Krishnas - Differential Equations Vector Calculus by

h n a's
Kris TEXT BOOK on

D ifferential E quations
& V ector C alculus
As per Allahabad State University Allahabad (U.P.)
(w.e.f. 2018-2019)
By
A. R. Vasishtha
Retired Head, Dep’t. of Mathematics
Meerut College, Meerut (U.P.)
Shilpi Singh Manoj Kumar Mishra

M.Sc., M.Phil., NET M.Sc., Ph.D.
Astt. Prof., Dep’t. of Mathematics Dep’t. of Mathematics
H.N.B.(P.G.) College Naini Allahabad (U.P.) P.B.(P.G.) College, Pratapgarh (U.P.)
(Allahabad Edition)
KRISHNA Prakashan Media (P) Ltd.

KRISHNA HOUSE, 11, Shivaji Road, Meerut-250 001 (U.P.), India
Jai Shri Radhey Shyam
Dedicated
to
Lord
Krishna
Authors & Publishers
Preface
This book on Differential Equations & Vector Calculus has been specially
written according to the latest Allahabad State University Syllabus to meet the
requirements of the B.A. and B.Sc. Part-II Students of Allahabad State
University in Uttar Pradesh.
The subject matter has been discussed in such a simple way that the students
will find no difficulty to understand it. The proofs of various theorems and
examples have been given with minute details. Each chapter of this book
contains complete theory and a fairly large number of solved examples.
Sufficient problems have also been selected from various university examination
papers. At the end of each chapter an exercise containing objective questions has
been given.
We have tried our best to keep the book free from misprints. The authors
shall be grateful to the readers who point out errors and omissions which, inspite
of all care, might have been there.
The authors, in general, hope that the present book will be warmly received
by the students and teachers. We shall indeed be very thankful to our colleagues
for their recommending this book to their students.
The authors wish to express their thanks to Mr. S.K. Rastogi, M.D.,
Mr. Sugam Rastogi, Executive Director, Mrs. Kanupriya Rastogi, Director and
entire team of KRISHNA Prakashan Media (P) Ltd., Meerut for bringing
out this book in the present nice form.
The authors will feel amply rewarded if the book serves the purpose for
which it is meant. Suggestions for the improvement of the book are always
welcome.
— Authors
Syllabus
& Vector Calculus
As per Allahabad State University
(w.e.f. 2018-19)
B.A./B.Sc. Paper-II M.M. : 50
Unit-1: Differential equation, formation of a differential equation, order and degree,

methods of solution of a differential equation of first order and first degree,
method of separation of variables, solution of homogeneous equation, linear
equation and exact equations.
Unit-2: Linear differential equations with constant, coefficients, homogeneous linear

differential equations. Differential equations of the first order but not of the
first degree. Clairaut's equations and singular solutions, orthogonal trajec-
tories.
Unit-3: Simultaneous linear differential equations with constant coefficients, general

linear differential equations of the second order, method of solution in terms of
one unknown integral belonging to the complementary function. Method of
solution by transformation of the equation to the normal form,
transformation of the equation by changing the independent variable. Method
of operational factors, method of variation of parameters.
Unit-4: Vector differentiation and integration, gradient, divergence and curl, and their
properties.
Unit-5: Line integrals, surface and volume integrals, Gauss, Green and Stokes
theorems and problems based on these.
B rief C ontents
Dedication.........................................................................(v)
Preface ...........................................................................(vi)
Syllabus ........................................................................(vii)
Brief Contents ...............................................................(viii)
Chap-1: Differential Equations of First Order and First Degree..............D-01—D-54
Chap-2: Differential Equations of the First Order but not of the
First Degree..........................................................................D-55—D-82
Chap-3: Orthogonal Trajectories........................................................D-83—D-94
Chap-4: Linear Differential Equations with Constant Coefficients........D-95—D-136
Chap-5: Homogeneous Linear Differential Equations.......................D-137—D-150
Chap-6: Ordinary Simultaneous Differential Equations.....................D-151—D-172
Chap-7: Linear Equations of Second Order with Variable
Coefficients......................................................................D-173—D-214
Chap-8: Differentiation of Vectors...................................................D-215—D-240
Chap-9: Gradient, Divergence and Curl...........................................D-241—D-286
Chap-10: Integration of Vectors.........................................................D-287—D-296
Chap-11: Line Integrals....................................................................D-297—D-314
Chap-12: Green's, Gauss's and Stoke's Theorems...............................D-315—D-356

Krishna's
DIFFERENTIAL EQUATIONS
AND VECTOR CALCULUS
C hapters
Differential Equations of First Order

1. and First Degree
Differential Equations of The First

2. Order But Not of The First Degree
1. Orthogonal Trajectories
3.
Linear Differential Equations with

4. Constant Coefficients
Homogeneous Linear Differential
5. Equations (Cauchy-Euler Equations)
Ordinary Simultaneous Differential

6. Equations
Linear Equations of Second Order

7. with Variable Coefficients
1. Differentiation of Vectors
8.
1. Gradient, Divergence and Curl

9.
1. Integration of Vectors
10.
1. Line Integrals
11.
1. Green's, Gauss's and Stoke's Theorems

12.
D-3
1
D ifferential E quations of
F irst O rder and F irst D egree
1 Definitions
differential equation is an equation containing the dependent and independent
A variables and different derivatives of the dependent variables w.r.t. one or more
independent variables.
The order of a differential equation is the order of the highest derivative (or
differential coefficient) occurring in the equation.
(Lucknow 2007; Meerut 09B, 10B; Bundelkhand 10)
The degree of a differential equation is the degree of the highest derivative (or diff.
coeff.) which occurs in it, after the differential equation has been rationalized (i. e.,
made free from radicals and fractions so far as derivatives are concerned). A
differential equation is called ordinary, if the unknown function depends on only one
argument (independent variable).
(Lucknow 2007; Meerut 09B, 10B; Bundelkhand 10)
A differential equation is said to be partial if there are two or more independent
variables.
A differential equation is said to be linear if the dependent variable, say, ‘ y ' and all its
derivatives occur in the first degree, otherwise it is non-linear.
D-4
A function y = f ( x) is called a Solution (or the primitive) of a differential equation if,

when substituted into the equation, it reduces the equation to an identity and the
process of finding all the solutions is called integrating (or solving) the differential
equation.
General solution: (Lucknow 2007; Gorakhpur 09)
A solution of a differential equation, containing independent arbitrary constants

equal in number to the order of the differential equation is called its general solution.
Particular solution: (Lucknow 2007)
A solution obtained by giving particular values to the arbitrary constants in the
general solution is called a particular solution or particular integral.
Arbitrary Constants: The solution of a differential equation may contain as many
arbitrary constants as is the order of the differential equation i. e., the solution of an nth
order differential equation may contain n arbitrary constants.
Example 1: Find the differential equation of the family of curves y = Ae x + ( B / e x ), for different
values of A and B. (Kanpur 2015)
x −x
Solution: We have y = Ae + Be . …(1)
To obtain the required differential equation the constants A and B are to be
eliminated with the help of the given equation (1) and the two equations obtained by
differentiating (1) once and twice. Thus differentiating (1), we get
dy
= Ae x − Be − x . …(2)
dx
Now differentiating (2), we get
d2 y
= Ae x + Be − x . …(3)
dx2
d2 y
Eliminating A and B between (1), (2) and (3), we obtain = y, which is the
dx2
required differential equation.
Example 2: Find the differential equation of all circles of radius a, or By the elimination of the
constants h and k, find the differential equation of which ( x − h)2 + ( y − k )2 = a2 , is a solution.
Solution: The equation of all circles of radius a is given by

( x − h)2 + ( y − k )2 = a2 , …(1)
h and k being parameters (i. e., arbitrary constants).

Differentiating (1), we get
D-5
dy
2 ( x − h) + 2 ( y − k ) = 0. …(2)
dx
2
d2 y  dy 
1 + ( y − k) +   = 0. …(3)
dx2  dx 
From (2) and (3), we obtain
[1 + (dy / dx)2 ]
( x − h) = − ( y − k ) (dy / dx) and ( y − k ) = − .
d2 y / dx2
Substituting these values in (1) and simplifying, we obtain
3 2
  dy  2   2 
2 d y
1 +    = a  2  ,
  dx    dx 
which is the required differential equation.
Comprehensive Exercise 1
1. Find the differential equation of the family of curves

y = Ae2 x + Be −2 x ,
for different values of A and B. (Bundelkhand 2003)
2. Find the differential equation corresponding to

y = ae2 x + be −3 x + ce x ,
where a, b, c are arbitrary constants.
3. Find the differential equation of the family of curves
y = e x ( A cos x + B sin x).
where A and B are arbitrary constants.
4. By eliminating the constants a and b obtain the differential equation of which
xy = ae x + be − x + x2 is a solution. (Purvanchal 2014)
5. Find the differential equation corresponding to the family of curves

y = c ( x − c )2 , where c is an arbitrary constant.
6. Show that Ax2 + By2 = 1 is the solution of

 d2 y  dy  2  dy
xy 2 +    − y = 0.
 dx  dx   dx
7. Show that v = (a / r) + B is a solution of

d2 v 2 dv
2
+ = 0.
dr r dr
D-6
A nswers 1
d2 y d3 y dy
1. 2
=4y 2. 3
−7 +6y =0
dx dx dx
2
d y dy d2 y dy
3. 2
−2 +2y =0 4. x 2
+2 − xy + x2 − 2 = 0
dx dx dx dx
3
dy  dy 
5. 8 y2 = 4 xy − 
dx  dx 
2 Differential Equations of First Order and First Degree

The differential equations of first order and of first degree can always be written in the
form
M + N ( dy / dx) = 0 or M dx + N dy = 0,
where M and N are some functions of x and y or are constants.
Here the differential equation being of 1st order, its general solution (or primitive) will
contain only one arbitrary constant. All differential equations of the first order cannot
be always solved. However they can be solved by suitable methods if they belong to
any one of the following standard forms :
(a) Variables separable,
(b) Homogeneous equations,
dy ax + by + c
(c) Equations of the form = ,
dx a1 x + b1 y + c1
(d) Linear equations,
(e) Equations reducible to the linear form,
(f) Exact differential equations,
(g) Equations reducible to exact form,
(h) Equations which by suitable change of variable can be transformed to any of the
above forms.
2.1 Variables Separable

If a differential equation of the first order and of the first degree is of the form
f1 ( x) dx = f2 ( y) dy, …(1)
where f1 ( x) is a function of x only and f2 ( y) is a function of y only, then we say that the
variables are separable in the differential equation.
In such equations it is possible to get dx and all the terms involving x on one side and dy
along with all the terms involving y on the other side.
D-7
To solve such a differential equation integrate the two sides and add an arbitrary constant
of integration to any one of the two sides.
Thus, integrating both the sides of (1), we get its solution as
∫ f1 ( x) dx = ∫ f2 ( y) dy + c,
where c is an arbitrary constant. The arbitrary constant can be chosen in any form
suitable for the answer, i. e., we can replace it by log c, tan−1 c, sin c , e c , etc.
Example 3: Solve (1 + x2 ) dy = (1 + y2 ) dx (Meerut 2003)

2 2
Solution: The given equation can be written as dy /(1 + y ) = dx /(1 + x ), in which the
variables have been separated.
∴ integrating, we get tan−1 y = tan−1 x + tan−1 c, where c is an arbitrary constant
or tan−1 y − tan−1 x = tan−1 c
y−x y−x
or tan−1 = tan−1 c, or =c
1 + yx 1 + yx
or ( y − x) = c (1 + yx), which is the required solution.
Example 4: Solve (1 + e x ) y dy = ( y + 1) e x dx.
y ex
Solution: Here dy = dx, (the variables being separated)
y +1 1+ ex
 1  ex
or 1 −  dy = dx.
 y + 1 1+ ex
∴ integrating, we get
y − log ( y + 1) = log (1 + e x ) + log c , (c being an arbitrary constant)
or y = log [c ( y + 1) (1 + e x )]
or c ( y + 1) (1 + e x ) = e y, which is the required solution.
Example 5: Solve dy / dx = ( x + y)2 . (Avadh 2011; Kanpur 15)
Solution: Here the variables are not separable but some suitable substitution will
reduce the differential equation to a form in which the variables are separable. Here we
put x + y = v.
Differentiating both sides w.r.t., ‘x ’, we have
dy dv dy dv
1+ = or = − 1.
dx dx dx dx
By these substitutions the given equation reduces to
dv dv dv
− 1 = v2 , or = v2 + 1, or 2
= dx.
dx dx v +1
D-8
Thus the variables being separated and so integrating, we get

dv
tan−1 v = x + c
2
v +1 ∫= dx + c ,
∫ or
or tan−1 ( x + y) = x + c, [∵ v = x + y]
or x + y = tan ( x + c ) as the required solution where c is an arbitrary constant.

Example 6: Solve dy / dx = sin ( x + y) + cos ( x + y).
Solution: Let x + y = v. Then differentiating, we have
dy dv dy dv
1+ = , or = − 1.
dx dx dx dx
Substituting these values in the given equation, we get
dv dv
− 1 = sin v + cos v, or = 1 + sin v + cos v
dx dx
dv
or = dx, separating the variables
(1 + cos v) + sin v
1 1
sec2 v
dv 2 2 dv = dx.
or = dx or
1 1 1 1
2 cos2
v + 2 sin v cos v 1 + tan v
2 2 2 2
1
Now integrating, we get log (1 + tan v) = x + c
2
1
or log {1 + tan ( x + y)} = x + c,
2
which is the required solution containing an arbitrary constant c.
Solve the following differential equations :

1. (i) (1 + x) y dx + (1 − y) x dy = 0 . (Garhwal 2011)
2
(ii) (1 − x ) (1 − y) dx = xy (1 + y) dy.
2. (i) x2 ( y + 1) dx + y2 ( x − 1) dy = 0 .
 dy   dy 
(ii)  y − x  = a  y2 +  ⋅
 dx   dx  (Purvanchal 2006; Avadh 10; Gorakhpur 08, 11)
2 2
3. (i) sec x tan y dx + sec y tan x dy = 0. (Garhwal 2003; Agra 06; Meerut 09B)
x− y 2 −y
(ii) dy / dx = e +x e . (Agra 2005; Avadh 07; Meerut 09; Purvanchal 10, 11;
Rohilkhand 10; Bundelkhand 04; Kumaun 14)
y y
4. (i) (e + 1) cos x dx + e sin x dy = 0. (Rohilkhand 2010)
x x 2
(ii) 3 e tan y dx + (1 − e ) sec y dy = 0 .
(Meerut 2008; Bundelkhand 04; Kumaun 14)
D-9
5. (i) (ds / dx) + x2 = x2 e3 s .

(ii) (dy / dx) tan y = sin ( x + y) + sin ( x − y). (Garhwal 2002; Kumaun 13)
6. (i) log (dy / dx) = ax + by.

(ii) xy (dy / dx) = (1 + y2 ) (1 + x + x2 ) /(1 + x2 ).
7. (i) If dy / dx = e x + y and it is given that for x = 1, y = 1, find y when x = − 1.
(ii) ( x − y)2 (dy / dx) = a2 .
8. (i) cos ( x + y) dy = dx. (Lucknow 2005; Gorakhpur 09; Purvanchal 10; Avadh 14)
(ii) cos ( x + y) dx = dy. (Gorakhpur 2005)
−1
9. (i) sin (dy / dx) = x + y.
(ii) (dy / dx) + 1 = e x + y. (Kumaun 2006)
2
10. (i) dy / dx = (4 x + y + 1) . (Lucknow 2006)
 x + y − a  dy  x + y + a 
(ii)   = ⋅
 x + y − b  dx  x + y + b 
11. ( y − x) (dy / dx) = a2 . (Meerut 2006B)
A nswers 2
1. (i) xy = ce y− x
1 2 1 2
(ii) log{ x (1 − y)2 } = x − y −2y + c
2 2
1 2
2. (i) ( x + y2 ) + ( x − y) + log { c ( x − 1)( y + 1)} = 0
2
(ii) y = c (a + x) (1 − ay)
1 3
3. (i) tan x tan y = c (ii) e y = ex + x +c
3
4. (i) sin x (e y + 1) = c (ii) tan y = c (1 − e x )3
3
5. (i) (e3 s − 1) = c1 e(3 s + x ), where c1 = e3 c (ii) sec y = c − 2 cos x
− by ax
6. (i) − (1/ b) e = (1/ a) e +c
1
(ii) log (1 + y2 ) = log x + tan−1 x + log c
2
a  x − y − a
7. (i) y = −1 (ii) y+c= log  
2  x − y + a
1 1
8. (i) y = c + tan ( x + y) (ii) tan ( x + y) = x + c
2 2
1
9. (i) [− 2 /( x + c )] = 1 + tan ( x + y)] (ii) ( x + c ) e( x + y) + 1 = 0
2
D-10
10. (i) 4 x + y + 1 = 2 tan (2 x + k )

(ii) (b − a) log {( x + y)2 − ab} = 2( x − y + c )
11. y + a2 log (a2 + x − y) = c
2.2 Homogeneous Equations

dy f ( x, y)
Definition: A differential equation of the form = 1 , where f1 ( x, y) and
dx f2 ( x, y)
f2 ( x, y) are homogeneous functions of x and y of the same degree, is called a
homogeneous equations. (Bundelkhand 2005)
To solve such an equation, put

dy dv
y = vx ; whence =v+ x ⋅
dx dx
These substitutions transform the given equation into an equation of the form
dv dv
v+ x = f (v) i. e., x = f (v) − v.
dx dx
The variables are now separable. Separating the variables and integrating, we get
dv
∫ f (v) − v
= log x + c , where c is an arbitrary constant.
Now replacing v by ( y / x) after integration, we get the required solution.

Working Rule: To get the solution of a homogeneous differential equation proceed as
follows :
dy dv
(i) Put y = vx and =v+ x .
dx dx
(ii) The equation thus obtained will be of the form in which the variables are
separable. Separate the variables and integrate.
(iii) After integration replace v by y / x and get the required solution.
Example 7: Solve ( x2 − y2 ) dx + 2 xy dy = 0 . (Gorakhpur 2008; Kumaun 13)
Solution: The given equation can be written as

dy y2 − x2
= …(1)
dx 2 xy
This is a homogeneous differential equation as each term in the N r and Dr of R.H.S. is
of the second degree in x and y.
dy dv
∴ putting y = vx and consequently =v+ x in (1), we get
dx dx
dv v2 x2 − x2 v2 − 1
v+ x = =
dx 2 vx2 2v
D-11
dv v2 − 1 v2 − 1 − 2 v2 −1 − v2
or x = −v= =
dx 2v 2v 2v
2 v dv dx
or =− , in which the variables have been separated.
v2 + 1 x
log (v2 + 1) = − log x + log c
or log {(v2 + 1) ⋅ x} = log c or x (v2 + 1) = c
or x {( y / x)2 + 1} = c [putting v = y / x]
2 2
or ( y + x ) = cx, which is the required solution.
Example 8: Solve y2 dx + ( xy + x2 ) dy = 0 . (Avadh 2010)

dy y2
=− , which is homogeneous.
dx xy + x2
∴ putting y = vx , this equation reduces to

2 2
 v + x dv = − v x v2
  2
=−
 dx  x . vx + x v +1
dv v2 v2 + v2 + v 2 v2 + v
or x =− −v=− =− ,
dx v +1 v +1 v +1
in which the variables are separable.

dx (v + 1) dv (v + 1)
∴ − = 2
= dv .
x (2 v + v) v (2 v + 1)
Integrating, we get
v +1
− log x + log c =
∫ v (2v + 1) dv , where c is an arbitrary constant.
v +1 A B
Now let = + . Then A = 1, B = − 1.
v (2 v + 1) v 2 v + 1
v +1 1 1  1
∴
∫ v (2v + 1) dv = ∫  v − 2v + 1 dv = log v − 2 log (2v + 1) .
Hence the solution is
1
− log x + log c = log v − log (2 v + 1)
2
1
or log (2 v + 1) = log v + log x − log c
2
or log (2 v + 1)1 /2 = log (vx / c )
or {2 ( y / x) + 1}1 /2 = y / c [∵ v = y / x]
2 2 2 2
or (2 y + x) / x = y / c or c (2 y + x) = xy .
D-12
Example 9: Solve x2 y dx − ( x3 + y3 ) dy = 0 .
(Gorakhpur 2007; Garhwal 07; Agra 08; Kumaun 08; Kanpur 15)
dy x2 y
= 3 . …(1)
dx x + y3
This is a homogeneous equation as each term in the N r and Dr on the R.H.S. is of the
same degree (3rd degree).
dy dv
∴ putting y = vx and =v+ x in (1), we get
dx dx
dv x2 . vx v dv v v4
v+ x = 3 3 3
= or x = − v = −
dx x + v x 1 + v3 dx 1 + v3 1 + v3
dx (1 + v3 ) dv dx 1 1
or =− or = − 4 dv − dv .
x v4 x v v
log x + log c = 1/(3 v3 ) − log v
or log (cxv) = 1/(3 v3 )
or 3 log (cxv) = 1/ v3 or log (cxv)3 = 1/ v3
3 3
/ y3
or c 3 x3 v3 = e1 / v or c 3 y3 = e x [∵ v = y / x]
3 3
or y3 = ke x / y
, which is the required solution.
Example 10: Solve ( x3 − 3 xy2 ) dx = ( y3 − 3 x2 y) dy.
(Meerut 2007; Kumaun 09; Garhwal 13)

dy x3 − 3 xy2
= 3 , …(1)
dx y − 3 x2 y
which is homogeneous.
dy dv
∴ putting y = vx and =v+ x in (1), we get
dx dx
dv x3 − 3 x . v2 x2 1 − 3 v2
v+ x = 3 3 2
= 3
dx v x − 3 x . vx v − 3 v
dv 1 − 3 v2 1 − 3 v2 − v4 + 3 v2 1 − v4
or x = 3 −v= 3
= 3
dx v − 3 v v − 3v v − 3v
dx v3 − 3 v  1 1 2v 
or = dv =  + − 2  dv ,
x 1 − v4  2 (v + 1) 2 (v − 1) v + 1
(by partial fractions)
1 1
log x + log c = log (v + 1) + log (v − 1) − log (v2 + 1)
2 2
D-13
or log (cx) = log [{(v + 1)1 /2 (v − 1)1 /2 } /(v2 + 1)]

or cx (v2 + 1) = (v2 − 1)1 /2 or c 2 x2 (v2 + 1)2 = v2 − 1
or c 2 x2 {( y2 / x2 ) + 1}2 = {( y2 / x2 ) − 1}, [∵ v = y / x]
or c 2 ( y2 + x2 )2 = ( y2 − x2 ), which is the required solution.
 y  y
Example 11: Solve x  cos  ( y dx + x dy) = y  sin  ( x dy − y dx).
 x  x
 y  dy   y   dy 
x cos   y + x  = y sin   x − y
 x  dx   x   dx 
dy  y y 2 y y
or x  x cos − y sin  = − y sin − xy cos
dx  x x x x
dy y { y sin ( y / x) + x cos ( y / x)}
or = …(1)
dx x{ y sin ( y / x) − x cos ( y / x)}
dy dv
Now putting y = vx and =v+ x in (1), we get
dx dx
dv vx (vx sin v + x cos v)
v+ x =
dx x (vx sin v − x cos v)
dv v (v sin v + cos v)
or v+ x =
dx (v sin v − cos v)
dv v2 sin v + v cos v − v2 sin v + v cos v
or x = , variables separable
dx v sin v − cos v
 − v sin v + cos v 2 dx
or −  dv = , separating the variables.
 v cos v  x
− log (v cos v) = 2 log x + log c
or log {1/(v cos v)} = log (c x2 ) or c x2 = 1/(v cos v)
or c x2 v cos v = 1 or c x2 . ( y / x)cos ( y / x) = 1
or cxy cos ( y / x) = 1, which is the required solution.

