + c1
dv 1 c
 p= = 2 ( x cos x + sin x)(x sin x + cos x) + 12 (x sin x + cos x)
dx x x
dy 1 1 1 1
dx
= sin x cos x cos 2x + 2 sin x cos x + c1
x x x
sin x  2 cos x
x
Integrating,
v=
1
2
cos2 x I 1
x
cos 2x dx + I 1
2x 2 sin 2x dx + c1 I 1
x
1
sin x  2 cos x dx
x
1 1 c1
= cos2 x sin 2x cos x + c2
2 2x x
x 1
y = vx = cos2 x sin 2x c1 cos x + c2 x.
2 2
d2 y dy
Example 37. Solve: (1 x2) +x y = x(1 x2)3/2.
dx 2 dx
d2 y x dv 1
2 + 2
 y = x(1 x2)1/2
dx 1 x dx 1  x2

Here P + Qx = 0.  y = x is a part of C.F. Linear Differential
Putting y = vx, the equation reduces to
Higher Order
d 2v x 2 dv x(1  x 2 ) 1/2
dx 2
+ 
1  x2 x
dx
=
x NOTES
dp x 2 dv
or
dx
+ 
1  x2 x
p= 1  x 2 , where p =
dx
I.F. = e
I

x 2
 dx
1  x2 x
= e
 1
2 log (1  x 2 )  2 log x

x2
1  x2
 p.
x
1  x2
= I x2 dx + c1 =
x3
3
+ c1
dv 1 2 c1
 p= = x 1 x  2 1  x2
dx 3 x
1 1
x 1  x 2  c1 (1  x 2 ) 1/ 2 . 2
=
3 x
Integrating,
1
v = (1 x2)3/2 + c1 (1 x2)1/2 
9
1
x
 c1

I dx
1  x2
+ c2
1 c1
= (1 x2)3/2 (1 x2)1/2 c1 sin1 x + c2
9 x
1
y = vx = x(1 x2)1/2 c1 {x sin1 x + 1  x 2 } + c2 x.
9
d2 y dy
Example 38. Solve cot x (1 cot x) y = ex sin x.
dx 2 dx
Sol. From the above equation, we have P + Q + 1 = 0
 y = ex is a part of C.F.
dy dv x
 Putting y = vex so that = e + v. ex
dx dx
d2 y d 2v x dv x
and 2
= 2
e +2 e + vex
dx dx dx
d2v dv
We have, + (2 cot x) = sin x
dx 2 dx
dp dv
or + (2 cot x) p = sin x, where p =
dx dx
which is linear equation.
(2  cot x ) dx e2 x
I.F. = e
I
 e 2 x  log sin x 
sin x
 p
e2x
sin x
= I e2x
sin x
1
sin x dx + c1 = e2x + c1
2

dv 1
Equations  p = sin x + c1e2x sin x
dx 2
1 c1 2x
Integrating, v= cos x + e ( 2 sin x cos x) + c2
NOTES 2 5
1 x c1 x
y = vex = e cos x e (2 sin x + cos x) + c2 ex.
2 5
d2 y dy
Example 39. Solve: (x + 2) 2 (2x + 5) + 2y = (x + 1) ex.
dx dx
Sol. The above given equation may be written as
d2 y 2 x  5 dy 2 x1 x
. + .y= e
dx 2
x2 dx x2 x2
P Q
Here, + 2 +1=0
2 2
 y = e2x is a solution of this equation.
 Putting y = ve2x, the equation reduces to
d 2v dv
(x + 2)
2
+ (2x + 3) = (x + 1) ex
dx dx
d2v 2 x  3 dv x  1 x
+ = e
dx 2 x  2 dx x2
dp 2x  3 x  1 x dv
or + p= e , where p =
dx x2 x2 dx
I.F. = e
I 2x  3
x2
dx
e
I
2


1
dx
x  2
=e
2 x  log ( x  2)

e2 x
x2
 p.
e2 x
x2
= I x1
( x  2) 2
ex dx + c1
= I %&' 1

1
x  2 ( x  2) 2
e x
dx  c1
()

*
ex
x2
+ c1
dv
 p= = ex + c1 e2x (x + 2).
dx
1 1
Integrating, v = ex c e2x(x
2 1
+ 2) c e2x
4 1
+ c2
1
= ex c
4 1
(2x + 5)e2x + c2
1
y = ve2x = ex c
4 1
(2x + 5) + c2e2x.
d2 y dy
Example 40. Solve : x 2
(2x + 1) + (x + 1)y = (x2 + x 1)e2x.
dx dx
d2 y 1 dy 1
1 2x
dx 2
2
x dx
+ 1
x
y = x  1
xe

Here 1 + P + Q = 0  y = ex is a part of C.F. Linear Differential
Putting y = vex the equation reduces to
Higher Order
d 2v 1 dv
1 xs
dx 2

x dx
= x  1
x
e
NOTES
dp 1 1
e, dv
or
dx
p = x  1
x x
x where p =
dx
I.F. = e
 I 1
x
dx 1
 e  log x 
x
 p.
1
x
= I x  1
1 x 1
x

e . dx + k
x
= I ex 
1 x
x
1

e  2 e x dx + k = ex +
x
ex
x
+k
dv
 p= = xex + ex + kx,
dx
k 2 k
Integrating, v = xex + x + c2 or v = xex + c1x2 + c2, where c1 =
2 2
y = vex = xe2x + c1x2ex + c2ex.
EXERCISE E
d2 y dy d2 y dy
1. x 2
 (3  x ) + 3y = 0 2. x 2
 (2x  1) + (x 1)y = 0
dx dx dx dx
d2 y dy 1
3. x2 x y = 0, given that y = x + is a solution.
dx 2 dx x
4. (2 + x)y (9 + 4x)y + (7 + 3x)y = 0
5. x(x cos x 2 sin x)y + (x2 + 2)sin x . y 2(x sin x + cos x)y = 0
d2 y dy d2 y dy
6. x  ( x  2) + 2y = x3. 7. x  2( x  1) + (x + 2)y = (x 2)ex.
dx 2 dx dx 2 dx
d2 y dy d2 y dy
8. 2
 (1  x ) + xy = x 9. ( x  1) 2  2( x  3) + (x + 5) = ex
dx dx dx dx
10. (x x2)y (1 2x)y + (1 3x + x2)y = (1 x)3
Answers
1. y = c1(x3 + 3x2 + 6x + 6) + c2ex 2. y = (c1 log x + c2)ex
c1  1
3. y=  c2  x   4. y = c1(2x + 3)e3x + c2ex
x  x
5. y = c1 sin x + c2x2 6. y = c1(x2 + 2x + 2) + c2ex x3
1
1 2 x 1 3 x x  x2
7. y=  x e + xex + c1x e + c2ex 8. y = c1e x  e 2 dx + c2ex + 1
2 3
1 x 1 x 1
9.
5
y =  xe  c1e ( x  1) + c2ex 10. y = c1x 2e  x + c2ex x
4 5 2

Equations REMOVAL OF THE FIRST DERIVATIVE
(Ruduction to Normal Form)
NOTES If the part of the complementary function is not obvious by inspection, it is
sometimes useful to reduce the given equation into the form in which the term
containing the first derivative is absent. For this we will change the dependent
variable in the equation.
d2 y dy
2 + P dx = Qy + R ...(1)
dx
By putting y = uv, where u is some function of x, so that
dy dv du
=u + v
dx dx dx
d2 y d 2v du dv d 2u
and 2 = u + 2 . + v
dx dx 2 dx dx dx 2
 Equation (1) reduces to
d 2v du dv + d u  P du  Qu v = R
2
u
dx 2
+ Pu  2
dx dx dx dx 2
d v P  2 du dv  1 d u  P du  Q
2

2 R
+
or
dx 2 u dx dx u dx u dx u
v=
2 ...(2)
2 du
Let us choose u such that P+ =0
u dx
du P du 1
or = u or = Pdx
dx 2 u 2
1
2
 Pdx
 u= e
 From equation (2), we have
1  u dP  P du  P du  Q"# v = R e I
d2v
1
P dx
! u 2 dx 2 dx u dx $
2
2
+
dx
d v2 1 dP  P  P u  P  P u  Q"# v = R e I 1
P dx
! 2 dx 2u 2 u 2 $
+ 
or 2
2
dx
d v 1 dP  1 P "# v = R. e I
2
2
1
P dx
! 2 dx 4 $
+ Q
or 2
2
dx
d 2v
or + Xv = Y ...(3)
dx2
where X=Q
1 dP 1 2
2 dx 4
 P and Y = Re 2

1
P dx I
The equation (3), may easily be integrated. Equation (3) is said to be the normal
form of the equation (1).
Note. Remember equation (3), and the values of u, X and Y.

SOLVED EXAMPLES Linear Differential
Higher Order
d2 y 1 dy 1 1 6 y = 0.
Example 41. Solve:
dx 2
+
x 1/3
dx

4x 2/3 4/3  2
6x x NOTES
1 1 6
Sol. Here P = x1/3, Q =   and R=0
4 x2/ 3 6 x4 / 3 x 2
On putting y = uv, the given equation reduces to the normal form
d2v
+ Xv = Y
dx 2
where u = e

1
2 I P dx
e

1
2 I x 1/3 dx
e

1 2/ 3
3
x
1 dP 1 2
X=Q  P
2 dx 4
1 1 6 1 1 1 6
=
4 x2/ 3

6x 4 /3
2
x 2
 x 4 / 3 x2/3 = 2
3 4 x
and Y = Re 2 =0
1
I P dx
 The normal form of the given equation is

d 2v 6 d2v
v=0 or x2 6v = 0 ...(1)
dx 2 x2 dx 2
which is a homogeneous linear equation.
dx
Putting x = ez, so that = ez = x
dz
dv dv dx dv
 = . =x
dz dx dz dx
d d
 x 
dx dz
d
Let D stands for , then
dz
d dv d 2v dv
x
dx
x
dx
= x2
dx 2
+x
dx
d2v d dv
or
dx 2
= x
dx
1 x x2
dx
= (D 1) Dv

 From equation (1), we get [(D 1) D 6] v = 0
or (D2 D 6) v = 0 ...(2)
Now, A.E. is m2 m 6 = 0 or m = 3, 2
Solution of (2) is v = c1e3z + c2e2z = c1x3 + c2 x2
 The solution of the given equation is
3 2/3
 x
4
y = uv = e (c1x3 + c2 x2).
d2 y dy
Example 42. Solve: 2
2(x2 + x) x2
+ (x2 + 2x + 2)y = 0.
dx dx
d2 y 1 dy 2 2 y = 0
dx 2
2 1
x dx
+ 1  2
x x
Ordinary Differential 1 2 2
Equations Here P = 2 1 
x
,Q=1+  2
x x
and R=0
d2v
Putting y = uv, the normal form is + Xv + Y
NOTES dx 2
where u= e
 1
2 I Pdx
=e
I
1+

1
dx
x
 e x  log x = xex
1 dP 1 2 2 2 1 2 1 1 2
X=Q  P = 1 +  2  . 2  4 1
2 dx 4 x x 2 x 4 x =0
1 Pdx
Y = Re2
I
and =0
2
d v
 The normal form is =0
dx 2
Integrating, v = c1x + c2
 The solution of the given equation is y = uv = xex (c1x + c2).
d2 y dy
Example 43. Solve 2
2 tan x + 5y = ex sec x.
dx dx
Sol. Here P = 2 tan x, Q = 5 and R = ex sec x
Putting y = uv is the given equation, the equation reduces to
d 2v  1 Pdx
+ Xv = Y, where u = e
I
2
2
dx
log sec x
tan x dx
=e =e I
= sec x
1 dP 1 2
X=Q  P
2 dx 4
1 1
=5+ 2 sec2 x . 4 tan2 x = 6
2 4
1 Pdx
Y = Re2 = ex
I
d2v
 The reduced equation is + 6v = ex
dx 2
where C.F. = c1 cos 6 x + c2 sin 6x
1 ex
and P.I. = ex =
D2  6 7
ex
 v = c1 cos 6 x + c2 sin 6x +
7
1 x .
y = uv = sec x c1 cos 6 x  c2 sin 6 x 
7
e

d2 y dy x2
Example 44. Solve 2
4x + (4x2 3)y = e .
dx dx
x2
Sol. Here P = 4x, Q = 4x 3, R = e
2
Putting y = uv, the normal form is,

d 2v  1 Pdx x2
+ Xv = Y, where u = e 2 = e
I
2
dx
1 dP 1 1 1
X=Q  P2 = 4x2 3 ( 4) (16x2) = 1
2 dx 4 2 4
 1 Pdx
Y = Re
I
and 2
=1
d 2v Linear Differential
 The normal form is v=1 Equations of Second and
dx 2 Higher Order
where C.F. = c1 ex + c2 ex
1 NOTES
and P.I. = 2 . 1 = (1 D2)1. 1 = 1
D 1
 v = c1ex + c2ex 1
2
y = uv = e x (c1ex + c2ex 1).
d2 y dy x2
Example 45. Solve 2
4x + (4x2 1) y = 3 e sin 2x.
dx dx
x2
Sol. Here P = 4x, Q = 4x2 1 and R = 3 e sin 2x
d2v
Putting y = uv, the equation reduces to + Xv = Y
dx 2
1 2
 Pdx x
where u= e =e
I
2
1 dP 1 2 1 1
X=Q  P = 4x2 1 ( 4) 16x2 = 1
2 dx 4 2 4
1 Pdx
Y = Re 2
I
= 3 sin 2x.
d2v
 The reduced equation is + v = 3 sin 2x
dx 2
whose C.F. = c1 cos x + c2 sin x
1 3
and P.I. = 2 ( 3 sin 2x) = sin 2x = sin 2x
D 1  22  1
 v = c1 cos x + c2 sin x + sin 2x
2
x
y = uv = e (c1 cos x + c2 sin x + sin 2x).
d2 y 2 dy 2
Example 46. Solve
dx 2

x dx
+ 1  2 y = xex.
x
2 2
Sol. Here P = , Q = 1 + 2 and R = xex
x x
d2v
Putting y = uv, the normal form is + Xv = Y
dx 2
 1 Pdx 1/ x dx log x
where u= e =e =e =x
I I
2
1 dP 1 2 1 Pdx
 P = 1 and Y = R e 2
I
X=Q
2 dx 4
= xex e  dx/x = xex e  log x = ex
I
 The normal form of the given equation is

d2v
+ v = ex whose C.F. = c1 cos x + c2 sin x
dx 2
1 ex
and P.I. = 2 ex =
D 1 2

 v = c1 cos x + c2 sin x + 2
ex
Equations
1
y = uv = x(c1 cos x + c2 sin x + 2
ex).
NOTES Example 47. Solve
d2 y dy
x2 (log x)2 2 2x log x + [2 + log x 2(log x)2] y = (log x)3 x2.
dx dx
d2 y 2 dy
 2
2
 2
1 2
 2 y = log x
"#
2
dx x log x dx x (log x) 2
!
x log x x $
2 2 1 2
Here, P= ,Q= 2  
x log x x (log x) 2 x 2 log x x 2
and R = log x
Putting y = uv, the given equation is transformed to
d 2v
+ Xv = Y
dx 2
where u = e

1
2 I P dx
e
I 1
x log x
dx
 e (log log x)  log x
1 dP 1 2
X=Q  P
2 dx 4
2 1 2 1 2 (log x  1) 1 4 2
= 2 2
 2
2 2
2
= 2
2 4
x (log x) x log x x 2 x (log x) ( x log x) x
1 Pdx
Y = Re
I
and =12
 The transformed equation is

d2v 2
2
2 v=1
dx x
2
d v
x2 2v = x2 ...(1)
dx 2
Putting x = ex, we get {D(D 1) 2} v = e2x or (D2 D 2) v = e2x
A.E. is m2 m 2 = 0, m = 2, 1
C.F. = c1e2z + c2ez = c1x2 + c2x1
1 1
and P.I. = 2 e2z = e2z
D D2 (D  2)(D  1)
1 1 1 1
= (e2z . 1) = e2z .1
3 D2 3 D22
1 2z 1 1 1
=
3
e
D 3 3
. 1 = ze2z = x2 log x
 The solution of equation (1) is

1 2
v = c1x2 + c2x1 + x log x
3

 The solution of the given equation is Linear Differential
1 Higher Order
y = uv = (log x) (c1x2 + c2 x1) + (x log x)2.
3
d2 y dy
Example 48. Solve + 2x + (x2 + 1)y = x3 + 3x. NOTES
dx 2 dx
Sol. Here P = 2x, Q = x2 + 1 and R = x3 + 3x
d 2v
Putting y = uv, the equation is transformed to + Xv = y,
dx 2
where u= e

1
e x /2
2 I P dx 2
1 dP 1 2
X=Q  P =0
2 dx 4
and Y = Re

1
2 I P dx
= (x3 + 3x) e x
2/ 3
d2v 2/3
 The transformed equation is, 2
= (x3 + 3x) e x
dx
Integrating,
dv
dx
= I x3 e x
2
/2
dx + 3 I xex
2
/2
dx + c1
= I x2 (x e x
2
/2
) dx + 3 e x
2
/2
+ c1
= x2 e  x
2
2
/2
2 I
2
xex
2
/2
2
dx + 3 e x
2
/2
+ c1
2
= x2 e x /2
2ex /2
+ 3 ex /2
+ c1 = (x2 + 1) e x /2
+ c1
Integrating again = I x2 e x
2
/2
dx + I ex
2
/2
dx + c1 x + c2
=
I x (x e x
2
/2
) dx + I ex
2
/2
dx + c1x + c2 = x e x
2
/2
+ c1x + c2
2
y = uv = x + (c1x + c2) e x /2
d2 y 1 dy 1
Example 49. Solve + ( 8 + x + x) y = 0.
dx 2
x dx 4x 2
1 1
Sol. Here, P= ,Q ( 8 + x + x) and R = 0
x 4x2
d 2v
+ Xv = Y
dx 2
where u= e

1
2 I P dx
e

1
2 I 1
x
dx
= e
x
1 dP 1 2 2
X=Q  P = 2
2 dx 4 x
Y = Re
1
2 I P dx
=0

Ordinary Differential  The transformed equation is
Equations
d2v 2 d 2v
2
2v=0 or x2 2v = 0
dx x dx 2
which is a homogeneous linear equation.
NOTES
{D(D 1) 2} v = 0
(D2 D 2) v = 0
A.E. is m2 m 2 = 0
m = 2, 1
 v = c1e2z + c2ez = c1x2 + c2 x1.
x
y = uv = e . (c1x2 + c2x1).
d2 y dy  1 x2
Example 50. Solve 2 + 2x + (x2 + 5) y = x e 2 .
dx dx
 1 x2
Sol. Here P = 2x, Q = x2 + 5, R = x e 2 .
d 2v
+ Xv = Y
dx 2
 1 Pdx  1 2 x dx 2
where u= e e e x /2
I I
2 2
1 dP 1 2
X=Q  P = x2 + 5 1 x2 = 4
2 dx 4
1 Pdx
and Y = Re 2 = x.
I

d 2v
+ 4v = x
dx 2
A.E. is m2 + 4 = 0  m =  2i
C.F. = c1 cos (2x + c2)
11 D2 1
x
P.I. = 2
D 4
x=
4
1
4 x=
4
2
 The solution is y = uv = e  x /2
[c1 cos (2x + c2) + 1
4 x].
EXERCISE F
d2 y 2 dy 2 2 y = 0 d2 y dy
1.
dx2

x

+ a  2
dx x 2. (x3 2x2)
dx2
+ 2x2
dx
12(x 2) y = 0
d2 y dy d2 y 2 dy
3. 2 tan x + 5y = 0 4. + + n2y = 0
dx 2 dx dx2 x dx
d2 y dy d  dy  dy
5. 2
2bx + b2x2y = 0 6. x x  y   2x + 2y + x2y = 0
dx dx dx  dx  dx
d  2 dy   d2 y   dy 
7.  cos x  + y cos2x = 0 8.  2  y  cot x  2   y tan x  = sec x
dx  dx   dx   dx 

1 x2 Linear Differential
d2 y dy ( x2  2x ) d2 y dy 2

9.  2 x + (x2 + 2)y = e 2 10.  2x + (x2 8)y = x e 2
dx 2 dx dx 2
dx Higher Order
11. x y 2x(1 + x)y + 2(1 + x)y = x3 (x > 0)
2 12. y (2 cot x)y + (1 + 2 cot2x)y = 0
2
13. y 4xy + (4x2 1)y = ex (5 3 cos 2x) NOTES
Answers
1. y = xc1 cos (ax + c2) 2. y = (c1x4 + c2 x3)/(x 2)
1
3. y = sec x(c1 cos 6 x + c2 sin 6x) 4. y = c cos (nx + c2)
x 1
1 bx 2
5. y = c1 e 2 cos ( 6 x + c2) 6. y = x(c1 cos x + c2 sin x)
1
7. 
y = sec x c1 cos 2x  c2 sin 2x  8. y =
2
(sin x  x cos x ) + (c1x + c2)cos x
x2 1 2 x2
1 (x  3x 2 
  1
2 x ) 
3 x
9. y= ez c1 cos 3x  c2 sin 3x  e 2 10. y = e 2
c1e  c2e   x 2  
4  9 9 
x2
11. y = (c1e2x + c2)x 12. y = (c1 + c2x) sin x
2
2
13. y = ex (c1 cos x + c2 sin x + 5 + cos 2x)
TRANSFORMATION OF THE EQUATION BY CHANGING

THE INDEPENDENT VARIABLE
Sometimes the equation is transformed to an integrable form by changing the

independent variable.
Let the equation be
d2 y dy
2 +P + Qy = R ...(1)
dx dx
Let the independent variable be changed from x to z, where z is a function of x.
dy dy dz d2 y d dy = d dy . dz

dx
= .
dz dx
and
dx 2
=
dx dx dx dz dx
d 2 y dz dy d 2 z 2
dz 2 dx
= + .
dx dx 2
Substituting in equation (1), we have
dz 2
d2 y
d2 z dz dy
dx dz 2
+
dx 2
P
dx dz
+ Qy = R
d2 y dy
or 2 + P1 + Q1y = R1 ...(2)
dz dz
d2z dz
P
dx 2
dx Q R
where P1 =
dz
2
, Q1 =
dz
2 and R1 =

dz
2

dx dx dx
P1, Q1, R1 are functions of x but may be expressed as functions of z by the given
relation between z and x.

Ordinary Differential dy
Equations We choose z to make the co-efficient of zero, i.e., P1 = 0
dx
d2z
d2 z dz dx 2
NOTES i.e., +P =0 or =P
dx 2 dx dz
I
dx
dz dz  P dx
Integrating, log= Pdx or = e
I
dx dx
Then the equation (2) is reduced to
d2 y
+ Q1y = R1
dz 2
which can be solved easily provided Q1 comes out to be a constant or a constant multi-
1
plied by .
z2
Again, if we choose z such that
Q
Q1 =
dz 2 = a2 (constant)
dx
dz 2
dz
i.e., a2
dx =Q or a
dx
= Q
 az =
Then equation (2) is reduced to
I Q dx
d2 y
dx
+ a2y = R1
2 + P1
dz dz
which can be solved easily provided P1 comes out to be a constant.
Note. It is advised to remember the equation (2) and the values of P1, Q1 and R1.
SOLVED EXAMPLES
d2 y 2 dy a 2
Example 51. Solve : +  y = 0.
dx 2 x dx x 4
Sol. Choose z such that
dz 2
a2
dx =Q=
x4
dz a a
 = 2 ,z=
dx x x
a
Changing the independent variable from x to z by the relation z = , we get
x
d2 y dy
+ P1 + Q1y = R1
dz 2 dz
d2z dz 2a 2 a
dx 2
P
dx

x 3

x
 2
x
where P=

dz
2
a
2 =0

dx
 2
x
Q R Linear Differential
Q1 = and R1 = =0
dz 2
dz 2 Equations of Second and
dx dx Higher Order
d2 y NOTES
 The transformed equation is +y=0
dz 2
a a
 y = c1 cos z + c2 sin z = c1 cos + c2 sin .
x x
d2 y dy
Example 52. Solve : + 4x3y = x5.
x
dx 2 dx
d2 y 1 dy
2 + 4x2y = x4
dx x dx
dz
dz 2
Choosing z, such that
dx
dx
= 2x 
= Q = 4x2
z = x2 or
Now changing the independent variable from x to z by the relation z = x2, we get
d2 y dy
2 + P 1 dz + Q1 y = R1
dz
d2 y dz 1
where P1 = dx 2
P
dx
=
2 
x
2x
=0

dz
2
(2 x) 2
dx
Q R x4 x2 1
Q1 =
dz 2 =1 and R1 =
dz 2 =
(2 x) 2 =
4
 z
4
dx dx
The given equation is transformed to
d2 y 1
2 +y= z
dz 4
whose C.F. = c1 cos z + c2 sin z
1 1 1 1 1
and P.I. = . 2 z = (1 + D2)1 z = 4
(1 D2 + D4 .....) z = 4
z
4 D 1 4
1
 y = c1 cos z + c2 sin z + 4
z
1
or y = c1 cos x2 + c2 sin x2 + 4
x2.
d2 y dy
Example 53. Solve : 2 + cot x + 4y cosec2 x = 0.
dx dx
Sol. Choosing z, such that
dz 2
dx = 4 cosec2 x, so that
dz x
= 2 cosec x or z = 2 log tan .
dx 2

Ordinary Differential Now changing the independent variable from x to z by the relation
Equations
x
z = 2 log tan , we get
2
d2 y dy
NOTES 2 + P1 + Q1y = R1
dz dz
d2z dz
P
dx 2 dx 2 cosec x cot x  2 cot x cosec x
where P1 =

dz
2 =
(2 cosec x) 2
=0

dx
Q R
Q1 =

dz
2 = 1 and R1 =

dz
2 =0

dx
dx
 The given equation is transformed to
d2 y
+y=0
dz 2
 y = c1 cos z + c2 sin z or y = k1 cos (z + k2)
1
or

y = k1 cos 2 log tan
2
x  k2 .
d2 y dy
Example 54. Solve: (1 + x2)2 2
2 + 2x(1 + x ) dx + 4y = 0.
dx
d2 y 2 x dy 4
2
+ + y=0
dx 1  x 2 dx (1  x) 2
dz 2
4 dz x
dx =Q=
1  x2

dx
=
1  x2
or z = 2 tan1 x
Changing the independent variable from x to z by the relation z = 2 tan1 x,
we get
d2 y dy
2 + P1 + Q1y = R1
dz dz
d z 2
dz  4x 2x 2
P 2 2
 .
2
dx  (1  x ) 1  x 1  x2
2
where P1 = dx =0
dz
2
4
dx (1  x 2 ) 2
Q R
Q1 =
dz
2 = 1
and R1 =
dz
2 = 0

dx dx
d2 y
 The transformed equation is +y=0
dx 2
 y = c1 cos z + c2 sin z
or y = c1 cos (2 tan1 x) + c2 sin (2 tan1 x)

2x +c Linear Differential
= c1 cos tan 1 1  x2
2
sin tan 1
2x
1  x2
Higher Order
1  x2 2x
= c1 2
 c2
1 x 1  x2 NOTES
or y(1 + x2) = c1 (1 x2) + 2c2 x.
d2y dy 1
Example 55. Solve x6 2 + 3x
5 + a2y = 2 .
dx dx x
d2 y 3 dy a2 1
+ + 6 y= 8
dx 2 x dx x x
dz 2
a2
dx =Q=
x6
dz a a
 = 3 or z= .
dx x 2 x2
a
Changing the independent variable from x to z by the relation z = , we get
2x2
d2 y dy
2 + P 1 dz + Q1 y = R1
dz
d2z dz
P
2
dx Q
where P1 = dx = 0, Q1 = =1

dz
2
dz 2

dx dx
R 1 2z
and R1 = = 2 2  3

dz
2
a x a
dx
d2 y 2z
+y= 3 whose C.F. = c1 cos z + c2 sin z
dz 2 a
a  c a
= c1 cos 
2x 2
2x 2 sin 
2
a a
= c1 cos + c2 sin
2x2 2 x2
1
2z 2
and P.I. = 2
D 1 a
 3 = 3 (1 + D2)1 z
a
2 2z 1
= 3 (1 D2 + D4 ....) z = 3 = 2 2
a a a x
a a 1
 y= c1 cos 2
+ c2 sin 2 + 2 2
.
2x 2x a x
d2 y
dy
Example 56. Solve : cos x 2y cos3 x = 2 cos5 x.
2
+ sin x
dx dx
d2 y dy 2 5
2 + tan x dx (2 cos x) y = 2 cos x
dx

dz 2
Equations Choosing z such that
dx = cos2 x
dz
 = cos x or z = sin x
dx
NOTES
Changing the independent variable from x to z by the relation z = sin x, we have
d2 y dy
2 + P1 dz + Q1 y = R 1
dz
2
d z dz
P
2
dx  sin x  tan x cos x 1
where P1 = dx  = 0, Q1 = =2

dz
2
cos 2 x dz 2

dx dx
R
and R1 = = 2 cos2 x = 2(1 z2)

dz
2

dx
d2 y
2y = 2(1 z2)
dz 2
whose C.F. = c1 e 2 z  c2 e  2z
1 D
= 1 
2 1
and P.I. =
D2  2
. 2(1 z2)
2 (1 z2)
D 2
D4
= 1  ..... (1 z )
2 
4 2
1
= (1 z2) + 2
(+2) = z2
 y = c1 e
2z
 c2 e  2z
+ z2
Required solution is = c1 e
2 sin x
 c2 e 2 sin x
+ sin2 x.
d2y 1 dy
Example 57. Solve:
dx 2 + 1
x dx
+ 4x2e2xy = 4(x2 + x3) e3x.
Sol. Choosing z, such that

dz 2
dz
dx = 4x2 e2x
dx

= 2xex or z = 2(x + 1) ex.
Changing the independent variable from x to z by the relation,

x = 2(x + 1) ex, we have
d2 y dy
2 + P1 + Q1y = R1
dz dx
d2z dz 1
where P1 = dx
2
P
dx

2(1  x) e  x  1 
x
2 xe  x

dz
2
4 x 2 e 2 x

dx
Q R 1
Q1 =

dz
2 =1 and R1 =

dz
2 = (1 + x) ex = 2
z

dx
dx

 Transformed equation is Linear Differential
d2 y 1 Higher Order
2 + y = 2 z
dz
whose C.F. = c1 cos z + c2 sinz NOTES
1 1 1
and P.I. = 2
D 1 2 2
 z   (1 + D2)1z

1 1
= 2
(1 D2 + D4 .....) z = 2
z
1
 y = c1 cos z + c2 sin z 2
z
y = c1 cos {2(x + 1)ex} c2 sin {2(x + 1) ex} + (x + 1) ex.
d2 y dy
Example 58. Solve : x 2
4x3y = 8x3 sin x2.
dx dx
d2 y 1 dy
4x2y = 8x2 sin x2
dx 2 x dx
dz 2
dz
dx = 4x2 or
dx
= 2x  z = x2
Changing the independent variable from x to z by the relation z = x2, we get

d2 y dy
+ P1 + Q1y = R1
dx 2 dx
d2z dz
P
dx 2
dx Q
where P1 =
dz
2 = 0, Q1 =
dz
2 = 1

dx dx
R
and R1 = = 2 sin x2 = 2 sin z
(dz/ dx) 2
d2y/dz2 y = 2 sin z
whose C.F. = c1ez + c2ez
1 1
and P.I. = (2 sin z) = 2 sin z = sin z
D2  1 1 1
 y = c1ez + c2ez sin z
2 2
y = c1 e x + c2 e  x sin x2.
EXERCISE G
d2 y dy d2 y dy
1. x 2 + (4x2 1) + 4x3y = 2x3 2. x4 + 2x3 + n2 y = 0
dx dx dx2 dx
Ordinary Differential d2 y dy d2 y dy
Equations 3. 2 cot x y sin2 x = 0 4. 2 + tan x + y cos2 x = 0
dx dx dx dx
d2 y dy
5. (1 + x)2 2  (1  x ) + y = 4 cos [log(1 + x)]
dx dx
NOTES
d2 y dy
6.  (3 sin x  cot x ) + 2y sin2 x = ecos x sin2x
dx 2 dx
d2 y dy
7.  cot x y sin2x = cos x cos3x
dx 2 dx
d2 y dy
8. 2
 (tan x  3 cos x ) + 2y cos2 x = cos4 x
dx dx
x
9. y (1 + 4ex)y + 3e2xy = e2(x + e ) 10. y (8e2x + 2)y + 4e4xy = e6x
Answers
x 2 1 n  x
1. y = e (c1x2 + c) + 2. y = c cos
2 x
3. y = c1 e  cos x  c2 ecos x 4. y = c1 sin (sin x + c2)
5. y = c1 cos log(1 + x) + c2sin log(1 + x) + 2 log(1 + x) sin log(1 + x)
METHOD OF VARIATION OF PARAMETERS
Here we shall explain the method of finding the complete primitive of a linear
equation whose complimentary function is known.
Let y = A(x) + B(x) be the complimentary function of the linear equation of
second order
d2 y dy
2 + P + Qy = R ...(1)
dx dx
where A and B are constants and (x) and (x) are functions of x
Since, y = A(x) + B (x)
d2 y dy
Satisfies the equation 2 +P + Qy = 0
dx dx
 [A(x) + B(x)] + P [A(x) + B(x)] + Q [A (x) + B (x)] = 0
or A [ (x) + P (x) + Q(x)] + B [ (x) + P(x) + Q(x)] = 0
  (x) + P (x) + Q(x) = 0 ...(2)
and  (x) + P (x) + Q (x) = 0 ...(3)
Now let us assume that
y = A(x) + B(x) ...(4)
is the complete primitive of (1) where A and B are not constants but functions of x to be
so chosen that (1) will be satisfied.
dy dA dB
 = A (x) + B (x) +  (x) +  (x)
dx dx dx
Let A and B satsify the equation,
dA dB
(x) + (x) =0 ...(5)
dx dx

dy Linear Differential
 = A (x) + B (x) Equations of Second and
dx
Higher Order
d2 y dA dB
and = A (x) + B (x) +  (x) +  (x).
dx 2 dx dx
Substituting in (1), we have
NOTES
A(x)  B (x)  dA (x)  dB  (x)"#

! dx dx $
+ P [A (x) + B (x)] + Q[A (x) + B (x)] = R
or A[ (x) + P (x) + Q (x)] + B[ (x) + P (x) + Q (x)]
dA dB
+  (x) +  (x) =R
dx dx
Since the co-efficient of A and B are zero [by (2) and (3),] we have
dA dB
 (x) +  (x) =R ...(6)
dx dx
dA dB
Solving (5) and (6) for and , we get
dx dx
dA  R ( x )  R ( x )
= 
dx ( x ) ( x )  ( x ) ( x ) W
dB R ( x ) R ( x )
and = 
dx ( x ) ( x )  ( x ) ( x ) W
( x ) ( x )
where W= is called the Wronskian of (x) and (x).
( x ) ( x )
R ( x ) R ( x )
Integrating (7), A = 
W
dx  c1 , B =  W
dx  c2
Substituting these values of A and B in (4), we get the complete solution of (1).
Note 1. As the solution is obtained by varying the arbitrary constants of the
complementary function, the method is known as variation of parameters.
2. Method of variation of parameters is to be used if instructed to do so.
SOLVED EXAMPLES
Example 59. Apply the method of variation of parameters to solve:

dy d2 y
x2 x
 y = x2ex.
2
dx dx
Sol. The given equation in standard form is
d2 y 1 dy 1
  y = ex ...(1)
dx 2 x dx x 2
1 1
Here P= , Q = 2 , R = ex
x x
Now to find the C.F. of (1) i.e., the solution of the equation
d2 y 1 dy 1
2
  2 y =0
dx x dx x
dy d2 y
Equations or x x2 y =0 ...(2)
dx 2 dx
which is a homogeneous equation, put x = ez so that z = log x.
NOTES d
Let D  , then equation (2) becomes
dz
[D(D 1) + D 1]y = 0 or (D2 1)y = 0
Its A.E. is m2 1 = 0 so that m = ± 1
 Solution of (2) is y = c1ez + c2ez = c1x + c2x1
1
 Parts of C.F. of (1) are (x) = x and (x) =
x
Wronskian of (x) and (x) is
1
x
( x ) ( x ) x 1 1 2
W=  =   
( x )  ( x ) 1 x x x
1 
x2
B
Let y = A(x) + B(x) = Ax + be the complete solution of (1) where A and B are
x
functions of x determined as follows:
1 x
.e 1 x 1 x
R (x )
A =  dx  c1    x dx  c1 =  e dx  c1  e  c1
W 2 2 2

x
R (x ) ex . x
and B=  W
dx  c2  
2
dx  c2

x
1 2 x 1 1
x e dx  c2 =  x 2  2x  2  e x  c2 =  x 2 e x  ( x  1) e x  c2
2
= 
2 2
Hence, the complete solution of (1) is
y = A(x) + B(x)
1   1  1
=  e x  c1  x   x 2e x  ( x  1)e x  c2  .
2   2  x
c2 1 x 1 x  x 1 x
= c1 x   xe  xe   e
x 2 2  x 
 c21
or  1   e x
y = c1 x 
x  x
where c1 and c2 are arbitrary constants of integration.
Example 60. Using method of variation of parameters, solve:
d2 y
dy
x2  2x
 12 y = x3 log x.
2
dx dx
d2 y 2 dy 12
  y = x log x ...(1)
dx 2 x dx x 2

2 12 Linear Differential
Here P= , Q =  2 , R = x log x Equations of Second and
x x Higher Order
Now to find the C.F. of (1) i.e., the solution of the equation
d2 y 2 dy 12 NOTES
2
  y =0
dx x dx x 2
d2 y
dy
or x2 2
 2x
 12 y = 0 ...(2)
dx dx
which is a homogeneous equation, put x = ez so that z = log x.
d
Let D  , the equation (2) becomes
dz
[D(D 1) + 2D 12]y = 0 or (D2 + D 12)y = 0
Its A.E. is m2 + m 12 = 0 or (m + 4) (m 3) = 0
 m = 3, 4
 Solution of (2) is y = c1e3z + c2e4z = c1x3 + c2x4
1
 Parts of C.F. of (1) are (x) = x3 and (x) =
x4
1
x3
( x ) ( x ) x4 4 3 7
W= = =    .
( x )  ( x ) 4 x 2
x 2
x2
3x 2  5
x
B
Let y = A(x) + B(x) = Ax3 + be the complete solution of (1) where A and B
x4
are functions of x determined as follows:
1
x log x .
R  (x ) x 4 dx  c
A=  dx  c1 =   1
W 7

x2
1 1 1 (log x )2 1
=
7  (log x ) . dx  c1 = .
x 7 2
 c1 
14
(log x )2  c1
R (x ) x log x . x 3
and B=  W
dx  c2 =  7
dx  c2
 2
x
1 1  x7 1 x7 
 .
7
=  (log x ) . x 6 dx  c2 =  (log x ) . dx   c2
7  7 x 7 
x7 1 x7 x7  1
=  log x  .  c2    log x    c2
49 49 7 49  7
y = A (x) + B (x)
1   x7  1  1
=  (log x )2  c1  x 3     log x    c2   4
14   49  7  x
c 3 3
3 x x  1
= c1 x  24  (log x )2    log x   .
x 14 49  7

Ordinary Differential Example 61. Apply the method of variation of parameters to solve:
Equations
d2 y
dy
x2  2x (1 + x)
+ 2(1 + x)y = x3.
2
dx dx
NOTES
d2 y
2(1  x ) dy 2(1  x )
2
  y =x ...(1)
dx x dx x2
2(1  x ) 2(1  x )
Here, P=  , Q= ,R=x
x x2
2(1  x ) 2(1  x )
Since P + Qx =  =0
x x
Now to find the C.F. of (1), i.e., the solution of the equation
d2 y 2(1  x ) dy 2(1  x )
  y =0 ...(2)
dx 2 x dx x2
dy dv d2 y d2v dv
Put y = vx so that= x+v and = x2
dx dx dx 2 dx 2 dx
d 2v dv 2(1  x )  dv  2(1  x )
x 2
2  x  v  · vx = 0
dx dx x  dx  x2
2
d v dv dv 2(1  x ) 2(1  x )
or x 2
2  2(1  x )  v + v =0
dx dx dx x x
d2v dv d2v dv
or x 2
 2x
= 0 or 2
2 =0
dx dx dx dx
Its A.E. is m2 2m = 0 so that m = 0, 2
 v = c1e0x + c2e2x = c1 + c2e2x
 Solution of (2) is y = vx = c1x + c2x e2x
 Parts of C.F. of (1) are (x) = x and (x) = xe2x
( x ) ( x ) x xe2 x
W= = = 2x2e2x
( x )  ( x ) 1 (1  2x )e2 x
Let y = A (x) + B(x) = Ax + Bxe2x be the complete solution of (1) where A and
B are functions of x determined as follows:
R (x ) x . xe 2x
A=  dx  c1 =  2x 2 e2x dx  c1
W
1 x
2
= 
dx  c1 =   c1
2
R (x ) x.x 1 2 x 1
and B=  dx  c2 =  2 2x dx  c2 =  e dx  c2 =  e 2x  c2
W 2x e 2 4
 x   1 
y = A (x) + B(x) =    c1  x    e 2 x  c2  xe 2x
 2   4 
x2 x
or y = c1x + c2xe2x

2 4
Example 62. Using method of variation of parameters, solve Linear Differential
2 Equations of Second and
dy d y
(1 x)  y = (1 x)2.
2
+x Higher Order
dx dx
NOTES
d2 y x dy 1
2
 .  y =1x ...(1)
dx 1  x dx 1  x
x 1
Here, P= , Q=  , R=1x
1x 1x
x x
Since, P + Qx =  =0
1x 1x
Now to find the C.F. of (1), i.e., the solution of the equation
d2 y x dy 1
2
  y =0 ...(2)
dx 1  x dx 1  x
dy dv d2 y d2v dv
Put y = vx so that = x v and 2
= 2
x2
dx dx dx dx dx
d 2v dv x  dv  1
x 2
2  x  v  .vx = 0
dx dx 1  x  dx  1x
d 2v dv x 2 dv
or x 2  =0
dx 2 dx 1  x dx
d2v2 x  dv
or  
2  =0
dx  x 1  x  dx
dp  2 1  dv
or    1  p = 0 where p =
dx  x 1  x  dx
dp  1 2
or = 1    dx
p  1x x
Integrating, log p = x + log (1 x) 2log x + log c1
c1 (1  x )e x
= log
x2
c1 (1  x )ex dv c1 (1  x )e x
or p= or =
x2 dx x2
 1 1
or dv = c1  2   e x dx Form [ f ( x )  f ( x )] e x
x x
 1
Integrating, v = c1    e x  c2 | f(x)ex
 x
 Solution of (2) is y = vx = c1ex + c2x
 Parts of C.F. of (1) are (x) = ex and (x) = x
( x ) ( x )  ex x
Wronskian of (x) and (x) is W= = = (x 1)ex
( x )  x  e x
1

Ordinary Differential Let y = A(x) + B(x) = A( ex) + Bx be the complete solution of (1) where A and
Equations B are functions of x determined as follows:
R ( x ) (1  x )x
A=  dx  c1 =   dx  c1   xe  x dx  c1
NOTES W ( x  1)e x
= x (  e  x )  1 . (  e  x ) dx  c1 = xex ex + c1
= (x + 1)ex + c1
R (x ) (1  x ) (  e x )
and B=  W
dx  c2 =  ( x  1)e x
dx  c2   dx  c2
= x + c2
y = A(x) + B(x)
= [ (x + 1)ex + c1] ( ex) + (x + c2)x
or y = c1ex + c2x + x + 1 + x2
EXERCISE H
Using method of variation of parameters, solve the following differential equations:
d2 y dy d2 y dy 1
1. x2 2
 4x  2 y  ex 2. x 2  4x  2y  x 2  2
dx dx dx 2 dx x
d2y dy d2 y dy
3. x2 2
x  9 y  48 x 5 4. x 2 x  y  x log x
dx dx dx 2 dx
d2y dy d2 y dy
5. x2 2
 4x  6 y = sin (log x) 6. x 2 x  x 3ex
dx dx dx 2 dx
d2 y dy d2 y dy
7. x2 x  y  x 2 log x 8. (1  x ) x  y  2(1  x )2 e  x
dx 2
dx dx 2 dx
Answers
c1 c2 1 c1 c2 1 2 1
1. y=  2
 2
ex 2. y =   x  2 log x
x x x x x2 12 x
x
3. y = c1 x3 + c2x3 + 3x5 4. y = c1 x log x + c2x + (log x )3
6
1
5. y = c1 x2 + c2x3 + [(sin (log x) + cos (log x)]
10
c2 1 3 4
6. y = c1 + c2 x2 + (x 1)ex 7. y = c1x   x log x  x 2
x 3 9
1 
8. y = c1x + c2ex +   x  e  x
2 

UNIT IV Power Series Solutions
NOTES
5. POWER SERIES SOLUTIONS
STRUCTURE
Introduction
Definitions
Power series solution, when x = 0 is an ordinary point of the equation
d2 y dy
2
 P(x ) + Q(x)y = 0
dx dx
Frobenius Method: Series solution when x = 0 is a regular singular point of
the differential equation
d2 y dy
2
 P(x ) + Q(x)y = 0
dx dx
INTRODUCTION
The solution of ordinary linear differential equations of second order with variable
coefficients in the form of an infinite convergent series is called solution in series or
integration in series.
The series solution of certain differential equations give rise to special functions
such as Bessels function, Legendres polynomials, Laguerres polynomial, Hermites
polynomial, Chebyshev polynomials. These special functions have wide applications
in engineering.
In this unit, we will discuss methods of solution of second order linear differential
equations with variable coefficients in series along with Bessels function, Legendres
polynomial and their properties.
DEFINITIONS
Power Series
An infinite series of the form


n0
an (x x0)n = a0 + a1 (x x0) + a2 (x x0)2 + ...
is called a power series in ascending powers of x x0.

Ordinary Differential In particular, a power series in ascending powers of x is an infinite series
Equations


n0
an xn = a0 + a1x + a2x2 + ...
NOTES 
xn x 2 x 3 ...
e.g., ex = 
n0
n!
 1 x  
2! 3!
+
Analytic Function
A function f(x) defined on an interval containing the point x = x0 is called analytic at x0

f ( n) ( x0 )
if its Taylor series 
n0
n!
(x x0)n exists and converges to f(x) for all x in the
interval of convergence of Taylors series.

Note 1. A rational function is analytic except at those values of x at which its denominator
x
is zero. e.g., Rational function is analytic everywhere except at x = 2 and x = 3.
x2  5 x  6
Note 2. All polynomial functions ex, sin x, cos x, sinh x and cosh x are analytic everywhere.
Ordinary Point
A point x = x0 is called an ordinary point of the equation
d2 y dy
 P( x) + Q(x) y = 0 ...(1)
dx 2 dx
if both the functions P(x) and Q (x) are analytic at x = x0.
Regular and Irregular Singular Points

If the point x = x0 is not an ordinary point of the differential equation (1), then it is
called a singular point of equation (1). There are two types of singular points:
(i) Regular singular point.
(ii) Irregular singular point.
A singular point x = x0 of the differential equation (1) is called a regular singular
point of (1) if both (x x0) P(x) and (x x0)2 Q(x) are analytic at x = x0.
A singular point which is not regular is called an irregular singular point.
Remark 1. When x = 0 is an ordinary point of equation (1), its every solution can be
expressed as a series of the form

y = a0 + a1x + a2x2 + a3x3 + ... = a
n0
n xn .
Remark 2. When x = 0 is a regular singular point of equation (1), at least one of its
solution can be expressed as

y = a0xm + a1xm+1 + a2xm+2 + ... = xm (a0 + a1x + a2x2 + ...) = a
n0
n xm  n
where m may be a positive or negative integer or a fraction.

Remark 3. If x = 0 is an irregular singular point of equation (1), then discussion of
solution of the equation is beyond the scope of this book.

Power Series Solutions
POWER SERIES SOLUTION, WHEN x = 0 IS AN
ORDINARY POINT OF THE EQUATION
G \ 3 [ G\ NOTES
G[ G[ + Q(x)y = 0
Steps for solution:
1. Assume its solution to be of the form y = a0 + a1x + a2x2 + ... + anxn + ...
...(1)
dy d2 y
2. Find (or y) and (or y) from y.
dx dx 2
dy d2 y
3. Substitute the values of y, and in the given differential equation.
dx dx 2
4. Equate to zero the coefficients of various powers of x and find
a2, a3, a4, a5, ... in terms of a0 and a1.
5. Equate to zero, the coefficient of xn. The relation so obtained is called the
recurrence relation. It helps us in finding the values of other constants
easily.
6. Give different values to n in the recurrence relation to determine various
ais in terms of a0 and a1.
7. Substitute the values of a2, a3, a4, ... in assumed solution (1) above to get
the series solution of the given equation having a0 and a1 as arbitrary
constants.
SOLVED EXAMPLES
Example 1. Solve in series the differential equation
d2 y
+ xy = 0.
dx 2
d2 y dy
Sol. Comparing the given equation with the form + P(x) + Q(x) y = 0,
dx 2 dx
we get P(x) = 0, Q(x) = x
At x = 0, both P(x) and Q(x) are analytic, hence x = 0 is an ordinary point.
Assume its solution to be
y = a0 + a1x + a2x2 + a3x3 + ... + anxn + ... ...(1)
dy
Then, = a1 + 2a2x + 3a3x2 + 4a4x3 + ... + n anxn1 + ...
dx
d2 y
and = 21 a2 + 32 a3x + 43 a4x2 + 54 a5x3 + ... + n(n 1) anxn2 + ...
dx 2
Substituting these values in the given differential equation, we get
[21 a + 32  a x + 43  a x2 + 54  a x3 + ... + n(n 1) a xn2 + ...]
2 3 4 5 n
+ x [a0 + a1x + a2x2 + a3x3 + ... + anxn + ...] = 0
21 a2 + (32 a3 + a0) x + (43  a4 + a1)x2 + (54  a5 + a2)x3 + ...
+ {(n + 2) (n + 1) a + a }xn + ... = 0
n+2 n1

Ordinary Differential Equating to zero, the various powers of x as,
Equations
Coefficient of x0 = 0
 21  a2 = 0  a2 = 0
NOTES Coefficient of x = 0
 32  a3 + a0 = 0
a0 a0
 a3 =  a3 =
32 3!
 43  a4 + a1 = 0
a1 2 a1
 a4 = or a4 =
4 3 4!
 54  a5 + a2 = 0
a2
 a5 = or a5 = 0
54
 65  a6 + a3 = 0
a3 a0 4 a0
 a6 =  or a6 =
6 5 6 53 ! 6!
and so on.
Coefficient of xn = 0
 (n + 2) (n + 1) an+2 + an1 = 0
an  1
 an+2 =
(n  2) (n  1)
which is the recurrence relation.
Putting n = 5, 6, 7, ..., successively in recurrence relation, we obtain
5  2a1  74
a7 = , a8 = 0, a9 = a0 and so on.
7! 9!
Substituting these values in (1), we get
a 2a 4a 5.2 a1 7 7.4
y = a0 + a1x 0 x 3  1 x 4  0 x 6  x a x9 + ...
3! 4! 6! 7! 9! 0


y = a0 1 
x 3 1.4 6 1.4.7 9 ...
 x  x 
"#
+ a1 x  x 

2 4 2.5 7 ...
x 
"#
! 3! 6! 9! #$ 4! !7! $
where a0 and a1 are constants.
Example 2. Solve in series the differential equation
d2 y dy
(1 + x2)
2
x y = 0 about the point x = 0.
dx dx
Sol. Comparing the given differential equation with the form
d2 y dy
 P( x) + Q(x) y = 0, we get
dx 2 dx
x 1
P(x) = and Q(x) = .
1 x 2 1  x2
Both P(x) and Q(x) are analytic at x = 0 Power Series Solutions
 x = 0 is an ordinary point of the given differential equation.
Assume the solution to be
y = a0 + a1x + a2x2 + a3x3 + ... + anxn + ... ...(1) NOTES
dy
Then, = a1 + 2a2x + 3a3x2 + ... + nanxn1 + ...
dx
d2 y
and = 21  a2 + 32  a3x + ... + n(n 1)an xn2 + ...
dx 2
Substituting these values in given equation, we get
(1 + x2) [21  a + 32  a x + 43  a x2 + ... + n(n 1) a xn2 + ...]
2 3 4 n
+ x [a1 + 2a2x + 3a3x2 + 4a4x3 + ... + n anxn1 + ...]
[a + a x + a x2 + a x3 + ... + a xn + ...] = 0
0 1 2 3 n
0
a0
 2.1. a2 a0 = 0  a2 =
2
Coefficient of x = 0
 3.2 a3 + a1 a1 = 0  a3 = 0
 2.1 . a2 + 4.3.a4 + 2a2 a2 = 0
 4.3 a4 + 3a2 = 0
a2 a a0
 a4 =  0 or a4 =
4 8 8
 5.4. a5 + 3.2. a3 + 3a3 a3 = 0
 20a5 + 8a3 = 0  a5 = 0
4
 6.5 . a6 + 4.3 . a4 + 4a4 a4 = 0
 30a6 + 15a4 = 0
a4 a a0
 a6 = = 0 or a6 =
2 16 16
Similarly, a7 = 0, a9 = 0, a11 = 0 and so on.
Also, Coefficient of xn = 0
(n + 2) (n + 1) an+2 + n(n 1)an + nan an = 0
n  1 a '
 an+2 =
n  2 n | n+10
Putting n = 6, 8, 10, ..., we get

5 5a0
a8 = a6  
8 128
7 7a0
a10 = a8  and so on.
10 256

Ordinary Differential Substituting these values in (1), we get
Equations
a0 2 a0 4 a0 6 5a0 8 7a0 10 ...
y = a0 + a1x + x  x  x  x  x
2 8 16 128 256
NOTES x 2 x 4 x 6 5x 8 7x 10 ...

y = a0 1 
2

8
  
16 128 256
 + a1x
Example 3. Solve: (1 x2)y xy + 4y = 0 in series.
y + P(x) y + Q(x) y = 0, we get
x 4
P(x) = 2 , Q(x) =
1 x 1  x2
Since both P(x) and Q(x) are analytic at x = 0, hence x = 0 is an ordinary point of
the given equation.
y = a0 + a1x + a2x2 + a3x3 + ... + anxn + ... ...(1)
Then, y = a + 2.a x + 3.a x2 + ... + na xn1 + ...
1 2 3 n
and y = 2.1. a2 + 3.2.a3x + ... + n(n 1)an xn2 + ...
(1 x2) [2.1. a2 + 3.2. a3x + 4.3. a4x2 + ... + n(n 1) an xn2 + ...]
x [a + 2a x + 3a x2 + ... + na xn1 + ...] + 4 [a + a x + a x2 + a x3 + ... + a xn + ...]
1 2 3 n 0 1 2 3 n
=0
 2.1. a2 + 4a0 = 0  a2 = 2a0
a1
 3.2 a3 a1 + 4a1 = 0  a3 =
2
Coefficient of x2 =0
 4.3. a4 2.1 a2 2a2 + 4a2 =0  a4 = 0
 5.4 a5 3.2 a3 3a3 + 4a3 =0

a3 1  a1 a
 a5 =
4

4 2  1
8
a1
 a5 = and so on.
8
Substituting these values in assumed solution (1), we get
a1 3 a1 5 ...
y = a0 + a1x 2a0x2 x x +
2 8
x 2 x 4 ...
 y = a0 (1 2x2) + a1x 1  2

8
–

Example 4. Find the power series solution of the following differential equation Power Series Solutions
about x = 0
d2 y dy
(1 x2)  2x + 2y = 0.
2
dx dx
NOTES
d2 y dy
 P( x) + Q(x)y = 0, we get
dx 2 dx
 2x 2
P(x) = 2 , Q(x) =
1 x 1  x2
Since both P(x) and Q(x) are analytic at x = 0, hence x = 0 is an ordinary point of
the given equation.
y = a0 + a1x + a2x2 + a3x3 + ... + anxn + ... ...(1)
Then, y = a + 2a x + 3a x2 + 4a x3 + ... + n a xn1 + ...
1 2 3 4 n
and y = 2.1. a2 + 3.2. a3 x + 4.3. a4 x2 + ... + n(n 1)an xn2 + ...
(1 x2) [2.1. a2 + 3.2. a3x + 4.3. a4x2 + ... + n(n 1) anxn2 + ...]
2x [a1 + 2a2x + 3a3x2 + 4a4x3 + ... + n an xn1 + ...]
+ 2 [a + a x + a x2 + a x3 + ... + a xn + ...] = 0
0 1 2 3 n
 2.1. a2 + 2a0 = 0  a2 = a0
 3.2. a3 2a1 + 2a1 = 0  a3 = 0
 4.3.a4 2.1. a2 4a2 + 2a2 = 0
a2 a a0
 12a4 4a2 = 0  a4 =  0  a4 =
3 3 3
 5.4.a5 3.2. a3 6a3 + 2a3 = 0
 20a5 10a3 = 0  a5 = 0
 6.5.a6 4.3. a4 8a4 + 2a4 = 0
3 a0
 30a6 18a4 = 0  a6 = a4  a6 =
5 5
Also, a7 = 0, a9 = 0 and so on.
Substituting these values in assumed solution (1), we get
a0 4 a0 6 ......
y = a0 + a1x a0x2 x  x
3 5
x 4 x 6 ...

y = a0 1  x 2 
3

5
 + a1x

Ordinary Differential Example 5. Solve in series the differential equation
Equations
d2 y dy
(1 x2) 2
 2x + p(p + 1)y = 0.
dx dx
NOTES  2x p ( p  1)
Sol. Here, P(x) = , Q(x) =
1 x 2 1  x2
Since both P(x) and Q(x) are analytic at x = 0  x = 0 is an ordinary point of
the given differential equation.