1. x + y (dy / dx) = 2 y. (Gorakhpur 2005)
2. y − x (dy / dx) = x + y (dy / dx). (Gorakhpur 2010; Garhwal 08)

2 2
3. ( x + y ) dx − 2 xy dy = 0 . (Avadh 2006; Bundelkhand 03)
D-14
4. ( x2 + y2 ) (dy / dx) = xy. (Kanpur 2002; Kumaun 12)

2 2
5. dy / dx = y /( xy − x ). (Kumaun 2015)
6. x2 dy + y ( x + y) dx = 0 .
7. x ( x − y) dy = y ( x + y) dx.
dy y2
8. x + = y.
dx x
9. x dy − y dx = √ ( x2 + y2 ) dx. (Meerut 2008; Lucknow 10)
10. (1 + e x / y) dx + e x / y{1 − ( x / y)} dy = 0 . (Meerut 2006)

dy y y
11. = + tan .
dx x x
dy y
12. x = y − x cos2 .
dx x
 dy 
13. x   = y + x e y / x .
 dx 
dy  y  y
14. x cos   = y cos   − x.
dx  x  x
dy
15. x − y = x √ ( x2 + y2 ).
dx (Garhwal 2010)
2 2
dy x + 3 y
16. + = 0.
dx 3 x2 + y2 (Garhwal 2009)
A nswers 3
1
1. log ( y − x) = c + x /( y − x) 2. log ( x2 + y2 ) + tan−1 ( y / x) = log c
2
2
/(2 y2 )
3. x2 − y2 = cx 4. cy = e x
5. y = ke y / x 6. ( y + 2 x) = c 2 x2 y
7. c 2 xy = e − x / y
8. cx = e x / y
9. y + √ ( y2 + x2 ) = cx2 10. x + ye x / y
=c
11. sin ( y / x) = cx 12. tan ( y / x) = log (c / x)
 y
13. e − y / x + log x = c 14. sin   + log x = c
 x
15. y = x sinh ( x + c ) 16. log ( x + y) + 2 xy ( x + y)−2 = c
D-15
2.3 Equations Reducible To Homogeneous Form

A differential equation of the form
dy ax + by + c a b
= , where ≠
dx a1 x + b1 y + c1 a1 b1
can be reduced to homogeneous form by taking new variables X and Y such that
x = X + h and y = Y + k , where h and k are constants to be so chosen as to make the
given equation homogeneous. With the above substitutions we get dx = dX and
dy = dY , so that dy / dx = dY / dX . Hence the given equation becomes
dY a ( X + h) + b (Y + k ) + c aX + bY + (ah + bk + c )
= = .
dX a1 ( X + h) + b1 (Y + k ) + c1 a1 X + b1Y + (a1h + b1k + c1)
Now choose h and k such that

ah + bk + c = 0, and a1h + b1k + c1 = 0 .
Then the differential equation becomes
dY aX + bY
= , which is homogeneous.
dX a1 X + b1Y
Now this equation can be solved as in article 2.2 substituting Y = vX . Finally by

replacing X by ( x − h) and Y by ( y − k ) we shall get the solution in original variables x
and y.
If, however, a / a1 = b / b1 = m (say) then the differential equation becomes of the form
dy m (a1 x + b1 y) + c
= .
dx a1 x + b1 y + c1
To solve such a differential equation put v = a1 x + b1 y, get rid of y and then the
transformed equation will be such that the variables are separable.
dy y − x +1
Example 12: Solve = .
dx y + x +5 (Kumaun 2011)
a b
Solution: Here ≠ i. e., the coefficients of x and y in the numerator and
a1 b1
denominator of the expression for (dy / dx) are not proportional. Such equations can be
reduced to homogeneous form by taking new variables X and Y such that x = X + h
and y = Y + k . where h and k are constants to be taken at our choice. With these
substitutions the given equation reduces to
dY Y + k − X − h + 1 Y − X + (k − h + 1)
= = . …(1)
dX Y + k + X + h + 5 Y + X + (k + h + 5)
Choose h and k such that
k − h + 1= 0 and k + h + 5 = 0.
D-16
Solving these, we get k = − 3, h = − 2.

With these values of h and k the equation (1) becomes
dY Y − X
= . …(2)
dX Y + X
Now (2) is a homogeneous equation, so putting Y = vX , we have
dv v − 1 dv v − 1 − 1 − v2
v+ X = , or X = −v=
dX v + 1 dX v + 1 v +1
v +1 dX v dv dX
or 2
dv = − or 2
dv + 2
=−
v +1 X v +1 v +1 X
2v 2 2
or 2
dv + 2 dv = − dX .
v +1 v +1 X
Integrating, we have
log (v2 + 1) + 2 tan−1 v = − 2 log X + c
or log {(v2 + 1) X 2 } = − 2 tan−1 v + c
or log (Y 2 + X 2 ) = − 2 tan−1 (Y / X ) + c , [∵ v = Y / X ]
or log [( y + 3)2 + ( x + 2)2 ] + 2 tan−1 {( y + 3) / ( x + 2)} = c ,
[∵ Y = y − k = y + 3, X = x − h = x + 2 ]
which is the required solution.
dy 2 x + 2 y − 2
Example 13: Solve = ⋅
dx 3x + y − 5 (Lucknow 2005)
Solution: Here (a / a1) ≠ (b / b1), therefore putting x = X + h and y = Y + k the given

equation reduces to
dY 2 ( X + h) + 2 (Y + k ) − 2 2 X + 2Y + (2 h + 2 k − 2)
= = . …(1)
dX 3 ( X + h) + (Y + k ) − 5 3 X + Y + (3 h + k − 5)
Now choose h and k such that
2h + 2k − 2 = 0 and 3h + k − 5 = 0.
Solving these, we get h = 2 and k = − 1.
With these values of h and k, the equation (1) becomes
dY 2 X + 2Y
= ⋅
dX 3X + Y
This is a homogeneous equation so putting Y = vX , we get
dv 2 X + 2 vX 2 + 2 v
v+ X = =
dX 3 X + vX 3+v
dv 2 + 2 v 2 + 2 v − 3 v − v2 2 − v − v2
or X = −v= =
dX 3+v 3+v 3+v
dX 3 + v 3 + v
or − = 2 dv = dv
X v + v−2 (v + 2) (v − 1)
D-17
 4 1 
= −  dv , by resolving into partial fractions
3 (v − 1) 3 (v + 2)
dX  1 4 
or 3 = − dv.
X 
 v + 2 v − 1
Integrating, we get
3 log X = log (v + 2) − 4 log (v − 1) + log c
or log { X 3 (v − 1)4 } = log { c (v + 2)}
or X 3 [(Y / X ) − 1]4 = c {(Y / X ) + 2}, [∵ v = Y / X ]
4
or (Y − X ) = c (Y + 2 X )
or {( y + 1) − ( x − 2)}4 = c {( y + 1) + 2( x − 2)}
or ( y − x + 3)4 = c (2 x + y − 3), which is the required solution.
Example 14: Solve (2 x + y + 1) dx + (4 x + 2 y − 1) dy = 0 .
(Lucknow 2011; Kanpur 12, 15)
dy 2x + y + 1
Solution: We have =− .
dx 4x + 2 y − 1
a b
Here = .
a1 b1
∴ let 2 x + y = v so that 2 + (dy / dx) = dv / dx.
With these substitutions the given equation reduces to
dv v +1
−2= − ,
dx 2v − 1
dv v + 1 3v − 3
or =2− =
dx 2v − 1 2v − 1
2v − 1 2 (v − 1) + 1 1 
or 3 dx = dv = dv = 2 +  dv.
v −1 v −1  v − 1
∴ integrating, we have
3 x + c = 2 v + log (v − 1)
or 3 x + c = 2 (2 x + y) + log (2 x + y − 1), [∵ v = 2 x + y]
or x + 2 y + log (2 x + y − 1) = c , which is the required solution.

1. (2 x + y − 3) dy = ( x + 2 y − 3) dx.(Bundelkhand 2010; Lucknow 09; Rohilkhand 11)
2. dy / dx = (2 x − y + 1) /( x + 2 y − 3).
3. (2 x + 3 y − 5) (dy / dx) + (3 x + 2 y − 5) = 0 .
4. ( x − y − 2) dx + ( x − 2 y − 3) dy = 0.
5. ( x + y − 1) dy = ( x + y) dx.
D-18
6. ( x + y) (dx − dy) = dx + dy. (Gorakhpur 2008)
7. ( x − y − 2) dx = (2 x − 2 y − 3) dy. (Kashi 2011)

dy x − y +3
8. = .
dx 2 x − 2 y + 5 (Garhwal 2002, 10)
9. (2 x + 4 y + 1) dy = (2 y + x + 1) dx.
10. (2 x + y + 1) dx + (4 x + 2 y − 1) dy = 0 . (Lucknow 2011)
A nswers 4
1. ( x + y − 2) = c 2 ( x − y)3
2. (5 y − 7)2 + (5 x − 1) (5 y − 7) − (5 x − 1)2 = c
3. 3 x2 + 4 xy + 3 y2 − 10 x − 10 y = c
4. log { c 2 (2Y 2 − X 2 )} = (1/ √ 2) log {(Y √ 2 − X ) /(Y √ 2 + X )},

where X = x − 1 and Y = y + 1.
5. 2 ( y − x) − log (2 x + 2 y − 1) = c 6. ( y − x) + log ( x + y) = c
7. log ( x − y − 1) = x − 2 y − c 8. x − 2 y + log ( x − y + 2) = c
9. 4 (2 y − x) − log (4 x + 8 y + 3) = c
10. x + 2 y + log (2 x + y − 1) = c
2.4. Linear Differential Equations

Definition: A differential equation is said to be linear when the dependent variable y and all
its derivatives occur in the first degree only and are not multiplied together.
An equation of the form
dy
+ Py = Q,
dx
where P and Q are functions of x only is called a linear differential equation of the first
order with y as the dependent variable. To solve such an equation, multiply both the
sides by e ∫ P dx . The equation then becomes
e∫
P dx
(dy / dx) + Pye ∫ P dx = Qe ∫ P dx
d
or
dx
{ ye∫ P dx} = Qe∫ P dx .
Integrating both sides w.r.t., ‘x’, we get
ye ∫ P dx = P dx
∫ Qe∫ dx + c ,
which is the solution of the differential equation.

D-19
Note 1: The factor e ∫ P dx , on multiplying by which the L.H.S. of the differential

equation becomes a differential coefficient of some function of x and y, is called an
integrating factor (I.F.) of the differential equation. (Agra 2007)
Note 2: Sometimes a given differential equation becomes linear if we take y as the

independent variable and x as the dependent variable.
dx
Then it is of the form + Px = Q,
dy
where P and Q are some functions of y only. The integrating factor (I.F.) in this case is
e ∫ P dy and the solution is
x ⋅ e ∫ P dy = P dy
∫ Q ⋅ e∫ dy + c .
Working Rule :
(i) Write the given equation in the form
dy dx
+ Py = Q or + Px = Q as the case may be.
dx dy
(ii) Find the integrating factor e ∫ P dx or e ∫ P dy.
(iii) The solution of the differential equation is either
y ⋅ (I. F. ) =
∫ {Q ⋅ (I. F. )} dx + c
or
∫
x ⋅ (I. F. ) = { Q ⋅ (I. F. )} dy + c as the case may be.
dy 2
Example 15: Solve + 2 xy = e − x .
dx (Meerut 2001, 03, 10B; Kumaun 06)
dy 2
Solution: The given differential equation is + 2 xy = e − x , which is linear with y as
dx
the dependent variable.
2
Here P = 2 x, and Q = e − x .
1 2
= x2 .
We have
∫ P dx =
∫ 2 x dx = 2 ⋅ 2 x
2
P dx
Therefore I.F. = e ∫ = ex .
Hence the solution is y. (I.F.) =

∫ {Q ⋅ (I. F. )} dx + c, where c is an arbitrary constant
2
− x2 x2
y ⋅ ex =
or
∫ (e . e ) dx + c
2 2
y ⋅ ex ye x = x + c .
or
∫
= dx + c or
Example 16: Solve ( x2 − 1) (dy / dx) + 2 xy = 1. (Meerut 2010)

D-20

dy 2x 1  dy 
+ y= 2 From + Py = Q
dx x2 − 1 x −1  dx 
2x 2 x dx 2
Here P =
x2 − 1
, and so
∫ P dx =
∫ x2 − 1 = log ( x − 1).
2
P dx
∴ integrating factor (I.F.) = e ∫ = e log ( x − 1)
= ( x2 − 1).
1  1 
the solution is y. ( x2 − 1) = 2
∴
∫ ( x2 − 1) ⋅ ( x − 1) dx + c , ∵ Q = 2


( x − 1)
y ( x2 − 1) =
or
∫ dx + c = x + c.
Hence y ( x2 − 1) = x + c is the required solution.
Example 17: Solve x (dy / dx) + 2 y = x2 log x.

dy 2  dy 
+ y = x log x,  from + Py = Q
dx x  dx 
2
Here P = (2 / x) and so
∫ P dx =
∫ (2 / x) dx = 2 log x = log x .
P dx x2
∴ I.F. = e ∫ = e log = x2 .
∴ the solution is
y (I. F. ) =
∫ Q ⋅ (I. F. ) dx + c
yx2 = 2 [∵ Q = x log x]
i. e.,
∫ ( x log x) x dx + c,
yx2 = x3 log x dx + c
or
∫
1 x4 x4 [integrating by parts]
yx2 = (log x) ⋅
or ⋅
x 4
dx + c ,
4
−
∫
1 1 1
or yx2 = x4 log x − ⋅ ( x4 ) + c
4 4 4
1
Hence 4 yx2 = x4 log x − x4 + k is the required solution.
4
Example 18: Solve cos2 x (dy / dx) + y = tan x.
(Meerut 2003, 13; Avadh 05; Garhwal 06)
dy
Solution: The given equation can be written as + sec2 x . y = tan x sec2 x, which is
dx
linear with y as the dependent variable.
Here P = sec2 x, and Q = tan x sec2 x.
2
We have
∫ P dx = ∫ sec x dx = tan x.
D-21
∴ I. F. = e ∫ P dx = e tan x .
the solution is y . e tan x 2

x . e tan x
∴ =
∫ tan x sec dx + c
ye tan x t 2
or =
∫ t e dt + c, putting tan x = t and sec x dx = dt
= [t e t − e t dt] + c , integrating by parts
∫
= t e t − e t + c = tan x e tan x
− e tan x
+c
tan x tan x
or ye =e (tan x − 1) + c .
Example 19: Solve (1 + y2 ) dx = (tan−1 y − x) dy. (Agra 2005; Avadh 07; Rohilkhand 10;
Bundelkhand 03, 04; Meerut 13B; Garhwal 11)
dx tan−1 y − x dx 1 tan−1 y
= or + ⋅x = .
dy 1 + y2 dy 1 + y 2
1 + y2
This is linear equation with y as independent variable and x as dependent variable,

Here P = 1/(1 + y2 ).
2 −1
∴
∫ P dy = ∫ {1/(1 + y )} dy = tan y.
−1
P dy
Hence I.F. = e ∫ = e tan y
.
Now the solution is x . (I. F. ) =

∫ {Q . (I. F. ) } dy + c
−1 tan−1 y −1  tan−1 y 
x e tan y
⋅ e tan y
i. e., =
∫ 1 + y2
dy + c , ∵ Q =


1 + y2 
t
dt + c , putting tan−1 y = t and dy /(1 + y2 ) = dt
=
∫te
= [t e t − 1 . e t dt] + c , integrating by parts
∫
= t e t − e t + c = (t − 1) e t + c
−1 −1
or x . e tan y
= (tan−1 y − 1) . e tan y
+ c.
Example 20: Solve (dy / dx) + 2 y tan x = sin x, given that y = 0 when x = π / 3.
dy
Solution: We have + 2 tan x . y = sin x, which is linear with y as dependent
dx
variable.
2
Here P = 2 tan x and so
∫ P dx = ∫ 2 tan x dx = 2 log sec x = log sec x.
sec2 x
∴ I.F. = e ∫ P dx = e log = sec2 x.
D-22
∴ the solution is y . (I. F. ) =

∫ [Q . (I. F. )] dx + c
y sec2 x = sin x . sec2 x dx + c =
i. e.,
∫ ∫ tan x sec x dx + c
or y sec2 x = sec x + c . …(1)

1
Now it is given that y = 0 at x = π.
3
Therefore, from (1)
1 1 1
0 . sec2 π = sec π + c or c = − sec π = − 2.
3 3 3
Hence from (1), y sec2 x = sec x − 2 is the required solution.

1. (dy / dx) + (1/ x) y = x n.
2. x2 (dy / dx) + y = 1.
3. x (dy / dx) + y = x2 + 3 x + 2 (Avadh 2010)
4. (sin x) (dy / dx) + 3 y = cos x. (Meerut 2004)
2 2
5. ( x + 1)(dy / dx) + 2 xy = 4 x . (Agra 2008; Kumaun 10)
2
6. (1 + x ) (dy / dx) + 2 xy = cos x. (Meerut 2009; Kumaun 09, 12)
2 2 3
7. x ( x + 1) (dy / dx) = y (1 − x ) + x log x. (Meerut 2007B)
8. (dy / dx) + (2 y / x) = sin x.

9. (dy / dx) + ( y / x) = sin x2
10. ( y sin x − 1) dx + cos x dy = 0. (Bundelkhand 2008)
11. sin 2 x (dy / dx) − y = tan x. (Agra 2007)
12. (1 + x) (dy / dx) − xy = 1 − x.

13. ( x + 2 y3 ) dy = y dx. (Rohilkhand 2006; Avadh 11)
14. ( x + y + 1) (dy / dx) = 1. (Gorakhpur 2006; Rohilkhand 08)

−1
15. (1 + y2 ) dx + ( x − e tan y
) dy = 0 . (Gorakhpur 2007; Lucknow 09; Purvanchal 11;
Bundelkhand 04)
A nswers 5
1. xy = x n + 2 / (n + 2) + c 2. y = 1 + ce1 / x
D-23
3. 6 xy = 2 x3 + 9 x2 + 12 x + c
1 1 1 1
4. y tan3 ( x) = 2 tan ( x) − x − tan3 ( x) + c
2 2 3 2
2 4 3
5. y (1 + x ) = x + c 6. y (1 + x2 ) = sin x + c
3
1 1
7. y ( x2 + 1) = cx + x3 log x − x3
2 4
1
8. yx2 = − x2 cos x + 2 x sin x + 2 cos x + c 9. yx = − cos x2 + c
2
10. y sec x = tan x + c 11. y = tan x + c √ (tan x)
12. y (1 + x) = x + ce x 13. x = y3 + cy
−1 1 2 tan −1
14. x = ce y − y − 2 15. xe tan y
= e y
+c
2
2.5. Equations Reducible to the Linear Form

Some equations can be reduced to the linear form by making suitable substitutions,
and hence can be solved easily.
(a) Bernoulli’s Equation: (Gorakhpur 2008)

dy
+ Py = Q y n,
dx
where P and Q are functions of x alone is called Bernoulli’s equation.
To solve this equation dividing both sides by y n, we get
dy
y −n + Py − n + 1 = Q
dx
[Note that the R.H.S. is now a function of x alone.]
dy dv
Now putting y − n +1 = v, so that (1 − n) y − n = , the equation (1) transforms to
dx dx
1 dv
+ Pv = Q
1 − n dx
dv
or + (1 − n) P . v = (1 − n) Q,
dx
which is a linear differential equation with v as dependent variable and can be solved
by the method discussed in article 2.4.
(b) Equations of the form
dy
+ Pφ ( y) = Q f ( y),
dx
where P and Q are functions of x alone.
To reduce such an equation to the linear form, dividing both sides by f ( y) and thus
getting rid of f ( y) from the R.H.S., we have
D-24
1 dy φ ( y)
⋅ + P⋅ = Q.
f ( y) dx f ( y)
φ ( y)
Now we try the substitution = v⋅
f ( y)
dv d  φ ( y)  1 dy
If =   = K⋅ ,
dx dx  f ( y) f ( y) dx
1 dv
where K is some constant, the equation reduces to the form + Pv = Q, or
K dx
dv
+ K Pv = KQ, which is a linear differential equation, with v as the dependent
dx
variable.
dy 1
Example 21: Solve + y = x2 y6 .
dx x
Solution: The given equation on dividing out by y6 , becomes
1 dy 1 1
+ ⋅ = x2 . …(1)
y6 dx x y5
1
Put = v, so that
y5
5 dy dv 1 dy 1 dv
− = or =− .
y6 dx dx y 6 dx 5 dx
With these substitutions the equation (1) becomes
1 dv 1 dv 5
− + ⋅ v = x2 or − v = − 5 x2 .
5 dx x dx x
This is a linear equation with v as the dependent variable. Here P = − 5 / x and
Q = − 5 x2 .
We have
−5
= log (1/ x5 ).
∫ P dx = ∫ (− 5 / x) dx = −5 log x = log x
5
∴ I.F. = e ∫ P dx = e log (1 / x )
= 1/ x5 .
∴ the solution is
v ⋅ (I. F. ) =
∫ {Q ⋅ (I. F. )} dx + c
v ⋅ (1/ x5 ) = − 5 x2 ⋅ (1/ x5 ) dx + c = −5 x −3 dx + c
i. e.,
∫ ∫
5 −2
i. e., (1/ y5 ) ⋅ (1/ x5 ) = x + c. [Note that v = 1/ y5 ]
2
5
Hence 1/( x5 y5 ) = ⋅ (1/ x2 ) + c is the required solution.
2
D- 25
dy  x 
Example 22: Solve +  y = x √ y.
dx 1 − x2 
Solution: Dividing both sides of the given equation by 2 y1 /2 , we have
1 dy 1  x  1
+   √ y = x. …(1)
2 √ y dx 2 1 − x2  2
1 −1 / 2
Now put y1 /2 = v, so that y (dy / dx) = (dv / dx).
2
dv 1  x  1
+  2  ⋅ v = x,

dx 2 1 − x  2
which is linear with v as the dependent variable.
1 1
Here P = { x /(1 − x2 )} and Q = x.
2 2
1 x 1 − 2x
We have
∫
P dx =
∫
2 1 − x2
dx = −
4 1 − x2
dx
∫
1
=− log (1 − x2 ) = log (1 − x2 )−1 /4 = log{1/(1 − x2 )1 /4 }.
4
2 1 /4
∴ I.F. = e ∫ P dx = e log{1 /(1− x ) }
= 1/(1 − x2 )1 /4 .
∴ the solution is
1
v /(1 − x2 )1 /4 = { x /(1 − x2 )1 /4 } dx + c
2 ∫
1 −1 / 4 putting (1 − x2 ) = t so that − 2 x dx = dt
v /(1 − x2 )1 /4 = −
or
4
t dt,
∫
1 4 [∵ v = √ y]
or √ y /(1 − x2 )1 /4 = − ⋅ t3 /4 + c ,
4 3
1 [∵ t = (1 − x2 )]
or √ y /(1 − x2 )1 /4 = − (1 − x2 )3 /4 + c .
3
Example 23: Solve x (dy / dx) + y = y2 log x.

(Bundelkhand 2006; Gorakhpur 10; Purvanchal 07, 08, 10; Kanpur 06, 15)

dy 1 y2
+ y= log x, dividing throughout by x
dx x x
1 dy 1 1 log x
or + ⋅ = , …(1)
y2 dx x y x
dividing throughout by y2 .
Now put 1/ y = v, so that (− 1/ y2 )(dy / dx) = (dv / dx).
With these substitutions equation (1) becomes

dv 1 log x
− + v=
dx x x
D-26
dv 1 log x
or − ⋅v= − .
dx x x
This is linear with v as the dependent variable.
Here P = − 1/ x and Q = − (log x) / x.
∴ I.F. = e ∫ P dx = e ∫ (−1 / x) dx = e − log x
= e log (1 / x) = (1/ x).
v (1/ x) =
∫ {− (log x) / x} ⋅ (1/ x) dx + c
v / x = − (1/ x2 ) log x dx + c
or
∫
∫
= − [(− 1/ x) log x − (1/ x)(−1/ x) dx] + c,
integrating by parts taking 1/ x2 as the second function

2
= (1/ x) log x −
∫ (1/ x ) dx + c = (1/ x) log x + (1/ x) + c
or 1/( xy) = (1/ x)(1 + log x) + c , [∵ v = 1/ y]
or 1 = y (1 + log x) + cxy.
dy y y
Example 24: Solve + log y = 2 (log y)2 .
dx x x (Kumaun 2011)
2
Solution: Dividing both sides of the given equation by y (log y) , we have
1 dy 1 1 1
+ ⋅
2
= 2 …(1)
y (log y) dx x log y x
Now put 1/ log y = v, so that
− {1/(log y)2 } ⋅ (1/ y) ⋅ (dy / dx) = dv / dx.
dv v 1 dv 1 1
− + =+ 2 or − ⋅v= − 2 .
dx x x dx x x
Here P = − 1/ x and Q = − 1/ x2 ,
∴ I.F. = e ∫ P dx = e − ∫ (1 / x)dx = e − log x
= e log(1 / x) = 1/ x.
2
∴ the solution is v / x =
∫ (−1/ x ).(1/ x)dx + c
(1/ log y)(1/ x) = − x −3 dx + c ,
or
∫ [∵ v = 1/ log y]
1 1
or = + c.
x log y 2 x2
dy 1 e y
Example 25: Solve + = .
dx x x2
Solution: On dividing out by e y, the given equation becomes
D- 27
dy 1 1
e− y + e− y ⋅ = 2 . …(1)
dx x x
Now put e − y = v, so that − e − y (dy / dx) = (dv / dx).
With these substitutions equation (1) becomes
dv 1 1 dv 1 1
− + ⋅v= 2 or − ⋅v= − 2 .
dx x x dx x x
Here P = − 1/ x and Q = − 1/ x2 .
∴ I.F. = e ∫ P dx = e ∫ −(1 / x) dx = e − log x
= e log (1 / x) = 1/ x.
∴ the solution is v. (I.F.) =

∫ {Q . (I. F. )} dx + c
1 1 1 1
i. e., v.
x
=−
∫ x2 ⋅ x dx + c = 2 x2 + c
or e − y . (1/ x) = 1/(2 x2 ) + c [∵ v = e − y]
or e − y(1/ x) = (1 + 2 cx2 ) /(2 x2 )
or 2 x = (2 cx2 + 1) e y.