Let the solution be y = a0 + a1x + a2 x2 + ...... + an xn + ...... = a
n0
n xn
...(1)

dy
 
dx n  0 
n an x n  1 ...(2)

d2 y
dx 2
  n(n  1) a
n0
n xn 2 ...(3)
Substituting the above values in the given equation, we get

  
(1 x2)  n(n  1) a
n0
n xn  2  2x  na
n0
n x n  1  p( p  1) a
n0
n xn = 0
 

n0
 n(n  1) an x n  2 
n0
 an[n(n 1) + 2n p(p + 1)]xn = 0
 
 
n0
n(n  1) an x n  2   a (n p) (n + p + 1)x
n0
n
n =0
This is an identity in x.
Coefficient of xn = 0
 (n + 2) (n + 1) an+2 (n p) (n + p + 1) an = 0
(n  p) (n  p  1)
 an+2 = an
(n  2) (n  1)
Putting n = 0, 2, 4, ..... etc., we get
 p ( p  1)
a2 = a0
2.1
(2  p) (3  p) ( p  2)( p) ( p  1) ( p  3)
a4 = a2  a0 etc.
4 .3 4!
Again, putting n = 1, 3, 5, ...... etc., we get
(1  p) ( p  2) ( p  1) ( p  2)
a3 = a1   a1
3. 2 3!
(3  p) ( p  4) ( p  3) ( p  1) ( p  2) ( p  4)
a5 = a3  a1 etc.
5.4 5!

Substituting these values in eqn. (1), we get Power Series Solutions

y = a0 1 
p ( p  1) 2 ( p  2) p ( p  1)( p  3) 4 ......
x  x 
"#
! 2! 4! $

+ a1 x 
( p  1)( p  2) 3 ( p  3)( p  1)( p  2)( p  4 ) 5 ......
x  x 
"# NOTES
! 3! 5! $
Note. Above method is an aliter to the method of solution in series discussed before and
preferred when, we get the recurrence relation in between an and an+2.
Example 6. Solve the differential equation y + (x 1)2 y 4(x 1) y = 0

in series about the ordinary point x = 1.
Sol. Put x = t + 1 (or x 1 = t)
dy dy dt dy ' dt
 .
dx dt dx
=
dt dx
1

d d
 
dx dt
d 2 y d dy
d dy d2 y

dx 2 dx dx


dt dt

dt 2
 The given equation becomes,
d2 y
+ t2y 4ty = 0 ...(1)
dt 2
Now, t = 0 is an ordinary point. | given
y = a0 + a1t + a2t2 + a3t3 + ... + antn + ... ...(2)
then y = a1 + 2a2t + 3a3t2 + ... + n an tn1 + ...
and y = 2a2 + 3.2. a3t + ... + n (n 1) an tn2 + ...
Substituting these values in eqn. (1), we get
[2a2 + 3.2. a3t + 4.3. a4 t2 + ... + n(n 1) an tn2 + ...]
+ t2 [a1 + 2a2t + 3a3 t2 + 4a4t3 + ... + n antn1 + ...]
4t [a0 + a1t + a2t2 + a3t3 + ... + antn + ...] = 0
Coefficient of t0 = 0
 2a2 = 0  a2 = 0
Coefficient of t = 0
2a0
 3.2. a3 4a0 = 0  a3 =
3
 4.3. a4 + a1 4a1 = 0
a1
 12a4 = 3a1  a4 =
4
 5.4. a5 + 2a2 4a2 = 0  a5 = 0

Ordinary Differential Coefficient of t4 = 0
Equations
 6.5. a6 + 3a3 4a3 = 0
a3 2a0 a0
 a6 =  a6 =
6.5 6.5.3 45
NOTES
Now, Coefficient of tn = 0
 (n + 2) (n + 1) an+2 + (n 1) an1 4an1 = 0
(n  5)
 an+2 = an  1
(n  2) (n  1)
Putting n = 5, 6, 7, 8, ..., we get
a7 = 0
1
a8 = a5  0
8.7
2  2 a0 a0
a9 = a6  
9.8 9.8 45 1620
and so on.
Substituting these values in (2), we get
2 a a a0
y = a0 + a1t + a0 t 3  1 t 4  0 t 6  t9 + ...
3 4 45 1620
2 3 1 6 1
t4

= a0 1 
3
t 
45
t 
1620
t 9  ...  a1 t 
4
 2 1 1   ( x  1)4 
 y = a0 1  ( x  1)3  ( x  1)6  ( x  1)9  ...  a1 ( x  1)  
 3 45 1620   4 
EXERCISE A
Solve the following equations in series: [Dashes denote differentiation w.r.t. x]

d2 y
1.  y0 2. y + x2y = 0
dx2
3. (i) y + xy + y = 0 (ii) y xy + y = 0
4. (i) y xy + x2y = 0 (ii) y + xy + x2y = 0
5. (1 x2) y + 2xy + y = 0 6. (2 + x2) y + xy + (1 + x) y = 0
7. (x2 + 1) y + xy xy = 0
8. (i) (x2 1) y + 4xy + 2y = 0 (ii) (x2 1) y + xy y = 0
9. (i) y + xy + (x2
+ 2) y = 0 (ii) (x2 1)y + 3xy + xy = 0 ; y(0) = 4, y(0) = 6.
10. (i) y xy + 2y = 0 near x = 1 (ii) y + (x 3) y + y = 0 near x = 2.
Answers
x  x  x  ...  a x  x  x  ... = a cosh x + a
2 4 6 3 5
1. 2 ! 4 ! 6 ! 3 ! 5 !
y = a0 1  1 0 1 sinh x
1  x  x  ...  a x  x  x  ...
4 8 5 9
2. y = a0
3 . 4 3.4 .7.8 4 .5 4 .5.8.9
1

x  x  x  ...  a x  x  x  x  ...
2 4 6 3 5 7 Power Series Solutions
3. 2 2.4 2.4 .6 3 3.5 3.5.7
(i) y = a0 1  1
x  x  3 x  3.5 x  ... + a x
(ii) y = a 1 
2 4
0
2! 4! 6! 8! 6 8
1
NOTES
x  x  ...  a x  x  x  x  ...
(i) y = a 1 
4 6 3 5 7
4. 0
12 90 6 40 144 1
x  x  ...  a x  x  x  ...
(ii) y = a 1 
4 6 3 5
0
12 90 6 40 1
x  x  ...  a x  x  x  ...
y = a 1 
2 4 3 5
5. 0
2 8 2 40 1
x  x  5x  ...  a x  x  x  ...
y = a 1 
2 3 4 3 4
6. 0
4 12 96 6 24 1
3 5
y = a 1  x  3 x  ...  a x  x  x  3 x  ...
3 4 5
7. 0 6 40 6 12 40 1
8. (i) y = a0 (1 + x2 + x4 + ...) + a1 (x + x3 + x5 + ...)

x2 x4

(ii) y = a0 1  2  4  ... + a1x
1  x x4 x3 3 5
9. (i) y = c0 2

4
 ... + c1 x 
2

40
x  ...
11 3 1 4 11 5
(ii) y = 4 + 6x + x + x + x + ...
3 2 4
1 " 1 "
10. (i) y = a0 1  ( x  1)2  ( x  1)3  ...#  a1 ( x – 1)  ( x  1) 2  ...#
! 3 $ ! 2 $
1 2 1 3 1 4 "
(ii) y = a0 1  ( x  2)  ( x  2)  ( x  2)  ...#
! 2 6 12 $
1 2 1 3 1 4 "
+ a1 ( x  2)  ( x  2)  ( x  2)  ( x  2)  ...#
! 2 6 6 $
FROBENIUS METHOD: SERIES SOLUTION WHEN X = 0

IS A REGULAR SINGULAR POINT OF THE
G \ 3 [ G\
DIFFERENTIAL EQUATION + Q(x)y = 0
G[ G[
Steps for solution:
1. Assume y = a0 xm + a1xm+1 + a2xm+2 + ... ...(1)
dy d 2 y
2. Substitute from (1) for y, , in given equation.
dx dx 2
3. Equate to zero the coefficient of lowest power of x. This gives a quadratic
equation in m which is known as the Indicial equation.
4. Equate to zero, the coefficients of other powers of x to find a1, a2, a3, ... interms
of a0.

Ordinary Differential 5. Substitute the values of a1, a2, a3,...in (1) to get the series solution of the
Equations given equation having a0 as arbitrary constant. Obviously, this is not the
complete solution of given equation since the complete solution must have
two independent arbitrary constants.
NOTES The method of complete solution depends on the nature of roots of the
indicial equation.
Case I. When Roots are distinct and do not differ by an integer

1
e.g., m1 =
, m2 = 1
2
Let m1 and m2 be the roots then complete solution is
y = c1 ( y) m1  c2 ( y) m2
SOLVED EXAMPLES
Example 7. Solve in series the differential equation:

d2 y dy
2x (1 x)  (5  7x) 3y = 0.
dx 2 dx
d2 y dy
Sol. Comparing the given equation with  P( x) + Q(x)y = 0, we get
dx 2 dx
5  7x 3
P(x) = , Q(x) =
2 x (1  x) 2 x (1  x)
At x = 0, Both P(x) and Q(x) are not analytic, hence x = 0 is a singular point.
5  7x
Now, x P(x) =
2 (1  x)
 3x
x2 Q(x) =
2 (1  x)
At x = 0, both x P(x) and x2 Q(x) are analytic, hence x = 0 is a regular singular
point.
Let us assume
y = a0 xm + a1xm+1 + a2 xm+2 + a3xm+3 + ... ...(1)
Then, y = ma0xm1 + (m + 1) a1 xm + (m + 2) a2 xm+1 + (m + 3) a3 xm+2 + ...
and y = m(m 1) a0 xm2 + (m + 1) m a1 xm1 + (m + 2) (m + 1) a2 xm
+ (m + 3) (m + 2) a3 xm+1 + .....
2x (1 x) [m(m 1) a0 xm2 + (m + 1) ma1 xm1
+ (m + 2) (m + 1) a2 xm + (m + 3) (m + 2) a3 xm+1 + ...]
+ (5 7x) [m a0 xm1 + (m + 1) a1 xm + (m + 2) a2 xm+1 + (m + 3) a3 xm+2 + ...]
3 [a xm + a xm+1 + a xm+2 + a xm+3 + ...] = 0
0 1 2 3
Now, coefficient of lowest power of x = 0
 Coefficient of xm1 = 0
 2m (m 1) a0 + 5m a0 = 0

 (2m2 + 3m) a0 = 0 Power Series Solutions
  2m2 + 3m = 0 '
( a0  0)
This is called indicial equation
m(2m + 3) = 0 NOTES
 m = 0, 3/2
Roots are distinct and do not differ by an integer.
Now, Coefficient of xm = 0
 2(m + 1) m a1 2m (m 1) a0 + 5(m + 1) a1 7ma0 3a0 = 0
 (m + 1) (2m + 5) a1 = (2m2 2m + 7m + 3) a0
(m  1) (2m  3)
a1 = a0
(m  1) (2m  5)
2m  3
 a1 = a0
2m  5
Coefficient of xm+1 = 0
 2(m + 2) (m + 1) a2 2(m + 1) m a1 + 5 (m + 2) a2 7(m + 1) a1 3a1 = 0
 (m + 2) (2m + 7) a2 = (2m2 + 2m + 7m + 7 + 3) a1
= (2m2 + 9m + 10) a1 = (2m + 5) (m + 2) a1
2m  5 2m  5 2m  3
 a2 = a1  . a
2m  7 2m  7 2m  5 0
2m  3
 a2 = a
2m  7 0
2m  7 2m  7 2m  3
Similarly, a3 = a2 = . a0
2m  9 2m  9 2m  7
2m  3
 a3 = a0
2m  9
and so on.
Hence, from (1),

y = xm a0 
2m  3
a0 x 
2m  3
a0 x 2 
2m  3
a0 x 3  ...
"#
! 2m  5 2m  7 2m  9 $
 y = a0 xm
1  2m  3 x  2m  3 x  2m  3 x  ..."#
2 3
! 2m  5 2m  7 2 m  9
...(2)
$
Now, y1 = (y)m=0
3 3 3
x  x 2  x 3  ...
"#
y1 = a0 1 
! 5 7 9 $ ...(3)
Also, y2 = (y)m= 3/2 = a0 x3/2 (1 + 0 . x + 0 . x2 + 0 . x3 + ...)

y2 = a0 x 3/2 ...(4)
Hence the complete solution is given by
3 3 3
y = c1 y1 + c2 y2 = c1 a0 1 
5 7 9
x  x 2  x 3  ... + c2a0 x 3/2

3 3 3
Equations  y = A 1
5
x  x 2  x 3  ... + Bx3/2
7 9
where A and B are constants.
NOTES Example 8. Solve in series the differential equation
d2 y dy
2x2 2
 (2x 2  x)  y  0.
dx dx
d2 y dy
Sol. Comparing the given equation with 2
 P( x) + Q(x) y = 0, we get
dx dx
2x 2  x 1 1
P(x) = 2
 1 and Q(x) =
2x 2x 2x2
1 1
Now, x P(x) = x and x2 Q(x) =
2 2
Since both x P(x) and x2 Q(x) are analytic at x = 0, hence x = 0 is a regular
singular point.
Let us assume
y = a0 xm + a1 xm+1 + a2 xm+2 + a3 xm+3 + ... ...(1)
Then, y = m a0 x m1 m
+ (m + 1) a1 x + (m + 2) a2 x m+1 + (m + 3) a3 xm+2 + ...
and y = m(m 1) a0 xm 2 + (m + 1) m a1 xm1 + (m + 2) (m + 1) a2 xm
+ (m + 3) (m + 2) a3 xm+1 + ...
2x2 [m(m 1) a0 xm2 + (m + 1)m a1 xm1 + (m + 2) (m + 1) a2xm
+ (m + 3) (m + 2) a3 xm+1 + ...] + (2x2 x) [m a0 xm1 + (m + 1) a1 xm
+ (m + 2) a2 xm+1 + (m + 3) a3 xm+2 + ...]
+ [a0 xm + a1 xm+1 + a2 xm+2 + a3 xm+3 + ...] = 0
Now, Coeff. of lowest power of x = 0 i.e., Coeff. of xm = 0
2m (m 1) a0 m a0 + a0 = 0
 (2m2 3m + 1) a0 = 0
 (2m 1) (m 1) = 0 (since a0  0)
which is indicial equation.
1
Its roots are m = 1,
2
Roots are distinct and donot differ by an integer.
Now, Coefficient of xm+1 = 0
 2m (m + 1) a1 + 2m a0 (m + 1) a1 + a1 = 0
 (2m2 + m) a1 + 2m a0 = 0
2
 a1 =
2m  1
a0 | ' m0
 2(m + 2) (m + 1) a2 + 2(m + 1) a1 (m + 2) a2 + a2 = 0
 (2m2 + 5m + 3) a2 + 2(m + 1) a1 = 0
 (2m + 3) (m + 1) a2 + 2(m + 1) a1 = 0

2 ( 2) ( 2) Power Series Solutions
 a2 = a1 = . a0
2m  3 2m  3 2m  2
4
 a2 = a0
(2m  1) (2m  3) NOTES
Similarly, we can find
8
a3 = a0
(2m  1) (2m  3) (2m  5)
16
a4 = a0
(2m  1) (2m  3) (2m  5) (2m  7)
and so on.
 2 4
 y = a0 xm 1  x x2
 2m  1 (2m  1) (2m  3)
8 
 x 3  ... ...(2)
(2m  1) (2m  3) (2m  5) 
Now, y1 = (y)m=1
y1 = a0 x 1 
2 x 4 x 2

8
x 3  ...
"#
! 3 3.5 3.5.7 $
or y1 = a0
2 2 x
x 1  x 
2
2

23
x3

 ... ...(3)
3 3.5 3.5.7
and y2 = (y)m=1/2
1 2 1 3 "#
!
y2 = a0 x1/2 1  x 
2
x  x  ...
6 $ ...(4)
Hence the complete solution is

y = c1 y1 + c2 y2
2
= c1a0 x 1  2 x  2 x 2  2
3
3 3.5 3.5.7
x 3  ...
1 2 1 3
 c2 a0 x 1  x 
2
x  x  ...
6
2
y = Ax 1  2 x  2 x 2  2
3
1 2 1 3

3 3.5 3.5.7 2
x 3  ... + B x 1  x  x  x  ...
6
EXERCISE B
Solve in series:
d2 y dy d2 y dy
1. 9x (1 x)  12 + 4y = 0 2. x (2 + x2) 2
 6xy = 0
2 dx dx dx
dx
d2 y dy d2 y dy
3. 3x 2  y0 4. 2x2 x + (1 x2) y = 0
dx 2 dx dx2 dx

Ordinary Differential d2 y dy
Equations 5. 2x2 y + xy (x + 1) y = 0 6. 2x (1 x)  (1  x) + 3y = 0
dx2 dx
d2 y dy 1 1
7. 2x2 x  ( x  5) y  0 8. y + y  y0
dx 2 dx 4x 8 x2
NOTES
d2 y dy d2 y dy
9. 2x2 2
x + (x2 + 1) y = 0 10. 4x  2 (1  x)  y  0.
dx dx dx 2 dx
11. 2x2y + 7x (x + 1)y 3y = 0 12. 2x2y + x(2x + 1) y y = 0
Answers
 1 1.4 2 1.4 .7 3   8 8.11 2 8.11.14 3 

1. y = A 1  x  x  x  ...  + Bx7/3 1  x x  x  ... 
 3 3.6 3.6.9   10 10.13 10.13.16 
 3 1 6   3 2 3.1 4 5.3.1 6 
2. y = A 1  3 x 2  x 4  x  ...  + Bx3/2 1  x  x  x  ... 
 5 15   8 8.16 8.16.24 
x x2 x3 x x2 x3
3. y = A 1  
2 20 480
 ... + Bx1/3 1   
4 56 1680
 ...
x2 x4 x2 x4
4. y = Ax 1  
2.5 2.4 .5.9
 ... + Bx1/2 1  
2.3 2.4 .3.7
 ...
5. y = Ax 1  1 x  1 x 2  ...   Bx  1/2  1 2 
1  x  2 x  ... 
 5 70   
 2 3 4 
6. y = A 1  3x  3x  3x  3x  ...  + B x (1  x)
 1.3 3.5 5.7 
 
x x2 x3 x x2 x3
7. y  c1x5 / 2 1   
9 198 7722
 ...  c2 x 1 1   
5 30 90
 ...
8. y = A x + Bx1/4
x2 x4 x2 x4
9. y = Ax 1  
10 360
 ...  Bx1/2 1 
6

168
 ...
x x2 x3 1  x  x  x  ... . 2 3
10. y = A 1 

 2  3
2 .1 ! 2 .2 ! 2 .3 !
 ... + B x

1.3 1.3.5 1.3.5.7
 7 49 2  3  21 49 2 
11. y = A x  1  x x  + Bx  1  x x 
 18 264   5 5 
 x2 x3   2 x 4 x 2 8x 3 
12. y = Ax1/2 1  x    ....  + Bx 1 
 
   .... 

 2 6   5 35 315 
Case II. When Roots are Equal e.g., m1 = m2 = 0

Complete solution is
y
y = c1 ( y) m  c2
1
m m1

SOLVED EXAMPLES Power Series Solutions
d2 y dy
Example 9. Solve in series: x (x 1) 2
 (3x  1)  y  0.
dx dx
NOTES
Sol. Comparing the given equation with
d2 y dy
2
 P( x) + Q(x) y = 0, we get
dx dx
3x  1 1
P(x) = and Q(x) =
x ( x  1) x ( x  1)
3x  1 x
Now, x P(x) = and x2 Q(x) =
x1 x1
Both x P(x) and x2 Q(x) are analytic at x = 0, hence x = 0 is a regular singular
point.
Let us assume
y = a0 xm + a1xm+1 + a2 xm+2 + a3 xm+3 + ... ...(1)
Then, y = m a0 xm1 + (m + 1) a1 xm + (m + 2) a2 xm+1 + (m + 3) a3 xm+2 + ...
and y = m (m 1) a0 xm2 + (m + 1)m a1 xm1
+ (m + 2) (m + 1) a2 xm + (m + 3) (m + 2) a3 xm+1 + ...
x (x 1) [m (m 1) a0 xm2 + (m + 1) m a1 xm1 + (m + 2) (m + 1) a0 xm
+ (m + 3) (m + 2) a3 xm+1 + ...]
+ (3x 1) [m a0 xm1 + (m + 1) a1 xm + (m + 2) a2 xm+1 + (m + 3) a3 xm+2 +...]
+ [a0 xm + a1 xm+1 + a2 xm+2 + a3 xm+3 + ...] = 0
Now, Coefficient of lowest power of x = 0
 Coefficient of xm1 = 0
 m (m 1) a0 m a0 = 0  m2 a0 = 0
m2 = 0 '
( a0  0)
which is Indicial equation
Its roots are m = 0, 0

Roots are equal.
Now, Coefficient of xm = 0
 m(m 1) a0 (m + 1) m a1 + 3m a0 (m + 1) a1 + a0 = 0
 (m + 1)2a0 (m + 1)2 a1 = 0
 a1 = a0 '
( m  1)
 (m + 1) m a1 (m + 2) (m + 1) a2 + 3(m + 1) a1 (m + 2)a2 + a1 = 0
 (m + 2)2 a1 (m + 2)2 a2 = 0
 a2 = a1 ( '
m  2)
 a2 = a0
Similarly, we can show that
a3 = a0
a4 = a0 and so on.

Ordinary Differential  y = a0xm (1 + x + x2 + x3 + ...) | From (1)
Equations
Now, y1 = (y)m=0 = a0 x0 (1 + x + x2 + x3 + ...) = a0 (1 + x + x2 + x3 + ...)
y
NOTES
y2 =
m m0
= [a0 (1 + x + x2 + x3 + ...) xm log x]m=0
= a0 log x(1 + x + x2 + x3 + ...)

y = c1y1 + c2y2 = c1a0 (1 + x + x2 + x3 + ...) + c2a0 log x(1 + x + x2 + x3 + ...)
y = (A + B log x) (1 + x + x2 + x3 + ...)
d2 y dy
Example 10. Solve in series the differential equation: x   y  0.
2
dx dx
d2 y dy
Sol. Comparing with the equation 2
 P( x) + Q(x) y = 0, we get
dx dx
1 1
P(x) =
and Q(x) =
x x
Since at x = 0, both P(x) and Q(x) are not analytic  x = 0 is a singular point.
Also, x P(x) = 1 and x2 Q(x) = x
Both x P(x) and x2 Q(x) are analytic at x = 0  x = 0 is a regular singular
point.
Let us assume
y = a0 xm + a1 xm+1 + a2xm+2 + a3 xm+3 + ... ...(1)
Then, y = m a0 xm1 + (m + 1) a1 xm + (m + 2) a2 xm+1 + (m + 3) a3 xm+2 + ...
and y = m(m 1) a0 xm2 + (m + 1) m a1 xm1
+ (m + 2) (m + 1) a2 xm + (m + 3) (m + 2) a3 xm+1 + ...
Substituting these values in the given equation, we get
x [m (m 1) a0 xm2 + (m + 1)m a1 xm1 + (m + 2) (m + 1) a2xm
+ (m + 3) (m + 2) a3 xm+1 + ...]
+ [m a0 xm1 + (m + 1) a1 x + (m + 2) a2 xm+1 + (m + 3) a3 xm+2 + ...]
m
[a0 xm + a1 xm+1 + a2 xm+2 + a3xm+3 + ...] = 0

Now, Coefficient of xm1 = 0
 m(m 1) a0 + ma0 = 0
 m2a0 = 0  m2 = 0 ( a0  0) '
which is Indicial equation.
Its roots are m = 0, 0 which are equal.

Coefficient of xm = 0
 (m + 1) ma1 + (m + 1)a1 a0 = 0  (m + 1)2 a1 = a0
a0
 a1 =
(m  1) 2
 (m + 2) (m + 1) a2 + (m + 2) a2 a1 = 0  (m + 2)2 a2 = a1

a1 a0
 a2 =  a2 =
(m  2) 2 (m  1) 2 (m  2) 2
a0
Similarly, a3 = and so on. NOTES
(m  1) (m  2) 2 (m  3) 2
2
 From (1),

y = a0xm 1 
x

x2

x3
 ...
"#
! (m  1) 2 (m  1) 2 (m  2) 2 (m  1) 2 (m  2) 2 (m  3) 2 $
...(2)
Now, y1 = (y)m=0 = a0 1  x 
x2

x3
 ...
"# ...(3)
! (2 !) 2 (3 !) 2 $
To get the second independent solution, differentiate (1) partially w.r.t. m.
y  x x2 x3 
= a0xm log x 1  2
 2 2
 2 2 2
 ...
m  (m  1) (m  1) (m  2) (m  1) (m  2) (m  3) 
+ a0xm 
2x %& 1  1 () x

2 2
(m  1) ( m  1) (m  2) ' m  1 m  2 *
! 3 2 2

2 %& 1  1  1 () x  ..."# 3
( m  1) ( m  2) (m  3) ' m  1 m  2 m  3 *
2 2
$ 2
y = a log x 1  x  x  x  ..."#
The second solution is y =
2 3
m
2
! (2 !)
m0 (3 !) $0 2 2
1 1  1 x  1 1  1  1 x  ..."#

! (2 !) 2 (3 !) 2 3
2 3
2a x  0 2
$ 2
1  1  1 x  1 1  1  1 x  ..."#

! (2 !) 2 (3 !) 2 3
2 3
= y log x 2a x 
1 0 2
$ 2
y = c1y1 + c2y2 = (c1a0 + c2a0 log x) 1  x 

x2

x3
 ...
"#
! (2 !) 2 (3 !) 2 $
1 1  1 x 1 1  1  1 x "#
2 2 3
2 3
2c2a0 x    ...
! (2 !) 2
(3 !) 2
$
 y = (A + B log x) 1  x 
x 2

x
 ...
3 "#
! (2 !) 2
(3 !) 2 $
1 1  1 x 1 1  1  1 x "#
2 2 3
2 3
2B x    ...
where c1a0 = A, c2a0 = B.

! (2 !) 2
(3 !) 2
$
EXERCISE C
Solve in series:
d2 y dy
1. (i) xy + (1 + x) y + 2y = 0 (ii) x 2
  xy  0
dx dx

Ordinary Differential d2 y dy d2 y dy
Equations 2. x2 2
 x ( x  1)  (1  x) y  0 3. (x x2)  (1  5x) 4y = 0
dx dx 2 dx
dx
4. (x x2) y + (1 x) y y = 0 5. x2 y x (1 + x) y + y = 0
6. xy + y + x2y = 0
NOTES
7. xy + y + xy = 0. (Bessels equation of order zero)
Answers
 3 2 4 3    13 2 
1. (i) y = A 1  2x  x  x  ...   B  y1 log x  a0  3x  4 x  ...  
 2! 3!    
 x2 x4   x2 3x 4 
(ii) y = (A + B log x) 1  2  2 2  ...  B  2   ... 

 2 2 .4  2 2.4 3

 

y = Ax + B x log x  x 
x2
 ...
"#
2.
! 4 #$
3. y = A (12 + 22x + 32x2 + 42x3 + ...) + B [y1 log x 2a0 (1.2x + 2.3 x2 + 3.4 x3 + ...)]
 2 2.5 3    2 14 3 
4. y = A 1  x  x 2  x  ...  + B  y1 log x  a0   2x  x  x  ...  
 4 4.9    27 
 1 1 3    3 
5. y = Ax 1  x  x 2  x  ...   B  y1 log x  a0 x 2   1  4 x  ...  
 2 2.3    
y = A 1
x3

x6

x9 "#
 ...
6.
! 3 2 4
3 (2 !) 2 6
3 (3 !) 2
#$
  x
3
1  1 6  
+ B  y1 log x  2 a0  3  5 1  2  x  ... 
  3 3 (2 !)2    
 x2 x4 x6    x 2 1  1
7. y = A 1  2  2 2  2 2 2  ...  + B  y1 log x  a0  2  2 2 1   x 4
  2
 2 2 .4 2 .4 .6    2 2 .4 
1 1 1 6 
+ 2 1    x  ... .
2 2
2 .4 .6  2 3  

Case III. When Roots are Distinct, Differ by Integer and Making a
Coefficient of y Infinite
Let m1 and m2 be the roots such that m1 > m2.
In this case, if some of the coefficients of y become infinite when m = m2, we
modify the form of y by replacing a0 by b0 (m m2).
Complete solution is
y
y = c1 ( y) m  c2
1
m m2
.
Remark. We can also obtain two independent solutions by putting m = m2 (value of m

y
for which some coefficients of y become infinite) in modified form of y and . The result of
m
putting m = m1 in y will give a numerical multiple of that obtained by putting m = m2.

SOLVED EXAMPLES Power Series Solutions
Example 11. Obtain the series solution of the Bessels equation of order two
d2 y dy
x2
2
x + (x2 4) y = 0 near x = 0. NOTES
dx dx
Sol. Comparing the given equation with the form
d2 y dy
2
 P( x) + Q(x)y = 0, we get
dx dx
1 x2  4 4
P(x) = and Q(x) =  1 2
x x 2
x
At x = 0, both P(x) and Q(x) are not analytic.
Therefore x = 0 is a singular point.
Also, x P(x) = 1 and x2 Q(x) = x2 4
Both x P(x) and x2 Q(x) are analytic at x = 0
 x = 0 is a regular singular point.
Let us assume,
y = a0xm + a1 xm+1 + a2 xm+2 + a3 xm+3 + ... ...(1)
dy
Then, = m a0 xm1 + (m + 1) a1 xm + (m + 2) a2 xm+1 + (m + 3) a3 xm+2 + ...
dx
d2 y
and = m(m 1) a0 xm2 + (m + 1) ma1 xm1 + (m + 2) (m + 1) a2 xm
dx 2
+ (m + 3) (m + 2) a3xm+1 + ...
x2 [m(m 1) a0 xm2 + (m + 1) ma1 xm1
+ (m + 2) (m + 1) a2 xm + (m + 3) (m + 2) a3 xm+1 + ...]
+ x [m a0 xm1 + (m + 1) a1 xm + (m + 2) a2 xm+1 + (m + 3) a3 xm+2 + ...]
+ (x2 4) [a0 xm + a1 xm+1 + a2 xm+2 + a3 xm+3 + ...] = 0
Now, Coefficient of lowest power of x = 0
 Coefficient of xm = 0
 m(m 1)a0 + m a0 4a0 = 0  (m2 4) a0 = 0
 m2 4 = 0 (Indicial equation) | ' a0  0
m = 2, 2
Roots are distinct and differ by integer.

(m + 1)m a1 + (m + 1) a1 4a1 = 0
 (m2 + 2m 3) a1 = 0  (m + 3) (m 1) a1 = 0
 a1 = 0 Since m  1, and
m3
 (m + 2) (m + 1) a2 + (m + 2) a2 + a0 4a2 = 0
 (m2 + 4m) a2 + a0 = 0

Ordinary Differential  a0
Equations  a2 =
m (m  4)
NOTES  (m + 3) (m + 2) a3 + (m + 3) a3 + a1 4a3 = 0
 (m + 1) (m + 5) a3 = a1
 a3 = 0 ' a1  0
Also, coefficient of xm+4 = 0

(m + 2) (m + 6) a4 + a2 = 0
 a2 a0
 a4 = 
(m  2) (m  6) m (m  2) (m  4) (m  6)
a0
 a4 =
m (m  2) (m  4) (m  6)
Similarly, a5 = a7 = a9 = ... = 0
 a0
a6 = etc.
m (m  2) ( m  4) 2 (m  6) (m  8)
Substituting above obtained values in assumed y given by eqn. (1), we get
y = a0xm 1 
x2

x4
! m (m  4) m (m  2) (m  4) (m  6)

x6
 ...
"# ...(2)
m (m  2) (m  4) 2 (m  6) (m  8) $
Putting m = 2 (the greater of the two roots) in (2), the first solution is
x2 x4 x6

y1 = a0x2 1   
2.6 2.4.6.8 2.4.6 2 .8.10
 ...
If we put m = 2 in (1), the coefficients become infinite due to the presence of
the factor (m + 2) in the denominator. To overcome this difficulty, let a0 = b0 (m + 2) so
that
 (m  2)x 2 x4 x6 
y = b0xm (m  2)    2
 ...
 m (m  4) m (m  4)(m  6) m (m  4) (m  6) (m  8) 
Differentiating partially w.r.t. m, we get
y
= b0xm log x (m  2) 
(m  2) x 2

x4
 ...
"#
m ! m (m  4) m (m  4) (m  6) $
(m  2) 1 %&1 1 () x
+ b0xm 1   
! ' *
2
m (m  4) m  2 m m  4
+
1 %& 
1

1

1
x 4 ...
() "#
'
m (m  4) (m  6) m m4 m6 * $
 y 
The second solution is y2 =  
 m m   2

= b0x2 log x
x  4
x 6 "
...#
! ( 2) (2) (4) ( 2) (2) (4) (6) $

2
1  x  1 1  1  1 x ..."#
2
! ( 2) (2) ( 2) (2) (4) 2 2 4 $

2 4
+b x 0 NOTES
= b0x2
1  x ..."#  b x 1  x  x  ..."#
log x 
2
2
2 4
! 2 . 4 2 . 4 .6 $
2 3 0
! 2 2 .4 $ 2 2 2

y = c1y1 + c2 y2
 x2 x4 x6    1 x2 
= Ax 2 1  2.6  2.4.6.8   ...    B  x 2 log x   2  3 ... 
2
2.4.6 .8.10   
     2 .4 2 .4.6 
x2 x4 "#

+ x 2 1  
22 22 . 4 2
 ... #$
where A = c1a0, B = c2b0.
d2 y dy
Example 12. Solve in series the differential equation x2 2
 5x  x2 y  0 .
dx dx
d2 y dy
 P( x) + Q(x) y = 0, we get
dx 2 dx
5
P(x) = , Q(x) = 1
x
At x = 0, since P(x) is not analytic  x = 0 is a singular point.
Also, x P(x) = 5
x2 Q(x) = 0
Since both x P(x) and x2 Q(x) are analytic at x = 0  x = 0 is a regular singular
point.
Let us assume
y = a0 xm + a1xm+1 + a2 xm+2 + a3 xm+3 + ... ...(1)
dy
 = m a0 xm1 + (m + 1) a1 xm + (m + 2) a2 xm+1 + ... ...(2)
dx
d2 y
and = m (m 1) a0 xm2 + (m + 1) m a1 xm1 + (m + 2) (m + 1) a2 xm + ...
dx 2
...(3)
Substituting the above values in given equation, we get
x2 [m(m 1) a0 xm2 + (m + 1) ma1 xm1 + (m + 2) (m + 1) a2 xm + ...]
+ 5x [ma0 xm1 + (m + 1) a1 xm + (m + 2) a2 xm+1 + ...]
+ x2 [a0xm + a1 xm+1 + a2 xm+2 + ...] = 0 ...(4)
Equating the coefficient of lowest power of x to zero, we get
m(m 1) a0 + 5ma0 = 0 [Coeff. of xm = 0]
 (m2 + 4m) a0 = 0
 m(m + 4) = 0 (Indicial equation) '
( a0  0)
 m = 0, 4
Ordinary Differential Hence the roots are distinct and differing by an integer. Equating to zero, the
Equations coefficients of successive powers of x, we get
(m + 1) m a1 + 5(m + 1) a1 = 0
NOTES
 (m + 5) (m + 1)a1 = 0  a1 = 0 ...(5)
| ' m  5, 1
Coefficient of xm+2 =0
(m + 2) (m + 1) a2 + 5(m + 2)a2 + a0 = 0
(m + 2) (m + 6) a2 + a0 = 0
 a0
a2 = ...(6)
(m  2) (m  6)
Again, Coefficient of xm+3 = 0

(m + 3) (m + 2) a3 + 5(m + 3) a3 + a1 = 0
(m + 3) (m + 7) a3 + a1 = 0
 a1
 a3 =
(m  3) (m  7)
 a3 = 0 ...(7)
Similarly, a5 = a7 = a9 = ... = 0
Now, Coefficient of x m+4 =0
(m + 4) (m + 3) a4 + 5(m + 4) a4 + a2 = 0
 (m + 4) (m + 8) a4 = a2
 a2 a0
a4 =  etc. ...(8)
(m  4) (m  8) (m  2) (m  4) (m  6) (m  8)
 x2 x4 
These give y = a0 xm 1    ...
 (m  2)(m  6) (m  2)(m  4)(m  6)(m  8) 
...(9)
Putting m = 0 in (9), we get
 2 4 
y1 = (y)m=0 = a0 1  x  x  ... ...(10)
 2.6 2.4.6.8 
If we put m = 4 in the series given by eqn. (9), the coefficients become infinite.
To avoid this difficulty, we put a0 = b0 (m + 4), so that
 (m  4) x 2 x4 
y = b0 xm (m  4)    ... ...(11)
 ( m  2)( m  6) ( m  2)( m  6)( m  8) 
y  m2  8m  20 2 (3m2  32m  76) 

Now, = y log x + b0xm 1  x  x 4  ...
m 2
 (m  8m  12)
2 3 2
(m 16m  76m  96)2

Second solution is given by
y x 2 x4
y2 = m m 4

= (y)m= 4 log x + b0 x4 1 
4

4
 ...

x4 x6 "#x 2 x4 Power Series Solutions
= b0x 4 log x 0  0 
!

( 2) (2) (4) 16
 ... + b0 x 4 1 
$4

4
 ...
 x
log x
4
x6
x2 x4
= b0x 4
16 
16
 ...  b0 x  4 1 
4

4
 ... NOTES
y = c1y1 + c2 y2
x2 x4 x4 x6
= c1 a0 1  
12 384
 ... + c2 b0x 4 log x  
16 16
 ...
x 2 x4
+ c2b0x 4 1  4

4
 ...
x2 x4 x2 x4
 y = A 1 
12 384
 ...  B x  4 1 
4

4
 ...
1  x  ... 2
B log x 16 16
where A = c1a0 and B = c2b0.
EXERCISE D
Solve in series:
d2 y dy d2 y dy
1. x (1 x) 2
 3x  y0 2. x2 2
x + (x2 1) y = 0
dx dx dx dx
(Bessels equation of order one)
d2 y dy d2 y dy
3. (x + x2 + x3) 2
 3 x2  2y  0 4. x (1 x)  (1  3 x)  y  0.
dx dx 2 dx
dx
Answers
1. y = (A + B log x) (x + 2x2 + 3x3 + 4x4 + ...) + B (1 + x + x2 + x3 + ...)
x2 x4
x2 x4
2.
y = Ax 1   2
2.4 2.4 .6
 ... + Bx1 log x 
2 2  ...
2 .4

1  x 2

3
x 4  ...
"#
+ Bx 1
! 2 2 2
2 .2 3
#$
1 2 1 3 "
3. y = Ax 1  x  x  x  ...# + B log x (2x + 2x2 x3 + ...) + B(1 x 5x2 x3 + ...)
! 2 2 $
4. y = (A + B log x) (1.2 x2 + 2.3x3 + 3.4x4 + ...) + B ( 1 + x + 5x2 + 11x3 + ...).
Case IV. When Roots are Distinct, Differ by Integer and Making One or
More Coefficients Indeterminate
Let the roots be m1 and m2. If one of the coefficients (suppose a1) become indeterminate
when m = m2, the complete solution is given by putting m = m2 in y which then contains
two arbitrary constants.
Note. The result contained by putting m = m1 in y merely gives a numerical multiple of
one of the series contained in the first solution. Hence we reject the solution obtained by putting
m = m1.

Ordinary Differential SOLVED EXAMPLES
Equations
Example 13. Solve in series the differential equation: xy + 2y + xy = 0.
NOTES
d2 y dy
2
 P( x) + Q(x) y = 0, we get
dx dx
2
P(x) = and Q(x) = 1
x
At x = 0, P(x) is not analytic  x = 0 is a singular point.
Also, xP(x) = 2 and x2 Q(x) = x2
At x = 0, since x P(x) and x2 Q(x) are analytic  x = 0 is a regular singular
point.
Let us assume
y = a0 xm + a1xm+1 + a2 xm+2 + a3 xm+3 + ... ...(1)
dy
Then, = ma0 xm+1 + (m + 1) a1 xm + (m + 2) a2 xm+1 + (m + 3) a3 xm+2 + ...
dx
d2 y
and = m (m 1) a0 xm2 + (m + 1) m a1 xm1 + (m + 2) (m + 1) a2 xm
dx 2
+ (m + 3) (m + 2) a3xm+1 + ...
x [m (m 1) a0 xm2 + (m + 1) m a1 xm1 + (m + 2) (m + 1) a2xm
+ (m + 3) (m + 2) a3 xm+1 + ...]
+ 2 [m a0 xm1 + (m + 1) a1 xm + (m + 2) a2 xm+1 + ...]
+ x [a0 xm + a1 xm+1 + a2 xm+2 + a3 xm+3 + ...] = 0
Now, Coefficient of xm1 = 0
 m (m 1) a0 + 2m a0 = 0
(m2 + m) a0 = 0
 m2 + m = 0 (Indicial equation) | ' a0  0
 m = 0, 1
Hence roots are distinct and differ by an integer.

Coefficient of xm = 0
 (m + 1) m a1 + 2(m + 1) a1 = 0
 (m + 1) (m + 2)a1 = 0
 (m + 1)a1 = 0 | ' m+20
0
Since m + 1 may be zero, hence a1 is arbitrary (or takes the form ). In other
0
words, a1 becomes indeterminate.
Hence the solution will contain a0 and a1 as arbitrary constants. The complete
solution will be given by putting m = 1 in y.
 (m + 2) (m + 1) a2 + 2(m + 2) a2 + a0 = 0
 (m + 2) (m + 3) a2 + a0 = 0
 a0
a2 =
(m  2) (m  3)
Coefficient of xm+2 = 0 Power Series Solutions
 (m + 3) (m + 2) a3 + 2(m + 3) a3 + a1 = 0
(m + 3) (m + 4) a3 + a4 = 0
– a1 NOTES
a3 =
(m  3)(m  4)
 (m + 4) (m + 3) a4 + 2(m + 4) a4 + a2 = 0
 (m + 4) (m + 5) a4 = a2
 a2
 a4 =
(m  4) (m  5)
a0
 a4 =
(m  2) (m  3) (m  4) (m  5)
(m + 5) (m + 4) a5 + 2(m + 5) a5 + a3 = 0
(m + 5) (m + 6) a5 = a3
a1
a5 =
(m  3) (m  4) (m  5) (m  6)
and so on.
Substituting these values in eqn. (1), we get
 a0 a1 a0
y = xm  a0  a1 x  x2  x3  x4
 (m  2)( m  3) ( m  3)(m  4) ( m  2)( m  3)(m  4)( m  5)
a1
x 5  ...
"#
+
(m  3) (m  4) (m  5) (m  6) $
% x2 x4 ()
&
y = xm a0 1    ...
! ' (m  2) (m  3) (m  2) (m  3) (m  4) (m  5) *
%&
+ a1 x 
x3

x5 ()"#
 ...
' (m  3) ( m  4) ( m  3) (m  4) (m  5) ( m  6) *#$
x2 x4 x  x  x  ... "#
3 5
! 2.3 2.3.4.5 #$
Now, (y)m=1 = x1 a0 1    ...  a1
1.2 1.2.3.4
= x1 [a0 cos x + a1 sin x]

Hence complete solution is given by
y = (y)m= 1
1
 y= (a cos x + a1 sin x).
x 0
Note. All those problems, in which x = 0 was an ordinary point of y + P(x) y + Q(x) y =
0, can also be solved by Frobenius method as given in Art. 2.4.4 and explained in above illustrative
example.

Ordinary Differential EXERCISE F
Equations
Solve in series:
d2 y dy d2 y dy
1. x2 2
 4x  ( x2  2) y  0 2. (1 x2) x  4y  0
NOTES dx dx dx 2 dx
d2 y dy
3. (1 x2)  2x  n (n  1) y  0.
dx2 dx
Answers
x3 x 5 x7
1. y = x2 (a0 cos x + a1 sin x) 2. y = a0 (1 2x2) + a1 x  2

8

16
 ...
n (n  1) 2 (n  2) n (n  1) (n  3) 4 "
3. y = a0 1  x  x  ...#
! 2! 4! $
(n  1) (n  2) 3 (n  3) (n  1) (n  2) (n  4) 5 "
+ a1 x  x  x  ...#
! 3! 5! $

Legendres Differential
Equations Solutions
6. DIFFERENTIAL EQUATIONS NOTES
STRUCTURE
Introduction
Legendres Function of First Kind Pn(x)
Legendres Function of Second Kind Qn(x)
Solution of Legendres Equation
Generating Function for Pn(x)
Rodrigues Formula
Recurrence Relations
Beltramis Result
Orthogonality of Legendre Polynomials
Laplaces Integral of First Kind
Laplaces Integral of Second Kind
Cristoffels Expansion Formula
Cristoffels Summation Formula
Expansion of a Function in a Series of Legendre Polynomials
(Fourier-Legendre Series)
INTRODUCTION
dy d2 y
The differential equation (1 x2)
+ n(n + 1)y = 0
2
 2x ...(1)
dx dx
where n is real number, is called Legendres differential equation. This equation
is of considerable importance in applied mathematics, particularly in boundary value
problems involving spherical configurations.
Though n is a real number, in most physical applications, only integral values
of n are required. Also, equation (1) can be solved in series of ascending or descending
powers of x. The solution in descending powers of x is more important than the one in
ascending powers.

Let y= 
k0
ak x m  k

 d2 y
then
dy
dx
  (m  k) a k x m k1
and
dx 2

k0

(m  k) (m  k  1) ak x m  k  2
k0

dy d2 y
Equations Substituting for y, and in (1), we get
dx dx 2
 
(1 x2)  (m  k) (m k 1) ak xmk2 2x  (m  k) a k xmk1
NOTES k0 k0

+ n(n + 1) 
k0
ak xmk = 0
 
or 
k0
( m  k) (m k 1) ak xmk2  [(m  k) (m k 1)
k0
+ 2(m k) n (n + 1)]ak xmk = 0
 
or 
k0
( m  k) (m k 1) ak xmk2  (m  k)
k0
2 n2 + (m k) n] ak xmk = 0
 
or  (m  k) (m k 1) a k xmk2 
k0
[(m k n) (m k + n + 1) ak xmk = 0.
k0
Equating to zero the coefficient of highest power of x, i.e., xm, we get the indicial
equation
(m n) (m + n + 1) a0 = 0
whence m = n or m = (n + 1) since a0  0
Equating to zero the coefficient of the next lower power of x, i.e., xm1, we get
(m + n) (m n 1) a1 = 0 or a1 = 0,
since (m + n) and (m n 1) are not zero for m = n or (n + 1).
Equating to zero the coefficient of xmk, we get the recurrence relation
[m (k 2)] [m (k 2) 1] ak2 (m k n) (m k + n + 1) ak = 0
(m  k  2) (m  k  1)
or ak = ak  2 ...(2)
(n  m  k) (n  m  k  1)
Since a1 = 0, therefore, from (2), we get a3 = a5 = a7 = ...... = 0.

Case I. When m = n, the recurrence relation (2) reduces to
(n  k  2) (n  k  1)
ak = ak  2
k (2n  k  1)
n(n  1)
Putting k = 2, 4, 6, ......, we get a2 = a0 ,
2 (2n  1)
(n  2) (n  3) n (n  1) (n  2) (n  3)
a4 = a2  a0, etc.
4 (2n  3) 2.4.(2n  1)(2n  3)
Therefore, one solution of Legendres equation is given by

y1 = a0 x n 
n(n  1) n 2 n(n  1)(n  2)(n  3) n  4
x  x  ...
"# ...(3)
! 2(2n  1) 2.4.(2n  1) (2n  3) $
Case II. When m = (n + 1), the recurrence relation (2) reduces to
(n  k  1)(n  k)
ak = ak2
k (2n  k  1)

Putting k = 2, 4, 6, ....., we get Legendres Differential
Equations Solutions
(n  1) (n  2)
a2 = a0 ,
2 (2n  3)
(n  3) (n  4) (n  1)(n  2)(n  3) (n  4) NOTES
a4 = a2  a0, etc.
4 (2n  5) 2 . 4 .(2n  3)(2n  5)
Therefore, the second solution of Legendres equation is given by
 
y2 = a0  x  n 1  ( n  1)( n  2) x  n  3  ( n  1)( n  2)( n  3)( n  4) x  n  5  ... ...(4)
 2(2 n  3) 2.4. (2n  3)(2 n  5) 
LEGENDRES FUNCTION OF FIRST KIND Pn(x)
1. 3 . 5 ...... (2n  1)
When n is a positive integer and a0 = ,
n!
the first solution given by (3) is denoted by Pn (x) and is called Legendres function of
first kind.
Thus,
1.3.5 ...... (2n  1)  n n ( n  1) n  2 n ( n  1) . ( n  2)( n  3) n  4 

Pn (x) =
n!  x  2(2n  1) x 
2.4 . (2n  1)(2n  3)
x  ...
 
Pn(x) is a terminating series. RHS is known as Zonal Harmonic. Pn(x) gives
Legendres polynomials for different values of n such that Pn (1) = 1.
Now, two cases arise:
Case I. When n is even:
n
No. of terms in the series within bracket = 1
2
{n (n  1) (n  2) (n  3) ...... 2 . 1}
Last term = (1)n/2 . .
(2 . 4 . 6 ... n) {(2n  1) (2n  3) ... (n  1)}
Case II. When n is odd:
n1
No. of terms in the series within bracket =
2
n –1
n (n  1) (n  2)(n  3) ... 3 . 2
Last term = ( 1) 2 . .
{ 2 . 4 . 6 ... (n  1)} {(2n  1)(2n  3) ... (n  2)}
LEGENDRES FUNCTION OF SECOND KIND Qn (x)
n!
When n is a positive integer and a0 = ,
1. 3 . 5 ... (2n  1)
the second solution is denoted by Qn (x) and is called Legendres function of second
kind.

Equations n!
x n  1 
(n  1) (n  2)  n  3
x
Thus, Qn (x) =
1. 3 . 5 ... (2n  1) ! 2 . (2n  3)
NOTES ( n  1)( n  2)( n  3)( n  4)  n  5 ...