1. (dy / dx) + ( y / x) = y2 . (Bundelkhand 2007; Garhwal 09, 11)
2. (dy / dx) + { y /( x − 1)} = xy1 /3 . (Meerut 2005B)

3 2
3. ( x y + xy) dx = dy. (Kanpur 2002)
3 3
4. dy / dx = x y − xy. (Gorakhpur 2005; Meerut 13B)
3 2
5. 3 (dy / dx) + 2 y /( x + 1) = x / y .
6. y (2 xy + e x ) dx − e x dy = 0 . (Kanpur 2001)
3
7. ( xy2 − e1 / x ) dx − x2 y dy = 0 . (Lucknow 2007)
x− y x y
8. dy / dx = e (e − e ).
9. (dy / dx) − y tan x = − y2 sec x. (Meerut 2009B; Garhwal 10)
10. cos x dy = (sin x − y) y dx. (Rohilkhand 2007)

3 2
11. (dy / dx) + x sin 2 y = x cos y. (Gorakhpur 2007, 11; Purvanchal 08; Avadh 14)
2 2
12. (dy / dx) + y cot x = y sin x.
13. 2 (dy / dx) − y sec x = y3 tan x.
D-28
dy
14. (1 − x2 ) + xy = xy2 .
dx (Purvanchal 2006)
dy
15. x + y log y = xye x .
dx (Bundelkhand 2005)
2 3 3
16. Solve xy (dy / dx) − 2 y = 2 x , given that y = 1 when x = 1.
dy dφ dφ
17. Solve + y = φ ( x) , where φ is some function of x only.
dx dx dx
A nswers 6
1. xy (c − log x) = 1
2 3
2. y2 /3 ( x − 1)2 /3 = x ( x − 1)5 /2 − ( x − 1)8 /3 + c
5 20
1 2
1 2 − x 2
3. 1 = 2 y (1 − x ) − cy e 2 4. (1/ y2 ) = ce x + x2 + 1
2
1 2 1
5. y3 ( x + 1)2 = x6 + x5 + x4 + c 6. e x + y (c + x2 ) = 0
6 5 4
3 x
7. 3 y2 = 2 x2 e1 / x + 3 cx2 8. e y = e x − 1 + ce −(e )
9. sec x = y (tan x − c ) 10. sec x = y (tan x − c )

2
11. 2 tan y = ( x2 − 1) + 2 ce − x 12. (1/ y) = (cos x − c ) sin x
2
13. − (sec x + tan x) = y (sec x + tan x − x + c )
14. 1/ y n−1 = 2 sin x − {2 /(1 − n)} + ce(n − 1)sin x 15. x log y = e x ( x − 1) + c
16. y3 = − 2 x3 + 3 x6 17. ye φ = φ
∫φe dφ + c
2.6. Exact Differential Equations (Lucknow 2006)

A differential equation is said to be exact if it can be obtained from its primitive (solution) directly
by differentiation, without involving any subsequent process of multiplication, elimination, etc.
Thus the differential equation of the form M dx + N dy = 0, where M and N are some
functions of x and y, is exact if it can be obtained directly by differentiating an
equation of the form u = c , where u is some function of x and y and c is an arbitrary
constant.
Theorem: The necessary and sufficient condition for the ordinary differential equation M
dx + N dy = 0 to be exact is that (Lucknow 2006, 09)
∂M / ∂y = ∂N / ∂x .
The condition is necessary: Suppose the differential equation
M dx + N dy = 0 …(1)
D- 29
is exact. Let the primitive of (1) be u = c , where u is some function of x and y and c is an
arbitrary constant.
Since u is a function of x and y, therefore from partial differentiation, we have
∂u ∂u
du = dx + dy .
∂x ∂y
∂u ∂u
Now u = c ⇒ du = 0 ⇒ dx + dy = 0. …(2)
∂x ∂y
Since the equation (1) is exact, therefore equation (2) must be identical with (1),
therefore, we have
∂u
=M …(3)
∂x
∂u
and = N. …(4)
∂y
Differentiating (3) and (4) partially with respect to y and x respectively, we obtain
∂2 u ∂M ∂2 u ∂N
= , = .
∂y ∂x ∂y ∂x ∂y ∂x
∂2 u ∂2 u ∂M ∂N
But = . Therefore = .
∂y ∂x ∂x ∂y ∂y ∂x
Hence the condition is necessary.
The condition is sufficient: We have to show that if ∂M / ∂y = ∂N / ∂x, then
M dx + N dy = 0 must be an exact differential equation. Let
∫ M dx = P , then
∂P ∂2 P ∂M ∂N
= M, so that = = , by hypothesis.
∂x ∂y ∂x ∂y ∂x
∂N ∂2 P ∂N ∂2 P ∂N ∂  ∂P 
Now = ⇒ = ⇒ =  
∂x ∂x ∂y ∂x ∂x ∂y ∂x ∂x  ∂y 
∂P
⇒ N = + f ( y), where f ( y) is some function of y alone.
∂y
∂P ∂P
Now putting M = and N = + f ( y) in M dx + N dy, we have
∂x ∂y
∂P  ∂P 
M dx + N dy = dx +  + f ( y) dy
∂x  ∂y 
∂P ∂P
= dx + dy + f ( y) dy = dP + f ( y) dy
∂x ∂y
= d [ P + F ( y)], where d [ F ( y)] = f ( y) dy

= an exact differential of some function of x and y.
Therefore the differential equation M dx + N dy = 0 is exact and hence the condition
is sufficient.
D-30
Working rule for solving an exact differential equation: (Meerut 2008)
To solve a differential equation of the form M dx + N dy = 0, first ascertain with the

help of the condition ∂M / ∂y = ∂N / ∂x whether the equation is exact or not. If the
equation is exact, then
(i) Integrate M with respect to x treating y as constant.
(ii) Integrate w.r.t. y only those terms of N which do not contain x.
(iii) Equate the sum of these two integrals [found in (i) and (ii)] to an arbitrary
constant and thus get the required solution.
Thus if the differential equation M dx + N dy = 0 is exact, its solution is
∫ M dx +
∫ N dy =c
treating y as constant take only those terms in

N which do not contain x.
Example 26: Solve (ax + hy + g) dx + (hx + by + f ) dy = 0. (Meerut 2003)
Solution: Here M = ax + hy + g and N = hx + by + f .

∂M ∂N
We have = h and = h . Thus we observe that ∂M / ∂y = ∂N / ∂x and therefore the
∂y ∂x
given differential equation is exact. Hence its solution is
∫ M dx +
∫ N dy =c
treating y as a constant taking only those terms in

N which do not contain x.
i. e.,
∫ (ax + hy + g) dx +
∫ (by + f ) dy = c
treating y as a constant
1 2 1
i. e., ax + hxy + gx + by2 + fy = c
2 2
i. e., ax2 + 2 hxy + by2 + 2 gx + 2 fy + c = 0 ,
where c is an arbitrary constant. We have replaced − 2c by c.
Example 27: Solve (4 x + 3 y + 1) dx + (3 x + 2 y + 1) dy = 0 .

(Purvanchal 2009; Meerut 11)
Solution: Here M = 4 x + 3 y + 1 and N = 3 x + 2 y + 1 .

∂M ∂N ∂M ∂N
We have = 3 and = 3 i. e., = , and hence the given equation is exact.
∂y ∂x ∂y ∂x
D- 31
Integrating M , i. e., 4 x + 3 y + 1 with respect to x, treating y as a constant, we obtain

1
4 ⋅ x2 + 3 xy + x i. e., 2 x2 + 3 xy + x .
2
Again, the only terms in N which do not contain x are 2 y + 1 .
Integrating (2 y + 1) with respect to y, we obtain
1
2 ⋅ y2 + y i. e., y2 + y.
2
Hence the solution of the given differential equation is
2 x2 + 3 xy + x + y2 + y = c , where c is an arbitrary constant
i. e., 2 x2 + 3 xy + y2 + x + y = c .
Example 28: Solve ( x2 − ay) dx − (ax − y2 ) dy = 0 .
Solution: Here M = x2 − ay and N = − (ax − y2 ).
We have
∂M ∂N ∂M ∂N
= − a and = − a i. e., =
∂y ∂x ∂y ∂x
and hence the given equation is exact.

2 1 3
Now
∫ M dx (regarding y as a constant) =
∫ (x − ay) dx =
3
x − ayx
…(1)
and
∫ N dy (taking in N only those terms which do not contain x)
2 1 3
=
∫y dy =
3
y s …(2)
∴ the required solution is,

(1) + (2) = c (an arbitrary constant)
1 3 1 3
or x − ayx + y = c,
3 3
or x3 + y3 − 3 axy = 3 c .
x dy − y dx
Example 29: Solve x dx + y dy + =0 .
x2 + y2 (Bundelkhand 2007)

 y   x 
x − 2 2
dx +  y + 2 2
dy = 0 .
 x + y   x + y 
Here
y x
M=x− 2 2
and N = y + .
x + y x + y2
2
D-32
We have
∂M 1 ⋅ ( x2 + y2 ) − y ⋅ 2 y x2 − y2 y2 − x2
=0 − = = ,
∂y ( x2 + y2 )2 ( x2 + y2 )2 (x2 + y2 )2
∂N 1 ⋅ ( x2 + y2 ) − x ⋅ 2 x y2 − x2
and =0 + = .
∂x ( x2 + y2 )2 ( x2 + y2 )2
Thus ∂M / ∂y = ∂N / ∂x and hence the given equation is exact. Therefore its solution is
 y 
∫  x − x2 + y2  dx +
∫ y dy =c
regarding y as a constant taking only those terms in

N which do not contain x
x2 1 x y2
or − y ⋅ tan−1 + =c
2 y y 2
or x2 − 2 tan−1 ( x / y) + y2 = 2 c = k .

1. dy / dx = (2 x − y) /( x + 2 y − 5).
2. (2 ax + by) y dx + (ax + 2 by) x dy = 0 .
3. (1 + 4 xy + 2 y2 ) dx + (1 + 4 xy + 2 x2 ) dy = 0 . (Bundelkhand 2004)
x dy − y dx
4. x dx + y dy = a2 ⋅
x2 + y2(Agra 2006; Meerut 08; Gorakhpur 11; Bundelkhand 11)
5. [1 + e x / y] dx + e x / y[1 − ( x / y)] dy = 0 . (Meerut 2004B, 06)
y y
6. (e + 1) cos x dx + e sin x dy = 0 . (Agra 2006)
2x
7. (sin x cos y + e ) dx + (cos x sin y + tan y) dy = 0.
8. ( y sin 2 x) dx − (1 + y2 + cos2 x) dy = 0 .
9. [ y {1 + (1 / x)} + cos y] dx + [ x + log x − x sin y] dy = 0 .
A nswers 7
1. x2 − y2 − xy + 5 y = c . 2. ayx2 + bxy2 = c
3. x + 2 x2 y + 2 xy2 + y = c 4. x2 + y2 + 2 a2 tan−1 ( x / y) = c
D- 33
5. x + y ex / y
=c 6. (e y + 1) sin x = c
1 2x 2 3
7. e − cos x cos y + log sec y = c 8. y cos 2 x + 2 y + y =c
2 3
9. y ( x + log x) + x cos y = c
3 Integrating Factors
(Lucknow 2007)
A differential equation which is not exact can sometimes be made exact by multiplying
by some suitable function of x and y. Such a function is called an integrating factor
(I.F.) of the equation.
Methods used to find the integrating factors:
Method 1: Integrating factor found by inspection: Sometimes integrating factors
are found by inspection.
The students should remember the following exact differentials. These help us in
finding the integrating factors.
 x y dx − x dy
(i) d ( xy) = x dy + y dx , (ii) d  = ,
 y y2
 y  x dy − y dx  x2  2 yx dx − x2 dy
(iii) d   = , (iv) d  = ,
 x x2  y y2
 y2  2 xy dy − y2 dx  x y dx − x dy
(v) d   = , (vi) d  tan−1  = ,
 x  x2  y x2 + y2
 y  x dy − y dx  x y dx − x dy
(vii) d  tan−1  = , (viii) d  log  = ,
 x x2 + y2  y xy
 y  x dy − y dx  ex  ye x dx − e x dy
(ix) d  log  = , (x) d  = ,
 x xy  y y2
x dx + y dy
(xi) d {log √ ( x2 + y2 )} = .
x2 + y2
Example 30: Solve (1 + xy) y dx + (1 − xy) x dy = 0 .

( y dx + x dy) + xy2 dx − x2 y dy = 0 ,
or d ( yx) + xy2 dx − x2 y dy = 0 . [ From (i) of article 3 ]
Dividing out by x2 y2 , we have
D-34
d ( yx) 1 1
2 2
+ dx − dy = 0 . …(1)
x y x y
Now putting xy = v in (1), we have
(1/ v2 ) du + (1/ x) dx − (1/ y) dy = 0 .
Now integrating each term, we get
− (1/ v) + log x − log y = c , where c is a constant
or −(1/ xy) + log x − log y = c [∵ v = xy]
or log ( x / y) = c + (1/ xy), is the required solution.
Example 31: Solve y dx − x dy + (1 + x2 ) dx + x2 sin y dy = 0 . (Lucknow 2005)

2
Solution: Dividing each term by x , the given equation becomes
y dx − x dy  1 
2
+  2 + 1 dx + sin y dy = 0
x x 
x dy − y dx 1
or − 2
+  2 + 1 dx + sin y dy = 0
x x 
y 1
or − d   +  2 + 1 dx + sin y dy = 0 , [ Refer article 3, result (iii)]
 x x 
Integrating each term, we have
y 1
− − + x − cos y = − c
x x
y 1
or + − x + cos y = c .
x x

1. x dy − y dx = xy2 dx . 2. x dy − y dx + 2 x3 dx = 0 .
3. ( y2 e x + 2 xy) dx − x2 dy = 0 . (Gorakhpur 2008)

x x
4. y (axy + e ) dx − e dy = 0 .
5. y (2 x2 y + e x ) dx − (e x + y3 ) dy = 0 . (Gorakhpur 2009; Avadh 09)

2 2
6. y sin 2 x dx = (1 + y + cos x) dy .
7. x dy − y dx = ( x2 + y2 ) dx .
8. x dx + y dy + ( x2 + y2 ) dy = 0 .
9. ( x2 + y2 + a2 ) y dy + ( x2 + y2 − a2 ) x dx = 0 . (Rohilkhand 2006, 07)

D- 35
A nswers 8
1. yx2 + 2 x = 2 cy . 2. y + x3 = cx .
1 2
3. e x + ( x2 / y) = c . 4. ax + (e x / y) = c .
2
2 3 1 2 1 3
5. x − y + (e x / y) = c . 6. − y cos2 x = y + y +c.
3 2 3
7. tan−1 ( y / x) = x + c . 8. x2 + y2 = e c − 2 y.
9. ( x2 + y2 )2 + 2 a2 ( y2 − x2 ) = c .
Method 2: If the equation M dx + N dy = 0 is of the form

[ f ( x, y)] y dx + [ F ( x, y)] x dy = 0,
and Mx − Ny ≠ 0 , then
1/( Mx − Ny) is an integrating factor.
Note: If Mx − Ny = 0 , then Mx = Ny or M/ y = N/x
i. e., the differential equation M dx + N dy = 0 reduces to y dx + x dy = 0 whose
solution is xy = c .
Example 32: Solve ( x3 y3 + x2 y2 + xy + 1) y dx + ( x3 y3 − x2 y2 − xy + 1) x dy = 0 .
Solution: Here
Mx − Ny = ( x3 y3 + x2 y2 + xy + 1) y ⋅ x − ( x3 y3 − x2 y2 − xy + 1) xy
= 2 ( x3 y3 + x2 y2 ) = 2 x2 y2 ( xy + 1).
∴ I.F. = 1 / {2 x2 y2 ( xy + 1)}.
The given equation may be written as
[ x2 y2 ( xy + 1) + ( xy + 1)] y dx + [( x3 y3 + 1) − xy ( xy + 1)] x dy = 0 .
Now multiplying by I.F. this becomes
[ x2 y2 ( xy + 1) + ( xy + 1)] y dx [( xy + 1){( x2 y2 − xy + 1) − xy}] x dy
+ =0
2 x2 y2 ( xy + 1) 2 x2 y2 ( xy + 1)
 x2 y2 + 1 ( x2 y2 − 2 xy + 1)
or  2 2  y dx +   x dy = 0
 x y   x2 y2 
y dx + x dy 2 x2 y
or ( y dx + x dy) + − dy = 0
x2 y2 x2 y
D-36
d ( xy) 2
or d ( xy) + 2 2
− dy = 0 .
x y y
Now integrating each term, we get
xy + (−1/ xy) − 2 log y = c .

1. ( xy2 + 2 x2 y3 ) dx + ( x2 y − x3 y2 ) dy = 0 .
2. ( x2 y2 + xy + 1) y dx + ( x2 y2 − xy + 1) x dy = 0 .
3. ( xy sin xy + cos xy) y dx + ( xy sin xy − cos xy) x dy = 0 . (Garhwal 2009)
A nswers 9
1. x2 = cy e1 /( xy) 2. xy + log x − log y − 1/( xy) = c
3. x sec ( xy) = cy
Method 3: When Mx + Ny ≠ 0 and the equation is homogeneous, then the

integrating factor of M dx + N dy = 0, is 1/( Mx + Ny) .
Example 33: Solve x2 y dx − ( x3 + y3 ) dy = 0 .
Solution: Here Mx + Ny = x3 y − x3 y − y4 = − y4 ≠ 0
and the equation is homogeneous.

∴ I.F. = 1/( Mx + Ny) = −1/ y4 .
Multiplying the given equation by the I.F., we get
x2 x3 + y3 dy x2 x3
− dx + dy = 0, or = 3 dx − 4 dy
y3 y4 y y y
dy 1  x3 
or = d .
y 3  y3 
∴ Integrating each term, we get

D- 37
1 x3
log y = + log c
3 y3
3
/3 y3
or y = ce x .
1  ∂M ∂N 
Method 4: If  −  is a function of x alone, say f ( x), then the integrating
N  ∂y ∂x 
f ( x) dx
factor for M dx + N dy = 0 is e ∫ .
1 3 1 2 1
Example 34: Solve ( y + y + x ) dx + ( x + xy2 ) dy = 0 .
3 2 4 (Lucknow 2007)
1 3 1 2 1
Solution: Here M = y + y + x and N = ( x + xy2 ) .
3 2 4
∂M ∂N 1
∴ = 1 + y2 and = (1 + y2 ) .
∂y ∂x 4
1  ∂M ∂N  4 1
∴  −  = {(1 + y2 ) − (1 + y2 )}
N  ∂y ∂x  x (1 + y2 ) 4
4 3 3
= ⋅ (1 + y2 ) = ,
x (1 + y2 ) 4 4
which is a function of x alone, say f ( x).
f ( x) dx x3
∴ I.F. = e ∫ = e ∫ (3 / x) dx = e3 log x
= e log = x3 .
Multiplying the given diff. equation by the I.F. x3 , we get
1 1 1
( x3 y + x3 y3 + x5 ) dx + ( x4 + x4 y2 ) dy = 0 ,
3 2 4
which is an exact diff. equation and its solution is
∫ M dx (treating y as a constant)
∫
+ N dy (taking in N terms without x) = c
x4 y x4 y3 x6
i. e., + + = c,
4 12 12
or 3 x4 y + x4 y3 + x6 = 12 c = k .
1  ∂N ∂M 
Method 5: If  −  is a function of y alone, say f ( y), then the integrating
M  ∂x ∂y 
f ( y) dy
factor for M dx + N dy = 0 is e ∫ .
D-38
Example 35: Solve ( xy3 + y) dx + 2 ( x2 y2 + x + y4 ) dy = 0 . (Lucknow 2006)
Solution: Here M = xy3 + y and N = 2 ( x2 y2 + x + y4 ).

∂M ∂N
∴ = 3 xy2 + 1 and = 2 (2 xy2 + 1) .
∂y ∂x
1  ∂N ∂M  1
∴  −  = {4 xy2 + 2 − 3 xy2 − 1}
M  ∂x ∂y  xy3 + y
1
= ( xy2 + 1)
y ( xy2 + 1)
1
= ,
y
which is a function of y alone, say f ( y).
f ( y) dy
∴ I.F. = e ∫ = e ∫ (1 / y) dy
= e log y
= y.
Multiplying the given equation with the I.F. y, we have
( xy4 + y2 ) dx + 2 ( x2 y3 + xy + y5 ) dy = 0 ,
which is an exact differential equation and its solution is
∫
+ N dy (taking in N only those terms which do not contain x)
= c, (a constant)
4 2 5
i. e.,
∫ ( xy + y ) dx ( y constant) +
∫2 y dy = c
1 2 4 1 6
or x y + xy2 + y =c.
2 3

1. ( x2 y − 2 xy2 ) dx − ( x3 − 3 x2 y) dy = 0 . (Lucknow 2011)
2 2
2. ( x + y ) dx − 2 xy dy = 0 . (Gorakhpur 2008)
2 4 3 3 2
3. (3 x y + 2 xy) dx + (2 x y − x ) dy = 0 .
4. ( xy2 − x2 ) dx + (3 x2 y2 + x2 y − 2 x3 + y2 ) dy = 0 .
D- 39
A nswers 10
1. ( x / y) − 2 log x + 3 log y = c 2. x2 − y2 = cx
3. x3 y2 + ( x2 / y) = c
1 1 1 1 1
4. e6 y[ x2 ( y2 − x) + ( y2 − y+ )] = 0
2 3 6 18 108
Method 6: If the equation is of the form

x a y b (my dx + nx dy) + x c y d ( py dx + qx dy) = 0 , …(A)
where a, b, m, n, c , d, p, q are constants, then the integrating factor is x h y k , where h, k can

be obtained by applying the condition that after multiplication by x h y k , the equation
(A) must become exact.
Example 36: Solve ( y2 + 2 x2 y) dx + (2 x3 − xy) dy = 0 .

y ( y dx − x dy) + 2 x2 ( y dx + x dy) = 0 .
This equation is of the form (A) , as mentioned above. So let the possible integrating
factor be x h y k .
Multiplying the given equation by the proposed I.F. x h y k , we have
( x h y k + 2 + 2 x h + 2 y k +1) dx + (2 x h + 3 y k − x h +1 y k +1) dy = 0. …(1)
h k +1 h+2 k
Now (∂M / ∂y) = (k + 2) x y + 2 (k + 1) x y
and (∂N / ∂x) = 2 (h + 3) x h + 2 y k − (h + 1) x h y k + 1.
If the equation (1) is exact, then
(∂M / ∂y) = (∂N / ∂x).
∴ equating the coefficients of x h y k +1 and x h + 2 y k on both sides, we get
k + 2 = − h −1 i. e., h+ k +3=0
and 2k + 2 = 2h + 6 i. e., h− k +2=0.
Solving these, we get h = − 5 / 2 and k = − 1/ 2.
∴ the integrating factor is x h y k = x −5 /2 y − 1 /2 .
With these values of h and k the equation (1) becomes

( x −5 /2 y3 /2 + 2 x −1 /2 y1 /2 ) dx + (2 x1 /2 y −1 /2 − x −3 /2 y1 /2 ) dy = 0 .
This differential equation is of the form M dx + N dy = 0 and is exact as can be seen by

verifying the condition ∂M / ∂y = ∂N / ∂x .
D-40
Hence its solution is
∫
+ N dy (taking in N only those terms which do not contain x)
= c, (a constant)
2 −3 / 2 3 / 2
i. e., − x y + 4 x1 /2 y1 /2 + 0 = c .
3
2 −3 / 2 3 / 2
Hence − x y + 4 x1 /2 y1 /2 = c is the required solution.
3 (Gorakhpur 2007)

1. (2 y dx + 3 x dy) + 2 xy (3 y dx + 4 x dy) = 0 .
2. (3 x + 2 y2 ) y dx + 2 x (2 x + 3 y2 ) dy = 0 .
3. x (3 y dx + 2 x dy) + 8 y4 ( y dx + 3 x dy) = 0 . (Lucknow 2010)
A nswers 11
1. x2 y3 + 2 x3 y4 = c 2. x3 y4 + x2 y6 = c 3. x3 y2 + 4 y6 x2 = c
4 Change of Variables
In some cases a suitable substitution (change of variable) reduces a given differential
equation to one or the other of the forms already discussed and hence the equation can
be solved.
x dx + y dy  a2 − x2 − y2 
Example 37: Solve =  2 2
 .
x dy − y dx  x + y 
(Rohilkhand 2005; Purvanchal 09; Avadh 09; Bundelkhand 03; Garhwal 10)
Solution: Put x = r cos θ and y = r sin θ . Then

D- 41
x2 + y2 = r2 , …(1)
and y / x = tan θ. …(2)
2 x dx + 2 y dy = 2 r dr or x dx + y dy = r dr .
x dy − y dx
= sec2 θ dθ
x2
or x dy − y dx = x2 sec2 θ dθ = r2 cos2 θ sec2 θ dθ = r2 dθ.
Substituting these values in the given differential equation, we get

r dr  a2 − r2  √ (a2 − r2 )
=   = .
r2 dθ  r
2
 r
dr
∴ = √ (a2 − r2 )
dθ
dr
or = dθ. [Separating the variables]
√ (a − r2 )
2
Integrating, we get
sin−1(r / a) = θ + c
or r = a sin (θ + c )
or √ ( x2 + y2 ) = a sin {tan−1( y / x) + c }.
[∵ r = √ ( x2 + y2 ) and θ = tan−1 ( y / x)]
5 Geometrical Problems
Remember the following formulae of the differential calculus:
Length of the tangent = y √ {1 + (dx / dy)2 } .
Length of the subtangent = y (dx / dy) = y / (dy / dx).