+ x  
2  4  (2n  3)(2n  5) 
It is a non-terminating series so there is no last term.
SOLUTION OF LEGENDRES EQUATION
Since y = Pn (x) and y = Qn (x) both are the solutions of the given equation hence the
most general solution is given by
y = APn (x) + BQn (x)

where A and B are arbitrary constants.
GENERATING FUNCTION FOR Pn(x)
We shall show that Pn(x) is the coefficient of hn in the expansion of (1 2xh + h2)1/2 in
ascending powers of h.

i.e., (1 2xh + h2)1/2 =  P (x) . h

n0
n
n
Using Binomial theorem,

1 3 1 3 5
. . .
1 2 2 2 2 2 2 t3 + ...
(1 t)1/2 =1+ t  t 
2 2! 3!
1 1 . 3 2 1 . 3 . 5 3 ... 1 . 3 . 5 ...... (2n  1) n
=1+ t t  t   t +
2 2.4 2.4.6 2 . 4 . 6 ...... 2n
...
 (1 2xh + h2)1/2 = [1 h(2x h)]1/2
1 h (2 x  h)  1 . 3 2
=1+ h (2x h)2 + ...
2 2.4
1 . 3 . 5 ...... (2n  5) n2
+ h (2x h)n2
2 . 4 . 6 ...... (2n  4)
1 . 3 . 5 ...... (2n  3) n 1
+ h (2 x  h) n 1
2 . 4 . 6 ...... (2n  2)
1 . 3 . 5 ...... (2n  1) n
+ h (2x h)n + ...
2 . 4 . 6 ...... (2n)
1 . 3 . 5 ...... (2n  1) n
Now, the coefficient of hn in h (2x h)n is
2 . 4 . 6 ...... (2n)
1 . 3 . 5 ...... (2n  1) 1 . 3 . 5 ...... (2n  1) 1 . 3 . 5 ...... (2n  1) n
= (2 x) n  (2 x) n  x
2 . 4 . 6 ...... (2n) 2 n (n) ! n!

1 . 3 . 5 ...... (2n  3) n1 Legendres Differential
The coefficient of hn in h (2x h)n1 is Equations Solutions
2 . 4 . 6 ...... (2n  2)
1 . 3 . 5 ...... (2n  3) 1 . 3 . 5 ...... (2n  3)
= [ n1C1 (2x)n2] = (n 1) 2n2 . xn2
2 . 4 . 6 ...... (2n  2) 2 . 4 . 6 ...... (2n  2)
NOTES
1 . 3 . 5 ...... (2n  3)
= (n 1) 2n2 . xn2
2 n  1 (n  1) !
1 . 3 . 5 ...... (2n  3) (2n  1) n (n  1) n  2
= . x
2 (n) ! 2n  1
1 . 3 . 5 ...... (2n  1) n (n  1) n2
= . x
n! 2 (2n  1)
1 . 3 . 5 ...... (2n  5) n2
Similarly, the coefficient of hn in h (2x h)n2 is
2 . 4 . 6 ...... (2n  4)
1 . 3 . 5 ...... (2n  1) n (n  1) (n  2) (n  3) n  4
= . x and so on.
n! 2 . 4 . (2n  1) (2n  3)
 The coefficient of hn in (1 2xh + h2)1/2 is given by

1 . 3 . 5 ...... (2n  1) n n (n  1) n  2 n (n  1) (n  2) (n  3) n  4
x  x  x  ... = Pn(x)
"#
n! ! 2 (2n  1) 2 . 4 . (2n  1) (2n  3) $
Thus, in the expansion of (1 2xh + h2)1/2, P1 (x), P2(x), P3(x), ......, Pn(x), ...... are
the coefficients of h, h2, h3, ......, hn, ..... respectively.

 (1 2xh + h2)1/2 = 1 + P1(x) . h + P2 (x) . h2 + ... + Pn(x) . hn + ... = P

n0
n ( x) . h n
The function (1 2xh + h2)1/2 is called the generating function for Pn(x).
SOLVED EXAMPLES
Example 1. Show that

(i) Pn(1) = 1 (ii) Pn ( x) = ( 1)n Pn (x) (iii) Pn( x) = ( 1)n + 1 Pn(x).

Sol. We know that
n0
h n
Pn ( x)  (1  2 xh  h2 )  1/2 ...(1)
(i) Putting x = 1 in eqn. (1), we get


h
n0
n
Pn (1)  (1  2 h  h2 )  1/ 2  (1  h)  1

=1+h+ h2 + ...... + hn + ...... = h
n0
n
Equating the coefficients of hn, we have Pn (1) = 1.

(ii) Replacing x by ( x) in eqn. (1), we get

 hn Pn (  x )  (1  2xh  h2 ) 1/2 ...(2)
n0
Again, replacing h by ( h) in eqn. (1), we have


n0
( h)n Pn(x) = (1 + 2xh + h2)1/2

Ordinary Differential 
Equations or  ( 1)
n0
n hn Pn(x) = (1 + 2xh + h2)1/2 ...(3)
 
NOTES
From (2) and (3),  hn Pn (  x )   (  1)n hn Pn ( x )
n0 n0
Equating the coefficients of hn, we have

Pn( x) = ( 1)nPn (x).
(iii) We have, Pn( x) = ( 1)n Pn(x) | Proved in (ii)
Differentiating w.r.t. x, we get
Pn( x) = ( 1)n Pn(x)
 Pn( x) = ( 1)n + 1 Pn(x).
Example 2. Show that:
2n !
(i) P2n(0) = ( 1)n 2n (ii) P2n+1(0) = 0.
2 (n !) 2

Sol. We know that h
n0
n
Pn ( x)  (1  2 xh  h2 )  1/ 2

Putting x = 0, we get h
n0
n
Pn (0)  (1  h2 )  1/ 2
1 2 1 . 3 4 ...... 1 . 3 . 5 ...... (2r  1)

h 
=1 h   (  1)r . h2r + ...
2 2.4 2 . 4 . 6 ...... (2r )
(i) Equating the coefficients of h2n on both sides, we get
1 . 3 . 5 ...... (2n  1) 1 . 2 . 3 . 4 ...... (2n  1)(2n )
P2n (0) = (1)n  ( 1)n
2 . 4 . 6 ...... (2n ) [2 . 4 . 6 ...... (2n )]2
n (2n ) ! ( 2n ) !
= (–1) n . 2
 (–1) n
[2 1 . 2 . 3 ...... n ] 2 (n !)2 2n
(ii) Equating the coefficients of h2n+1 on both sides, we get P2n+1 (0) = 0, since the
right-hand side contains only even power of h.
Example 3. Prove that:

1
(i)
n0
P n (x) 
2  2x
(ii) Pn (1) = (1)n.
Sol. We know that,


(1 2x h + h2)1/2 = h
n0
n
Pn ( x) ...(1)
(i) Put h = 1 in (1), we get


(1 2x + 1)1/2 = P
n0
n ( x)

1

2  2x
 P
n0
n ( x)
(ii) We have already proved in example 1 (ii) that

Pn ( x) = ( 1)n Pn (x)

Put x = 1 in above relation, we get Legendres Differential
Pn ( 1) = ( 1)n Pn(1) = (1)n | ' Pn(1) = 1
Equations Solutions


1  z2

(1  2xz  z 2 ) 3/2 n  0

(2n  1) Pn z n . NOTES
Sol. We know that


(1 2zx + z2)1/2 = z
n0
n
Pn ( x) ...(1)
Differentiating (1) w.r.t. z, we get


1

2
(1 2zx + z2)3/2 . (2z 2x) =  nz
n0
n1
Pn ( x)

 (x z) (1 2zx + z2)3/2 =  nz
n0
n1
Pn ( x) ...(2)
Multiplying both sides of eqn. (2) by 2z, we get


2z (x z) (1 2zx + z2)3/2 =  2nz
n0
n
Pn ( x) ...(3)
Adding (1) and (3), we get


(1 2zx + z2)3/2 (2zx 2z2 + 1 2zx + z2) =  (2n  1) z
n0
n
Pn ( x)

1  z2

(1  2 zx  z 2 ) 3 / 2

n0
 (2n  1) z n
Pn ( x) .

1 z 1
z 1  2xz  z 2
 
z n0 
(Pn  Pn  1 ) z n .
 
Sol. RHS = 
n0
Pn z n  P
n0
n1 zn
 
1
= P
n0
n zn 
z n0Pn  1 z n  1

1
= P
n0
n zn  (P z + P2z2 + P3z3 + ... + Pnzn + ...)
z 1

1
= P
n0
n zn 
z
{ P0 + P0z0 + P1z + P2z2 + ... + Pnzn + ...} | ' P0 = 1


1 1


1  P
1
=
n0
Pn z n  
z z n0
Pn z n = 1 
z n0
n zn 
z
...(1)
1 z 1
1 1
LHS =
z 1  2 xz  z 2

z
= 1
z
(1 2xz + z2)1/2
z
1
= 1  z

1
z
n
Pn  ...(2)
n0
z
Hence the proof. | Since LHS = RHS
Ordinary Differential Example 6. Prove that:
Equations
n (n  1)
(i) Pn (1) =
2
n(n  1) n (n  1)
(ii) Pn ( 1) = ( 1)n1 . Or Pn ( 1) = ( 1)n+1 . .
NOTES 2 2
Sol. Legendres differential equation is
d2 y dy
(1 x2) 2
– 2x + n(n + 1) y = 0
dx dx
y = Pn(x) is a solution of it. So,
(1 x2) Pn (x) 2x Pn (x) + n(n + 1) Pn (x) = 0 ...(1)
(i) Put x = 1 in (1), 2Pn (1) + n(n + 1) Pn (1) = 0
n (n  1)
 Pn (1) =
2
| ' Pn (1) = 1
(ii) Put x = 1 in (1),
2Pn (1) + n(n + 1) Pn(1) = 0
 2Pn (1) + n(n + 1) (1)n = 0 | See Ex. 3 (ii)
 2Pn (1) = n(n + 1) (1)n
Case I. 2Pn (1) = (1)1 . n(n + 1) (1)n = ( 1)n+1 . n(n + 1)
n (n  1)
 Pn (1) = (1)n+1 . .
2
Case II. 2Pn (1) = (1)1 . n(n + 1) . (1)n = (1)n1 . n(n + 1)
n (n  1)
 Pn (1) = (1)n1 .
2
RODRIGUES FORMULA
1 dn
The relation Pn(x) = (x2 1)n is known as Rodrigues formula.
2 n ! dx n
n
dv
To prove it, let v = (x2 1)n then v1 = = n(x2 1)n1 . 2x
dx
Multiplying both sides by (x2 1), we get
(x2 1)v1 = 2nx (x2 1)n = 2nxv
or (1 x2)v1 + 2nxv = 0
Differentiating (n + 1) times by Leibnitzs theorem, we have

(1  x 2 )vn  2  (n  1) ( 2 x)vn  1 
(n  1)n "#
( 2)vn + 2n [xvn+1 + (n + 1) vn] = 0
! 2! $
or (1 x2)vn+2 2xvn+1 + n(n + 1)vn = 0
d(vn )d 2 (vn )
or (1 x2)  2x
+ n(n + 1)vn = 0
2
dx dx
which is Legendres equation and vn is its solution. But the solutions of Legendres
equations are Pn (x) and Qn (x).
dn
Since vn = n
( x 2  1) n contains only positive powers of x, it must be a constant
dx
multiple of Pn(x).

i.e., vn = cPn (x) Legendres Differential
n Equations Solutions
d
or cPn(x) = (x2 1)n
dx n
dn
= [(x 1)n (x + 1)n] ...(1) NOTES
dx n
dn n1 d
n1
= (x 1)n n n
n (x + 1) + C1 . n(x 1) (x + 1)n + .......
dx dx n  1
dn
+ (x + 1)n (x 1)n
dx n
dn 1
= (x 1)n.n ! + nC1.n(x 1)n1 . (x + 1)n + ..... + (x + 1)n n !
dx n  1
= n ! (x + 1)n + terms containing powers of (x 1)
Putting x = 1 on both sides, we get
cPn (1) = n ! . 2n or c = 2n n !, since Pn(1) = 1
Substituting in (1), we get
1 dn
Pn(x) = (x2 1)n
2 n n ! dx n
Putting n = 0, 1, 2, 3, ...... in Rodrigues formula, we get Legendres polynomials.

Thus
P0(x) = 1
1 d
P1(x) = ( x 2  1)  x
2 dx
1 d2 1 d2
P2(x) = ( x 2  1) 2  (x4 2x2 + 1) = 1
(3x2 1)
2 2 (2) ! dx 2 8 dx 2 2
1 d3 1 d3
P3(x) = ( x 2  1) 3  (x6 3x4 + 3x2 1)
23 (3) ! dx 3 48 dx 3
1
= 2
(5x3 3x)
1
Similarly, P4(x) = (35x4 30x2 + 3)
8
1
P5(x) = (63x5 70x3 + 15x)
8
1
P6(x) = (231x6 351x4 + 105x2 5) etc.
6
SOLVED EXAMPLES
1
Example 7. Show that x4 = [8P4 (x) + 20P2 (x) + 7P0 (x)].
35
Sol. We know that P4 (x) = 81 (35x4 30x2 + 3)
1
P2(x) = 2
(3x2 1), P0(x) = 1
1 1
 [8P4(x) + 20P2(x) + 7P0(x)] = [35x4 30x2 + 3) + 10(3x2 1) + 7] = x4.
35 35

Ordinary Differential Example 8. Express f(x) = x3 5x2 + x + 2 in terms of Legendres polynomials.
Equations 1
Sol. We know that P3(x) = 2
(5x3 3x)
2 3
 x3 = P3 ( x)  x
NOTES 5 5
2 3 2 "# 2 8
P3 ( x)  ( x) 5x2 + x + 2 = P3 ( x)  5 x  x  2
 f(x) =
3 ! 5 5 $ 5
2 2 1 8 "#
= P3 ( x)  5
5 3 !
P2 ( x) 
3
 x+2
5 $
'
[ P2(x) = 1
2
(3x2 1)  x2 = 2
3
P2 (x) + 1
3
]
2 10 8 1
= P3 ( x)  P2 ( x)  x
5 3 5 3
2 10 8 1
= P3 ( x)  P2 ( x)  P1 ( x)  P0(x)
5 3 5 3
[ '
x = P1 (x) and 1 = P0 (x)]
Example 9. Prove that: I 1

1
Pn (x) dx 
0, n0
2, n0
. ()
*
1 dn
Sol. We know by Rodrigues formula, Pn(x) = {( x 2  1) n }
2 n n ! dx n
Integrating, we get
I1
1
Pn ( x) dx 
1
2n n ! I 1
1 dx n
dn
(x2 1)n dx
= n
1 dn  1 2
( x  1) n
"# 1
=0
2 n ! dx n  1 ! $ 1
When n = 0,
I 1
1
P0 (x) dx = I 1
1
1 dx = 2. |Q P0 (x) = 1
Example 10. Express 4x3 + 6x2 + 7x + 2 in terms of Legendres polynomials.

Sol. Let 4x3 + 6x2 + 7x + 2   P3 (x) +  P2 (x) +  P1 (x) +  P0 (x) ...(1)
5 x  3 x   3 x  1 +  (x) + (1)
3 2
 2 2
5 3 3 x    
x   
2 2
3 2
 x 
2 2
Equating the coefficients of like powers of x, we get
5 8
=4  =
2 5
3
6=  =4
2
3 12 47
7=  7=  =
2 5 5

2=  2=2  =4
2
Hence from (1),
8 47
4x3 + 6x2 + 7x + 2 = P3 (x) + 4P2(x) + P (x) + 4P0 (x).
5 5 1

Example 11. Prove that: Legendres Differential
1 + ... + P  1 P 1 .
Equations Solutions
1  P  1 P 1  P  1 P
Pn 
2 2 2 2 0 2n 1 2n  1
2 2 2
2n 0
Sol. We know that, NOTES


h
n0
n
Pn ( x)  (1  2 hx  h2 )  1/ 2 ...(1)
1
Put x = in (1),
2
h

1 = (1 h + h )
2
n
Pn 2 1/2 ...(2)
n0
1
Put x = in (1),
2
h

1 = (1 + h + h )
2
n
Pn – 2 1/2 ...(3)
n0
Replacing h by h2 in (3), we get


h 1 = (1 + h
2
2n
Pn  2 + h4)1/2 = [(1 + h2)2 h2]1/2
n0
= (1 + h2 + h)1/2 (1 + h2 h)1/2


1 .  h 
1
h n Pn 
2 2
n
= Pn | From (2) and (3)
n0 n0
h 1 = P  1  1  ...  1

2 
2n 2n  1
 Pn  0   2   h P1   2    h P2n 1   
n0      2
 1 
+ h2n P2n     ...
 2 
 1  1  1  1  
 P0  2   hP1  2   ......  h P2n 1    h2n P2n    ...
2 n 1
     2 2 
Equating the coefficients from both sides of the above equation, we get the
required result.
 1  1 1   1  1  ......  1 1 
Pn     P0   2  . P2n  2   P1   2  P2n 1 2  P2n   2  P0 2 .
 2            
Example 12. Let Pn (x) be the Legendre polynomial of degree n. Show that for
any function f(x), for which the nth derivative is continuous
I1
1
f(x) Pn(x) dx =
(  1)n
2 n!n I 1
1
(x2 1)n f (n)(x) dx.
Sol. I1
1
f(x) Pn(x) dx = I 1
1
f (x) .
1
2 n n ! dx n
dn
(x2 1)n dx
Using Rodrigue’s
formula
= n
1
2 n! I1
1
f(x) .
dn
dx n
(x2 1)n dx

 n 1 
1 
Equations 1   f ( x ) d n 1 d n 1 

2 2 n
= ( x  1)   f ( x ) ( x 1) dx
2n n !   dx n 1 1 1
 dx n  1 
 
Integrating by parts
I
NOTES
=
1
0
1
f  ( x)
dn 1
( x 2  1) n dx
"#
2n n ! ! 1 dx n  1 $
=
( 1)1
n
2 n! I
1
1
f  (x )
d n 1
dx n 1
( x 2  1)n dx
=
( 1) 2
2n n ! I 1
1
f  ( x)
dn 2
dx n  2
( x 2  1) n dx Integrating by parts again
=
( 1) n
n
2 n! I 1
1
f ( n) ( x) ( x 2  1) n dx .
Integrating by parts
(n  2) times
RECURRENCE RELATIONS
n Pn(x) = (2n 1) x Pn1(x) (n 1) Pn2(x)

Or
(n + 1) Pn+1 (x) = (2n + 1) x Pn(x) n Pn1(x)
We know that,

(1 2xh + h2)1/2 = h
n0
n
Pn ( x) ...(1)
Differentiating both sides w.r.t. h, we get


1

2
(1 2xh + h2)3/2 (2h 2x) =  0
n hn1 Pn (x)
 (x h) (1 2xh + h2)1/2 = (1 2xh + h2)  nh

0
n1
Pn ( x)
 
 (x h)  0
h n Pn ( x) = (1 2xh + h2)  nh
0
n1
Pn ( x)
Equating coefficient of hn1 on both sides,

x Pn1 (x) Pn2 (x) = n Pn (x) 2x (n 1) Pn1 (x) + (n 2) Pn2 (x)
 n Pn(x) = (2n 1) x Pn1 (x) (n 1) Pn2 (x)
Replacing n by (n + 1), we get the other form.
n Pn (x) = x Pn¢ (x) P¢n1 (x)


We know that, (1 2h x + h2)1/2 = h
n0
n
Pn ( x) ...(1)

Differentiating both sides of (1) w.r.t. h, we get Legendres Differential

Equations Solutions
1

2
(1 2xh + h2)3/2 . ( 2x + 2h) =  nh
0
n1
Pn ( x)
 NOTES
 (x h) (1 2hx + h2)3/2 = 
0
n h n  1 Pn ( x) ...(2)
Differentiating both sides of (1) w.r.t. x, we get


1
(1 2hx + h2)3/2 . ( 2h) =
2
h
0
n
Pn  ( x)
 (x h) (1 2hx + h2)3/2 = (x h) h
0
n1
Pn  ( x) ...(3)
Equating eqns. (2) and (3), we get

 
 nh n1
Pn ( x) = (x h) h
0
n1
Pn  ( x)
0
Comparing the coefficient of hn1 on both sides, we get
n Pn (x) = x Pn (x) Pn1 (x)
(2n + 1) Pn (x) = P¢n+1 (x) P¢n1 (x)

From Recurrence relation (1),
(2n + 1) x Pn(x) = (n + 1) Pn+1 (x) + n Pn1 (x)
(2n + 1) [x Pn (x) + Pn (x)] = (n + 1) Pn+1 (x) + n Pn1 (x) ...(1)
From Recurrence relation (2), x Pn (x) = n Pn(x) + Pn1 (x)
From (1),
(2n + 1) [n Pn(x) + Pn1 (x) + Pn (x)] = (n + 1) Pn+1 (x) + n Pn1 (x)
 (2n + 1) (n + 1) Pn(x) = (n + 1) Pn+1 (x) (n + 1) Pn1 (x)
 (2n + 1) Pn(x) = Pn+1 (x) Pn1 (x)
(n + 1) Pn (x) = P¢n+1 (x) x Pn¢ (x)

From Recurrence relation (3), we have
(2n + 1) Pn (x) = Pn+1 (x) Pn1 (x) ...(1)
n Pn (x) = x Pn (x) Pn1 (x) ...(2)
Subtraction yields, (n + 1) Pn(x) = Pn+1 (x) x Pn (x)
(1 x2) P¢n(x) = n[Pn1(x) xPn(x)]

Pn (x) xPn1 (x) = nPn1(x) ...(1)
x Pn (x) Pn1 (x) = n Pn (x) ...(2)
Multiplying (2) by x and subtracting from (1), we get
(1 x2) Pn (x) = n [Pn1 (x) xPn (x)]

Ordinary Differential (1 x2) P¢n(x) = (n + 1) [xPn(x) Pn+1(x)]
Equations
Recurrence relation (1) may be written as
(n  1  n) xPn(x) = (n + 1) Pn+1(x) + nPn1(x)
NOTES
 (n + 1) xPn(x) + nxPn(x) = (n + 1) Pn+1 (x) + nPn1(x)
or (n + 1) [xPn(x) Pn+1(x)] = n[Pn1(x) xPn(x)]
= (1 x2) Pn (x) | Using recurrence relation (5)
 (1 x ) Pn(x) = (n + 1) [xPn(x) Pn+1 (x)].
2
BELTRAMIS RESULT
(2n + 1) (x2 1) Pn¢ = n(n + 1) (Pn+1 Pn1)

n(Pn1 xPn) = (1 x2) Pn ...(1)
(n + 1)(xPn Pn+1) = (1 x2) Pn ...(2)
From eqn. (1), nPn1 nx Pn = (1 x2) Pn
nPn  1  (1  x 2 ) Pn
 xPn = ...(3)
n
(1  x 2 ) Pn
From eqn. (2) xPn Pn+1 =
n1
(1  x 2 ) Pn
 xPn = Pn+1 + ...(4)
n1
From (3) and (4),
nPn  1 – (1  x 2 ) Pn (1  x 2 ) Pn
 Pn  1 
n n1
(n  1) Pn  1  (1  x 2 ) Pn
=
n1
 (n + 1) {nPn1 (1 x2) Pn} = n{(n + 1) Pn+1 + (1 x2) Pn}
 (2n + 1) (1 x2) Pn = n (n + 1) {Pn1 Pn+1}
 (2n + 1) (x2 1) Pn = n(n + 1) {Pn+1 Pn1}.
ORTHOGONALITY OF LEGENDRE POLYNOMIALS
We shall show that
%K0 ,
I1
1
Pm ( x) Pn ( x) dx  &K 2 ,
' 2n  1
if m  n
if m  n

Case I. When m  n Legendres Differential
Equations Solutions
We know that Pm(x) and Pn(x) are the solutions of the equations
(1 x2)u 2xu + m(m + 1)u = 0 ...(1)
and (1 x2)v 2xv + n(n + 1)v = 0 ...(2) NOTES
Multiplying (1) by v and (2) by u and subtracting, we get
(1 x2) (uv vu) 2x(uv vu) + [m(m + 1) n(n + 1)] uv = 0
d
or [(1 x2)(uv vu)] + (m n)(m + n + 1)uv = 0
dx
d
or (n m)(n + m + 1) uv = [(1 x2) (uv vu)]
dx
Integrating w.r.t. x from 1 to 1, we get
I " 1
uv dx  (1  x ) (u v  v u)#
1
2
(n m) (n + m + 1) =0
1
! #$
I
1
1
Hence Pm(x) Pn(x) dx = 0, since m  n.
1
Case II. When m = n


We know that (1 2xh + h2)1/2 =
n0
h n
Pn ( x)
Squaring both sides, we get

   
(1 2xh + h2)1 = 
n0
[ h n Pn ( x)]2  
n0
h 2 n [Pn ( x)]2  2  
m0 n0
hm+n Pm(x) Pn(x)
( m  n)
Integrating w.r.t. x between the limits 1 to 1, we have

I
n0
1
1
h 2n [Pn ( x)]2 dx  2  

m0 n0
( m  n)

I
1
1
h m n Pm ( x) Pn ( x) dx  I 1
1
dx
1  2 xh  h 2
I I
 1 1 dx
or h 2n [ Pn ( x)] 2 dx 
1  1 1  2 xh  h 2
n0
| Since other integrals on the LHS vanish by Case I as m  n
=
1
log (1  2 xh  h 2 )
"# 1

1
[log (1 h)2 log (1 + h)2]
2h ! #$ 1
2h
1
= [log (1 + h) log (1 h)]
h
1 h  h  h  h  ......  h  h  h  h
2 3 4 2 3 4 "#
! 2 3 4 2 3 4 #$
=  ......
h
2 3 " 5
 ......#
h h
= h 
h! 3 5 $
or 
n0

h 2n I1
1
n
2 h  h  ......  h  ......
[P ( x)] dx  2 1 
3 5 2n  1
2 4 2n

Ordinary Differential Equating the coefficients of h2n on the two sides, we get
Equations
NOTES
I 1
1
[Pn ( x)]2 dx 
2 .
2n  1
LAPLACES INTEGRAL OF FIRST KIND
1 
Pn(x) =  {x  x 2  1 cos } n d 
 0
We know that,
I
0
 d
a  b cos 


a  b2
2
;a>b ...(1)
Replace a by (1 xz) and b by z x2  1

 a2 b2 = (1 xz)2 z2 (x2 1) = 1 2xz + z2
Then (1) becomes
I 
0
d
1  xz  z x  1 cos  2


1  2 xz  z 2
 I 
0
d
1  z { x  x  1 cos }2
= (1 2xz + z2)1/2
2
Let z{x  x  1 cos } = t, then
I0
 d
1 t

n0

z n Pn ( x)

...(2)

If | t | < 1, then (1 t)1 = t
n0
n
 From (2), I 
0
n0

 [z n
{ x  x 2  1 cos  } n ] d   z
n0

n
Pn ( x)
Equating the coefficient of zn on both sides, we get
Pn(x) =
1
 I0

{x  x 2  1 cos  } n d.
LAPLACES INTEGRAL OF SECOND KIND
1  d
Pn(x) =

0
.
{ x  x 2  1 cos }n 1

We know that Legendres Differential
I
Equations Solutions
 d 
 ;a>b ...(1)
0 a  b cos  a  b2
2
NOTES
Put a = xz 1 and b = z x 2  1 so that
a2 b2 = (xz 1)2 z2(x2 1) = 1 2xz + z2
With above substitutions, (1) becomes,
I 
0
d
2
xz  1  z x  1 cos 