Length of the normal = y √ {1 + (dy / dx)2 }.
Length of the subnormal = y (dy / dx).

Intercept made by the tangent on the x-axis is x − y (dx / dy), and on the y-axis is
y − x (dy / dx) .
Angle φ between the radius vector and the tangent is given by
tan φ = r (d θ / dr) or cot φ = (1/ r)(dr / dθ) .
Length of the polar subtangent = r2 (dθ / dr).
Length of the polar subnormal = (dr / dθ) .
Radius of curvature ρ = [1 + (dy / dx)2 ]3 /2 / (d2 y / dx2 ).
D-42
Example 38: Find the curves in which the polar subnormal is of constant length.
Solution: Let the constant length of the polar subnormal be a.
But the length of the polar subnormal = dr / dθ .
∴ as given in the question, dr / dθ = a .
This is the differential equation of the required curves.
To solve, put it into the form dr = a dθ. [Separating the variables]
∴ integrating, we get r = aθ + c , where c is an arbitrary constant.
This is the polar equation of the required family of curves.
Example 39: Find the curves for which the sum of the reciprocals of the radius vector and the polar
subtangent is constant.
Solution: The polar subtangent = r2 (dθ / dr).
1 1 dr
∴ as given in the question, + = λ , where λ is a constant.
r r2 dθ
This is the differential equation of the required curves. This equation may be written
as
dr 1
= r2  λ −  = r (λr − 1)
dθ  r
dr  λ 1
or dθ = = −  dr. [Separating the variables]
r (rλ − 1)  λr − 1 r
θ + c = log (λr − 1) − log r , where c is an arbitrary constant
or θ + c = log {(λr − 1) / r}, or (λr − 1) / r = e θ + c .
Hence λr − 1 = re θ + c is the equation of the required family of curves.
Example 40: Find the curves in which the cartesian subtangent varies as the abscissa.
Solution: The cartesian subtangent is y /(dy / dx).
∴ as given in the question y /(dy / dx) = λx, where λ is a constant
or λ dx = λx dy
or (dx / x) = λ (dy / y), separating the variables.
log x = λ log y + log c , where c is an arbitrary constant
Hence log x = log ( y λ . c )
or x = cy λ , is the required family of curves.
D- 43
Example 41: Show that the curve in which the angle between the tangent and the radius vector at
every point is half of the vectorial angle is a cardioid.
1
Solution: As given, φ = θ, where θ is the vectorial angle.
2
1
∴ tan φ = tan θ .
2
But tan φ = r (dθ / dr).
1 1
∴ r (dθ / dr) = tan θ or (dr / r) = (cot θ) dθ ,
2 2
in which the variables have been separated.
1
∴ integrating, we have log r = 2 log (sin θ) + log c ,
2
where c is an arbitrary constant
1 1
or log r = log (sin2 θ) + log c , or log (r / c ) = log (sin2 θ)
2 2
1 1 1
or (r / c ) = sin2 θ = (1 − cos θ) , or r = c (1 − cos θ)
2 2 2
1
or r = a (1 − cos θ), where a = c . This is a cardioid.
2
Example 42: Find the curve for which the tangent at each point makes a constant angle α with the
radius vector.
Solution: As given φ = α (a constant)
or tan φ = tan α .
But tan φ = r (dθ / dr) .
∴ r (dθ / dr) = tan α or (dr / r) = (cot α) dθ .
Now the variables have been separated. Therefore integrating, we have
log r = θ cot α + log c , where c is an arbitrary constant
or log r − log c = θ cot α
or log (r / c ) = θ cot α
or r / c = e θ cot α .
Hence r = ce θ cot α is the required curve.
Example 43: Show that the curve in which the slope of the tangent at any point equals the ratio of
the abscissa to the ordinate of the point is a rectangular hyperbola.
Solution: The slope of the tangent at any point ( x, y) = tan ψ = dy / dx .
dy abscissa x
∴ as given, = =
dx ordinate y
or y dy = x dx , in which the variables have been separated.
1 1
∴ integrating, we have y2 = x2 + c , where c is an arbitrary constant.
2 2
D-44
Hence y2 − x2 = 2 c = a2 (say) is the required curve which is the equation of a

rectangular hyperbola.
Example 44: Find the family of curves whose tangent forms an angle π / 4 with the hyperbola
xy = c .
Solution: Let m1 and m2 be the gradients of the tangents of the required family of curves
and the given hyperbola respectively. Since the angle between these tangents is given
1
to be π , therefore
4
m − m2 m − m2
tan (π / 4) = 1 or 1= 1
1 + m1m2 1 + m1m2
or 1 + m1m2 = m1 − m2 or m1 = (1 + m2 ) / (1 − m2 ) . …(1)
Now m1 = gradient of the tangent of the required family of curves = dy / dx
and m2 = (dy / dx) for hyperbola xy = c .
d c c ∵ y = c 
∴ m2 =   =− 2 ,
dx  x  x  x 
Hence from (1), we have
dy 1 + (− c / x2 ) x2 − c  2c 
= 2
= 2 = 1 − 2 
dx 1 − (− c / x ) x + c  x + c
 2c 
or dy = 1 − 2  dx, in which the variables have been separated.
 x + c
2c x
∴ integrating, we have y = x − tan−1   + c1,
√c  √ c
where c1 is an arbitrary constant. This is the required family of curves.
1. By the substitution y2 = v − x reduce the equation y3 (dy / dx) + x + y2 = 0 to

the homogeneous form and hence solve the equation.
2. Find the curve in which the polar subtangent is constant.
3. Find the equation of the family of curves for which the sum of the reciprocals of
the radius vector and the polar subnormal is constant.
4. Find the curve in which the cartesian sub-normal is equal to the abscissa.
5. Find the equation of the curve for which the cartesian subtangent varies as the
reciprocal of the square of the abscissa.
6. Show that the parabola is the only curve in which the sub-normal is constant.
D- 45
7. Show that the curve for which the normal at every point passes through a fixed
point is a circle.
8. Find the equation of the curve in which the perpendicular from the origin on
any tangent is equal to the abscissa of the point of contact.
9. The normal PN to a curve meets the x-axis in N. If the distance of N from the
origin is twice the abscissa of P, prove that the curve is a rectangular hyperbola.
10. The tangent of any point P of a curve meets the x-axis in Q. If Q is on the
positive side of the origin O and OP = OQ , show that the family of curves
having this property are parabolas whose common axis is the x-axis.
A nswers 12
1
1. tan−1 {( y2 + x) / x} − log { x2 + ( y2 + x)2 } = log c
2
2. r (θ + c ) + a = 0 3. θ = λr − log r + c
2 2 2 1
4. y − x = c 5. λ log y = x3 + c
3
6. y2 = 2 λ { x + (c / 2 λ )} 8. y2 + x2 = cx
O bjective T ype Q uestions
Multiple Choice Questions

Indicate the correct answer for each question by writing the corresponding letter from (a),
(b), (c) and (d).
1. The order of the differential equation
2
d4 y  d3 y  d2 y dy
−3 3  +4 2 −5 + 6 y = 0 is
dx4  dx  dx dx
(a) 3 (b) 6
(c) 4 (d) 2
dy
2. The integrating factor of the differential equation cos2 x + y = tan x is
dx
(a) sec2 x (b) e tan x
2
(c) cos x (d) tan x (Garhwal 2011)
dy
3. The integrating factor of the differential equation + y cot x = 2 cos x is
dx
(a) sin x (b) log sin x
(c) cot x (d) cos x (Kumaun 2007)
D-46
4. The differential equation M dx + N dy = 0 , where M and N are functions of x

and y, is exact if
∂M ∂N ∂M ∂N
(a) = (b) =
∂y ∂x ∂x ∂y
∂M ∂N ∂M ∂N
(c) + =0 (d) + =0
∂y ∂x ∂x ∂y
(Avadh 2005; Kumaun 06, 10, 11, 13; Kanpur 16)
5. The complete solution of the differential equation

( x2 − ay) dx − (ax − y2 ) dy = 0 is
(a) x3 + y3 + 3 axy = c (b) x3 + y3 − 3 axy = 4
(c) x3 + y3 − 3 axy = c (d) x3 + y3 − 3 axy = 0
dx
6. The integrating factor of differential equation + P ( y) x = Q( y) is
dy
(a) e ∫ P dy (b) e ∫ P dx
(c) e ∫ Q dx (d) e ∫ − P dx (Bundelkhand 2007)
dy
7. The differential equation ( x2 + y2 + a2 ) y + x ( x2 + y2 − a2 ) = 0 is
dx
(a) variables separable (b) linear
(c) homogeneous (d) exact (Rohilkhand 2007)
2 4
 d2 y   dy 
8. The order and degree of differential equation  2  + y =  
 dx   dx 
(a) 2, 4 (b) 4, 4
(c) 2, 2 (d) 4, 2 (Garhwal 2009, 13)
2 3 /2
  dy   d2 y
9. The order and degree of the differential equation 1 +    =k

 dx 
 dx2
(a) 3, 1 (b) 3, 2
(c) 3, 3 (d) 2, 2 (Garhwal 2010)
4
d4 y  d2 y 
10. The order and degree of the differential equation + 2 k  2  + k 4 y = 0 is
dx4  dx 
(a) 4, 4 (b) 4, 2
(c) 4, 1 (d) 1, 4 (Garhwal 2009)
−1
11. Order and degree of differential equation y ′′ + 2 sin ( y ′) − 3 y = 0 is
(a) order = 2, degree = 1
(b) order = 2, degree does not exist
(c) order does not exist and degree exists
(d) none of these (Garhwal 2012)
D- 47
d4 y  d2 y 
12. Degree of the differential equation 4
+ 2 k 2  2  + k 4 y = 0 is
dx  dx 
(a) 4 (b) 2
(c) 1 (d) 0 (Garhwal 2014)
2 3
 d2 y    dy  2 
13. Degree of the differential equation  2  − 1 +    = 0 is
 dx 
 dx   
(a) 2 (b) 3
(c) 4 (d) 5
(Kumaun 2006, 14; Garhwal 15)
14. The integrating factor of the differential equation ( x + 2 y3 ) dy = y dx is

1 1
(a) (b)
x y
x
(c) (d) xy
y (Garhwal 2009)
15. The integrating factor in the differential equation (1 + y2 ) dx = (tan−1 y − x) dy

is
1
(a) tan y
(b) e tan y
e
1
(c) e tan−1 y
(d) tan−1 y (Garhwal 2009)
e
16. In differential equation y (2 xy + e x ) dx − e x dy = 0 the integrating factor is

(a) e − x (b) e x
(c) e1 / x (d) e −1 / x
dy
17. The solution of the differential equation + y = e − x is
dx
(a) ye − x = x + c (b) ye x = x − c
(c) ye x = x + c (d) y = xe − x (Garhwal 2005)
18. Solution of the differential equation x dy − y dx − 2 x3 dx = 0 is

(a) y = x3 + cx (b) y + x3 + cx = 0
(c) y + x3 = c (d) y = x3 + c (Garhwal 2006)
dy 2
19. The solution of the differential equation + 2 xy = e − x is
dx
2 2
(a) ye − x = x + c (b) ye x = x + c
2 2
(c) xe x = y + c (d) xe y = x + c (Garhwal 2010)
D-48
20. The integrating factor in the differential equation

( x2 y − 2 xy2 ) dx + (3 x2 y − x3 ) dy = 0 is
1 1
(a) (b) 2
xy x y
1 1
(c) (d)
x2 y2 x3 y3 (Garhwal 2010)
21. The solution of the differential equation (e y + 1)cos x dx + e y sin x dy = 0 is

(a) sin y = (e x + 1) + c (b) sin x = (e x + 1) + c
(c) sin x (e y + 1) = c (d) none of these (Garhwal 2010)
−1 dy
22. The integrating factor in the differential equation (1 + y2 ) + ( x − e tan y
) =0
dx
is
1
−1 −1
(a) e tan x
(b) e tan x
tan−1 y
(c) e (d) 1 (Garhwal 2010)
dy
23. Integrating factor of differential equation x log x + y = 2 log x is
dx
1
(a) (b) x log x
x
1
(c) (d) log x
log x (Garhwal 2012)
dy
24. Integrating factor of the differential equation ( x + 2 y3 ) = y is
dx
(a) y (b) 1/ y
(c) 1 (d) −1/ y (Garhwal 2013)
dy
25. Solution of the differential equation + y2 = 0 is
dx
1 x3
(a) y = (b) y = − +c
x+c 3
(c) y = c e x (d) y = x2 (Kumaun 2015)
23
d2 y   dy  
26. The degree of the differential equation + 1 +    = 0 is
dx2 
 dx 

(a) 1 (b) 2
(c) 3 (d) 4 (Kumaun 2011)
2
27. The differential equation x dx + 2 y dy = 0 is
(a) exact (b) non-exact
(c) partial (d) homogeneous (Kumaun 2014)
D- 49
dy 3 sin x
28. Integrating factor for the differential equation + y = 3 will be
dx x x
(a) x2 (b) x3
(c) x (d) log x (Kumaun 2014)
3
d2 y  dy 
29. The order of the differential equation +   + y4 = e − t is
dt2  dt 
(a) 1 (b) 2
(c) 3 (d) none of these (Kumaun 2015)
dx
30. Which of the following is a solution of the differential equation + 3x = 0 ?
dt
(a) x = 3 e − t (b) x = 2 e − 3 t
3
(c) x = − t 2 (d) x = 3 t 2
2 (Kumaun 2015)
dy y
31. Solution of the differential equation = will represent
dx x
(a) family of circles (b) family of straight lines
(c) family of hyperbolas (d) none of these (Kumaun 2015)
dy  x  2
32. The differential equation +  y = x is
dx 1 − x2 
(a) linear (b) homogeneous
(c) exact (d) total (Kumaun 2013)
2
 dy 
33. The differential equation   + 5 y1 /3 = x is
 dx 
(a) linear of order 2
(b) non-linear of order 1 and degree 2
(c) non-linear of order 1 and degree 6
(d) none of these (Kanpur 2016)
2
d y dy
34. The degree of differential equation + 1+ = 0 is
dx2 dx
(a) 0 (b) 1
(c) 2 (d) 4 (Kanpur 2016)
35. The differential equation whose solution is of the form
y = A cos x + B sin x is
d2 y d2 y
(a) 2
+ y =0 (b) − y =0
dx dx2
d2 y d2 y
(c) + y = A cos x (d) + y = B sin x
dx2 dx2 (Kanpur 2016)
D-50
dy
36. The integrating factor of ( x + 1) − y = e3 x (1 + x)2 is
dx
(a) x + 1 (b) log ( x + 1)
1
(c) (d) e x +1
x +1 (Kanpur 2016)
37. The solution of y ′ + y tan x = cos x, y (0 ) = 0 is
(a) cos x (b) sin x
(c) x cos x (d) x sin x (Kanpur 2016)
38. Which one is the homogeneous equation ?
dy x2 + 2 y2 dy x3 + 2 y3
(a) = 2 (b) = 2
dx 3 x + 2 y3 dx 3 x + 2 y2
dy x2 + 2 y2 dy x2 + 2 y2
(c) = 2 (d) =
dx 3 x + 4 y2 dx 3 x2 + 2 y (Kanpur 2016)
x x 2
39. Solution of differential equation 3 e tan y dx + (1 − e )sec y dy = 0 is
x 3 x 2
(a) tan y = c (1 − e ) (b) tan y = c (1 − e )
x 2
(c) sin y = c (e − 1) (d) tan−1 y = c (1 − e x )3
(Kanpur 2016)
dy 3 y 1
40. Solution of differential equation + = 4 is
dx x x
(a) y x3 = log x + c (b) y x3 = x log x + c
c
(c) y x −3 = log x + (d) y x2 = log x + c
x (Kanpur 2016)
41. In solving (2 x − y + 2) dx + (6 x − 3 y + 4) dy = 0 substitution used is
(a) 2 x − y = z (b) 2 x + y = z
(c) x − y = z (d) x + y = z (Kanpur 2016)
2 2
42. The differential equation ( y2 e x y + 6 x) dx + (2 xye x y − 4 y) dy = 0 is
(a) non-linear, non-homogeneous and not exact
(b) non-linear, non-homogeneous and exact
(c) non-linear, homogeneous and exact
(d) linear, homogeneous and exact
∂M ∂N
43. If M dx + N dy = 0 and = then differential equation is
∂y ∂x
(a) linear (b) homogeneous
(c) non-homogeneous (d) exact differential equation
44. The differential equation M dx + N dy = 0 can be reduced to exact if
1  ∂M ∂N 
 −  is
N  ∂y ∂x 
(a) a function of x and y (b) a function of x alone
(c) a function of y alone (d) none of these
D- 51
1  ∂N ∂M 
45. The differential equation  −  is
M  ∂x ∂y 
(a) a function of x and y (b) a function of x alone
(c) a function of y alone (d) none of these
46. The differential equation (ay + x + x ) dx + ( y8 − y + bxy) dy = 0 is exact if
2 8
(a) 2a = b (b) a = b
(c) 2a ≠ b (d) a ≠ b
dy − x2
47. The integrating factor of + 2 xy = e is
dx
2 2
(a) e − x (b) e x
(c) e − x (d) e x
dy
48. The integrating factor of cos2 x + y = tan x is
dx
(a) e tan x (b) e − tan x
cot x
(c) e (d) e − cot x
dy 1
49. The integrating factor of + y = x2 y6 is
dx x
(a) x5 (b) x −5
1 1
(c) 5 (d) − 5
x x
Fill in the Blank(s)

Fill in the blanks “……” so that the following statements are complete and correct.
1. The order of the differential equation
d3 y d2 y dy
3
−2 2
+3 + 2 y = 0 is …… .
dx dx dx
3
 dy  dy
2. The degree of the differential equation   − 4 xy + 8 y2 = 0 is …… .
 dx  dx
2 3
 d2 y    dy  2 
3. The order and degree of the differential equation  2  − 1 +    = 0 is
 dx   dx 
 
…… and …… (Rohilkhand 2006)
dy f ( x, y)
4. A differential equation of the form = 1 , where f1 ( x, y) and f2 ( x, y)
dx f2 ( x, y)
are homogeneous functions of x and y of the same degree, is called a …… .
5. To solve the differential equation x2 y dx − ( x3 + y3 ) dy = 0 , we put…… .
dy 2x + y + 1
6. To solve the differential equation = , we put …… .
dx 4 x + 2 y − 1
7. A differential equation is said to be …… when the dependent variable y and all

its derivatives occur in the first degree only and are not multiplied together.
D-52
dy
8. The integrating factor of the differential equation + Py = Q , where P and Q
dx
are functions of x only, is …… .
dy
9. The integrating factor of the differential equation ( x2 − 1) + 2 xy = 1 is …… .
dx
dy
10. The integrating factor of the differential equation + 2 y tan x = sin x is …… .
dx
(Kumaun 2009)
11. The ordinary differential equation M dx + N dy = 0 , where M and N are
functions of x and y, is exact if and only if …… .
(Meerut 2003; Bundelkhand 08; Rohilkhand 10)
12. A differential equation which is not exact can sometimes be made exact by
multiplying by some suitable function of x and y. Such a function is called an
…… of the equation.
dx 1 tan−1 y
13. The differential equation + 2
⋅x=
dy 1 + y 1 + y2
is linear with …… as dependent variable. (Kumaun 2010)
14. If the differential equation M dx + N dy = 0 is homogeneous and if

Mx + Ny ≠ 0 , then the integrating factor of this differential equation is …… .
(Bundelkhand 2007)
2
15. The differential equation ( x + y) dy = a dx is linear with …… as independent
variable.
16. A solution of differential equation of three order has …… constants.
(Agra 2007)
17. The differential equation y2 dx + ( xy + x2 ) dy = 0 is …… . (Kumaun 2007)
dy
18. Differential equation + Py = Q is …… .
dx (Kumaun 2008)
True or False
Write ‘T’ for true and ‘F’ for false statement.
1. The differential equation y2 dx + ( xy + x2 ) dy = 0 is homogeneous.
2
d3 y  dy 
2. The order of the differential equation − 7   + 6 y = 0 is 2.
dx3  dx 
2
d2 y  dy 
3. The degree of the differential equation + 4   − 5 y = 0 is 2.
dx2  dx 
dy
4. The differential equation x2 + y = 1 is linear with y as dependent variable.
dx
dy
5. The integrating factor of the differential equation + y sec x = tan x is
dx
(sec x + tan x).
D- 53
dy
6. The differential equation ( x + y + 1) = 1 is linear with y as dependent
dx
variable.
7. The differential equation (4 x + 3 y + 1) dx + (3 x + 2 y + 1) dy = 0 is exact.

dy y
8. The integrating factor of the differential equation + = sin x2 is log x.
dx x
dy 1
9. The differential equation + y = x2 y6 is linear with y as dependent
dx x
variable.
dy
10. The differential equation x + y = x2 y4 can be made linear by putting
dx
1
= v.
y3
dy
11. A differential equation of the form + Py = Qy n,
dx
where P and Q are functions of x alone, is called Bernoulli’s equation.
(Meerut 2003)
12. If the differential equation M dx + N dy = 0 is exact, its solution is
∫ M dx +
∫ N dy = c.
treating y as constant taking only those
terms in N which
do not contain x (Meerut 2003)
A nswers

1. (c) 2. (b) 3. (a) 4. (a) 5. (c)
6. (a) 7. (d) 8. (c) 9. (d) 10. (c)
11. (b) 12. (c) 13. (a) 14. (b) 15. (c)
16. (b) 17. (c) 18. (a) 19. (b) 20. (c)
21. (c) 22. (c) 23. (d) 24. (b) 25. (a)
26. (a) 27. (a) 28. (b) 29. (b) 30. (b)
31. (b) 32. (a) 33. (c) 34. (c) 35. (a)
36. (c) 37. (c) 38. (c) 39. (a) 40. (a)
41. (a) 42. (b) 43. (d) 44. (b) 45. (c)
46. (a) 47. (b) 48. (a) 49. (c)
D-54

1. 3 2. 3 3. 2, 2 4. homogeneous equation
5. y = vx 6. 2 x + y = v 7. linear 8. e ∫ P dx
∂M ∂N
9. x2 − 1 10. sec2 x 11. = 12. integrating factor
∂y ∂x
1
13. x 14. 15. y 16. 3
Mx + Ny
17. homogeneous equation 18. linear differential equation
True or False
1. T 2. F 3. F 4. T 5. T
6. F 7. T 8. F 9. F 10. T
11. T 12. T
¨
D-55
2
D ifferential E quations of
T he F irst O rder B ut N ot of
T he F irst D egree
1 Introduction
n the present chapter we shall discuss the solutions of differential equations which
I are of the first order but are of degree higher than one. Such differential equations
will contain only the first differential coefficient dy / dx but it will occur in a degree
higher than one. It is usual to denote dy / dx by p. The general form of such a differential
equation is then
pn + A1 pn − 1 + A2 pn − 2 + … + An − 1 p + An = 0 ,
where A1 , A2 , … , An are some functions of x and y.