1  2 xz  z 2

  1
= =  Pn ( x)
1 x  1 2 z n0
zn
z 1 2
z z
 I
0
 d
z { x  x 2  1 cos }  1


zn  1

 P (x)
n 0
n
 I 0
 d


t  1 zn  1 n0

 P ( x) n ...(2) where t = z {x ± x 2  1 cos  }
Now, LHS = I
0

t 1

d
1
t

 I0
 1
t

1
1
t

1
d
If
1
t
< 1, then LHS = 0
 1

t
1 1
t t t
1 
1   2  ......  n  ......  d =
 I
0
 
n0
t
1
n1
d
= I
0

n0

z n1
1
{ x  x  1 cos } n  1
2
d
1
Now, comparing and equating the coefficients of n1 on both sides of eqn.(2),
z
we get
Pn(x) =
1
 I
0

{ x  x 2  1 cos }n  1
d
.
CRISTOFFELS EXPANSION FORMULA
Pn¢(x) = (2n 1) Pn1(x) + (2n 5) Pn3(x) + (2n 9) Pn5(x) + ...... + Last term.
where Last term =

%&3P ; 1 when n is even
.
()
'P; 0 when n is odd *
From Recurrence relation (3), we know that,
(2n + 1) Pn = Pn+1 Pn1
 Pn+1 = (2n + 1) Pn + Pn1 ...(1)

Ordinary Differential Replace n by (n 1), then
Equations
Pn = (2n 1) Pn1 + Pn2 ...(2)
Now replace n by (n 2) in (2), we get
NOTES Pn2 = (2n 5) Pn3 + Pn4 ...(3)
 From (2), Pn = (2n 1) Pn1 + (2n 5) Pn3 + Pn4
Proceeding in this manner, for the last term, two cases arise:
Case I. When n is even:
P2 = 3P1 + P0 = 3P1 'P 0  1 and P0  0
so, last term = 3P1
Case II. When n is odd:
P3 = 5P2 + P1 'P 1  x  P1  1  P0
= 5P2 + P0
so, last term = P0.
CRISTOFFELS SUMMATION FORMULA
The sum of first (n + 1) terms of the series


 n  1) [Pn  1 ( x ) Pn ( y) – Pn ( x ) Pn  1 ( y)]

m0
(2m + 1) Pm(x) Pm(y) =
x y
By Recurrence relation (1),
(2m + 1) xPm(x) = (m + 1) Pm+1(x) + mPm1(x) ...(1)
Similarly, (2m + 1) yPm(y) = (m + 1) Pm+1(y) + mPm1(y) ...(2)
Multiplying (1) by Pm(y) and (2) by Pm(x) and then subtracting (2) from (1), we
get
(2m + 1) (x y) Pm(x) Pm(y) = (m + 1) Pm+1(x) Pm(y) + mPm1(x) Pm(y)
(m + 1) Pm+1(y) Pm(x) mPm1(y) Pm(x)
= (m + 1) [Pm+1(x) Pm(y) Pm+1(y) Pm(x)]
m [Pm1(y) Pm(x) Pm1(x) Pm(y)]
Put 0, 1, 2, 3, ......, n for m in succession, we get
(x y) P0(x) P0(y) = P1(x) P0(y) P1(y) P0(x)
3(x y) P1(x) P1(y) = 2{P2(x) P1(y) P2(y) P1(x)} {P0(y) P1(x) P0(x) P1(y)}
5(x y) P2(x) P2(y) = 3 {P3(x) P2(y) P3(y) P2(x)} 2{P1(y) P2(x) P1(x) P2(y)}
# # #
# # #
(2n + 1) (x y) Pn(x) Pn(y) = (n + 1) [Pn+1(x) Pn(y) Pn+1(y) Pn(x)]

n [Pn1(y) Pn(x) Pn1(x) Pn(y)]
Adding simultaneously, we get
(x y) [P0(x) P0(y) + 3P1(x) P1(y) + 5P2(x) P2(y) + ...... + (2n + 1) Pn(x) Pn(y)]
= (n + 1) [Pn+1(x) Pn(y) Pn+1(y) Pn(x)]
 P0(x) P0 (y) + 3P1(x) P1(y) + ... + (2n + 1) Pn(x) Pn(y) Legendres Differential
n  1 {P Equations Solutions

=
x – y n+1(x) Pn(y) Pn+1(y) Pn(x)}
 Sum of first n terms of the series NOTES


(n  1) { Pn  1 ( x) Pn ( y)  Pn  1 ( y) Pn ( x)}

m0
(2m + 1) Pm(x) Pm(y) =
x y
.
EXPANSION OF A FUNCTION IN A SERIES OF

LEGENDRE POLYNOMIALS (FOURIER-LEGENDRE
SERIES)
The orthogonal property of Legendre polynomials enables us to expand a function f(x),

defined from x = 1 to x = 1 in a series of Legendre polynomials.

Let f(x) = a
n0
n Pn ( x) = a P (x) + a P (x) + a P (x) + ...
0 0 1 1 2 2 ...(1)
To determine an, multiplying both sides of (1) by Pn(x) and integrating w.r.t. x
from 1 to 1, we have
I 1
1
f ( x) Pn ( x) dx  an I 1
1
Pn 2 ( x) dx  an
2
2n  1
 an = n 

1
2

I
1
1
f ( x) Pn ( x) dx
Expansion of f(x) given by (1) is known as Fourier-Legendre series.
SOLVED EXAMPLES
Example 13. Prove that: I

1
1
(1  x 2 ) Pm Pn dx = 0
where m and n are distinct positive integers and m  n.
Sol.
I1
1
(1  x 2 ) Pm Pn dx
1
  1 d 
= (1  x 2 ) Pm Pn    Pn  {(1  x 2 ) Pm }  dx
   1 1  dx 
| Integrating by parts
= I 1
1
Pn
d
dx
{(1 x2) Pm} dx
= I 1
1
Pn { m (m + 1) Pm} dx | From Legendres differential equation
= m(m + 1) I1
1
Pn Pm dx = m(m + 1) . 0 = 0

Equations Example 14. Prove that: I 1
1
x 2 Pn  1 (x) Pn+1(x) dx =
2n (n  1)
(2n  1)(2n  1)(2n  3)
.
Sol. Recurrence relation (1) is

NOTES
(n + 1) Pn+1 = (2n + 1) xPn n Pn1
 (2n + 1) xPn = (n + 1) Pn+1 + nPn1 ...(1)
Replacing n by n + 1 and n 1 respectively in (1), we get
(2n + 3) xPn+1 = (n + 2) Pn+2 + (n + 1) Pn ...(2)
and (2n 1) xPn1 = nPn + (n 1) Pn2 ...(3)
Multiplying (2) and (3) and integrating within limits 1 and 1, we get
(2n + 3) (2n 1) I 1
x 2 Pn+1Pn1 dx
I I
1
= n (n + 1)
1
Pn 2 dx + n(n + 2)
I 1
Pn Pn  2 dx + (n2 1)
1
1
Pn  2 Pn dx
I
1 1
1
+ (n + 2) (n 1) Pn  2 Pn 2 dx
1
2
= n(n + 1) . [Using orthogonal properties]
2n  1
 I1
1
x 2 Pn  1 Pn  1 dx 
2n (n  1)
(2n  1)(2n  1) (2n  3)
.
Example 15. Prove that: I0

1
Pn2 (x) dx 
1
2n  1
.
Sol. We know that by orthogonal property,

I–1
1
Pn2 ( x) dx 
2
2n  1
 I1
0
Pn2 ( x) dx 
Put x = y in first integral, then dx = dy

I0
1
Pn2 ( x) dx 
2
2n  1
I
0
1
Pn2 ( y) dy  I0
1
Pn2 ( x) dx 
2
2n  1
 I
0
1
Pn2 ( x) dx  I0
1
Pn2 ( x) dx 
2
2n  1
 I 0
1
( 1) 2n Pn2 ( x) dx  I1
0
Pn2 ( x) dx 
2
2n  1
 2 I0
1
Pn2 ( x) dx 
2
2n  1
 I0
1
Pn2 ( x) dx 
1
2n  1
.
I1
1
Pn (x) (1  2xt  t 2 )1/2 dx 
2t n
2n  1
where n is a positive integer.

Sol.
I 1
Pn ( x) (1  2 xt  t 2 ) 1/2 dx
Equations Solutions
I
1
1
= Pn ( x) {  t n Pn ( x)} dx
1
= tn
I 1
1
Pn2 ( x) dx
I
All other terms vanish since
1
1
Pm ( x) Pn ( x) dx  0 ; m  n
NOTES
2
= tn . . | By II orthogonal property
2n  1

I
1
1
x m Pn (x) dx  0 , if m < n.
Sol. I 1
1
x m Pn ( x) dx  I1
1
xm
1
2 n n ! dx n
dn
(x2 1)n dx (Using Rodrigues Formula)
2n n !
=
1
I 1
1
xm
dn
dx n
(x2 1)n dx
Integrating by parts, we get

% "#
()
I
1
dn  1 dn  1
1
&x
1
m
( x 2  1) n mx m  1 ( x 2  1) n dx
=
2 n
n! '
! dx n1
* 1

1 dx n1
#$
=0
m
2n n ! I1
1
xm 1
dn 1
dx n  1
(x2 1)n dx
Similarly,
1 I 1 m(m  1)
x m Pn ( x) dx  ( 1) 2
2n n !
Integrating (m 2) times in all, we get
I 1
1
xm 2
dn  2
dx n  2
(x2 1)n dx
I = ( 1)m
m(m  1) ... 1
2n n ! I 1
1
dn m
dx n  m
(x2 1)n dx
=
( 1) m m !
2n n ! I 1
1
dn m
dx n  m
(x2 1)n dx
=
( 1) m m ! d n  m  1
( x 2  1) n
"# 1
2n n ! dx n  m  1 ! $ 1
= 0.

Pn  1  Pn  1
2n  1
 I Pn dx  c .
Sol. From Recurrence relation (3), we have

Pn+1 Pn1 = (2n + 1) Pn
Integrating both sides w.r.t. x, we get
Pn  1  Pn  1
2n  1
 I Pn dx  c .
Example 19. Prove that: xPn = nPn + (2n 3) Pn2 + (2n 7) Pn4 + ...
Sol. From Recurrence relation (2),
xPn = n Pn + Pn1 ...(1)

Ordinary Differential From Recurrence relation (3),
Equations
Pn+1 = (2n + 1) Pn + Pn1 ...(2)
Replacing n by n 2 in (2), we get
NOTES Pn1 = (2n 3) Pn + Pn3 ...(3)

Replacing n by n 4 in (2), we get
Pn3 = (2n 7) Pn4 + Pn5 ...(4)
 From (1), (3) and (4),
xPn = nPn + (2n 3) Pn + (2n 7) Pn4 + ...
Example 20. Prove that: I1

1
xPn Pn dx 
2n
2n  1
.
I I
Sol.
1 1
Pn ( x Pn ) dx = Pn [nPn  ( 2n  3) Pn  2  ( 2n  7) Pn  4  ... ] dx
I I I
1 1
1 1 1
= nPn2 dx  ( 2n  3) Pn Pn  2 dx  ( 2n  7) Pn Pn  4 dx  ...
1 1 1
2
=n. + 0 + 0 + ... | Using orthogonal property
2n  1
2n
=
2n  1
Example 21. Prove that: I1

1
(x 2  1) Pn  1 Pn dx 
2n(n  1)
(2n  1) (2n  3)
.

n(Pn1 xPn) = (1 x2) Pn
 (x2 1) Pn = n(xPn Pn1) ...(1)
Now,
I 1
( x 2  1) Pn  1 Pn dx = I 1
{( x 2  1) Pn } Pn  1 dx
I I I
1 1
1 1 1
= n ( xPn  Pn  1 ) Pn  1 dx = n xPn Pn  1 dx  n Pn  1 Pn  1 dx
I ' I
1 1 1
1 1
=n xPn Pn  1 dx ...(2) Pn  1 Pn  1 dx  0
1 1
From Recurrence relation (1),

(2n + 1) xPn = (n + 1) Pn+1 + nPn1
(n  1) Pn  1  nPn  1
 xPn =
(2n  1)
 From (2), I
1
1
( x 2  1) Pn+1 Pn dx = n I !
1
1
(n  1) Pn  1  n Pn  1
2n  1
"# . P
$ n+1 dx
=
n (n  1)
2n  1 I1
1
P n  1 dx 
n2
2n  1 I1
1
Pn  1 Pn  1 dx
n (n  1) 2 2n (n  1)
= . +0= .
(2n  1) 2 (n  1)  1 (2n  1) (2n  3)



xn  1 1 1 x Legendres Differential
Equations Solutions
n0
n1
Pn (x)  log
2 1 x
.

Sol. We know that
 NOTES
h
n0
n Pn(x) = (1 2xh + h2)1/2
Integrating both sides w.r.t. h from 0 to h, we get


n0
hn  1
n1
Pn ( x)  I 0
h dh
1  2hx  h 2
= I h
0
dh
(h  x) 2  (1  x 2 )
; if | x | < 
(h  x)  h2  2hx  1
= log
1 x
Putting h = x in the expression, we get
 1 – x2 1 1  x

xn  1
P (x) = log = 2 log 1  x .
n0
n1 n 1 x
0 ,  1 x  0 ()
Example 23. If f(x) =
x , 0 x1
show that:
*
1 1 5 3
f(x) = P0 (x)  P1 (x)  P2 (x)  P (x) + ...
4 2 16 32 4
Sol. We know that
f(x) =
n0

 an Pn ( x) ...(1) where an = n 

1
2

I
–1
1
f ( x)Pn ( x) dx
=
2n  1
2 ! I 0
1
f ( x) Pn ( x) dx  I 1
0
f ( x) Pn ( x) dx
"#
$
=
2n  1
2 I
0
1
f ( x) Pn ( x) dx ...(2)
Putting n = 0, 1, 2, 3, 4, ... successively in (2), we get
I 1 1
1 1 1 x2
a0 =
2 0
x P0 ( x) dx  4 2 2 0
I 3x 3 1
3 1 1
x P ( x) dx  
2 3
a1 = 1
2 0 2 0
a2 =
5
2 I
0
1
x P ( x) dx 
2
5
2
x I
3x  1 dx  5
2 16
1
0
2
a3 =
7
2 I
0
1
x P ( x) dx 
3
7
2
x
2 I
5x  3 x dx = 0
0
1 3
a4 =
9
2 I
1
0
x P ( x) dx 
4
9
2
x
I
35x  30 x  3 dx   3 and so on.
8
1
0
32
4 2
Putting these values in (1), we get

1 1 5 3
f(x) = P0 ( x)  P1 ( x)  P2 ( x)  P (x) + ...
4 2 16 32 4
Ordinary Differential Example 24. Compute the first three non-vanishing terms in the Fourier-Legendre
Equations series over the interval ( 1, 1) of the function
%K 1 ,
f(x) = &K 2 |x| 
.
NOTES  |x| 1 '0 ,
Sol. Let the Fourier-Legendre series be

f(x) = a n Pn ( x) ...(1)
I I
n0
2n  1 1 2n  1 
where, an = f ( x) Pn ( x) dx = f ( x) Pn ( x) dx ...(2)
2 1 2 
Putting n = 0, 1, 2, 3, ... successively in (2), we get
a0 =
1
2 I 1
  2
P0 ( x) dx =
1
4
(2) =
1
2
|Q P0 (x) = 1
3  1
a1 =
2   2
Therefore an = 0 for all n odd.
I
P1 ( x) dx =
3
4 I 

x dx = 0
Now, a2 =
5
2 I 1
  2
P2 ( x) dx =
5
4 I 
 3x2  1
2

dx =
5
8

x3  x


5 3 5
 a2 = ( ) = (2 1)
4 4
Also, a4 =
9  1
2   2 I
P4 ( x) dx
=
9
4 I 

35 x 4  30 x 2  3
8
dx =
9
32

7 x 5  10 x 3  3 x



9
 a4 = (75 102 + 3)
16
and so on ......
Hence the Fourier-Legendre series is
1 5 9
f(x) = P (x) + (2 1) P2(x) + (75 102 + 3) P4 (x) + .....
2 0 4 16
n
Example 25. Prove that:  (2r  1) P
r0
r = Pn + Pn+1

(2n + 1) Pn = Pn+1 Pn1
Putting n = 1, 2, 3, ..., n, we get
3P1 = P2 P0
5P2 = P3 P1
7P3 = P4 P2
: :
(2n 1) Pn1 = Pn Pn2
(2n + 1) Pn = Pn+1 Pn1

Adding simultaneously, we get Legendres Differential
Equations Solutions
3P1 + 5P2 + 7P3 + ... + (2n + 1) Pn = Pn+1 + Pn P0 P1
= Pn+1 + Pn 1 = Pn+1 + Pn P0
 ...
P0 + 3P1 + 5P2 + 7P3 + + (2n + 1) Pn = Pn+1 + Pn
n
NOTES
  (2r  1) P
r0
r = Pn+1 + Pn.
Example 26. If Pn(x) is a Legendre polynomial of degree n and  is such that

Pn() = 0. Show that Pn1 () and Pn+1() are of opposite signs.
(2n + 1) xPn(x) = (n + 1) Pn+1(x) + nPn1(x) ...(1)
Given that Pn() = 0 ...(2)
 Put x =  in (1) and using (2), we get
(2n + 1) . 0 = (n + 1) Pn+1() + nPn1()
Pn  1 () n
  ...(3)
Pn  1 () n1
Since n is a positive integer so RHS of (3) is negative. Hence (3) shows that
Pn+1() and Pn1() are of opposite signs.
Example 27. Show that all the roots of Pn(x) = 0 are real and lie between
1 and 1.
Sol. Let f(x) = (x2 1)n = (x 1)n (x + 1)n ...(1)
From (1), we see that f(x) vanishes for x = 1 and x = 1 hence by Rolles theorem,
f (x) must vanish at least once for some value  of x lying between 1 and 1.
From (1), we have
f (x) = n(x 1)n1 (x + 1)n + n(x 1)n (x + 1)n1
which shows that f (x) vanishes at x = 1 and x = 1.
Again, we have already shown that f (x) vanishes at x = . Now applying Rolles
theorem to f (x) two times, we conclude that f (x) must vanish at x =  between
 and 1.
Proceeding in this manner, we conclude that f (n)(x) = 0 must have n real roots
lying between 1 and 1.
By Rodrigues formula and eqn. (1), we have
1
Pn(x) = f (n)(x)
n
2 n!
Hence we see that Pn(x) = 0 must have n real roots lying between 1 and 1.

Ix
1
Pn (x) dx = (Pn1 Pn+1)/(2n +1).

(2n + 1) Pn(x) = Pn+1(x) Pn+1(x)
1 d
 Pn(x) = [P (x) Pn1(x)]
2n  1 dx n+1

Ordinary Differential Integrating w.r.t. x between limits x to 1, we get
I "#
Equations 1
1 1
Pn ( x) dx  Pn  1 ( x)  Pn – 1 ( x)
x 2n  1 ! #$ x
NOTES
1
= [{Pn+1(1) Pn1(1)} {Pn+1(x) Pn1(x)}]
2n  1
=
1
{P (x) Pn+1(x)}.
' Pn  1 (1)  1
2n  1 n1 and Pn  1 (1)  1
Example 29. Prove that: I 1
1
x 2 Pn2 dx 
1

3

1
8 (2n  1) 4(2n  1) 8 (2n  3)
.

(2n + 1)xPn = (n + 1) Pn+1 + nPn1
Squaring both sides, we get
(2n + 1)2 x2Pn2 = (n + 1)2 P2n+1 + n2 P2n1 + 2n(n + 1) Pn1 Pn+1
Integrating w.r.t. x between limits 1 to 1, we get
(2n + 1)2 I
1
1
x 2 Pn 2 dx = (n + 1)2
I1
1
Pn2  1 dx  n2 I1
1
Pn2  1 dx + 2n (n + 1) I
1
1
Pn – 1 Pn  1 dx
2 2
= (n + 1)2 .  n2 . +0
[2 (n  1)  1] [2 (n  1)  1]

I 1
1
x 2 Pn2 dx =
2
(2n  1) 2
2n  3

(n  1) 2
!

n2
2n  1
"#
$
1 3 1
=  
8 (2n  1) 4 (2n  1) 8 (2n  3)
(on resolving into partial fractions)
Example 30. Show that:

I 1
1
xPn (x) Pn 1 (x) dx 
2n
4n 2  1
.
Sol. Recurrence relation (1) is

nPn = (2n 1) x Pn 1 (n 1) Pn 2 ...(1)
1
From (1), x Pn 1 = [nPn + (n 1) Pn 2]
2n  1
Therefore, I1
1
xPn ( x) Pn  1 ( x) dx
=
1
2n  1 I 1
1
[ nPn 2  (n  1) Pn Pn 2 ] dx
n 2
=

2n  1 2n  1 | Using orthogonal property
2n
= .
4n2  1

I 1
[Pn  (x)] 2 dx  n(n  1) .
Equations Solutions
I
1
1
2
Sol. [Pn  ( x)] dx
1
= I1
1
[(2n  1) Pn  1  (2n  5) Pn  3  (2n  9) Pn  5  ...]2 dx
| By Cristoffels expansion formula

NOTES
= I1
1
(2n  1) 2 P 2 n  1 dx  I 1
1
(2n  5) 2 P 2 n 3 dx  I 1
1
(2n  9) 2 P 2 n  5 dx  ...
+2
I 1
1
(2n  1) (2n  5) Pn  1 Pn 3 dx  2
2 2
I1
1
(2n  1) (2n  9) Pn  1 Pn  5 dx  ...
= (2n 1)2 .  (2n  5) 2 .

2 (n  1)  1 2 (n  3)  1
2
+ (2n 9)2 . + ...... + 0 + 0 + 0 + ...
2 (n  5)  1
| By orthogonal properties
2 2 2
= (2n 1)2 .  (2n  5) 2 .  (2n  9) 2 . + ...
2n  1 2n  5 2n  9
= 2 [(2n 1) + (2n 5) + (2n 9) + ... + 1]
Here l = a + (N 1)d
n1
1 = (2n 1) + (N 1) ( 4)  N=
2
n1
 No. of terms in above series =
2
1 n1 ' n
 Sn 1  2 .
2 2
[(2n 1) + 1] = n(n + 1) Sn 
2
(a  l)
I
2
1
Hence, [Pn  ( x)]2 dx = n(n + 1).
1
Example 32. Prove that (1 2xz + z2) 1/2 is a solution of the equation
2    
z (z )  (1  x 2 ) = 0.
z 2 x  x 
 
Sol. We have,  = (1 2xz + z2)1/2 = 
n0
z n Pn ( x) or z = 
n0
z n  1Pn ( x)
2 

 z
z 2
( z) 
n0

(n  1) n zn Pn(x) ...(1)


Also, 
x n  0 
z n Pn  ( x)
 %&   ()  "#
 (1  x 2 )
'  (1  x 2 ) z n Pn  ( x)
*  #$
x x x n0 !  
= (1 x2) 
n0
z n Pn ( x)  2 x z
n0
n
Pn ( x) ...(2)

Ordinary Differential Substituting in LHS of the given equation, we have
Equations 

n0
[(n + 1)n zn Pn(x) + (1 x2) zn Pn(x) 2xzn Pn(x)]

NOTES
= z
n0
n [(1 x2) Pn (x) 2x Pn (x) + n(n + 1) Pn(x)]
= 0. | ' Pn(x) is a solution of Legendres equation
EXERCISE
1. Show that:
1 2 2 3
(i) x2 = P0 ( x)  P2 ( x) (ii) x3 = P3 ( x)  P1( x)
3 3 5 5
8 7 27 "
5
(iii) x = P5 ( x)  P3 ( x)  P1 ( x) #
63 ! 2 8 $
2. Express the following in terms of Legendres polynomials:

(i) 1 + x x2 (ii) x4 + 3x3 x2 + 5x 2
(iii) 5x3 + x (iv) x3 5x2 + 6x + 1
(v) 4x3 2x2 3x + 8 (vi) x4 + 2x3 + 2x2 x 3
3 2
(vii) 2x + 2x x 3
3. Expand x4 3x2 + x in a series of the form CrPr(x).
0,  1 x  0 ()
4. Expand f(x) in a series of Legendre polynomials if f(x) =
2x  1 , 0  x  1
.
*
5. Obtain the Fourier-Legendre expansion of f(x) defined as
0 ,  1 x  0 ()
f(x) = 1 , 0x1 . *
6. Express the function f(x) = ()
0 ,  1  x  0 in Fourier-Legendre expansion.
x2 , 0x1 *
x
7. Expand f (x) = cos in Fourier-Legendre series.
2
8. Prove that:
(i)
I 1
P2n ( x) P2n  1 ( x) dx 
I 1
P2n ( x) P2 n  1 ( x) dx (ii)
I 1
x6 P4 ( x) dx 
16
I
0 0 2 231
1
(iii) P2n  1 ( x ) dx  ( 1)n .
0
9. Prove that:
(i)
I 1
Pn ( x) dx 
1 P
n1
n1 (0) (ii)
I 1
P32 ( x) 
2
7
I
–1
I
0
1 1
(iii) P n  1 ( x) P  n  1 ( x) dx  n (n  1) (iv) P2n ( x) dx  0
–1 0
Answers
2 2
2. (i) P0(x) + P1(x) P2(x)
3 3
8 6 2 34 224
(ii) P (x) + P3 ( x)  P2 ( x)  P1 ( x)  P (x) (iii) 2P3 (x) + 4P1(x)
35 4 5 21 5 105 0
2 10 33 2
(iv) P (x) P2 ( x)  P1 ( x)  P0(x)
5 3 3 5 3

22 3 4 8 Legendres Differential
(v) P0(x) P (x) P (x) + P (x) Equations Solutions
3 5 1 3 2 5 3
8 4 40 1 224
(vi) P4 ( x)  P3 ( x)  P2 ( x)  P1 ( x)  P0 ( x)
35 5 21 5 105
NOTES
4 4 1 7
(vii) P3 ( x )  P2 ( x )  P1 ( x )  P0 ( x )
5 3 5 3
4 10 8 7 5 7
3. P (x) + P1(x) P2 ( x)  P (x) 4. P0(x) + P (x) + P2 ( x) – P (x) + ...
5 0 7 35 4 4 1 8 16 3
1 3 7 1 3 1 7
5. P0 ( x)  P1( x)  P (x) + ... 6. P (x) + P1(x) + P2 ( x)  P (x) + ...
2 4 16 3 6 0 8 3 48 3
x
7. cos = 0.6366 P0 0.6871 P2 + .0518 P4 .0013 P6 + ...
2

Equations
NOTES
7. BESSELS DIFFERENTIAL
EQUATION
STRUCTURE
Introduction
Solution of Bessels equation
Series Representation of Bessel functions
Recurrence Relations for Jn(x)
Generating function for Jn(x)
Integral Form of Bessel Function
Equations Reducible to Bessels Equation
Modified Bessels Equation
BER and BEI Functions
Orthogonality of Bessel Functions
Fourier-Bessel Expansion of F(x)
INTRODUCTION
The differential equation
d2 y dy
x2 2
x + (x2 n2)y = 0
dx dx
is called Bessels differential equation of order n, where n is a positive constant.
This equation can also be put in the form
d dy + (x

x x 2 n2) y = 0
dx dx
The particular solutions of this equation are called Bessels functions of order n.
SOLUTION OF BESSELS EQUATION
Bessels equation is
d2 y dy
x2 2
x + (x2 n2)y = 0 ...(1)
dx dx

d2 y dy Bessels Differential
Comparing equation (1) with the form  P( x) + Q(x) y = 0, we get Equation
dx 2 dx
1 n2
P(x) =
and Q(x) = 1 2
x x NOTES
At x = 0, both P(x) and Q(x) are not analytic  x = 0 is a singular point.
Also, xP(x) = 1 and x2 Q(x) = x2 n2
Since both xP(x) and x2 Q(x) are analytic at x = 0
 x = 0 is a regular singular point.