Now we shall consider the various methods of solving the differential equations of the
above type.
2 Equations Solvable for p

Suppose a differential equation of first order and of degree n can be solved for p i. e., it
can be resolved into n linear factors in p of the type
{ p − f1( x, y)} { p − f2 ( x, y)}...{ p − f n ( x, y)} = 0 .
D-56
We can equate each factor to zero and the resulting differential equations of the first
order and first degree can be solved. Let their solutions be
φ1 ( x, y, c1) = 0 , φ2 ( x, y, c2 ) = 0 , ..., φn( x, y, c n) = 0 ,
where c1, c2 , … , c n are arbitrary constants. There is no loss of generality if we replace the
arbitrary constants c1, c2 , … , c n by a single arbitrary constant c because in any of the
above n solutions c is free to take any real value. Thus the n solutions of the given
differential equation are
φ1 ( x, y, c ) = 0 , φ2 ( x, y, c ) = 0 , … , φn ( x, y, c ) = 0 .
Combining the above equations, we get a single composite solution as
φ1 ( x, y, c ) φ2 ( x, y, c ) … φn ( x, y, c ) = 0 .
Example 1: Solve p2 − 7 p + 12 = 0 . (Bundelkhand 2008; Kanpur 04)
Solution: Resolving into linear factors, the given differential equation can be written as
( p − 3) ( p − 4) = 0 .
Its component equations are p = 3, p = 4 .
Solving the differential equation p = 3 i. e., dy / dx = 3, we get y = 3 x + c . Also the
solution of the differential equation p = 4 is y = 4 x + c .
So the solutions of the given differential equation are
y = 3 x + c, y = 4 x + c .
The single combined solution is
( y − 3 x − c) ( y − 4 x − c) = 0 .
Example 2: Solve p2 + 2 py cot x = y2 . (Gorakhpur 2005; Rohilkhand 09; Kanpur 08;
Bundelkhand 04)
Solution: The given differential equation is

p2 + 2 py cot x − y2 = 0 .
Solving for p , we get
dy − 2 y cot x ± √ (4 y2 cot2 x + 4 y2 )
p= =
dx 2
= − y cot x ± y cosec x = y (− cot x ± cosec x) .
Thus the component equations are
dy
= y (− cot x + cosec x) , …(1)
dx
dy
and = − y (cot x + cosec x) . …(2)
dx
In each of the above differential equations, the variables are separable.
From (1), separating the variables, we have
D-57
dy
= (− cot x + cosec x) dx .
y
Integrating, we get
1
log y − log c = − log sin x + log tan x
2
 tan 1 x   sin 1 x / cos 1 x 
 y  2   2 2 
or log   = log   = log 
c sin x 1 1 
   2 sin x cos x 
   2 2 
 
 1   1 
= log   = log  .
2 1 1 + cos x
 2 cos x
 2 
∴ y / c = 1/(1 + cos x)
or y = c /(1 + cos x). …(3)
From (2), separating the variables, we have
dy
= − (cot x + cosec x) dx .
y
Integrating, we get
1
log y − log c = − (log sin x + log tan x)
2
1
or log ( y / c ) = − log{(sin x)(tan
x)}
2
1 1 1 1 1
= − log {2 sin x cos x ⋅ (sin x /cos x)} = − log (2 sin2 x)
2 2 2 2 2
= − log (1 − cos x) = log (1 − cos x)−1 = log {1/(1 − cos x)} .
∴ y / c = 1/(1 − cos x)
or y = c /(1 − cos x). …(4)
Thus the solutions of the given differential equation are given by (3) and (4). The
single combined solution is
 c   c 
y−  y−  =0.
 1 + cos x   1 − cos x 
Example 3: Solve ( p − xy) ( p − x2 ) ( p − y2 ) = 0 .
Solution: Equating each factor to zero, the component equations are

p − xy = 0 , …(1)
2
p− x =0 , …(2)
2
and p− y =0 . …(3)
From (1), p = xy
or dy / dx = xy
or (1/ y) dy = x dx.
D-58
Integrating, we get
1 2
log y = x + log c
2
1
or log ( y / c ) = x2
2
1 2
x
or y / c = e2
1 2
x
or y= ce 2 . …(4)
From (2),
p = x2 or dy / dx = x2 or dy = x2 dx .
Integrating, we get
1 3 1
y= x + c
3 3
or 3 y − x3 = c . …(5)
From (3), p = y2
or dy / dx = y2
or (1/ y2 ) dy = dx.
− (1 / y ) = x + c
or xy + cy + 1 = 0 . …(6)
Now (4), (5) and (6) are the solutions of the given differential equation. The single
combined solution is
 1 x2 
 y − ce2  (3 y − x3 − c ) ( xy + cy + 1) = 0 .
 
Example 4: Solve p ( p − y) = x ( x + y) .
Solution: The given differential equation can be written as
p2 − py − ( x2 + xy) = 0 .
Solving for p, we get
dy y ± √ ( y2 + 4 x2 − 4 xy) y ± √ {( y + 2 x)2 } y ± ( y + 2 x)
p= = = = .
dx 2 2 2
Thus the component equations are
dy / dx = { y + ( y + 2 x)} / 2 = y + x, …(1)
and dy / dx = ( y − y − 2 x) / 2 = − x . …(2)
From (1), we have (dy / dx) − y = x , which is a linear differential equation with y as
the dependent variable.
Integrating factor is
− dx
e∫ = e− x .
D-59
Therefore its solution is

ye − x = −x
∫ xe dx + c
ye − x = − xe − x − −x
or
∫ (− e ) dx + c , integrating by parts
or ye − x = − xe − x − e − x + c
or ye − x = − e − x ( x + 1) + c
or y = − ( x + 1) + ce x
or y + x + 1 − ce x = 0 . …(3)
From (2), we have
dy = − x dx .
Integrating, we get
1 2 1
y=− x + c
2 2
or 2 y + x2 − c = 0 . …(4)
Thus (3) and (4) are the required solutions of the differential equation. The single
combined solution is
( y + x + 1 − ce x ) (2 y + x2 − c ) = 0 .
 y4  2 y y2
Example 5: Solve 1 − y2 + 2
 p −2 p+ 2 =0 .
 x  x x
Solution: Multiplying throughout by x2 , the given differential equation becomes

( x2 − x2 y2 + y4 ) p2 − 2 xyp + y2 = 0
or x2 p2 − 2 xyp + y2 = p2 ( x2 y2 − y4 )
or ( xp − y)2 = p2 y2 ( x2 − y2 )
∴ xp − y = ± py √ ( x2 − y2 )
or p [ x ± y √ ( x2 − y2 )] = y
1 dx x ± y √ ( x2 − y2 )
or = = .
p dy y
Putting x = vy so that (dx / dy) = v + y (dv / dy), we have
dv vy ± √ (v2 y2 − y2 )
v+ y = = v ± y √ (v2 − 1)
dy y
dv
or y = ± y √ (v2 − 1)
dy
dv
or = ± √ (v2 − 1)
dy
or dy = ± [1/ √ (v2 − 1)] dv, separating the variables.
D-60
Integrating, we get y + c = ± cosh −1 v

or y + c = ± cosh −1( x / y)
or cosh −1 ( x / y) = ± ( y + c )
or x / y = cosh { ± ( y + c )} .
Since cosh (− x) = cosh x , therefore, we have
x = y cosh ( y + c ) and x = y cosh ( y + c ), as the required solutions.
The single combined solution is [ x − y cosh ( y + c )]2 = 0 .

1. p2 − 5 p + 6 = 0 . (Rohilkhand 2006; Kanpur 02; Kumaun 12)
2 2 2
2. y + xyp − x p = 0 . (Meerut 2007B)
2 2 2
3. xyp − ( x + y ) p + xy = 0 . (Gorakhpur 2011)
2 2 2
4. xyp + p (3 x − 2 y ) − 6 xp = 0 .
5. x (dy / dx)2 + ( y − x)(dy / dx) − y = 0 . (Meerut 2006B; Bundelkhand 09, 10)
2
6. y (dy / dx) + ( x − y)(dy / dx) − x = 0 . (Meerut 2005B)
A nswers 1
1. ( y − 2 x − c) ( y − 3 x − c) = 0
2. ( y2 − c x1+√5 )( y2 − c x1−√5 ) = 0
3. ( x2 − y2 − c ) ( y − cx) = 0
4. ( y − cx2 ) ( y2 + 3 x2 − c ) = 0
5. ( y − x − c ) ( xy − c ) = 0
6. ( y − x − c ) ( x2 + y2 − c 2 ) = 0
3 Equations Solvable for y

Suppose the given differential equation is solvable for y. Then it can be put in the form
y = f ( x, p). …(1)
Differentiating (1) w.r.t. x and denoting dy / dx by p, we obtain
D-61
 dp
p = φ  x, p,  , …(2)
 dx 
which is a differential equation in two variables x and p. Suppose it is possible to solve
the differential equation (2). Let its solution be
F ( x, p, c ) = 0, …(3)
where c is the arbitrary constant.
Eliminating p between (1) and (3), we get the required solution of (1) in the form
ψ ( x, y, c ) = 0 .
If it is not easily practicable to eliminate p between (1) and (3), we may solve (1) and
(3) to get x and y in terms of p and c in the form
x = f1 ( p, c ), y = f2 ( p, c ) , …(4)
which give us the required solution of (1) in the form of parametric equations, the
parameter being p.
Special case: Equations that do not contain x: In this case the equation has the
form f ( y, p) = 0 . If it is solvable for p, it will give
p = φ ( y) i. e., dy / dx = φ ( y) ,
which can be easily solved by separating the variables.
If it is solvable for y, it will give y = ψ ( p), which can be solved by the method just
explained in article 3.
Example 6: Solve y + px = p2 x4 .
(Rohilkhand 2006; Avadh 08; Agra 08; Purvanchal 11; Lucknow 11; Garhwal 10)
Solution: Solving for y, the given differential equation can be written as

y = − px + p2 x4 . …(1)
Differentiating (1) w.r.t. x and denoting dy / dx by p, we get
dp dp
p= − p− x + 2 p x4 + 4 p2 x3
dx dx
dp
or 2 p − 4 p2 x3 + x (1 − 2 x3 p) = 0
dx
dp
or 2 p (1 − 2 x3 p) + x (1 − 2 x3 p) = 0
dx
 dp
or (1 − 2 x3 p) 2 p + x  = 0 .
 dx 
dp
∴ 2p + x = 0, …(2)
dx
D-62
and 1 − 2 x3 p = 0 .
From (2), dp / dx = −2 p / x
or dp / p = −(2 / x) dx .
Integrating, we get
log p = − 2 log x + log c
or log px2 = log c
or px2 = c or p = c / x2 .
Substituting this value of p in (1), we get
y = − x (c / x2 ) + (c 2 / x4 ). x4
or y = − (c / x) + c 2
or xy = c 2 x − c , as the required solution of (1).
Note: If we eliminate p between (1) and (3), we get
y = − (1/ 2 x3 ) x + (1/ 4 x6 ). x4 = − (1/ 2 x2 ) + (1/ 4 x2 ) = −1/ 4 x2 ,
which is also a solution of (1) because it satisfies (1). This solution itself does not
contain any arbitrary constant. Also it cannot be obtained from the general solution
xy = c 2 x − c by giving any particular value to c. Such a solution is called the singular
solution and we shall discuss it in details later on.
Example 7: Solve y = 2 px − p2 . (Meerut 2001, 04B; Kanpur 07, 15; Kumaun 09, 11)

y = 2 px − p2 . …(1)
Differentiating (1) w.r.t. x and denoting dy / dx by p, we get
dp dp
p = 2p + 2x − 2p
dx dx
dp
or p+2 ( x − p) = 0
dx
dx
or p + 2x = 2p
dp
dx 2
or + x = 2, …(2)
dp p
which is a linear differential equation with x as dependent variable and p as the

independent variable.
2
∴ integrating factor = e ∫ (2 / p) dp = e2 log p = e log p = p2 .
2
the solution of (2) is xp2 = 2 3
∴
∫ 2 p dp + c = 3 p +c
2
or x= p + cp−2 …(3)
3
D-63
Here it is not easily practicable to eliminate p between (1) and (3). So putting the value
2
of x from (3) in (1), we get y = 2 p ( p + cp−2 ) − p2
3
1 2
or y = p + 2 cp−1. …(4)
3
The equations (3) and (4), which express x and y in terms of a parameter p, constitute
the required solution of (1).
2
dy  dy 
Example 8: Solve y − x = x +  .
dx  dx  (Meerut 2008; Kumaun 08)
Solution: Denoting dy / dx by p, the given differential equation can be written as

y − x = xp + p2 .
Solving it for y, we get y = x (1 + p) + p2 . …(1)
Differentiating (1) w.r.t. x and writing p for dy / dx, we get
dp dp
p = 1+ p + x + 2p
dx dx
dp dp
or 1+ x + 2p =0
dx dx
dx
or + x = − 2 p, …(2)
dp
which is a linear differential equation.
Here the I.F. = e ∫ 1 dp = e p .
∴ the solution of (2) is
x ep = p
∫ − 2 pe dp + c
or x e p = −2 ( pe p − e p ) + c , integrating by parts
or x e p = − 2 e p ( p − 1) + c
or x = − 2 ( p − 1) + ce − p . …(3)
Substituting this value of x in (1), we get
y = (1 + p) { − 2 ( p − 1) + ce − p } + p2
or y = c (1 + p) e − p + 2 − p2 . …(4)
the required solution of (1).
Example 9: Solve y = a √ (1 + p2 ).

y = a √ (1 + p2 ) . …(1)
1 a dp
p = a ⋅ (1 + p2 )−1 /2 ⋅ 2 p (dp / dx) or = dx.
2 √ (1 + p2 )
D-64
Integrating, we get
x = a log { p + √ (1 + p2 )} + c . …(2)
Now from (1),
y2 = a2 + a2 p2 or a2 p2 = y2 − a2 or p = √ ( y2 − a2 ) / a .
Putting √ (1 + p2 ) = y / a and p = √ ( y2 − a2 ) / a in (2), we get
 √ ( y2 − a2 ) y 
x = a log  + +c
 a a
or x = a log { y + √ ( y2 − a2 )} − a log a + c
or x = a log { y + √ ( y2 − a2 )} + c ,
writing c for − a log a + c because c is an arbitrary constant.
Hence the required solution is
x = log { y + √ ( y2 − a2 )} + c .

1. y = 2 px + p4 x2 . (Meerut 2006; Avadh 09)
3
2. 4 p + 3 xp = y .
3. x2 + p2 x = yp .
4. y − 2 x p = f ( xp2 ) . (Meerut 2007)

2
5. y = x { p + √ (1 + p )} . (Kanpur 2009; Rohilkhand 09)
6. y = sin p − p cos p. (Gorakhpur 2006)
7. By differentiating with respect to x the equation p3 + xp2 = y, obtain its general
solution in the form x = f ( p), y = φ ( p) .
A nswers 2
1. ( y − c 2 )2 = 4 cx
12 2 8
2. x=− p + cp−3 /2 , y = − p3 + 3 cp−1 /2
7 7
1 1 1
3. x = − p3 + cp1 /2 , y = (− p2 + cp1 /2 )2 / p + p2 (− p2 + cp1 /2 )
3 3 3
4. y = 2 c √ x + f (c 2 )
5. x2 + y2 − 2 xc = 0
D-65
6. x = c − cos p , y = sin p − cos p

1
(2 c + 3 p2 − 2 p3 ) 2 cp2 + 2 p3 − p4
7. x= 2 ,y=
2
( p − 1) 2 ( p − 1)2
4 Equations Solvable for x

Suppose the given differential equation is solvable for x. Then it can be put in the form
x = f ( y, p) . …(1)
Differentiating (1) w.r.t. y and writing 1 / p for dx / dy, we get
1  dp
= φ  y, p,  , …(2)
p  dy 
which is a differential equation in two variables y and p. Suppose it is possible to solve

the differential equation (2). Let its solution be
F ( y, p, c ) = 0, …(3)
Eliminating p between (1) and (3), we get the required solution of (1) in the form
ψ ( x, y, c ) = 0 .
In case it is not easily practicable to eliminate p between (1) and (3), we may solve (1)
and (3) to get x and y in terms of p and c in the form
x = f1( p, c ), y = f2 ( p, c ), …(4)
which give us the required solution of (1) in the form of parametric equations, the
parameter being p.
Special Case: Equations that do not contain y:
In this case the equation has the form f ( x, p) = 0 .
If it is solvable for p, it will give
p = φ ( x)
i. e., dy / dx = φ ( x) , which can be easily integrated.
If it is solvable for x, it will give x = ψ ( p), which can be solved by the method just
explained in article 4.
Example 10: Solve y = 2 px + y2 p3 .

(Purvanchal 2007, 10; Lucknow 10; Avadh 07; Kumaun 07)
Solution: Solving for x, the given differential equation can be written as

2 px = y − y2 p3
D-66
y y2 p2
or x= − . …(1)
2p 2
Differentiating (1) w.r.t. y and writing 1/p for dx / dy, we obtain
1 1 y dp 2 yp2 y2 dp
= − 2 − − ⋅2p
p 2 p 2 p dy 2 2 dy
1  y  dp
or + yp2 +  2 + py2  =0
2p 2 p  dy
 1   1  dp
or p  2 + py + y  2 + py =0
2 p  2 p  dy
 1  dp
or  + py  p + y  = 0 .
2 dy 
2 p 
dp 1
∴ p+ y =0 or + py = 0 .
dy 2 p2
1
The equation + py = 0 will give us the singular solution of (1).
2 p2
dp
From p + y = 0, we have
dy
dp p
=− or (1/ p)dp = −(1/ y) dy.
dy y
Integrating, we get
log p = − log y + log c or log py = log c
or py = c or p = c / y.
Substituting this value of p in the given differential equation, we get
y = 2 x ⋅ (c / y) + y2 (c 3 / y3 ) or y = 2 cx / y + c 3 / y
or y2 = 2 cx + c 3 , which is the required solution.
Example 11: Solve the differential equation

p3 − 4 xyp + 8 y2 = 0 . (Garhwal 2008)
Solution: Solving the given differential equation for x, we get

p2 2 y
4 xyp = p3 + 8 y2 or x= + . …(1)
4y p
Differentiating (1) w.r.t. y and writing 1/ p for dx / dy, we get
1 p2 2 p dp 2 2 y dp
=− 2
+ + −
p 4y 4 y dy p p2 dy
dp  p 2 y p2 1
or  − 2  = −
dy 2 y 2 p
 p  4 y
D-67
dp  p 2 y p  p 2 y
or  − 2  =  − 2 
dy  2 y p  2y 2 y p 
dp p cancelling the common factor on either side which
or = ,
dy 2 y corresponds to the singular solution of (1)
or (2 / p) dp = (1/ y) dy, separating the variables.
Integrating, we get
2 log p = log y + log c or log p2 = log cy
or p2 = cy . …(2)
Now we shall eliminate p between the given differential equation and the equation (2).
From the given differential equation, we have
8 y2 = p (4 xy − p2 )
or 8 y2 = (cy)1 /2 (4 xy − cy), substituting for p from (2)
or 8 y2 = c1 /2 y3 /2 (4 x − c ) or 8 y1 /2 = c1 /2 (4 x − c )
1 2
or 64 y = c (4 x − c )2 or 64 y = 16 c ( x − c) ,
4
1 1 1
or y= c ( x − c )2 or y = c ( x − c ) , writing c for c.
4 4 4
Hence the required solution is y = c ( x − c )2 .
Example 12: Solve ap2 + py − x = 0 . (Rohilkhand 2005; Meerut 13B)
Solution: Solving the given differential equation for x, we get

x = py + ap2 . …(1)
Differentiating (1) w.r.t. y and writing 1/ p for dx / dy, we get
1 dp dp
= p+ y + 2 ap
p dy dy
1 − p2 dp dp
or = y + 2 ap
p dy dy
1 − p2 dy
or − y = 2 ap, multiplying both sides by dy / dp
p dp
dy p 2 ap2
or + 2 y=− 2 , …(2)
dp p − 1 p −1

2 1 log ( p2 −1)
Here the I.F. = e ∫ {p /( p −1)}dp
= e2 = ( p2 − 1)1 /2 .
− 2 ap2 ( p2 − 1) + 1
y ( p2 − 1 )1 /2 = ( p2 − 1)1 /2 dp + c = − 2 a
∫ p2 − 1 ∫ √ ( p2 − 1)
dp + c
D-68
 1 
= − 2 a √ ( p2 − 1) +
∫
2  dp + c
√ ( p − 1)
1 1
= − 2 a [ p √ ( p2 − 1) − cosh −1 p + cosh −1 p] + c
2 2
= − ap √ ( p2 − 1) − a cosh −1 p + c
c − a cosh −1 p
or y= − ap. …(3)
√ ( p2 − 1)
Substituting this value of y in (1), we get
 c − a cosh −1 p  p (c − a cosh −1 p)
x = p  2
− ap + ap2 or x= . …(4)
 √ ( p − 1)  √ ( p2 − 1)
The equations (3) and (4) constitute the parametric equations of the required
solution.

1. y2 log y = xyp + p2 . (Agra 2007; Purvanchal 08)
2
2. p y + 2 px − y = 0 . (Meerut 2005; Purvanchal 06; Lucknow 08; Kumaun 10)
2 4
3. 4 ( xp + yp) = y .
4. (2 x − b) p = y − ayp2 . 5. xp3 = a + bp .
6. p = tan { x − p /(1 + p2 )} . 7. x + p / √ (1 + p2 ) = a .
A nswers 3
1. log y = cx + c 2 2. y2 − 2 cx + c 2 = 0
3. y = 4 c ( xyc + 1) 4. ac 2 + (2 x − b) c − y2 = 0
5. x = (a / p3 ) + (b / p2 ), y = (3 a / 2 p2 ) + (2 b / p) + c
6. x = p /(1 + p2 ) + tan−1 p, y = c − 1/(1 + p2 ) 7. ( x − a)2 + ( y + c )2 = 1
5 Clairaut’s Equation
(Meerut 2004; Lucknow 08)
(a) The differential equation
y = px + f ( p) , …(1)
D-69
is known as Clairaut’s equation. Here f ( p) is some function of p only. This differential

equation is very important and the students should note its form carefully. To solve
this differential equation, we shall obviously adopt the method explained in article 3.
Thus differentiating (1) w.r.t. x and writing p for dy / dx, we get
dp dp
p= p+ x + f ′ ( p)
dx dx
dp
or { x + f ( p)} = 0 .
dx
dp
∴ =0. …(2)
dx
or x + f ′ ( p) = 0 . …(3)
From (2), we have p = constant = c, (say). …(4)
Eliminating p between (1) and (4), we obtain
y = cx + f (c ) , …(5)
which is the required general solution of (1).
If we eliminate p between (1) and (3), we get the singular solution of (1).
Remember: To obtain the general solution of a differential equation in Clairaut’s form
simply replace p by c.
(b) To solve the differential equation
y = x f1 ( p) + f2 ( p). …(1)
This differential equation is not in Clairaut’s form. However it can be solved by the
method we adopted in solving Clairaut’s equation. Thus differentiating (1) w.r.t. x
and writing p for dy / dx, we get
dp dp
p = f1( p) + x f1 ′ ( p) + f2 ′ ( p)
dx dx
dp dp
or p − f1( p) = x f1 ′ ( p) + f2 ′ ( p)
dx dx
dx
or [ p − f1( p)] − x f1 ′ ( p) = f2 ′ ( p)
dp
dx f ′ ( p) f ′ ( p)
or + 1 ⋅x= 2 …(2)
dp f1( p) − p p − f1( p)
which is a linear differential equation with x as the dependent variable and p as the
independent variable.
Let the solution of (2) be
φ ( x, p, c ) = 0 . …(3)
Then eliminating p between (1) and (3), we get the required solution.
Example 13: Solve y = x (dy / dx) + (dy / dx)2 . (Meerut 2011; Kumaun 14; Kanpur 15)
D-70
Solution: Denoting dy / dx by p, the given differential equation is y = px + p2 , which is

in Clairaut’s form. So replacing p by the arbitrary constant c, the required solution is
y = cx + c 2 .
Example 14: Solve p = log ( px − y). (Gorakhpur 2007; Lucknow 09; Garhwal 09)
Solution: The given differential equation can be written as px − y = e p or y = px − e p ,

which is in Clairaut’s form. So replacing p by the arbitrary constant c in the given
differential equation, the required solution is c = log (cx − y) .
Example 15: Solve sin px cos y = cos px sin y + p.
(Meerut 2003; Rohilkhand 06, 08, 10; Lucknow 07)

sin px cos y − cos px sin y = p
or sin ( px − y) = p or px − y = sin−1 p
or y = px − sin−1 p, which is in Clairaut’s form.
So changing p to the arbitrary constant c, the required solution is
y = cx − sin−1 c .
Example 16: Solve p2 ( x2 − a2 ) − 2 pxy + y2 − b2 = 0 .

p2 x2 − 2 pxy + y2 = p2 a2 + b2
or ( y − px)2 = b2 + a2 p2 or y = px ± √ (b2 + a2 p2 ),
each of which is in Clairaut’s form. Hence the required solution is
( y − cx)2 = b2 + a2 c 2 .
Example 17: Solve y = 2 px + pn.

y = 2 px + pn. …(1)
p = 2 p + 2 x (dp / dx) + npn −1(dp / dx)
or p + 2 x (dp / dx) = − npn −1(dp / dx)
or p (dx / dp) + 2 x = − npn −1, multiplying both sides by dx / dp
dx 2
or + x = − npn − 2 , …(2)
dp p
2
Hence the I.F. = e ∫ (2 / p) dp = e2 log p = e log p = p2 .
xp2 = − n −2 2
∫np p dp + c
D-71
xp2 = − n pn dp + c = − n pn +1 /(n + 1) + c
or
∫
or x = cp−2 − { n /(n + 1)} pn −1. …(3)
Substituting this value of x in (1), we get
y = 2 p [cp−2 − { n / (n − 1)} pn −1] + pn
2n n n −1 n
or y = 2 cp−1 + pn − p = 2 cp−1 − p. …(4)
n +1 n +1
the required solution.
6 Equations Reducible to Clairaut’s Form

Some differential equations by suitable change of variables may be reduced to
Clairaut’s form.
Example 18: Use the transformation x2 = u, y2 = v to solve
( px − y) ( py + x) = h2 p . (Agra 2006; Lucknow 08, 10; Kanpur 15)

p2 xy + p ( x2 − y2 − h2 ) − xy = 0 . …(1)
Putting x2 = u and y2 = v, we have 2 x dx = du and 2 y dy = dv .
y dy dv dy x dv x dv
∴ = or = or p= .
x dx du dx y du y du
Putting this value of p in the differential equation (1), we get
2
x2  dv  x dv 2
xy   + ( x − y2 − h2 ) − xy = 0
y2  du y du
2
x  2  dv  2 2 2 dv 
or x   + (x − y − h ) − y2  = 0
y   du du 
or x2 (dv / du)2 + ( x2 − y2 − h2 ) (dv / du) − y2 = 0
or u (dv / du)2 + (u − v − h2 )(dv / du) − v = 0 ,
[putting u for x2 and v for y2 ]
or u P2 + (u − v − h2 ) P − v = 0 , where P = dv / du
or u P ( P + 1) − v ( P + 1) − h2 P = 0
or v ( P + 1) = uP ( P + 1) − h2 P
or v = uP − h2 P /( P + 1), which is in Clairaut’s form y = px + f ( p) .
∴ replacing P by the arbitrary constant c, we get the required solution as
v = uc − h2 c /(c + 1) or y2 = cx2 − h2 c / (c + 1).
D-72

1. y = px + a / p. (Rohilkhand 2006, 07; Agra 05; Gorakhpur 08, 11; Purvanchal 06)
2. y = px + ap (1 − p) . (Agra 2006)
3
3. y = px + p − p . (Kanpur 2010)
4. ( y − px) ( p − 1) = p. (Agra 2005; Gorakhpur 06; Lucknow 09, 11)
2
5. ( x − a) p + ( x − y) p − y = 0 . (Agra 2007; Garhwal 07, 08)
6. p2 ( x2 − a2 ) − 2 pxy + y2 + a2 = 0 .
7. y2 + x2 (dy / dx)2 − 2 xy (dy / dx) = 4(dx / dy)2 .
8. cos y cos px + sin y sin px = p
9. 9 ( y + x p log p) = (2 + 3 log p) p3 .
10. x2 ( y − px) = yp2 . (Gorakhpur 2006, 09)
[Hint. Put x2 = u and y2 = v]
11. e3 x ( p − 1) + p3 e2 y
=0. [Hint. Put e x = u and e y = v .]
A nswers 4
1. y = cx + a / c 2. y = cx + ac (1 − c )
3. y = cx + c − c 3 4. ( y − cx) (c − 1) = c
5. ( x − a) c 2 + ( x − y) c − y = 0 6. c 2 ( x2 − a2 ) − 2 cxy + y2 + a2 = 0
7. ( y − cx)2 = 4 / c 2 8. y = cx − cos −1 c
1 1
9. x = p2 + c p−1, y = − xp log p + (2 + 3 log p) p3
3 9
10. y2 = cx2 + c 2 11. e y = ce x + c 3
7 Geometrical Meaning of a Differential Equation

of the First Order
(Avadh 2010)
Let f ( x, y, p) = 0 , …(1)
be a differential equation of the first order, where p = dy / dx.
Suppose the general solution of (1) is
φ ( x, y, c ) = 0, …(2)
D-73
For each real value of c the equation (2) represents a curve. Thus (2) is the equation of a
family of curves; c being the parameter for the family. Hence every differential equation of
the first order represents a family of curves.
Let us take any point ( x1, y1) in the plane. If we substitute the coordinates of this point
in (1) and (2) and solve the resulting equations in p and c, the values of c so obtained are
the values of the parameter for the curves of the family (2) which pass through ( x1, y1)
and the values of p are the slopes of the tangents to these curves at the point ( x1, y1).
Naturally the degree of c in (2) must be equal to the degree of p in (1).
8 Singular Solutions
(Gorakhpur 2009)
Sometimes a differential equation (for degree higher than one) possesses a solution
which does not contain any arbitrary constant and which cannot be derived from the
general solution of the differential equation by giving a particular value to the
arbitrary constant. Such a solution is called a singular solution and it is generally not
included in the general solution of the differential equation.
The singular solution of a differential equation is given by the envelope of
family of curves represented by that differential equation.
Whenever the envelope of the family of curves
φ ( x, y, c ) = 0 , …(1)
represented by the general solution of the differential equation
f ( x, y, p) = 0 …(2)
exists, the equation of the envelope is the singular solution of the differential equation (2).
Suppose that the family of curves (1) possesses an envelope. For any point P ( x, y) on
the envelope, there exists a curve of the family (1), say φ ( x, y, c ) = 0, which touches the
envelope at ( x, y). The values of x, y, dy / dx for the curve at P satisfy the differential
equation (2). But the values of x, y, dy / dx at P for the envelope are the same as for the
curve. Hence the values of x, y, dy / dx at each point of the envelope satisfy the
differential equation (2). Consequently the envelope of (1) is also a solution of (2).
This solution does not contain any arbitrary constant and in general, cannot be
obtained from (1) by giving any particular value to the arbitrary constant c. Hence this
envelope is the singular solution of (2).
9 Determination of Singular Solution with the Help of

c-Discriminant and p-Discriminant Relations
The discriminant: Let F (c ) = 0 be an algebraic equation. The discriminant of this
equation is the simplest function of the coefficients the vanishing of which represents
the condition that the equation should have two equal roots. Thus the discriminant of
the quadratic Ac 2 + Bc + C = 0 is B2 − 4 AC. The equation
D-74
B2 − 4 AC = 0 is called the discriminant relation.

Let the given differential equation be
f ( x, y, p) = 0 . …(1)
and let its general solution be
φ ( x, y , c ) = 0 . …(2)
The envelope of the family of curves (2) is contained in the locus obtained on
eliminating c, between (2) and
∂φ ( x, y, c )
=0. …(3)
∂c
Let the c-eliminant between (2) and (3) be
ψ ( x, y) = 0 . …(4)
Since the equation (4) may sometimes represent other loci besides the envelope,
therefore only that part of the locus (4) is the singular solution which also satisfies the
given differential equation (1). Therefore if the equation (4) fails to satisfy (1), we
should resolve it into others that are simpler. Then we should try for each part whether
it satisfies the differential equation (1) or not. Only those parts will constitute the
singular solution which satisfy (1).
From our knowledge of the theory of equations we know that the c-eliminant between
(2) and (3) gives the condition for the equation (2) in c to have two equal roots.
Therefore we shall call this c-eliminant the c-discriminant relation. The singular
solution is contained in this c-discriminant relation. From the geometrical point of
view the c-discriminant relation (4) is the locus of the points ( x, y) such that two of the
curves of the family (2) through ( x, y) coincide. Now if two curves of the family
coincide, their tangents must also coincide and so the corresponding values of p i. e.,
(dy / dx) given by (1) must also coincide. The condition for the equation (1) in p to have
two equal roots is obtained on eliminating p between (1) and
∂f ( x, y, p)
=0. …(5)
∂p
The p-eliminant between (1) and (5) is called the p-discriminant relation and the
envelope and hence the singular solution is also contained in it.
Note 1: From the above discussion it is clear that if p occurs only in the first degree in the
differential equation, there will be no singular solution.
Note 2: A differential equation which possesses a singular solution is not considered

completely solved until the singular solution also has been found. Therefore while
solving a differential equation of the first order but higher degree we must find the
general solution and also discuss the singular solution.
D-75
10 Working Rule for Finding the Singular Solution

Let the given differential equation be f ( x, y, p) = 0 . …(1)
Find the general solution of (1) and let it be φ ( x, y, c ) = 0 . …(2)
Find the c-discriminant relation by eliminating c between
∂φ ( x, y, c )
φ ( x, y, c ) = 0 and =0.
∂c
Also find the p-discriminant relation by eliminating p between
∂ f ( x, y, p)
f ( x, y, p) = 0 and =0.
∂p
The locus common to both the c-discriminant and the p-discriminant relations gives
us the singular solution of (1) provided it satisfies (1).
However, we can find only one of these two discriminant relations and only that part
of the locus contained in it is the singular solution which also satisfies the given
differential equation (1).
11 The Singular Solution of Clairaut’s Equation

We know that the general solution of the Clairaut’s equation
y = px + f ( p) …(1)
is y = cx + f (c ) …(2)
Differentiating (2) partially w.r.t. c, we get
0 = x + f ′ (c ) . …(3)
The singular solution, which is the envelope of (2), is obtained by eliminating c
between (2) and (3).
Now differentiating (1) partially w.r.t.p, we get
0 = x + f ′ ( p). …(4)
The equation (1) and (4) differ from the equations (2) and (3) only in having p instead
of c. Therefore the c-eliminant between (2) and (3) and the p-eliminant between (1)
and (4) are identical and either of them gives us the singular solution. Hence the
singular solution of the Clairaut’s equation (1) is obtained by eliminating p
between (1) and (4).
Note: The equation (4) is the same as the equation (3) of article 5 (a).
Example 19: Find the general and singular solutions of

y2 − 2 pxy + p2 ( x2 − 1) = m2 . (Bundelkhand 2010; Kumaun 15)

D-76
( y − px)2 = p2 + m2
or y − px = ± √ ( p2 + m2 ) or y = px ± √ ( p2 + m2 ) ,
either of which is in Clairaut’s form. Hence the general solution is obtained on

replacing p by the arbitrary constant c in the given differential equation. Thus the
general solution is
( y − cx)2 = c 2 + m2
or c 2 ( x2 − 1) − 2 xyc + y2 − m2 = 0 . …(1)
Now from the given differential equation and from the general solution (1) both the p
and the c discriminant relations are
4 x2 y2 − 4 ( x2 − 1) ( y2 − m2 ) = 0
or x2 m2 + y2 − m2 = 0 or y2 + m2 x2 = m2 ,
which is the singular solution.

Example 20: Find the general and singular solution of the differential equation.
( xp − y)2 = p2 − 1 . (Bundelkhand 2008; Lucknow 06)
Solution: From the given differential equation, we have

xp − y = ± √ ( p2 − 1) or y = px ± √ ( p2 − 1)
either of which is in Clairaut’s form. Hence the general solution is obtained on

replacing p by the arbitrary constant c in the given differential equation.
Thus the general solution is
( xc − y)2 = c 2 − 1
or c 2 ( x2 − 1) − 2 xyc + y2 + 1 = 0 . …(1)
The envelope of the family of curves (1) is the singular solution.
From (1) the c-discriminant relation is
4 x2 y2 − 4 ( x2 − 1) ( y2 + 1) = 0
or − x2 + y2 + 1 = 0
or x2 − y2 = 1 ,
which is obviously also the p-discriminant relation. Since the c-discriminant relation
contains only one locus, therefore it gives us the envelope of the family of curves (1).
Hence the singular solution is x2 − y2 = 1 .
Example 21: Find the general and singular solution of

9 p2 (2 − y)2 = 4 (3 − y).
Solution: Solving the given differential equation for p, we have

dy 2 √ (3 − y)
p= =±
dx 3 2− y
D-77
3 2− y
or dx = ± dy , separating the variables.
2 √ (3 − y)
3 2− y 3 2 − (3 − t2 )
x+c=±
2 ∫ √ (3 − y)
dy = ±
2 ∫ t
(−2 t) dt ,
putting 3 − y = t2 so that − dy = 2 t dt
1
= ± 3 (t2 − 1) dt = ± 3 ( t3 − t) = ± t (t2 − 3)
∫ 3
= ± √ (3 − y) (− y), [∵ 3 − y = t2 ]
∴ ( x + c )2 = y2 (3 − y), …(1)
which is the general solution of the given differential equation.

Differentiating (1) partially w.r.t. c, we get
2 ( x + c) = 0 i. e., x + c =0 . …(2)
Eliminating c between (1) and (2), we get the c-discriminant relation as

y2 (3 − y) = 0 .
Now y = 0 gives dy / dx = p = 0 . Putting y = 0 and p = 0 in the given differential
equation, we see that these values of y and p do not satisfy it. Therefore y = 0 is not a
solution.
Again 3 − y = 0 gives y = 3 and so dy / dx = p = 0 . Putting y = 3 and p = 0 in the given
differential equation, we see that these values of y and p satisfy it. Therefore y = 3 is
the singular solution.
Example 22: Reduce the differential equation ( px − y) ( x − py) = 2 p to Clairaut’s form by

substituting x2 = u and y2 = v and find its complete primitive and the singular solution.
Solution: We put u = x2 and v = y2 , so that
du / dx = 2 x and dv / dx = 2 y (dy / dx) = 2 py.

dv dv / dx 2 py py
∴ = = =
du du / dx 2x x
or P = py / x, where P = dv / du
or p = xP / y.
Substituting this value of P in the given differential equation, we get
 x2 P  P
 − y ( x − x P) = 2 x or x ( x2 P − y2 ) (1 − P) = 2 x P
 x  y
or ( x2 P − y2 ) (1 − P) = 2 P or (u P − v) (1 − P) = 2 P
or u P − v = 2 P /(1 − P)
D-78
or v = uP − 2 P /(1 − P),
which is in Clairaut’s form.
Hence its general solution is v = uc − 2 c /(1 − c )
or y2 = x2 c − 2 c /(1 − c ) or y2 (1 − c ) = x2 c (1 − c ) − 2 c
or c 2 x2 − c ( x2 + y2 − 2) + y2 = 0 . …(1)
Now (1) is a quadratic in c. So the c-discriminant relation is
( x2 + y2 − 2)2 − 4 x2 y2 = 0
or ( x2 − 2 xy + y2 − 2) ( x2 + 2 xy + y2 − 2) = 0
or {( x − y)2 − 2} {( x + y)2 − 2} = 0
or ( x − y − √ 2)( x − y + √ 2)( x + y − √ 2)( x + y + √ 2) = 0 .
All the four equations x − y − √ 2 = 0 , x − y + √ 2 = 0 , x + y − √ 2 = 0 and
x + y + √ 2 = 0 satisfy the given differential equation and are therefore the singular
solutions.
1. Find the complete primitive and singular solution of y = px + √ (b2 + a2 p2 ).

Interpret your results geometrically. (Avadh 2010)
2. Find the general and singular solutions of y = px + a / p .

3. Solve and examine for singular solution the differential equation
x3 p2 + x2 yp + a3 = 0 . (Gorakhpur 2007)
3
4. Find the general and singular solution of 27 y − 8 p = 0 .
5. Solve and examine for singular solution the equation xp2 − ( x − a)2 = 0 .
6. Find the general and singular solution of

(a) ( y − px)2 + a2 p = 0 . (b) 3 xy = 2 px2 − 2 p2 .
(c) p2 + y2 = 1 .
7. Examine y2 (1 + p2 ) = r2 for singular solution. (Bundelkhand 2009)
2 2 2
8. Reduce the equation x p + py (2 x + y) + y = 0 , where p = (dy / dx) to
Clairaut’s form by putting u = y and v = xy and find its complete primitive and
also its singular solution. (Gorakhpur 2010)
9. Solve the differential equation ( px2 + y2 ) ( px + y) = ( p + 1)2 by reducing it to

Clairaut’s form and find its singular solution. (Purvanchal 2009)
D-79
10. Reduce the differential equation y = 2 px + y2 p3 to Clairaut’s form by putting

y2 = v and hence find its general and singular solutions.
11. Solve and test for singular solutions p3 − 4 pxy + 8 y2 = 0 .
A nswers 5
1. y = cx + √ (b2 + a2 c 2 ); x2 / a2 + y2 / b2 = 1
2. y = cx + a / c ; y2 = 4 ax
3. c 2 + cxy + a3 x = 0 ; x = 0 , xy2 − 4 a3 = 0
4. y2 = ( x + c )3 , y = 0
5. 9 ( y + c )2 = 4 x ( x − 3 a)2 , x = 0
6. (a) ( y − cx)2 + a2 c = 0 ; xy = a2 / 4 (b) (3 y + 2 c )2 = 4 cx3 ; x3 − 6 y = 0
(c) y = cos ( x + c ) ; y = ± 1
7. y2 + ( x + c )2 = r2 ; y = ± r
8. xy = cy + c 2 , y = 0 , y + 4 x = 0
9. c 2 ( x + y) − cxy − 1 = 0 ; x2 y2 + 4 ( x + y) = 0
4 3
10. y2 = cx + c 3 / 8 ; 27 y4 + 32 x3 = 0 11. y = c ( x − c )2 ; y = 0 , y = x
27

(b), (c) and (d).
1. Solution of the differential equation p2 − 8 p + 15 = 0 is
(a) p = 5 , p = 3 (b) ( y − 5 x − c ) ( y − 3 x − c ) = 0
(c) ( y + 5 x) ( y + 3 x + c ) = 0 (d) none of these
(Kumaun 2006, 08)
2. Solution of the equation y2 log y = xyp + p2 is
(a) log y = cx + x2 (b) log y = cx2 + e x
(c) log x = cy + y2 (d) none of these.
2
3. Solution of the equation y = px + p is
(a) y = e x + c (b) y = c x + c 2
(c) y = log cx (d) x = e y + c .
D-80
4. Which of the following equations is Clairaut’s equation ?

(a) x = py + f ( p) (b) y = px + f (c )
(c) y = px + f ( p) (d) y = x2 + c
5. Solution of the equation p2 − 5 p + 6 = 0 is
(a) y = 2 x + c (b) y = 3 x + c
(c) ( y − 2 x − c ) ( y − 3 x − c ) = 0 (d) none of these
(Garhwal 2007; Kanpur 16)
a
6. Solution of the equation y = px + is
p
a
(a) y = cx (b) y = cx +
c
a
(c) y + cx = a (d) cy = x +
c
(Garhwal 2008; Kumaun 14; Kanpur 16)
2
7. Solution of the equation ( x − a) p + ( x − y) p − y = 0 is
ac 2 ac 2
(a) y = cx − (b) y = cy +
c +1 x +1
ac 2 ac 2
(c) y = cx − (d) x = cy − (Garhwal 2009)
y +1 c +1
8. Solution of equation y = px + log p is
(a) y = e x + c (b) y = cx + log c
(c) y = log cx (d) x = e y + c (Garhwal 2011)
4 2
9. The solution of the equation y + px = x p is
(a) xy = − c + c 2 x (b) x2 = c − c 2 x
(c) y2 = cx + c (d) xy2 = cx − c (Garhwal 2010)
10. The solution of y = px + f ( p) is
(a) y = cx + f (c ) (b) y + cx = f (c )
(c) y − cx = f (c ) (d) y = f (c ) − cx
(Garhwal 2005, 06, 14)
11. Solution of the differential equation y = px + e p is
(a) y = − ax + e p (b) y = ax + e a
(c) y = ax + e − a (d) None of these
(Kumaun 2006; Garhwal 15)
12. General solution of differential equation sin ( y − px) = p is

(a) y = c x − sin−1 c (b) y = c x − cos −1 c
(c) y = c x + sin−1 c (d) y = c x2 + sin−1 c (Kanpur 2016)
a
13. Singular solution of y = px + represents
p
(a) ellipse (b) circle
(c) parabola (d) none of these (Kanpur 2016)
D-81
14. General solution of differential equation p = log ( px − y) is

(a) y = c x − e c (b) y = c x + e c
(c) y = c x2 + e 2 x (d) y = c x2 − e 2 x (Kanpur 2016)
2
 dy 
15. General solution of differential equation   − ax3 = 0 is
 dx 
(a) 25 ( y − c ) = 4 ax5 (b) 25 ( y + c ) = 4 ax5
(c) 25 ( y + c )2 = 4 ax5 (d) 25 ( y − c )2 = 4 ax5
(Kanpur 2016)
2
16. General solution of the differential equation xp − yp + a = 0 is
(a) xc 2 − yc + a = 0 (b) xc 2 + yc + a = 0
x y
(c) 2
− + a=0 (d) none of these
c c
dy
17. General solution of differential equation y = x + e (dy / dx) is
dx
(a) y = c x + ce (b) y = c x + e c
(c) y = c x + e 1 / c (d) none of these
dy
18. Let p = , then the solution of y = px + 1 − p2 is
dx
(a) y = c x (b) y = c x + 1 − c 2
(c) y = x2 + c (d) y = c1 + c2 x (Kanpur 2016)
3
19. General solution of the differential equation y = px + p − p is
(a) y = c x + c − c 3 (b) y = x + c − c 3
(c) y = x − c 2 + c 3 (d) none of these
Fill in The Blanks

dy
1. We usually denote by the letter …… .
dx
2. The differential equation y = px + f ( p), is known as …… .
(Meerut 2003; Avadh 05; Kumaun 10)
n n −1 n−2
3. p + A1 p + A2 p + ... + An − 1 p + An = 0 ,
where A1, A2 , … , An are some functions of x and y, is a differential equation of

…… order and …… degree.
4. Solution of the equation p2 − 7 p + 12 = 0 is …… .
5. Solution of the equation y = px + ap (1 − p) is …… .

D-82
6. Solution of the differential equation y = px + e p is …… .