Assume y=
k0
a k xm  k

dy
Then,
dx
 
k0
( m  k) ak x m  k 1
2 
d y
and
dx 2
 
k0
( m  k) (m + k 1) ak xm+k2
dy d2 y
Substituting for y, and in (1), we get
dx dx 2
 
x2 
k0
(m  k ) (m  k  1) ak x m  k  2  x 
k0
(m  k ) ak x m  k 1
+ (x2 n2) a
k0
k xm k = 0
 
or  [(m  k)
k0
2
 (m  k)  (m  k)  n 2 ] ak x m  k  a
k0
k xm k  2 = 0
 
or  [(m  k)
k0
2
 n 2 ] ak x m  k  a
k0
k xm  k  2 = 0
The lowest power of x is xm corresponding to k = 0. Equating to zero the coefficient

m
of x , we get the indicial equation
m2 n2 = 0, since a0  0 whence m = ± n
Equating to zero the coefficient of next term i.e., xm+1, we get
[(m + 1)2 n2]a1 = 0
 a1 = 0, since (m + 1)2 n2  0 for m = ± n
Equating to zero the coefficient of xm+k+2, we get the recurrence relation
[(m + k + 2)2 n2] ak+2 + ak = 0
ak
or ak+2 =
(m  n  k  2) (m  n  k  2)
Putting k = 1, 3, 5, ....., we get a3 = a5 = a7 = ...... = 0
Putting k = 0, 2, 4, ....., we get
a0
a2 =
(m  n  2) (m  n  2)
a2 a0
a4 = 
(m  n  4) (m  n  4) (m  n  4) (m  n  4)(m  n  2) (m  n  2)
and so on.

Equations  y = a0xm 1 
x2

x4 "#
 ...
! (m  2) – n 2 2 2 2
[(m  2)  n ][(m  4 )  n ] 2 2
#$ ...(2)
Depending upon the values of n, we get different types of solutions.

NOTES
Case I. When n  0 or n  an integer.
In this case, we get two independent solutions for m = n and m = n.
For m = n, we get
y1 = a0xn 1 
x 2

x "
 ......#
4
! 2 (2n  2) 2 4 ( 2n  2)( 2n  4 ) #$
.
= a0xn
1  (  1) 1 x 2
 (  1)
x 2
"
 ...#
4
! 2
2 (1) ! (n  1) 2 ( 2) ! (n  1)(n  2) #$
4

( 1) k
= a0xn 
k0 2 2 k ( k) ! ( n  1) ( n  2) ...... (n  k)
x2k

( 1) k  (n  1)
 y1 = a0xn 
k0
2k
2 . k !  (n  k  1)
x2k ...(3)
Replacing n by n, the second independent solution corresponding to m = n is


( 1)k (  n  1)
y2 = a0xn 
k0
22k ( k ) ! (  n  k  1)
x 2k ...(4)
 The complete solution of equation (1) is y = c1y1 + c2y2

Since a0 is arbitrary, we can choose it in any manner.
1
Choose a0 = n , then (3) takes the form
2 (n  1)
xn

2
(  1) k  ( n  1)


(  1) k
x
n  2k
y1 =
2 n  ( n  1) k0
2k
. k !  ( n  k  1)
x 2k 
k0
( k ) ! ( n  k  1) 2
This is called Bessel function of the first kind of order n and is denoted by
Jn(x). Thus,
 (k) ! (n  k  1) 2

 n  2k
( 1) k x
Jn(x) =
k0
The solution corresponding to m = n is


( 1) k x  n  2k
Jn(x) =
k0
(k) !  ( n  k  1) 2
which is called Bessel function of the first kind of order n.
When n is not an integer, Jn(x) is distinct from Jn(x). Hence the complete solution
of the Bessels equation may be expressed as
y = AJn(x) + BJn(x) , where A and B are arbitrary constants.

Case II. When n = 0, the Bessels equation (1) takes the form Bessels Differential
Equation
d 2 y dy
x   xy = 0.
dx 2 dx
This is called Bessels equation of order zero. NOTES
The two roots of the indicial equation are equal, each = 0.
From equation (2), putting n = 0, we have (assuming a0 = 1)

y = xm 1 
x2

x4

x6 "#
 ......
! (m  2) 2
(m  2) (m  4 ) 2 2 2
(m  2) (m  4 ) (m  6) 2 2
#$
which is a solution if m = 0.
The first solution is given by


( 1) k x 2k
J0(x) =
k0
(k !) 2 2 , since (k + 1) = k !
which is Bessel function of the first kind of order zero.
Now,
y
 x m log x 1 

x2

x4
 ......
"#
m !
(m  2 )2 (m  2) 2 (m  4 ) 2 #$
+ xm
x 2
.
2

x4 %& 2  2 ()  ......"#
! (m  2) 2 m  2 (m  2 )2 (m  4 )2 'm  2 m  4 * #$
y
The second independent solution is given by
m m0
1 x  1 1  1 x  1 1  1  1 x  ......"#

2 .4 2 2 .4 .6 2 3
2 4 6
= J0(x) log x +
!2 2 2 2
#$ 2 2 2
x 1 1  1 x  1 1  1  1 x  ......"#
2 4 6
= J (x) log x + 
0
! 2 (2 !) 2 2 (3 !) 2 3 2
2
#$ 2
= J (x) log x + 

(  1) 1  1  1  ......  1 x
k 1 2k
0
k 1 ( k !) 2 3 2 k 2
It is denoted by Y0(x) and is called Bessel function of the second kind of
order zero or Neumann function.
Thus the complete solution of the Bessels equation of order zero is
y = AJ0(x) + BY0(x)
Case III. When n is an integer, the two functions Jn (x) and Jn(x) are not
independent but are connected by the relation
Jn(x) = (1) n Jn(x).
Now, when n is an integer, y2 fails to give a solution for positive values of n and
y1 fails to give a solution for negative values of n. Let us find an independent solution
of Bessels equation (1), when n is an integer.
Let y = u(x) Jn(x) be a solution of (1) when n is integral.
dy d2 y
Then  u J n  uJ n and  u J n  2u J n  uJ n
dx dx 2
dy d2 y
Equations Substituting the values of y, and in (1), we get
dx dx 2
x2(uJn + 2uJn + uJn) + x(uJn + uJn) + (x2 n2)uJn = 0
NOTES or u[x2Jn + xJn + (x2 n2)Jn] + x2uJn + 2x2uJn + xuJn = 0

or x2uJn + 2x2uJn + xuJn = 0 since Jn is a solution of (1).
Dividing throughout by x2uJn, we get
u J 1
2 n  =0
u Jn x
Integrating w.r.t. x, we get
log (uJn2x) = log B
I
or u Jn2x = B
B dx
or u = or u=B A
xJ 2n xJ 2n
Substituting the value of u in the assumed solution y = u(x) Jn(x), we have
y= B

! I dx
xJ 2n ( x)
"#
 A Jn(x)
#$
or y = AJn (x) + BYn(x), where
xJ 2n ( x)
The function Yn(x) is called the Bessel function of the second kind of order
Yn(x) = Jn(x) I dx
n or Neumann function.
SERIES REPRESENTATION OF BESSEL FUNCTIONS
J (x) = 

(  1) k
x
n  2k
Since n
k0

k !  (n  k  1) 2
J (x) = 

( 1) x =  ( 1) x , Q
k 2k  k 2k
k !  ( k  1) 2 ( k !) 2
 0 2 (k + 1) = k !
k0 k0
x 1 x  1 x  ......
=1 
2 4 6
2 (2 !) 2 (3 !) 2
2 2
x2 x4 x6
= 1    ......
22 22 . 4 2 2 2 . 4 2 . 62


( 1) k x 1  2k


( 1) k x 1  2k
J1(x) =
k0
k ! ( k  2 ) 2 
k0
k ! ( k  1) ! 2
x 1 x
1
3
x 5
 ...... 
x x3 x5
......
= 
2 2! 2


2!3! 2  2  2 2
2 2 .4 2 .4 .6
In particular, J0(0) = 1 and J1(0) = 0.
The values of J0(x) and J1(x) are given in Jahnke Emdes tables to four decimal
places at intervals of 0.1.

Bessels Differential
RECURRENCE RELATIONS FOR Jn(x ) Equation
x Jn¢ = nJn x Jn+1

NOTES
We know that,
Jn =

( 1) r
 r ! n  r  1 2
x n  2r
r0


( 1)r ( n  2r ) 1 x n  2r  1
Jn =
r0
.
r ! n  r  1 2 2
Multiplying both sides by x and breaking it into two terms


( 1)r x n  2r


( 1)r x n  2r  1
xJn = n
r0
r ! n  r  1 2 +x
r 1
(r  1) ! n  r  1 2


( 1) s  1 x n  2s  1
= nJn + x
s0
s ! n  s  2 2 | Here r = s + 1
= nJn x

( 1) s
 s ! n  s  2 2
x ( n  1)  2s
s0
 x Jn = nJn xJn+1 ...(1)
xJn¢ = nJn + xJn1
We know that

( 1)r (n  2r ) x n  2r


( 1)r ( 2n  2r  n ) x n  2r
xJn = 
r0
r ! n  r  1 2 =
r0
r ! n  r  1 2
=n

(  1) r
 r ! n  r  1 2
x n  2r
+x


(  1) r ( n  r ) x
 r ! n  r  1 2
n  2r  1
r0 r0

( 1)r x n  2r  1
= nJn + x
r0
r ! n  r 2


( 1)r x n  1  2r
= nJn + x
r0
r ! (n  1)  r  1 2
xJn = nJn + xJn1 ...(2)
2Jn¢ = Jn1 Jn+1
Adding equations (1) and (2), we get

2xJn = x (Jn1 Jn+1)
 2Jn = Jn1 Jn+1 ...(3)

Ordinary Differential 2nJn = x(Jn1 + Jn+1)
Equations
Subtracting (2) from (1), we get
0 = 2n Jn x Jn1 xJn+1
NOTES  2nJn = x (Jn1 + Jn+1) ...(4)
d
(xn Jn) = xn Jn+1
dx
Multiplying eqn. (1) by xn1, we get

xn Jn = nxn1 Jn xn Jn+1
 xn Jn xn1 . nJn = xn Jn+1
d
(xn Jn) = xn Jn+1 ...(5)
dx
d
(xn Jn) = xn Jn1
dx
Multiplying eqn. (2) by xn1, we get
xn Jn = n xn1 Jn + xn Jn1
 xn Jn + nxn1 Jn = xn Jn1
d
 (xn Jn) = xn Jn1 ...(6)
dx
SOLVED EXAMPLES
Example 1. Prove that: Jn(x) = (1)n Jn(x).

Sol. Since  p is infinity (p > 0), we get terms in Jn(x) equal to zero till
r + 1 n  1 so that the series begins when r  n
Hence we can write,


( 1) r x  n  2r
Jn (x) =
rn
r!  n r 1 2
Putting r = n + s, we get


(  1) n  s
x
n  2s
Jn(x) =
s0
( n  s ) ! s  1 2


( 1)r x n  2r 
( 1)r
x
n  2r
= (1)n
r0
( n  r ) ! r  1 2 = (1)n 
r0

n  r  1 r ! 2
 Jn (x) = (1)n Jn(x).
Example 2. Prove that: J0 (x) = J1 (x).
d
Sol. We know that [xn Jn(x)] = xn Jn+1 (x)
dx
d
Putting n = 0, we get [J (x)] = J1(x)  J0(x) = J1(x).
dx 0

Example 3. Prove that: Bessels Differential
Equation
2 2
(i) J1/2(x) = sin x (ii) J1/2(x) = cos x .
x x
Sol. We know that, NOTES
xn 1  x  2
x "
 ......#
4
Jn(x) =
2n n  1 ! . .
2 ( 2n  2) 2 4 ( 2n  2)( 2n  4 ) #$ ...(1)
1
(i) Putting n = in (1),
2
x 1  x  x  ..."#
2 4
J1/2(x) =
2 3/2 ! 3! 5! #$
1 "
3 3
 ...# =
x x x 2
. x 
=
2.
1

x ! 3! 5! #$ x
sin x
2
1
(ii) Putting n = in (1),
2
x  1/ 2 1  x  x  ... =
2 4
J1/2(x) =
2  1/ 2 2 ! 4 !
1/2
2
x
cos x .

d n 2
n1 2 "#
[J 2(x) + J2n+1 (x)] = 2
dx n x
J n (x) –
!x
J n  1 (x) .
$
Sol. LHS = 2Jn Jn + 2Jn+1 Jn+1 ...(1)
But xJn = nJn xJn+1 | Recurrence relation (1)
n
 Jn = J Jn+1 ...(2)
x n
and also, xJn = nJn + xJn1 | Recurrence relation (2)
n
 Jn = J + Jn1
x n
n  1 J
or Jn+1 =
x n+1 + Jn ...(3)
Substituting these values of Jn and Jn+1 from (2) and (3) in eqn. (1), we get
n J n1
LHS = 2Jn
x n

 J n  1  2J n  1 
x
Jn  1  Jn

n 2 n  1 J
2
x
2
=2 Jn n 1 = RHS
x
Hence the result.
2 sin x  cos x 2  cos x  sin x .

(i) J3/2(x) =
x x (ii) J3/2(x) =
x x

Ordinary Differential Sol. By Recurrence relation (4), we have
Equations
2n Jn(x) = x [Jn1(x) + Jn+1(x)] ...(1)
2n
 Jn+1(x) = J (x) Jn1(x) ...(2)
x n
NOTES (i) Putting n = 1/2 in (2), we get
1
J3/2(x) = J (x) J1/2(x)
x 1/2

2 sin x "#
=
x x !  cos x
$ | Using results of Ex. 3
(ii) From equation (1),

2n
Jn1(x) =
J (x) Jn+1(x) ...(3)
x n
Putting n = 1/2 in (3), we get
1
J3/2(x) = J (x) J1/2(x)
x 1/2

2  cos x "#
=
x x!  sin x
$ | Using results of Ex. 3
2 3  x sin x  3 cos x "#

2
! x
Example 6. Prove that: J5/2 (x) =
x 2
x #$
2nJn(x) = x[Jn 1 (x) + Jn + 1 (x)]
2n
 Jn + 1 (x) = J (x) Jn 1 (x) ...(1)
x n
Putting n = 1/2, 3/2 in (1), we get
1
J3/2 (x) = J (x) J 1/2 (x) ...(2)
x 1/2
3
J5/2 (x) = J (x) J1/2 (x) ...(3)
x 3/2
From (2) and (3),

3 1 "#
J 1/ 2 ( x)  J  1/ 2 ( x) J1/2 (x)
J5/2 (x) =
!
x x $
3  1 J (x) 3 J (x) = 3  x 2 sin x  3 . 2
2
x
= 2
x 1/2 x x
1/2 x 2
x
cos x
2 3  x "
cos x # .
2
3
=
x ! x sin x 
2
x #$
48  8 J (x)  1  24 J (x) .
Example 7. Prove that: J (x) =
x x
4 3 1 2 0
x
Hence or otherwise find J6 (x) in terms of J0(x) and J1(x).
2nJn (x) = x [Jn 1 (x) + Jn + 1 (x)]
2n
 Jn + 1 (x) = J (x) Jn 1 (x) ...(1)
x n
Putting n = 1, 2, 3 in eqn. (1), we get Bessels Differential
Equation
1
J2 (x) =
[2J1 (x) xJ0 (x)] ...(2)
x
1
J3 (x) = [4J2 (x) xJ1 (x)] ...(3) NOTES
x
1
J4(x) = [6J3 (x) xJ2 (x)] ...(4)
x
From (2) and (3),
8 4
J3 (x) = 2
J 1 ( x)  J 0 ( x)  J 1 ( x)
x x
8  x J
=
2
4
x 2 1 ( x) 
x
J 0 ( x) ...(5)
Again from (4) and (5),

48  6 x J (x)  24 J ( x)  2 J
2
J4 (x) = x 3
x
1
x 2 0 1 ( x)  J 0 ( x)
=
48  8 J (x)  1  24 J (x) .
x x 3 x 1 2 0 ...(6)
Again, putting n = 4, 5 in eqn. (1), we get

8
J5 (x) = J (x) J3 (x) ...(7)
x 4
10
J6(x) = J (x) J4(x) ...(8)
x 5
From (7) and (8),
10 8 "#
J6(x) =
x x
J 4 ( x)  J 3 ( x) J4 (x)
! $
80  1 J (x) 10 J (x)
= x 2
x 4 3
=
80  1 48  8 J (x)  1  24 J (x)"#
x ! x x
2 x $
3 1 2 0
10 8
 1 J (x)  4x J (x)"#$ | Using eqn. (5) and
x ! x 2 1 0
(6)
3840  768  18 J (x) + 144  1920  1 J (x).
 J6(x) =
x x x 5 3 x x 1 2 4 0
Example 8. Prove that: I J 3 (x) dx  J 2 (x) 

2
x
J1 (x) = 0.
Sol. We know that,

d
[xn Jn (x)] = x n Jn + 1 (x)
dx
 I x  n J n  1 ( x) dx = x n Jn (x) ...(1)

Now, I J 3 ( x) dx  I x 2 [ x  2 J 3 ( x)] dx
I
Equations
= x2 . [ x2 J2 (x)] 2 x . [ x  2 J 2 ( x)] dx
NOTES = J2 (x) + 2
I x –1 J 2 ( x)dx
2
= J2 (x) + 2 [ x1 J1 (x)] = J2 (x) J 1 ( x)
I
x
2
or J 3 ( x) dx  J 2 ( x)  J 1 ( x) = 0.
x
Example 9. Prove that: I xJ02 (x) dx 

1 2 2
2
x [ J0 (x)  J12 (x) ]  c .
Sol. I x J 20 ( x) dx  J 20 ( x) .
x2
2
 I 2J 0 ( x) J 0 ( x) .
x2
2
dx  c
=
x2 2
2
J 0 ( x)  I x 2 J 0 ( x) { J 1 ( x)} dx  c [ ' J0 (x) = J1 (x)]
=
x2 2
2
J 0 ( x)  I xJ 1 ( x) . xJ 0 ( x) dx  c
=
x2 2
2
J 0 ( x)  I xJ 1 ( x) .
d
dx
[ xJ 1 ( x)] dx  c
| Using Recurrence relation
x2 2 [ xJ 1 ( x)]2 x2 2
= J 0 ( x)  c = [ J 0 ( x)  J 21 ( x)]  c .
2 2 2
Example 10. Prove that: 4 Jn (x) = Jn 2 (x) 2Jn (x) + Jn + 2 (x).
2Jn = Jn 1 Jn + 1 ...(1)
Differentiating, 2Jn = Jn 1 Jn + 1
or 4Jn = 2Jn 1 2Jn + 1 = (Jn 2 Jn) (Jn Jn + 2) | Using (1)
 4Jn = Jn 2 2Jn + Jn + 2
Example 11. Prove that: 4J0 (x) + 3J0 (x) + J3 (x) = 0.

Sol.We know that
J0 = J1
1
Differentiating gives, J0 = J1 = (J J2)
2 0
[By Recurrence relation 2Jn = Jn 1 Jn + 1]
1 1 1 1
Differentiating again, J0 = (J0 J2) = J0 + . [J1 J3]
2 2 2 2
1 1 1 1 1 1
=
2
J0  + J 1 J 3 =
4 4 2
J0 J0 J3
4 4
'
| J1 = J0
3 1
= J0 J3
4 4
 4J0 + 3J0 + J3 = 0.

d Bessels Differential
Example 12. Prove that: [xJn(x) Jn + 1 (x)] = x [Jn2 (x) J2n + 1 (x)]. Equation
dx
d
Sol. LHS = [ x  n J n ( x) . x n  1 J n  1 ( x)]
dx NOTES
d d
= x n Jn (x) [ x n  1 J n  1 ( x)]  x n  1 J n  1 ( x) . [ x  n J n ( x)]
dx dx
= xn Jn (x)  xn + 1 Jn (x) + xn + 1 Jn + 1 (x) [ xn Jn + 1 (x)]
d
dx
( x n J n ) = xn Jn 1 '
|
= xJn2 (x) xJ2n + 1 = x [Jn2 (x) J2n + 1] = RHS.
J n (x) 1
Example 13. Prove that: lim n
 n
; (n > 1).
x0 x 2  n1
Sol. We know that:
1  x 
xn x 2 "
 ......#
4
Jn (x) =
2  n 1 !n .
2 . ( 2n  2) 2 4 . ( 2n  2) ( 2n  4 ) #$
 lim
J (x)
n
 lim
1 1  x  x 2 "
– .....#
4
x0 x n
2  n 1 !
x0 n . . .
2 ( 2n  2) 2 4 ( 2n  2)( 2n  4 ) #$
1
= n .
2 n1
4 J 2
Example 14. Prove that: J2 (x) = 1 
x 2 1 (x) 
x
J0 (x) .
Sol. By Recurrence relation (2), we have

xJn = nJn + xJn 1 ...(1)
Putting n = 2, xJ2 = 2J2 + xJ1
2
 J2 =
J + J1 ...(2)
x 2
By Recurrence relation (1), we have
xJn = nJn xJn + 1 ...(3)
From (1) and (3), we have
nJn + xJn 1 = nJn xJn + 1
Putting n = 1, J1 + xJ0 = J1 xJ2
2
 J2 = J1 J0 ...(4)
x
2 2 4 2
 From (2), J2 =
x x
J 1  J 0 + J1 = 1  2 J 1  J 0 .
x x
2
Jn + 3 + J n + 5 =
(n + 4) Jn + 4 .
x
2nJn = x (Jn 1 + Jn + 1)

Ordinary Differential Replacing n by n + 4, we get
Equations
2
(n + 4) Jn + 4 = Jn + 3 + Jn + 5 .
x
Example 16. Prove that Jn (x) = 0 has no repeated root except at x = 0.
NOTES
Sol. Suppose, if possible,  is a double root of Jn (x) = 0
Then, Jn () = 0 and Jn () = 0 ...(1)
From Recurrence relations, we know that
n
Jn + 1 (x) = J (x) Jn (x)
x n
n
and Jn 1 (x) = J (x) + Jn (x)
x n
Using (1), we get Jn + 1 () = 0 and Jn 1 () = 0
which is inadmissible as power series cannot have the same sum function.
Hence Jn (x) has no repeated root except x = 0.
x2 Jn(x) = (n2 n x2) Jn (x) + x Jn + 1 (x) ; n = 0, 1, 2.
Sol. We have
xJn = nJn xJn + 1 ...(1) |By R.R. (1)
Diff., xJn + Jn = nJn xJn + 1 Jn + 1
 x2 Jn = (n 1) x Jn x2 Jn + 1 xJn + 1 ...(2)
By Recurrence relation (2),
xJn = nJn + xJn 1
 xJn + 1 = (n + 1) Jn + 1 + xJn ...(3)
 From (2), x2Jn = (n 1) [nJn xJn + 1] x [ (n + 1) Jn + 1 + xJn] xJn +1
 x2 Jn = (n2 n x2) Jn + x Jn + 1.
x
Example 18. Prove that: J n  (n  1) J n  1  (n  3) J n 3  (n  5) J n  5 – .....
2
2nJn = x(Jn 1 + Jn + 1)
 2(n + 1) Jn + 1 = x (Jn + Jn + 2) | Replacing n by (n + 1)
x x
 J n  (n  1) J n  1  J n  2 ...(1)
2 2
x x
 J n  2 = (n  3) J n  3  J n  4 ...(2)
2 2
 From (1) and (2),
x x
J n  (n  1) J n  1  (n  3) J n  3  J n  4
2 2
Continuing this way,
x
J n  ( n  1) J n  1  ( n  3) J n  3  ( n  5) J n  5 – .....
2

Example 19. If n > 1, show that: I0
x
x  n J n  1 (x) dx  n
2  n1
1
 x  n J n (x) .
Bessels Differential
Equation
Sol. We know that,

d
[ x  n J n ( x)]   x  n J n  1 ( x) | Recurrence relation NOTES
dx
Integrating it between 0 and x, we get
I0
x
x  n J n  1 ( x) dx   x  n J n ( x)
x
0
= xn Jn (x) + Lt
1
x0
J (x) "#
! x $
n
n
= xn Jn (x) + n
.
2 n1

2 n
J n  (n  2) J n  2  (n  4) J n  1 – ...... .
"#
Example 20. Prove that: Jn =
!
x 2 $
Sol. From Recurrence formula (2), we have
n
Jn =  J n  Jn  1
x
n 2
=  J n  [ nJ n  ( n  2 ) J n  2  ... ] |Using
x x
example 18
2 n
J n  (n  2) J n  2  ... .
"#
=
x 2 ! $
GENERATING FUNCTION FOR Jn(x)
x 1
z
2 z
The function e is called generating function.
Prove that
x 1 
z
z  n
(i) e2
z
= J n ( x)
n
x 1
z 
i.e., Jn (x) is the coefficient of zn in the expansion of e 2 z
.
x 1 
z 
(ii) e 2 z
=
n
 ( 1) n
J n ( x) z  n
x 1
z 
(1)n Jn (x) is the coefficient of zn in the expansion of e 2 z

i.e., .
Note. Above results are true if n is an integer.
Proof. We have,
x 1 xz x
z  
2 z
e = e2 e 2z
xz x 2
z 2 ..... "#
x x 2
1 ..... "#
= 1
! 2

2 2!
 1 
#$ !
2z 2z 2!

#$
Ordinary Differential (i) Coeff. of zn in this product
Equations
x n
1 x n2
1 x n4
1 . 1  ......
=
2 n!

2 .
( n  1) !

2 .
(n  2 ) ! 2!
NOTES


( 1) r x n  2r
=
r0
r ! ( n  r) ! 2 = Jn
(ii) Coeff. of zn in this product
x n
1 x n2
1 x n4
1 1
= 
2 n!
 –
2 .
(n  1) !
 
2 . .
(n  2) ! 2 !
 ......


( 1) r x n  2r
= (1) n
r0
r ! (n  r ) ! 2 = (1)n Jn.
INTEGRAL FORM OF BESSEL FUNCTION

x 1 
t
t  n
We know that e 2 t
= J n ( x)
n
= J0 (x) + tJ1 (x) + t2J2 (x) + t3J3 (x) + ...... + t1J1 (x) + t2 J2 (x) + t3J3 (x) + ......
= J0 (x) + tJ1 (x) + t2J2 (x) + t3J3 (x) + ...... t1J1 (x) + t2 J2 (x) t3J3 (x) + ......
'
[ Jn (x) = (1)n Jn (x)]
1 1
1

2 3
= J0 (x) + t  J1 (x) + t  2 J2 (x) + t  3 J3 (x) + ...... ...(1)
t t t
Put t = cos  + i sin 
1
 tn = cos n + i sin n and = cos n i sin n
tn
| By De-Moivres theorem
n 1 n 1
so that t  = 2 cos n and t  n = 2i sin n
tn t
Substituting these values in (1), we have
eix sin  = J0 (x) + 2i sin J1 (x) + 2 cos 2 J2 (x) + 3i sin 3 J3 (x) + ......
...(2)
Since eix sin = cos (x sin ) + i sin (x sin )
 Equating the real and imaginary parts in (2), we get
cos (x sin q) = J0 (x) + 2[J2 (x) cos 2q + J4 (x) cos 4q + ......] ...(3)
sin (x sin q) = 2[J (x) sin q + J (x) sin 3q + ........]
1 3 ...(4)
These are known as Jacobi series.
Multiplying both sides of (3) by cos n and integrating w.r.t.  between the
limits 0 and  (when n is odd, all terms on the RHS vanish; when n is even, all terms
on the RHS except the one containing cos n vanish), we get
I 
0
cos ( x sin ) cos n d 
%&0,
' J n
when n is odd
( x), when n is even

Similarly, multiplying (4) by sin n and integrating w.r.t.  between the limits Bessels Differential
0 and , we get Equation
I
0

sin ( x sin ) sin n d 
%&J
' 0,
n ( x), when n is odd
when n is even
NOTES
Adding, we get I 
[cos ( x sin ) cos n  sin ( x sin ) sin n] d = Jn (x)
I
0
1 
 Jn (x) = cos ( n  x sin ) d for all integral values of n.
 0
SOLVED EXAMPLES
Example 21. Use Jacobi series to prove that

[J0 (x)]2 + 2[J1 (x)]2 + 2[J2 (x)]2 + 2[J3 (x)]2 + ...... = 1.
Sol. The Jacobi series are
J0 (x) + 2J2 (x) cos 2 + 2J4 (x) cos 4 + ...... = cos (x sin ) ...(1)
and 2J (x) sin  + 2J (x) sin 3 + ...... = sin (x sin )
1 3 ...(2)
Squaring (1) and (2) and integrating w.r.t.  between the limits 0 and , and
remembering that if m, n are integers then
I0

cos 2 n d  I0

sin 2 n d 

2
and I0

cos m cos n d  I0

sin m sin n d = 0, m  n, we get
[J0 (x)]2  + 2[J2 (x)]2  + 2[J4(x)]2  + ...... = I 
0
cos 2 ( x sin ) d
Adding, we have
2[J1 (x)]2  + 2[J3 (x)]2  + ...... =
{[J0 (x)]2 + 2[J1(x)]2 + 2[J2(x)]2 + ......}

I 0

sin 2 ( x sin ) d

=
I 
0
[cos 2 ( x sin )  sin 2 ( x sin )] d 
[J0 (x)]2 + 2[J1(x)]2 + 2[J2(x)]2 + ...... = 1.

I
0

d   .
Example 22. Prove that: J0 (x) =

1
 I 0

cos (x cos  ) d .
Sol. We know that
x.
1 1
z 
1 1 1

J1  z 2  2 J 2  z 3  3 J 3 + ...

2 z
e  J0  z  ...(1)
z z z
1
Putting z = ei so that = ei
z
1 1
and z = 2 cos  ; z = 2i sin ,
z z

Ordinary Differential Eqn. (1) becomes,
Equations
e ix sin   J 0  ( 2i sin  ) J1  ( 2 cos 2 ) J 2  ( 2i sin 3 ) J 3  ( 2 cos 4 ) J 4  ......
Equating real parts,
NOTES
cos (x sin ) = J0 + 2 cos 2 J2 + 2 cos 4 J4 + ...... ...(2)

Putting  =   in (2), we get
2
   J
cos (x cos ) = J0 + 2 cos ( + 2)J2 + 2 cos 4 2 4 + .......
= J0 + ( 2 cos 2) J2 + (2 cos 4)J4 + ........
 I 0

cos ( x cos  ) d  J 0 I 0

d  2 J 2 I 
0
cos 2 d  2J 4 I0

cos 4  d  ... = J0
 J0 =
1
 I 
0
cos ( x cos ) d .
EXERCISE A
1. Show that:
2 3 sin x  3  x cos x"#

2
(i) J1/2 (x) = J1/2 (x) cot x (ii) J5/2 (x) = x
!x x #$2
2
(iii) [J1/2 (x)]2 + [J1/2 (x)]2 =
x
2 15  6 x2 15 "
(iv) J7/2(x) = 3 sin x  2
 1 cos x # .
x ! x x #
$
2. Show that:
1
(i) J0 = (J J0) (ii) J2 = J0 x1 J0
2 2
J 1 J  1
(iii) 2 =  0 (iv) J1(x) = J1(x) + J (x).
J1 x J0  x 2
3. Prove that:
d n d
(i) [ x J n (ax)] = ax n Jn 1 (ax) (ii) [J 2n ( x)] = x [ J 2 n  1 ( x)  J 2n  1 ( x)] .
dx dx 2n
4. Prove that:
(i) I J0 ( x) J1 ( x) dx  
1 2
J0 ( x) (ii) I x2 J0 ( x) J 1 ( x) dx 
1 2 2
x J 1 ( x)
I I
2 2
J 4 ( x) 1 2 4 8
(iii) dx   J 3 ( x)  2 J 2 ( x) (iv) J 5 ( x) dx   J 4 ( x)  J 3 ( x)  2 J 2 ( x)
I
x x x x x
(v) x 3 J 0 ( x) dx  x3 J 1 ( x)  2 x 2 J 2 ( x)
5.
(vi) I
0
Prove that:

x J 0 (x) dx 


J 1 ( ) .
(i) Jn (x) =
1 2
2 0 I
cos ( x sin   n) d
(ii) cos x = J0 2J2 + 2J4 ...... ; sin x = 2J1 2J3 + 2J5 .......
[Hint: Put  = /2 in Jacobi series]

(iii) cos (x cos ) = J0 2J2 cos 2 + 2J4 cos 4 ....... Bessels Differential
sin (x cos ) = 2J1 cos  2J3 cos 3 + 2J5 cos 5 ....... Equation

[Hint: Replace  by   in Jacobi series]
2
NOTES
(iv) J0 (x) =
1
 I 0

cos ( x sin ) d 
2
 I
0
/2
cos ( x sin ) d 
2
 I0
/2
cos ( x cos ) d
6.
(v) J0 + 2J2 + 2J4 + 2J6 + ..... = 1 (vi)
Show that Bessels function Jn (x) is an even function when n is even and is an odd
I 0
/2
x J 1/2 (2 x) dx = 1.
function when n is odd. [Hint: Jn ( x) = ( 1)n Jn(x)]

7. (i) Express J6 (x) in terms of J0 (x) and J1 (x).
(ii) Express J5 (x) in terms of J0 (x) and J1 (x).
8  1 J (x)  4 J (x) .
(iii) Show that J3(x) =
x 2
x
1 0
8. Show that:
(i) I x 2 J 1( x) dx = 2x J1 (x) x2J0 (x) + c
(ii) I x3 J3 ( x) dx = x3 J2 (x) 5x2 J1(x) 15 x J0 (x) + 15

I J0 ( x) dx .
9. Evaluate:
I x 4 J 1( x) dx .
Answers
3840  768  18 J (x) + 144  1920  1 J

7. (i) J6 (x) =
x x x
5 3 x x 1 2 4 0 (x).
384  72  1 J (x) + 12  192 J (x).

(x) =
(ii) J5
x 4
x 2
x x 1 3 0
9. (8x2 x4) J0(x) + (4x3 16x) J1(x).
EQUATIONS REDUCIBLE TO BESSELS EQUATION
A number of second order differential equations with variable coefficients can be

reduced to Bessels equation by a suitable transformation and, hence, can be solved in
terms of Bessel functions.
d2 y dy
x2
+ (1 2)x + [22 x2 + (2 n22)]y = 0 ...(1)
dx 2 dx
where , ,  and n are constants.
Putting X = x and Y = x y, equation (1) reduces to
d2 Y dY
X2 2
X  (X 2  n 2 ) Y  0 ...(2)
dX dX
which is Bessels equation.

Ordinary Differential When n is not an integer, the solution of (2) is
Equations
Y = c1Jn (X) + c2Jn (X)
and hence, the solution of (2) is

NOTES
x y = c1Jn (x) + c2Jn (x)
or y = x [c1Jn (x) + c2Jn(x)]
When n is an integer, the solution of (2) is

Y = c1Jn (X) + c2Yn (X)
and hence, the solution of (2) is
y = x [c1Jn (x) + c2Yn (x)].
Equation (1) is a general form of Bessels equation with , ,  and n as

parameters. Comparing the given equation with (1), we get specific values of the
parameters and hence the solution.
SOLVED EXAMPLES
Example 23. Solve the following differential equations in terms of Bessel

functions:
2 1

2
(i) y y + 4 x  2 y = 0 (ii) xy 3y + xy = 0.
x x
Sol. (i) The given equation is x2y 2xy + (4x4 4)y = 0
Comparing with the general form, we get
1 2 = 2, 22 = 4, 2 = 4, 2 n22 = 4
3 5
i.e.,  = 2
,  = 1,  = 2, n = 4
.
Here n is not an integer and the solution is
y = x3/2 [c1J5/4 (x2) + c2J5/4 (x2)]
(ii) Multiplying by x, the given equation becomes
x2y 3xy + x2y = 0
1 2 = 3, 22 = 1, 2 = 2, 2 n22 = 0
i.e.,  = 2,  =  = 1, n = 2
Here n is integer and the solution is y = x2[c1J2 (x) + c2Y2 (x)].
Example 24. Obtain in terms of Bessel functions, the solution of differential
equation
d2 y 20
dx 2
x
 9x  2 y = 0.

Sol. The given equation on multiplying by x2 is
d2 y
x2  (9 x 3  20) y  0
dx 2

Comparing this with the standard transformed equation Bessels Differential
Equation
x2 y + (1 2) xy + {22 x2 + (2 n22)} y = 0,
we get 1 2 = 0, 22 = 9, 2 = 3, and 2 n22 = 20
1 3 NOTES
This gives,  = ,  = ,  = 2, n = 3
2 2
Here n is an integer.
Hence the solution is y= x [c1J3(2x3/2) + c2Y3(2x3/2)]
Example 25. Solve the differential equation
y n2 y0
x
 4 x2  2
y 
x
in terms of Bessels functions.

x 2 y  xy  4 ( x 4  n 2 ) y  0 ...(1)
1 2 = 1 ...(2)
22 = 4 ...(3)
2 = 4 ...(4)
 2 m22 = 4n2 ...(5)
From (2), =0
From (4),  = 2
From (3), 2 = 1  =1
From (5), 0 m2 (4) = 4n2  m=n
when n is not an integer, solution to (1) is
y = x [c1 Jm (x) + c2Jm (x)]
= x0 [c1Jn (x2) + c2Jn (x2)] = c1Jn(x2) + c2Jn (x2)
when n is an integer, solution to (1) is
y = x [c1 Jm (x) + c2m (x)]
= x0 [c1Jn (x2) + c2Yn (x2)] = c1Jn (x2) + c2Yn (x2).
MODIFIED BESSELS EQUATION
2 dy d2 y
The differential equation x  (x 2  n2 ) y  0
2
x ...(1)
dx dx
is called modified Bessels equation of order n.
dy d2 y
Equation (1) can be re-written as x 2  (i 2 x 2 – n 2 ) y  0
2
x
dx dx
When n is not an integer, its solution is given by y = c1Jn (ix) + c2Jn (ix)



( 1) k ix
n  2k
Now, Jn (ix) =
k0

k !  (n  k  1) 2
 k!  (n  k  1) 2
 n  2k
1 x
=i
n
[' (1)k (i)2k = i2k . i2k = i4k = 1]
r0

Equations


1 x n  2k
The series
k0
k!  (n  k  1) 2 is a real function with all terms positive.
It is denoted by In (x) and is called the modified Bessel function of the first kind
NOTES of order n.
Thus, In(x) = inJn (ix)
Since in is a constant, In (x) is also a solution of (1).
If n is not an integer, a second independent solution of (1) is In (x),


1 x  n  2k
where In (x) =
r0
k !  ( n  k  1) 2
Thus, if n is not an integer, the complete solution of (1) is given by
y = c1In (x) + c2In (x)
If n is a non-zero integer, a second independent solution of (1) is given by
/2
Kn (x) =
[I (x) Jn (x)]
sin n n
and is called modified Bessel function of second kind of order n.
In this case, the complete solution of (1) is given by
y = c1In(x) + c2Kn(x).
BER AND BEI FUNCTIONS
d 2 y dy
Consider the differential equation x   ixy  0 ...(1)
dx 2 dx
Comparing it with equation (1) of Art. 4.7, we have
 = 0, n = 0,  = 1 and 2 = i or 2 = i3 so that  = i3/2.
Hence a solution of (1) is given by J0 (i3/2 x)
Replacing x by i3/2 x in the series for J0 (x), we have
i3 x 2 i6x 4 i9x 6 i12 x 8
J0 (i3/2 x) = 1     ......
22 ( 2!) 2 . 2 4 (3!) 2 . 2 6 ( 4 !) 2 . 2 8

= 1
x4

x8
– .....
"#
! 22 . 4 2 22 . 4 4 . 62 . 82 #$
+i
x 2

x6

x 10 "#
 .......
!2 2
2 . 4 2 . 62
2
2 2 . 4 2 . 6 2 . 82 . 10 2 #$
Thus J0 (i2/3 x) is a complex function for real values of x. The real and the
imaginary parts are denoted by ber (x) (Bessel-real) and bei (x) (Bessel-imaginary)
respectively. Thus

(  1) k x 4 k
ber(x) = 1 +
k 12

2 . 2 . 2 .......
4 6 ( 4k )2

(  1) k x 4 k  2
and bei(x) = 1 2
k 1
2 . 2 . 2 .......
4 6 ( 4k  2) 2
Hence a solution of (1) is y = J0(i3/2 x) = ber (x) + i bei (x).

Example 26. Show that: Bessels Differential
Equation
d d
(a) [x ber (x)] = x bei (x) (b) [x bei (x)] = x ber (x).
dx dx
Sol. We know that: NOTES
4 8
x x
ber (x) = 1    .......
22 . 4 2 22 . 4 2 . 62 . 82
x2 x6 x 10
bei (x) =    .......
22 22 . 4 2 . 6 2 2 2 . 4 2 . 6 2 . 82 . 102
x3 x7
(a) Now, ber (x) =    ........
22 . 4 22 . 4 2 . 62 . 8
x4 x8
 x ber (x) =    ........
22 . 4 22 . 4 2 . 62 . 8
d x3 x7
[x ber (x)] =  2  2 2 2  ........
dx 2 2 .4 .6
x 2
x6
= x 2 2

2 .4 .6
2 2 2
 ........ = x bei (x).
5
x x x9
(b) Also, bei (x) =  2 2  2 2 2 2  ........
2 2 . 4 . 6 2 . 4 . 6 . 8 . 10
x2 x6 x 10
 x bei (x) =  2 2  2 2 2 2  ........
2 2 . 4 . 6 2 . 4 . 6 . 8 . 10
d x5 x9
[x bei (x)] = x  2 2  2 2 2 2  ........
dx 2 .4 2 .4 .6 .8
x4 x8
= x 1 
2 2 . 4 2 22 . 4 2 . 62 . 8 2
 ........ = x ber (x).
ORTHOGONALITY OF BESSEL FUNCTIONS
If  and  are the roots of Jn (x) = 0, then
I0
1
x J n ( x) . J n ( x) dx 
%& 0 ,
'J
1
2
2
when   
n  1 () , when   
()
*
Consider the Bessels equations
x2u + xu + (2x2 n2)u = 0 ...(1)
and x2v + xv + (2x2 n2) v=0 ...(2)
Their solutions are u = Jn (x) and v = Jn (x) respectively.
v u
Multiplying (1) by and (2) by and subtracting, we get
x x
x(uv uv) + (uv uv) + (2 2) xuv = 0
d
or [x(uv uv)] = (2 2) xuv
dx

Ordinary Differential Integrating both sides w.r.t. x between the limits 0 and 1, we get
I "# "#
Equations 1
1
= u v  uv
(2 2)
0
xuv dx  x (u v  uv )
! $ 0 ! $ x1
...(3)
NOTES Since, u = Jn (x)

d d d (x)
 u = [J (x)] = ( J n (x)] .  J n  (x)
dx n d (x) dx
Similarly, v = Jn (x)  v = Jn (x)
Substituting for u, v, u and v in (3), we get
I
0
1
xJ n (x) J n (x) dx 
2   2
J n () J n ()  J n () J n ()
If  and  are distinct roots of Jn(x) = 0, then Jn () = 0 and Jn () = 0.

...(4)
Hence, from (4), we have
I0
1
xJ n (x) J n (x) dx  0
0
However, if  = , the value of the integral is , which is indeterminate.
0
To evaluate the integral, we assume that  is a root of Jn(x) = 0 so that Jn() = 0
and  is a variable approaching . Thus, from (4), we have
Lt
 0 I 1
xJ n (x) J n (x) dx  Lt

J n () J n ()
2   2
or I0
1
xJ 2n (x) dx  Lt

J n () J n ()
2
[by L-Hospitals rule]
1
= [J ()]2 ...(5)
2 n
But xJn (x) = nJn (x) xJn + 1 (x)
 xJn () = nJn () Jn + 1 () = Jn + 1 (), since Jn () = 0
 Jn () = Jn + 1 ()
Hence, from (5), we get
Note. If the interval is from 0 to a, it can be shown that

I0
1
xJ n 2 (x) dx 
1 2
2
J n  1 () .
I0
a
xJ n2 (x) dx 
a2 2
2
J n  1 (a) , where  is a root of Jn (a) = 0.
FOURIER-BESSEL EXPANSION OF f (x)
From the orthogonal property of Bessel functions, we can expand a function f(x) in
Fourier-Bessel series in the range 0 to a.

Let f(x) = c1Jn (1x) + c2Jn (2x) + ....... + cnJn (nx) + ........ = cJ
i1
1 n ( i x ) ...(1)
where 1, 2, ........ are the roots of the equation Jn (a) = 0.

To determine ci, we multiply both sides of (1) by xJn (ix) and integrate w.r.t. x Bessels Differential
between the limits 0 to a. From the orthogonal property of Bessel functions, all integrals Equation
on the right hand side will vanish except the one containing ci and we have
I a
xf ( x) J n ( i x) dx  ci I a
xJ n 2 ( i x) dx  ci .
a2 2
J n  1 ( i a) NOTES
I
0 0 2
2 a
 ci = 2 2
xf ( x) J n ( i x) dx
a J n1 ( ia ) 0
Putting i = 1, 2, 3, ....... we can find c1, c2, c3 , ........ and hence the function f(x).
SOLVED EXAMPLES
Example 27. If 1, 2, ..........., n are the positive roots of J0 (x) = 0, prove that

1 J0 (  n x)
 
2 n  1  n J1 (  n )
.
Sol. We know that if f(x) =  cJ

i1
i n ( i x) ...(1)
then ci = 2 2
a J n1
2
( i a) I
0
a
xf ( x) J n ( i x) dx
Taking f(x) = 1, a = 1 and n = 0, we get
I "# 1
2 1 2 xJ 1 ( i x) 2
ci = xJ 0 ( i x) dx  2 
2
J 1 ( i ) 0 J 1 ( i ) i ! $ 0
 i J 1 ( i )

2
 From (1), we have 1 = 
i1 i J 1 ( i )
J 0 ( i x)

1 J 0 ( n x)
or  
2 n  1  n J 1 ( n )
.
Example 28. Show that the Fourier-Bessel series in J2(nx) for f(x) = x2 (0 < x < a),
where n a are positive roots of J2 (x) = 0, is

J 2 (  n x)
x2 = 2a
2

n1
a n J 3 (  n a)
.
Sol. Let the Fourier-Bessel series representing f(x) = x2 be given by


x2 = cJ
n1
n 2 ( n x)
Multiplying both sides by xJ2(nx) and integrating w.r.t. x between the limits 0
to a, we get
I0
a
x 3 J 2 (  n x ) dx  cn I 0
a
xJ 22 (  n x ) dx
x J ( x) "#
3
3 n
a
a2 2
or
!  #$ n 0
= cn . J ( a)
2 3 n

Equations
a 3 J 3 ( n a) a2 2
or = cn . J ( a)
n 2 3 n
2a2 1
 cn = .
NOTES a n J 3 ( n a)

J 2 ( n x )
Hence, x2 = 2a2 
n1
a n J 3 ( n a)
.
EXERCISE B
1. Solve the following differential equations in terms of Bessel functions:

(i) xy + y = 0 (ii) xy y + 4x3y = 0
y = 0
1 1 1 1
(iii) y +
x
y + 4 1  2
x
(iv) y +
x
y + 3 
4 x2
y=0

20 y = 0.
(v) y + 9 
x 2 (vi) x2y xy + 4x2y = 0
1 1
(vii) y + y + 8  2
y = 0 (viii) 4y + 9xy = 0
x x
1 y  1 
(ix) xy + y + y = 0 (x) y +  1 y = 0
4 x  9x 2 
2. Expand f(x) = 1 over the interval 0 < x < 3 in terms of the functions J0 (nx), where n are
determined by J0(3) = 0.
3. Expand f(x) = 4x x3 over the interval (0, 2) in terms of Bessel functions of first kind of
order one which satisfy the condition [J1 (x)]x = 2 = 0.
4. If 1, 2, ........, n, ........ are the positive roots of J1 (x) = 0, prove that
 
J 0 ( n x)

1 1 J0 ( n x)
(i) x2 =
2
4
n1 n
2
J 0 ( n )
(ii) (1 x2)2 =
3
 64 
n1 n
2
J0 ( n )
.
5. If a is the root of the equation J0 (x) = 0, show that
(i) I
0
1
J 1 ( ax) dx 
1
a
(ii) I
0
a
J 1 ( x) dx  1 .
Answers
1. (i) y = x [ c1 J1 (2 x )  c2 Y1 (2 x )] (ii) y = x [c1J1/2 (x2) + c2J1/2 (x2)]
(iii) y = c1 J2 (2x) + c2 Y2 (2x) (iv) y = c1 J1/2 ( 3x) + c2 J1/2 ( 3x)
(v) y = x [c1J9/2 (3x) + c2J9/2 (3x)] (vi) y = x[c1J1(2x) + c2Y1(2x)]
(vii) y = c1J2 ( 4 2x ) + c2Y2 ( 4 2x ) (viii) y = x1/2 [c1J1/3 (x3/2) + c2J1/3 (x3/2)]
(ix) y = c1J0 ( x ) + c2Y0 ( x ) (x) y = c1J1/3 (x) + c2 J 1/3 (x)

 
J 0 ( n x) J 3 (2 n )
 
2
2. 1= . 3. 4x x3 = 8 . J1 (n x) .
3
n1
 n J 1 (3 n )
n1
2n J 22 (2 n )

Ordinary Differential Equations - BAMAT-201

Uploaded by

Copyright:

Available Formats

Ordinary Differential Equations - BAMAT-201

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ordinary Differential Equations - BAMAT-201

Uploaded by

Copyright:

Available Formats

What are differential equations and their types?

What are differential equations and their types?

How can differential equations be solved using separation of variables?

How can differential equations be solved using separation of variables?

ORDINARY DIFFERENTIAL

Self Learning Material

Directorate of Distance Education

SWAMI VIVEKANAND SUBHARTI UNIVERSITY

Copyright © Laxmi Publications Pvt Ltd.

2. EXACT DIFFERENTIAL EQUATIONS 70

3. LINEAR DIFFERENTIAL EQUATIONS OF THE FIRST ORDER

4. LINEAR DIFFERENTIAL EQUATIONS OF SECOND AND HIGHER ORDER

5. POWER SERIES SOLUTIONS

7. BESSEL’S DIFFERENTIAL EQUATION

Self-Instructional Material iii

Course Name: Ordinary Differential Equations (ODE) Course Code: BAMAT-201

1. Formulate Differential Equations for various Mathematical models.

An equation involving independent variables, dependent variables and at least one

Order and Degree of a Differential Equation

This equation is not free from radicals.

Linear Differential Equation

Example 1. Determine the order and degree, if defined, of the following

(iii) ( y)2 + (y)3 + (y)4 + y5 = 0 (iv) y + (y)2 + 2y = 0.

FORMATION OF A DIFFERENTIAL EQUATION WHOSE

If we have an equation between two variables, involving arbitrary constants,

I. Method of Forming a Differential Equation

Working Steps for the Formation of Differential Equations

Example 3. Form the differential equation of the following families of curves:

Example 4. Form the differential equation of the following families of curves:

This is the required differential equation.

(iii) xy = C cos x (iv) y =

SOLUTION OF A DIFFERENTIAL EQUATION

A solution of a differential equation is a functional relation between the variables

= bex(1  3 + 2) + ce2x(4  6 + 2) = bex(0) + ce2x(0) = 0.

Example 11. Show that y = A 2  x 2 , x  (A, A) is a solution of

A differential equation with given initial conditions is called an initial value

SOLUTION OF A DIFFERENTIAL EQUATION BY THE

This represents the general solution of the differential equation (1).

Example 14. Solve the differential equations:

y = log (ex + ex) + C.

    EQU  Z   UGE Z dx

Example 16. Solve the differential equation:

Example 17. Solve the differential equations:

Example 18. Solve the differential equations:

Example 19. Solve the differential equations:

Integrating, we get I  1  1 dx = 1

 log|tan x| = log |tan y| + C  log |tan x tan y| = C

 log |ex  1| = log |tan y| + C

 ezdz = dy (Variables are separate)

Now, y(0) = 2 implies log  2  = +C

Now, y(1) = 4 implies log |4| = log |1|+ C or C = log 4

Now, y  2  = 2 implies log sin

 The required particular solution is log |sin y| + sin x = 1.

Example 26. Solve the following initial value problems:

Solution. (i) We have

 (1)  0  1 = log |0  (1)| + C  1 = 0 + C  C = 1

This represents the general solution of the differential equation (1).

Example 27. Solve the differential equations:

Example 28. Solve the differential equations:

Solution. (i) We have

= bex(1 3 + 2) + ce2x(4 6 + 2) = bex(0) + ce2x(0) = 0.

Example 11. Show that y = A 2  x 2 , x  (A, A) is a solution of

y = log (ex + ex) + C.

 EQU Z UGE Z dx

Integrating, we get I 1  1 dx = 1

 log|tan x| = log |tan y| + C  log |tan x tan y| = C

 log |ex 1| = log |tan y| + C

Now, y 2 = 2 implies log sin

 (1)  0 1 = log |0 (1)| + C  1 = 0 + C  C = 1

 log |v| log |1 + v| = log C|x|

Integrating, we get I cot v dv = I dx

y y dx = x y sin y  x cos y dy.

2ev = log  y  + C  2ex/y + log | y | = C.

= log  y  + log C

 cos v = log y + C

Solve the following differential equations (Q. No. 125):

where V = Y/X, X = x h and Y = y k.

 y = e2x I e2x cos 3x dx + Ce2x ...(2)

 The solution is yemx = I emx emx dx + C.

 (2)  y etan x= (tan x 1)etan x + C

1 4 tan x  2x – 2 tan x "# + C = 2 tan x x 1 tan

I P dx = I EQV Z dx = 3 log |sin x| = 3 log sin x = log (sin x)3

10. (i) y log x = (log x)2 + C (ii) [G Z

= I (sin x)2x dx + I x2 cos x dx

This is a Bernoullis equation.

NOTES  (sec y)ex = I x  ex dx + C

Example 62. Solve: y dx + (x y3) dy = 0.

equation to variable separable form.

Example 1. Solve (5x4 + 3x2y2 2xy3) dx + (2x3y 3x2y2 5y4) dy = 0.