3
 dy   dy 
7. Solution of the equation y = x   +   is …… .
 dx   dx  (Kumaun 2007, 09)
8. Solution of the equation cos y cos px + sin y sin px = p is …… .
9. Solution of the equation ( x − a) p2 + ( x + y) p − y = 0 is …… .
True or False
1. Every differential equation of the first order represents a family of curves.
2. The singular solution of a differential equation is given by the envelope of the

family of curves represented by that differential equation.
3. Only those differential equations possess a singular solution in which

p (or dy / dx) occurs only in the first degree.
A nswers
1. (b) 2. (a) 3. (b) 4. (c) 5. (c)
6. (b) 7. (a) 8. (b) 9. (a) 10. (a)
11. (b) 12. (c) 13. (c) 14. (a) 15. (c)
16. (a) 17. (b) 18. (b) 19. (a)

1. p 2. Clairaut’s equation
3. first, nth 4. ( y − 3 x − c ) ( y − 4 x − c ) = 0
5. y = c x + ac (1 − c ) 6. y = cx + e c
7. y = cx + c 3 8. y = cx − cos −1 c
9. ( x − a) c 2 + ( x − y) c − y = 0
True or False
1. T 2. T 3. F
¨
D-83
3
O rthogonal T rajectories
1 Trajectory
(Lucknow 2005; Gorakhpur 06; Bundelkhand 08; Meerut 13, 13B)
efinition:. A trajectory of a given system of curves is defined to be a curve which cuts all the
D members of the family according to a given law. Here we propose to find the equation of
the trajectories (a family of curves) each member of which cuts each member of a given
family of curves at a constant angle. If the angle is a right angle, the trajectories are
called orthogonal trajectories, when it is other than a right angle they are called to be
oblique trajectories.
2 Trajectories
Cartesian Co-ordinates: To find the trajectories which cut every member of a given family of
curves at a constant angle.
Let the equation of the given family of curves be
f ( x, y, c ) = 0 , …(1)
c being the arbitrary parameter. Let the required trajectories cut the given curves (1) at
a constant angle α.
Differentiating (1) with respect to x, we have
D-84
Y
∂f ∂f dy
+ ⋅ =0. …(2)
∂x ∂y dx
Eliminating c between (1) and (2), we get the
differential equation of the given family of curves
(1). Let it be (x, y) Trajectory
dy  rve
 Cu α
φ  x, y,  = 0 . …(3)
 dx  ψ1
ψ2
Let ( x, y) be the coordinates of a point of O X
intersection of a member of the given curves (1)

and a member of the required trajectories. If the tangent to the trajectory at this point
makes an angle ψ1 with x-axis and if the tangent to the given curve at this point makes
an angle ψ2 with x-axis, we have
tan ψ1 − tan α
dy / dx given by (3) = tan ψ2 = tan (ψ1 − α) = .
1 + tan ψ1 tan α
Putting this value of (dy / dx) in (3), we get
 tan ψ1 − tan α 
φ  x, y,  =0. …(4)
 1 + tan ψ1 tan α 
Now dy / dx for the trajectory is tan ψ1. So replacing tan ψ1 in the equation (4) by
dy / dx, we get the differential equation of the required trajectories as
 dy / dx − tan α 
φ  x, y,  =0.
 1 + (dy / dx) tan α 
Solving this differential equation, we shall obtain the equation of the required
trajectories.
Orthogonal trajectories: If α is a right angle, we have
tan ψ1 × tan ψ2 = − 1 .
dy 1
∴ given by (3) = tan ψ2 = − .
dx tan ψ1
Putting this value of dy / dx in (3) of the last article, we get
 1 
φ  x, y, −  =0. …(5)
 tan ψ1 
Now (dy / dx) for the trajectory is tan ψ1. So replacing tan ψ1 in the equation (5) by
dy / dx, we get differential equation of the orthogonal trajectories of (1) as
 dx 
φ  x, y, −  = 0 .
 dy 
Solving this differential equation, we shall get the equation of the orthogonal
trajectories of the given family (1).
Remember: To obtain the differential equation of the orthogonal trajectories, we have to write
− dx / dy for dy / dx in the differential equation of the original family of curves.
D-85
3 Orthogonal Trajectories
Polar Co-ordinates: Let the equation of the given family of curves be
f (r, θ, c ) = 0 , …(1)
c being the parameter i. e., c can take any real value. Differentiating (1) w.r.t. θ and then
eliminating c, let the differential equation of the given family (1) be
 dr 
φ  r, θ,  = 0 . …(2)
 dθ 
Now if at a point of intersection P (r, θ), the tangents to the trajectory and the given
curve make angles φ1 and φ2 with the common radius vector OP of that point, then
tan φ1 − tan φ2
tan 90 ° = tan (φ1 − φ2) =
1 + tan φ1 tan φ2
Curve
Trajectory
and therefore tan φ1 tan φ2 = − 1 .
dθ P (r, θ)
Now r given by (2)
dr α
φ2 φ1
1
= tan φ2 = − ⋅
tan φ1
O X
∴ dr / dθ given by (2) = − r tan φ1 .
∴ putting this value of dr / dθ in (2), we have
φ (r, θ, − r tan φ1) = 0
dθ
But tan φ1 = r for the trajectory.
dr
∴ the differential equation of the required orthogonal trajectories is
 dθ 
φ  r, θ, − r2  =0.
 dr 
Solving this we shall get the required equation of the orthogonal trajectories.
Remember: The differential equation of the orthogonal trajectories is obtained from the
differential equation (in polar co-ordinates) of a given family of curves by writing
dθ dr dθ 1 dr
− r2 for or −r for
dr dθ dr r dθ
1 dr dθ
or − for r ⋅
r dθ dr
Example 1: Find the orthogonal trajectories of the family of curves y = ax n.

(Kumaun 2008, 12; Meerut 11; Kanpur 15)
Solution: The equation of the given family of curves is

y = ax n, a being the parameter .…(1)
D-86
Differentiating (1) with respect to x, we get

dy / dx = a . nx n−1. …(2)
Dividing (2) by (1) to eliminate a between (1) and (2), we have
dy / dx anx n−1 n
= = or x (dy / dx) = ny, …(3)
y ax n x
which is the differential equation of the given family of curves. Now to obtain the
differential equation of the orthogonal trajectories, write − dx / dy for dy / dx in (3).
Hence the orthogonal trajectories are given by the differential equation
x (− dx / dy) = ny. …(4)
Separating the variables, we have − x dx = ny dy.
1 1
Integrating, we have − x2 = n y2 + c , where c is constant of integration.
2 2
Thus x2 + ny2 = 2 c or x2 + ny2 = c 2 is the required family of orthogonal trajectories.
Example 2: Find the orthogonal trajectories of the family of parabolas y2 = 4 ax, where ‘a’ is the
variable parameter. (Meerut 2013B; Kanpur 09)
Solution: Differentiating the given equation y2 = 4 ax, …(1)
with respect to x, we get

2 y (dy / dx) = 4 a . …(2)
Eliminating a between (1) and (2), we get
2 x (dy / dx) = y, …(3)
which is the differential equation of the given family of parabolas.
So to obtain the differential equation of the orthogonal trajectories, write − dx / dy for
dy / dx in (3). Thus the differential equation of the required family of orthogonal
trajectories is
2 x (− dx / dy) = y or 2 x dx = − y dy.
Integrating, we get
1 2
x2 = − y + c2 or 2 x2 + y2 = k 2 ,
2
which is the required family of orthogonal trajectories.
Example 3: Find the orthogonal trajectories of the family of rectangular hyperbolas xy = c 2 .
(Meerut 2003; Rohilkhand 06, 07; Kanpur 07, 12; Bundelkhand 09; Avadh 10, 11)
Solution: The equation of the given family of curves is

xy = c 2 , c being the parameter. …(1)
Differentiating (1) w.r.t. x, we get
x (dy / dx) + y = 0 , …(2)
which is independent of the parameter c. Hence the equation (2) is the differential
equation of the given family of curves.
D-87
So to obtain the differential equation of the orthogonal trajectories, write − dx / dy for

dy / dx in (2).
Thus the differential equation of the required family of orthogonal trajectories is
x (− dx / dy) + y = 0
or x dx = y dy, in which the variables are separated.
1 1 2
∴ Integrating, x2 = y + c , where c is constant of integration
2 2
or x2 − y2 = 2 c = k 2 (say) is the required family of orthogonal trajectories.
Example 4: Find the orthogonal trajectories of the family of parabolas y2 = 4 a ( x + a), where ‘a’
is the parameter. (Meerut 2006; Purvanchal 06; Lucknow 06; Avadh 07, 09)
Solution: The equation of the given family of parabolas is

y2 = 4 a ( x + a) , …(1)
a being the parameter.
Differentiating (1) w.r.t. x, we get
2 y (dy / dx) = 4 a. …(2)
Now eliminating a between (1) and (2), we get
1
y2 = 2 y (dy / dx)[ x + y (dy / dx)]
2
or y = 2 x (dy / dx) + y(dy / dx)2 , …(3)
which is the differential equation of the given family of parabolas. So to obtain the
y = 2 x (− dx / dy) + y (− dx / dy)2
or y (dy / dx)2 + 2 x (dy / dx) = y. …(4)
Now we observe that the differential equation (4) of the orthogonal trajectories is the
same as the differential equation (3) of the given family of parabolas. Therefore the
given family of parabolas (1) is self-orthogonal i. e., the orthogonal trajectories of the
system belong to the system itself. Hence the equation of the orthogonal trajectories
of (1) is y2 = 4 c ( x + c ), c being the parameter.
Example 5: Find the orthogonal trajectory of the family of circles x2 + y2 = 2 ax, a being the
parameter. (Meerut 2005, 10)
Solution: Differentiating
x2 + y2 = 2 ax …(1)
with respect ot x, we get
2 x + 2 y (dy / dx) − 2 a = 0
or x + y (dy / dx) − a = 0. …(2)
Eliminating ‘a’ between (1) and (2), we get
D-88
x2 + y2 = 2 x { x + y (dy / dx)}
or x2 + 2 xy (dy / dx) − y2 = 0 , …(3)
which is the differential equation of the given family of circles. So to obtain the
x2 + 2 xy (− dx / dy) − y2 = 0
or dx / dy = ( x2 − y2 ) / 2 xy. …(4)
This is a homogeneous differential equation. To solve it putting x = vy, so that
dx / dy = v + y (dv / dy), in (4), we get
v + y (dv / dy) = y2 (v2 − 1) /(2 vy2 )
dv v2 − 1 1 + v2
or y = −v=−
dy 2v 2v
2 v dv dy
or =− .
1 + v2 y
log (1 + v2 ) = − log y + log b
or 1 + v2 = b / y
or 1 + ( x2 / y2 ) = b / y, since v = x / y.
Hence x2 + y2 = by is the equation of the required orthogonal trajectories.
Example 6: Find the orthogonal trajectories of the family of curves

{ x2 /(a2 + λ )} + { y2 /(b2 + λ )} = 1 ,
where λ is the parameter. (Meerut 2007B, 09B; Lucknow 10; Rohilkhand 10;
Gorakhpur 11; Purvanchal 07, 09, 11)
Solution: Differentiating the given equation with respect to x, we get
2 x /(a2 + λ ) + {2 y /(b2 + λ )} ⋅ (dy / dx) = 0
or x (b2 + λ ) + y (dy / dx)(a2 + λ ) = 0
or λ { x + y (dy / dx)} = − { b2 x + a2 y (dy / dx)} .
∴ λ = − { b2 x + a2 y (dy / dx)} ÷ { x + y (dy / dx)}.
Thus a2 + λ = (a2 − b2 ) x / { x + y (dy / dx)}
and b2 + λ = − (a2 − b2 ) y (dy / dx) ÷ { x + y (dy / dx)} .
Substituting these values of (a2 + λ ) and (b2 + λ ) in the given equation, we get the
differential equation of the given family of curves as
x2 { x + y (dy / dx)} y2 { x + y(dy / dx)}
− =1
(a2 − b2 ) x (a2 − b2 ) y (dy / dx)
D-89
 dy 1 
or x2 − y2 + xy  −  = a2 − b2 . …(1)
 dx dy / dx 
Hence putting − dx / dy for dy / dx in (1), the differential equation of the orthogonal
trajectories is
 dx dy 
x2 − y2 + xy  − +  = a2 − b2
 dy dx 
 dy 1 
or x2 − y2 + xy  −  = a2 − b2 ,
 dx dy / dx 
which is the same as (1). Therefore solving it we shall get
x2 / (a2 + µ ) + y2 (b2 + µ ) = 1 ,
µ being the parameter of the orthogonal trajectories. In other words the system of
given confocal conics x2 / (a2 + λ ) + y2 / (b2 + λ ) = 1 is self-orthogonal.
Example 7: Find the orthogonal trajectories of the cardioids r = a (1 − cos θ), a being the
parameter. (Meerut 2006B; Gorakhpur 05; Lucknow 11; Kashi 11)
Solution: The given family of curves is

r = a (1 − cos θ), …(1)
the parameter being a. Differentiating (1) w.r.t. θ, we get
dr / dθ = a sin θ . …(2)
Now we shall eliminate a between (1) and (2). So dividing (1) by (2), we get
r (dθ / dr) = (1 − cos θ) / sin θ, …(3)
which is the differential equation of the given family of curves (1).
To obtain the differential equation of the orthogonal trajectories, write − (1/ r) (dr / dθ)
for r (dθ / dr) in (3). Thus we get
1
2 sin2 θ
1 dr 1 − cos θ 2 1
− = = = tan θ ,
r dθ sin θ 1 1 2
2 sin θ cos θ
2 2
which is the differential equation of the orthogonal trajectories. To solve it, separating
the variables, we get
1 1 1
− sin θ − sin θ
dr 1 2 2 2
= − tan θ dθ = dθ = 2 ⋅ dθ ⋅
r 2 1 1
cos θ cos θ
2 2
1
log r = 2 log cos θ + log c
2
1
or log (r / c ) = log cos2 θ
2
D-90
1 1
or r / c = cos2 θ = (1 + cos θ)
2 2
1
or r= c (1 + cos θ)
2
or r = k (1 + cos θ),
which is the required family of orthogonal trajectories.
Example 8: Find the orthogonal trajectories of the system of curves r n = a n cos n θ, a being the
parameter. (Meerut 2007, 10B; Bundelkhand 05; Gorakhpur 06; Lucknow 08; Kanpur 08)
Solution: The given family of curves is r n = a n cos nθ , …(1)
the parameter being a. Taking logarithm of both sides of (1), we get

n log r = n log a + log cos nθ .
Now differentiating both sides w.r.t. θ, we get
n dr − n sin n θ
=0 +
r dθ cos n θ
1 dr
or = − tan nθ …(2)
r dθ
which is the differential equation of the family of curves (1).
dθ 1 dr
Now putting − r for in (2), the differential equation of the orthogonal
dr r dθ
trajectories is
dθ
−r = − tan nθ
dr
dr
or = cot nθ dθ .
r
Integrating, we get log r = (1 / n) log sin nθ + log c
or n log (r / c ) = log sin nθ
or log (r / c )n = log sin nθ
or r n = c n sin nθ,
which is the required family of orthogonal trajectories, the parameter being c.
1. Find the equation of the family of curves orthogonal to the family y = ax3 .
(Avadh 2005; Rohilkhand 09; Kanpur 11)
2. Find the orthogonal trajectories of the semi-cubical parabolas ay2 = x3 , where a
is the variable parameter. (Gorakhpur 2007; Kashi 13; Kumaun 15)
D-91
3. Find the orthogonal trajectories of the family of circles x2 + y2 = a2 , where ‘a’ is

the parameter. (Bundelkhand 2008; Kanpur 10; Kumaun 10, 13)
4. Find the orthogonal trajectories of the family of curves x2 /3 + y2 /3 = a2 /3 ,

where a is parameter.
5. Find the equation of the family of curves that is orthogonal to ax2 + y2 = 1 .
6. Find the orthogonal trajectories of the system of curves (dy / dx)2 = a / x.

(Bundelkhand 2007, 10; Agra 08; Kumaun 11; Purvanchal 14)
7. Find the differential equation of the family of curves given by the equation
x2 − y2 + 2 λxy = 1 , where λ is a parameter.
Obtain the differential equation of its orthogonal trajectories and solve it.
8. Show that the orthogonal trajectories of the family of conics
y2 − x2 + 4 xy − 2 cx = 0 consist of a family of cubics, with the common
asymptote x + y = 0 . (Meerut 2009)
2 2
9. Find the orthogonal trajectories of the family of circles x + y + 2 fy + 1 = 0 .
where f is the parameter. ( Meerut 2004; Gorakhpur 08, 11)
10. Find the orthogonal trajectories of the family of curves
x2 y2
+ = 1 , λ being parameter.
a2 b2 + λ
(Lucknow 2005; Meerut 05B; Purvanchal 14)
11. Find the orthogonal trajectories of the family of coaxial circles
x2 + y2 + 2 gx + c = 0 , where g is a parameter and c is a constant. (Agra 2007)
12. Find the orthogonal trajectories of
(i) rθ = a , (ii) r = a θ . (Rohilkhand 2008)
13. Find the orthogonal trajectories of the family of cardioids r = a (1 + cos θ) .
(Rohilkhand 2006; Lucknow 06; Purvanchal 08, 10; Gorakhpur 10)
14. Find the orthogonal trajectories of r = e aθ . (Kumaun 2009)
15. Find the orthogonal trajectories of r n sin nθ = a n. (Meerut 2001, 04B)
A nswers 1
3 2
1. x2 + 3 y2 = c 2 2. x2 + y = c2
2
3. x = cy 4. x4 /3 − y4 /3 = c 4 /3
2
+ y2 ) 9 2
5. y2 = ce( x 6. ( x3 /2 + c 3 /2 )2 = ay
4
1 2
7. ( x + y2 )2 + y2 − x2 = c 2 8. y3 + 3 x2 y + 4 x3 = k 3
2
9. y2 + x2 − 1 = c x 10. x2 + y2 − 2 a2 log x = c
D-92
11. x2 + y2 + 2 fy − c = 0 , f being parameter.

1 2
2 − θ
12. (i) r2 = c 2 ⋅ e θ (ii) r = ce 2
13. r = c (1 − cos θ) 14. θ2 + (log r)2 = c 2

15. r n cos nθ = c n

(b), (c) and (d).
dr
1. To find the orthogonal trajectories in polar form, we replace by
dθ
dr dθ
(a) − (b)
dθ dr
dθ dθ
(c) − r2 (d) r2
dr dr (Kumaun 2010)
2. Family of orthogonal trajectories for the curve xy = c is
(a) y2 − x2 = 2 c (b) y2 = 2 c + x
(c) y = x2 + c (d) e y − e x = c (Kumaun 2014)
2
 dy  a
3. Orthogonal trajectories of the system of curves   = is
 dx  x
(a) 9 a ( y − c )2 = 4 x3 (b) 9 a ( y + c )2 = 4 x3
(c) 9 a2 ( y + 2 c )2 = 4 x2 (d) 9 a ( y − 2 c )2 = 4 x2
(Kanpur 2016)
4. Orthogonal trajectories of cardioids r = a (1 − cos θ), a being a parameter, is
c c
(a) r = (1 + cos θ) (b) r = (1 − cos θ)
2 2
2 c c
(c) r = (1 − cos θ) (d) r = (a − sin2 θ)
2 2 (Kanpur 2016)
5. Orthogonal trajectories of the family of parabolas y2 = 4 ax, a being a
parameter, is
(a) x2 + y2 = k 2 (b) x2 − y2 = k 2
(c) 2 x2 + y2 = k 2 (d) x2 + 2 y2 = k 2
6. Equation of the family of curves orthogonal to the family y = ax3 is
(a) 3 x2 + y2 = c 2 (b) x2 + 3 y2 = c 2
(c) 3 x2 − y2 = c 2 (d) x2 − 3 y2 = c 2
D-93
7. Orthogonal trajectories of the semi-cubical parabolas ay2 = x3 , a is the

variable parameter, is
3 3 2
(a) x2 − y2 = c 2 (b) x2 + y = c2
2 2
3 2 3 2
(c) x − y2 = c 2 (d) x + y2 = c 2
2 2
8. Orthogonal trajectories of r θ = a is
(a) r 2 = c 2 e θ (b) r 2 = c 2 e − θ
2 2
(c) r 2 = c 2 e θ (d) r = c 2 e θ
9. Orthogonal trajectories of r = a θ is
2 2
/2 /2
(a) r = c e − θ (b) r = c e θ
3
(c) r = c e − θ /2 (d) r = c e θ /2
10. Orthogonal trajectories of r = e a θ is

(a) θ2 + log r = c 2 (b) θ2 + (log r)2 = c 2
(c) θ2 − log r = c 2 (d) θ2 − (log r)2 = c 2
11. Orthogonal trajectories of the family of curves y = a x n is

(a) x2 + ny2 = c 2 (b) nx2 + y2 = c 2
(c) x2 − ny2 = c 2 (d) nx2 − y2 = c 2

1. The differential equation of the orthogonal trajectories is obtained from the
differential equation of the given family of curves by writing …… for dy / dx.
2. The differential equation of the orthogonal trajectories is obtained from the
differential equation (in polar coordinates) of the given family of curves by
dr
writing …… for ⋅
dθ
3. Orthogonal trajectories of the system of curves r n = a n cos nθ, a being the
parameter, is ……
True or False
1. A family of curves is said to be self orthogonal if the differential equation of the
orthogonal trajectories is the same as the differential equation of the given
family of curves.
2. Family of curves { x2 / (a2 + λ )} + { y2 / (b2 + λ )} = 1 , where λ is the parameter,
is self orthogonal.
D-94
A nswers
1. (c) 2. (a) 3. (b) 4. (a) 5. (c)
6. (b) 7. (b) 8. (c) 9. (a) 10. (b)
11. (a)

dx dθ
1. − 2. − r2 3. r n = c n sin nθ
dy dr
True or False
1. T 2. T
¨
D-95
4
L inear D ifferential E quations
with C onstant C oefficients
1 Definitions
linear differential equation is an equation in which the dependent variable and its
A derivatives appear only in the first degree.
A linear differential equation of order n of the form
dn y d n −1 y dn − 2 y dy
n
+ a1 n −1
+ a2 n−2
+ ... + an−1 + an y = Q , …(1)
dx dx dx dx
where a1, a2 , … , an−1, an are constants and Q is any function of x is called a linear
differential equation with constant coefficients.
d d2 d3 dn
For convenience, the operators , 2 , 3 , … , n are also denoted by D,
dx dx dx dx
D2 , D3 , … , Dn respectively.
Thus the equation (1) can also be written as
Dn y + a1 Dn −1 y + ... + an −1 Dy + an y = Q
or [ Dn + a1 Dn −1 + … + an −1 D + an] y = Q . …(2)
If y = f ( x) is the general solution of

D-96
[ Dn + a1 Dn − 1 + … + an−1 D + an] y = 0 , …(3)
and y = φ ( x) is any particular solution of the equation (2) not containing any arbitrary
constant, then
y = f ( x) + φ ( x) ,
is the general solution of (2).
Thus the method of solving a linear equation is divided into two parts :
First, we find the general solution of the equation (3).
It is called the complementary function (C.F.). It must contain as many arbitrary
constants as is the order of the given differential equation.
Next, we find a solution of (2) which does not contain an arbitrary constant. This is
called the particular integral (P.I.).
If we add (C.F.) and (P.I.), we get the general solution of (2). Thus the general solution
of (2) is
y = C. F. + P. I. (Gorakhpur 2005)
2 Determination of Complementary Function (C.F.)

Consider a linear nth order differential equation with constant coefficients of the form
f ( D) y = 0 i. e.,
[ Dn + a1 Dn −1 + a2 Dn − 2 + … + an] y = 0 . …(1)
This is equivalent to
[( D − m1) ( D − m2 ) … ( D − mn)] y = 0 . …(2)
The solution of any one of the equations
( D − m1) y = 0 , ( D − m2 ) y = 0 , … , ( D − mn) y = 0 …(3)
is also a solution of (2) and we know that the general solution of ( D − m1) y = 0 is
y = Ae m1 x .
Hence we can assume that a solution of the equation (2) is of the form y = e mx . Then,
substituting e mx for y in (1), so that Dy = me mx , D2 y = m2 e mx , … , Dn y = m ne mx , we get
e mx (m n + a1m n − 1 + a2 m n − 2 + … + an) = 0
or m n + a1m n − 1 + a2 m n − 2 + … + an = 0 , because e mx ≠ 0 .
Hence e mx will be a solution of (1) if m has the value obtained from the equation
m n + a1m n − 1 + … + an = 0 . …(4)
The equation (4) is called the auxiliary equation (A.E.) and is obtained by putting
D = m in f ( D) = 0 .
It will give in general n roots, say, m1, m2 , m3 , … , mn .
D-97
Case I: If all the roots of the Auxiliary equation (A.E.) are distinct:
If the roots m1, m2 , m3 , …, mn are all distinct, then e m1 x , e m2 x , … , e mn x are all distinct and
linearly independent. So the general solution of (1) in this case is
y = c1e m1 x + c2 e m2 x + … + c n e mn x . …(5)
Case II: Auxiliary equation having equal roots :
If two roots are equal say m1 = m2 , then the solution (5) becomes
y = c1e m1 x + c2 e m1 x + c3 e m3 x + … + c ne mn x
or y = (c1 + c2 ) e m1 x + c3 e m3 x + … + c ne mn x .
Now (c1 + c2 ) can be replaced by single constant say c.
Therefore this solution has only (n − 1) arbitrary constants and so it is not the general
solution.
To obtain the general solution, consider the differential equation ( D − m1)2 y = 0 in
which the two roots are equal.
This can be written as ( D − m1) [( D − m1) y] = 0 . …(6)
Now putting ( D − m1) y = v, we get ( D − m1) v = 0 ,
or dv / dx = m1v or dv / v = m1 dx, (variables being separated).
∴ integrating,
log v = m1 x + log c1 or log (v / c1) = m1 x or v = c1 e m1 x .
Thus ( D − m1) y = v = c1 e m1 x
and putting it in (6), we get ( D − m1) y = c1e m1 x
dy ∵ D ≡ d 
or Dy − m1 y = c1e m1 x or − m1 y = c1e m1 x .
dx  dx 
This is a linear equation of the first order. Hence the I. F. = e ∫ − m1 dx = e − m1 x .
∴ The solution of this equation is
y ⋅ e − m1 x = m1 x − m1 x
∫ c1e ⋅ e dx + c2
∫
= c1 dx + c2 = c1 x + c2
or y = (c1 x + c2 ) e m1 x or y = (c2 + c1 x) e m1 x .
Hence the general solution of f ( D) y = 0 in this case is
y = (c1 + c2 x) e m1 x + c3 e m3 x + … + c ne mn x .
Similarly if three roots of the auxiliary equation are equal say, m1 = m2 = m3 , the
general solution of f ( D) y = 0 will be
y = (c1 + c2 x + c3 x2 ) e m1 x + c4 e m4 x + … + c ne mn x
and so on.
Case III: Auxiliary equation having complex roots: (Meerut 2003)
Let the two roots of the auxiliary equation be complex, say m1 = α + iβ and m2 = α − iβ ,
(where i = √ − 1).
D-98
The solution corresponding to these two roots will be

y = c1e(α + iβ) x + c2 e(α − iβ) x = c1e αx ⋅ e iβx + c2 e αx ⋅ e − iβx
= c1e αx (cos βx + i sin βx) + c2 e αx (cos βx − i sin βx)
= (c1 + c2 ) e αx cos βx + i (c1 − c2 ) e αx sin βx
or y = e αx [ A1 cos βx + A2 sin βx] , where A1 = c1 + c2 and A2 = i (c1 − c2 )
or y = e αx (c1 cos βx + c2 sin βx), chaning the constants.
If the imaginary roots are repeated, say α + iβ and α − iβ occur twice then the solution
will be
y = e αx [(c1 + c2 x)cos βx + (c3 + c4 x) sin βx] ,
and so on.
Note 1: The expression e αx (c1 cos βx + c2 sin βx) can also be written as
c1e αx sin (βx + c2 ) or c1e αx cos (βx + c2 ) .

Note 2: If a pair of the roots of the auxiliary equation are irrational i. e., they are α ± √ β ,
where β is positive, then the corresponding term in the C.F. will be
e αx (c1 cosh √ βx + c2 sinh √ βx),
or c1e αx sinh (√ βx + c2 ) or c1e αx cosh (√ βx + c2 ).
If these irrational roots are repeated twice, then the corresponding portion of the
solution will be
e αx {(c1 + c2 x) cosh √ βx + (c3 + c4 x) sinh √ βx} .
d2 y dy
Example 1: Solve 2
−7 + 12 y = 0 .
dx dx
Solution: The given differential equation is ( D2 − 7 D + 12) y = 0 .
∴ the auxiliary equation is m2 − 7 m + 12 = 0
or (m − 3) (m − 4) = 0 . ∴ m = 3, 4 .
y = c1e3 x + c2 e4 x .
Example 2: Solve ( D3 + 6 D2 + 11D + 6) y = 0 . (Meerut 2010)
Solution: The auxiliary equation is

m3 + 6 m2 + 11m + 6 = 0
or (m + 1) (m2 + 5 m + 6) = 0
or (m + 1) (m + 2) (m + 3) = 0 .
D-99
∴ m = − 1, − 2, − 3 .
y = c1e − x + c2 e −2 x + c3 e −3 x .
d2 x dx
Example 3: Solve 2
−3 + 2 x = 0 , given that when t = 0 , x = 0 and dx / dt = 0 .
dt dt
(Gorakhpur 2008)
2
Solution: The auxiliary equation is m − 3 m + 2 = 0
or (m − 1) (m − 2) = 0 or m = 1, 2 .
t 2t
Hence the solution is x = c1e + c2 e , …(1)
where c1 and c2 are arbitrary constants.
Now x = 0 when t = 0 ;
∴ 0 = c1 + c2 . …(2)
t 2t
Also dx / dt = c1 e + 2 c2 e , and dx / dt = 0 when t = 0 .
∴ 0 = c1 + 2 c2 . …(3)
Solving (2) and (3), we get c1 = 0 , c2 = 0 .
Now putting the values of c1 and c2 in (1), we get the required solution as x = 0 .
Example 4: Solve ( D3 − 3 D + 2) y = 0 .
Solution: The auxiliary equation is m3 − 3 m + 2 = 0
or (m − 1) (m2 + m − 2) = 0 or (m − 1) {(m − 1) (m + 2)} = 0 .

∴ m = 1, 1, − 2.
y = (c1 + c2 x) e x + c3 e −2 x .
d3 y
Example 5: Solve −8y =0.
dx3 (Meerut 2010B)
3
Solution: The auxiliary equation is m − 8 = 0
or (m − 2) (m2 + 2 m + 4) = 0 i. e., m − 2 = 0 and m2 + 2 m + 4 = 0 .

∴ m = 2 and m = − 1 ± i √ 3 .
y = e − x (c1 cos √ 3 x + c2 sin √ 3 x) + c3 e2 x .
d4 y d3 y dy
Example 6: Solve 4
−2 3
−2 − y =0.
dx dx dx
m4 − 2 m3 − 2 m − 1 = 0
or (m4 − 1) − 2 m (m2 + 1) = 0
or (m2 + 1) (m2 − 1) − 2 m (m2 + 1) = 0
D-100
or (m2 + 1) (m2 − 2 m − 1) = 0 i. e., m2 = − 1 or m2 − 2 m − 1 = 0 .

∴ m = 0 ± i ,1 ± √ 2 .
y = e0 x (c1 cos x + c2 sin x) + e x (c3 cosh √ 2 x + c4 sinh √ 2 x)
or y = c1 cos x + c2 sin x + e x { c3 cosh √ 2 x + c4 sinh √ 2 x} .
Example 7: Solve ( D4 + k 4 ) y = 0 . (Meerut 2003, 04B, 05B, 06)
Solution: The auxiliary equation is m4 + k 4 = 0
or (m2 + k 2 )2 − 2 k 2 m2 = 0
or (m2 + k 2 )2 − (√ 2 km)2 = 0
or (m2 + k 2 − √ 2 km) (m2 + k 2 + √ 2 km) = 0
or m2 − √ 2 km + k 2 = 0 and m2 + √ 2 km + k 2 = 0
√ 2 k ± √ (2 k 2 − 4 k 2 ) − √ 2 k ± √ (2 k 2 − 4 k 2 )
or m= and m =
2 2
k k k k
or m= ±i and − ±i ⋅
√2 √2 √2 √2
y = e kx / √2 { c1 cos (kx / √ 2) + c2 sin (kx / √ 2)}
+ e − kx / √2 { c3 cos (kx / √ 2) + c4 sin (kx / √ 2)} .

d2 y dy d2 y dy
1. 2
+ (a + b) + aby = 0 . 2. 2
−3 − 4 y = 0.
dx dx dx dx
3 2
d y d y dy d2 y dy
3. 3
+6 2
+3 − 10 y = 0 . 4. 2
−4 + 4 y = 0.
dx dx dx dx dx
3 2
5. ( D − 4 D + 5 D − 2) y = 0 .
d4 y d3 y d2 y dy
6. 4
+2 3
−3 2
−4 + 4 y = 0.
dx dx dx dx (Garhwal 2009)
4
d y
7. (i) ( D4 + 1) y = 0 . (Agra 2006) (ii) − k4 y = 0.
dx4
8. ( D4 + 8 D2 + 16) y = 0 .
9. ( D2 ± µ 2 ) y = 0 . (Avadh 2005, 14)
10. ( D3 − 2 D2 − 4 D + 8) y = 0 . (Garhwal 2010)

D-101
1
11. Solve (d2 y / dx2 ) + y = 0 , given y = 2 for x = 0 and y = − 2 for x = π.
2
A nswers 1
1. y = c1 e − ax + c2 e − bx
2. y = c1e − x + c2 e4 x
3. y = c1e x + c2 e −2 x + c3 e −5 x
4. y = (c1 + c2 x) e2 x
5. y = (c1 + c2 x) e x + c3 e2 x
6. y = (c1 + c2 x) e x + (c3 + c4 x) e −2 x
7. (i) y = e x / √2 { c1 cos ( x / √ 2) + c2 sin ( x / √ 2)}
+ e − x / √2 { c3 cos ( x / √ 2) + c4 sin ( x / √ 2)} .
(ii) y = c1e kx + c2 e − kx + c3 cos kx + c4 sin kx
8. y = (c1 + c2 x) cos 2 x + (c3 + c4 x) sin 2 x
9. y = c1 cos µx + c2 sin µx and y = c3 e µx + c4 e − µx
10. y = (c1 + c2 x) e2 x + c3 e −2 x
1
11. y = 2 √ 2 cos  x + π
 4 
3 The Particular Integral (P.I.)

As already shown in article 1 the complete solution of
( Dn + a1 Dn − 1 + a2 Dn − 2 + … + an − 1 D + an) y = Q …(1)
or F ( D) y = Q is y = C. F. + P. I.,
where the C.F. consists of the general solution of the differential equation
F ( D) y = 0 .
In article 2 we have discussed different methods of finding the complementary
function by taking the differential equation as F ( D) y = 0 . Methods of finding the
particular integral will be discussed now.
1
The particular integral of the differential equation F ( D) y = Q is Q. It is
F ( D)
obviously a function of x which when operated by F ( D) gives Q.
1 1
Now as F ( D) Q = Q, therefore can be regarded as the inverse operator of
F ( D) F ( D)
F ( D). Similarly D and 1/ D are inverse operations. If D stands for differentiation then
1/ D will stand for integration.
D-102
Hence the particular integral of the equation

1
F ( D) y = Q will be Q
F ( D)
because it satisfies the given equation.
4 Particular Integral in Some Special Cases

Case I: To find P.I. when Q is of the form e ax , where a is any constant and
F ( a) ≠ 0 .
By simple differentiation we know that D (e ax ) = ae ax ;
D2 (e ax ) = a2 e ax ; D3 (e ax ) = a3 e ax , … , Dn (e ax ) = a ne ax .
It suggests that F ( D) e ax = F (a) e ax . …(1)
Let F (a) ≠ 0 .
Operating on both sides of (1) with 1 / F ( D), we have
1 1
F ( D) e ax = { F (a) e ax }
F ( D) F ( D)
1
or e ax = F (a) ⋅ e ax , because F (a) is a constant
F ( D)
1 1
or e ax = e ax , because F (a) ≠ 0
F (a) F ( D)
1 1 ax
∴ P.I. = e ax = e , provided F( a) ≠ 0.
F ( D) F (a)
Working Rule : If P.I. = {1/ F ( D)} e ax , then put a for D in F ( D) and we get the P.I.,
provided F (a) ≠ 0 .
Examples on case I:
d2 y dy
Example 8: Solve 2
−3 + 2 y = e5 x .
dx dx
(Kumaun 2006; Avadh 08; Garhwal 10B; Purvanchal 11; Kanpur 12)
( D2 − 3 D + 2) y = e5 x , where d / dx ≡ D.
Here F ( D) = D2 − 3 D + 2 and Q = e5 x .
The auxiliary equation is

m2 − 3 m + 2 = 0 or (m − 1) (m − 2) = 0 .
∴ m = 1, 2 .
D-103
∴ C.F. = c1e x + c2 e2 x , the roots of the A.E. being distinct.

1 1 e5 x e5 x
and P. I. = e ax = 2 e5 x = 2 = ⋅
F ( D) D − 3D + 2 5 − 3 ⋅ 5 + 2 12
[We have put 5 for D in F ( D), because here a = 5]
Hence the complete solution is y = (C.F.) + (P.I.)
1 5x
or y = c1e x + c2 e2 x + e .
12
d2 y dy
Example 9: Solve 2
+ 31 + 240 y = 272 e − x .
dx dx (Meerut 2010B, 11)

m2 + 31m + 240 = 0 or (m + 15) (m + 16) = 0 ;
∴ m = − 15, − 16 .
∴ C. F. = c1e −15 x + c2 e −16 x .
1 1
and P. I. = (272 e − x) = 272 ⋅ e− x
D2 + 31D + 240 D2 + 31D + 240
1 272 − x 136 − x
= 272 ⋅ e− x = e = e .
(− 1)2 + 31 (− 1) + 240 210 105
Hence the general solution is y = (C. F. ) + (P. I. )
136 − x
i. e., y = c1e −15 x + c2 e −16 x + e .
105
Example 10: Obtain the complete solution of the differential equation
d2 y dy
−7 + 6 y = e2 x ,
dx2 dx
and determine the constant so that y = 0 when x = 0 .
Solution: The given equation is [ D2 − 7 D + 6] y = e2 x .
∴ auxiliary equation is m2 − 7 m + 6 = 0
or (m − 1) (m − 6) = 0 or m = 1, 6.
∴ C. F. = c1e x + c2 e6 x .
1 1 e2 x
and P. I. = 2
e2 x = 2
e2 x = − ⋅
D − 7D + 6 2 − 7 ⋅2 + 6 4
∴ the general solution is y = (C. F. ) + (P. I. )
1
i. e., y = c1e x + c2 e6 x − e2 x . …(1)
4
Now when y = 0 , x = 0 ;
1
∴ 0 = c1 + c2 − , from (1)
4
1
so that c2 =  − c1 .
4 
D-104
Hence from (1), we have

1 1 1
y = c1e x +  − c1 e6 x − e2 x = c1 (e x − e6 x ) + (e6 x − e2 x )
4  4 4
1 2x 4x
or y = c1 (e x − e6 x ) + e (e − 1) is the required solution.
4

d2 y dy
1. 2
− 2k + k2 y = e x .
dx dx
d2 y dy
2. 2
+2 + y = 2 e2 x .
dx dx (Bundelkhand 2001)
2
d y dy
3. + + y = e x.
dx2 dx
4. [ D2 + D + 1] y = e − x .
d3 y d2 y dy
5. 3
+6 + 11 + 6 y = e2 x .
dx dx2 dx
6. ( D3 + 1) y = (e x + 1)2 .
A nswers 2
2 2x
1. y = (c1 + c2 x) e kx + e x /(1 − k )2 2. y = (c1 + c2 x) e − x + e
9
1 1 1
3. y = e − x /2 { c1 cos ( x √ 3) + c2 sin ( x √ 3)} + e x
2 2 3
4. y = c1e − x /2 cos { x (√ 3 / 2) + c2 } + e − x
1 2x 1 2x
5. y = c1e − x + c2 e −2 x + c3 e −3 x + e 6. y= e + e x +1
60 9
Case II: To find P.I. when Q is of the form sin ax or cos ax and F ( − a2 ) ≠ 0 .
By simple differentiation we know that
D (sin ax) = a cos ax ; D2 (sin ax) = − a2 sin ax ;
D3 (sin ax) = − a3 cos ax ; D4 (sin ax) = (− a2 )2 sin ax,
......,( D2 )n sin ax = (− a2 )n sin ax.
It suggests that
F ( D2 ) sin ax = F (− a2 ) sin ax. …(1)
D-105
Let F (− a2 ) ≠ 0 .
Now operating both sides of (1) with 1/ F ( D2 ), we get
1 1
F ( D2 ) sin ax = { F (− a2 ) sin ax}
F ( D2 ) F ( D2 )
1
or sin ax = F (− a2 ) ⋅ sin ax.
F ( D2 )
1 1
Thus 2
sin ax = sin ax, provided F (− a2 ) ≠ 0 .
F( D ) F( − a2 )
1 1
Similarly cos ax = cos ax, provided F (− a2 ) ≠ 0 .
F( D2 ) F ( − a2 )
Working Rule: If P.I. = {1 / F ( D)} sin ax or cos ax, put − a2 for D2 , − a2 D for D3 ,
(− a2 )2 i. e., a4 for D4 , a4 D for D5 , − a6 for D6 etc. in F ( D) and calculate the P.I.
Note: Linear factors in D of the form ( pD ± q) appearing in the denominator are

removed by first multiplying the Nr. and Dr. by the conjugate factors ( pD ∓ q) and
then putting − a2 for D2 in the denominator. The operation left in the numerator
can be easily worked out because the operator D stands for differentiation with
respect to x.
Examples on Case II:

d2 y dy
Example 11: Solve 2
− − 2 y = sin 2 x.
dx dx (Meerut 2008, 10)
Solution: The given equation is [ D2 − D − 2] y = sin 2 x.
∴ auxiliary equation is m2 − m − 2 = 0
or (m + 1) (m − 2) = 0 ; ∴ m = − 1, 2 .
−x 2x
∴ C. F. = c1e + c2 e .
1 1
And P. I. = sin 2 x = sin 2 x,
D2 − D − 2 −4− D−2
putting − 22 i. e., − 4 for D2

1 ( D − 6) ( D − 6)
=− sin 2 x = − sin 2 x = − 2 sin 2 x
D+6 ( D + 6) ( D − 6) D − 36
( D − 6)
=− sin 2 x, putting −4 for D2
−4 − 36
1 1
= ( D − 6) sin 2 x = { D (sin 2 x) − 6 sin 2 x}
40 40
D-106
1 1 3
= (2 cos 2 x − 6 sin 2 x) = cos 2 x − sin 2 x .
40 20 20
∴ the complete solution is y = (C. F. ) + (P. I. )
1 3
or y = c1e − x + c2 e2 x + cos 2 x − sin 2 x .
20 20
Example 12: Solve ( D2 − 5 D + 6) y = sin 3 x . (Kanpur 2015)
2
Solution: Here the auxiliary equation is m − 5 m + 6 = 0 .
or (m − 2) (m − 3) = 0 ; ∴ m = 2, 3.
2x 3x
∴ C. F. = c1e + c2 e .
1 1 1
And P. I. = 2
sin 3 x = 2
sin 3 x = sin 3 x
D − 5D + 6 − 3 − 5D + 6 − 9 − 5D + 6
−1 − (5 D − 3) − (5 D − 3)
= sin 3 x = sin 3 x = sin 3 x
5D + 3 (5 D + 3)(5 D − 3) 25 D2 − 9
− (5 D − 3) 1
= 2
sin 3 x = {5 D (sin 3 x) − 3 sin 3 x}
25 ⋅ (− 3 ) − 9 234
1 1
= (5 ⋅ 3 cos 3 x − 3 sin 3 x) = (5 cos 3 x − sin 3 x) .
234 78
∴ the complete solution is y = (C. F. ) + (P. I. )
or y = c1e2 x + c2 e3 x + (1/ 78) (5 cos 3 x − sin 3 x) .
d2 y
Example 13: Solve + 9 y = cos 2 x + sin 2 x .
dx2
Solution:. The auxiliary equation is
m2 + 9 = 0 , ∴ m = ± 3i .
∴ C. F. = c1 cos 3 x + c2 sin 3 x or = c1 cos (3 x + c2 )

1 1 1
And P. I. = (cos 2 x + sin 2 x) = cos 2 x + sin 2 x
( D2 + 9) D2 + 9 D2 + 9
cos 2 x sin 2 x cos 2 x sin 2 x 1
= + = + = (cos 2 x + sin 2 x) .
− 22 + 9 − 22 + 9 5 5 5
Hence the complete solution is y = (C. F. ) + (P. I. )

1
or y = c1 cos (3 x + c2 ) + (cos 2 x + sin 2 x) .
5

1. (i) (d2 y / dx2 ) + 9 y = cos 4 x (ii) ( D2 + 4) y = sin 3 x (Kumaun 2013)
D-107
2. (i) ( D2 − 3 D + 2) y = sin 3 x (ii) ( D2 − 2 D + 5) y = sin 3 x

3. (i) ( D2 − 3 D + 2) y = cos 3 x (ii) ( D3 + D2 − D − 1) y = cos 2 x
(Kanpur 2006; Avadh 08)

2 2
d y dy d y dy
4. −2 + y = cos 3 x 5. −8 + 9 y = 40 sin 5 x
dx2 dx dx2 dx
d2 y dy
6. 2
−4 + y = a sin 2 x
dx dx (Meerut 2006B)
2
d y dy
7. 2
+2 + 10 y + 37 sin 2 x = 0
dx dx (Lucknow 2005; Gorakhpur 07)
3 3 2
8. ( D + 1) y = cos 2 x 9. ( D − 2 D + 3) y = cos x
A nswers 3
1
1. (i) y = c1 cos 3 x + c2 sin 3 x − cos 4 x
7
1
(ii) y = c1 cos 2 x + c2 sin 2 x − sin 3 x
5
1
2. (i) y = c1e x + c2 e2 x + (9 cos 3 x − 7 sin 3 x)
130
(ii) y = e2 x (c1 cos 2 x + c2 sin 2 x) + (1/ 26)(3 cos 3 x − 2 sin 3 x)
1
3. (i) y = c1e x + c2 e2 x − (7 cos 3 x + 9 sin 3 x)
130
(ii) y = (c1 + c2 x) e − x + c3 e x − (1/ 25) (2 sin 2 x + cos 2 x)
1
4. y = (c1 + c2 x) e x − (3 sin 3 x + 4 cos 3 x)
50
5
5. y = c1e4 x cosh ( x √ 7 + c2 ) + (5 cos 5 x − 2 sin 5 x)
29
6. y = c1e2 x cosh (√ 3 x + c2 ) + (1 / 73) a (8 cos 2 x − 3 sin 2 x)
37
7. y = e − x (c1 cos 3 x + c2 sin 3 x) + (2 cos 2 x − 3 sin 2 x)
26
  x √ 3  x √ 3  1
8. y = c1e − x + e x /2  c2 cos   + c3 sin  + (cos 2 x − 8 sin 2 x)
  2   2   65
  x √ 3  x √ 3  1
9. y = c1e − x + e3 x /2  c2 cos   + c3 sin  + (5 cos x − sin x)
  2   2  26
Case III: To find P.I. when Q is of the form x m, where m is a positive integer.
Consider first {1/( D − a)} x m . We have
D-108
1 1 1
xm = − xm = − xm
( D − a) (a − D) a {1 − ( D / a)}
−1
1 D
=− 1 −  xm
a a
1 D D2 
=− 1 + + 2 + … x m , expanding by the binomial theorem
a a a 
1 m 1 1
=−  x + mx
m −1
+ 2 m (m − 1) x m − 2 + … .
a a a 
Here we observe that in the expansion by the binomial theorem, the terms of the
expansion beyond the mth power of D need not be written since Dm + 1 x m = 0 ,
Dm + 2 x m = 0, etc.
Working Rule: In order to evaluate {1 / F ( D)} x m , bring out common the lowest degree
term in D from F ( D) so that remaining factor in the denominator is of the form
[1 + f ( D)] or [1 − f ( D)] which is taken in the numerator with a negative index. Next we
expand [1 ± f ( D)]−1 in powers of D by the binomial theorem and operate upon x m
with the expansion obtained.
This expansion should be done upto the term Dm , since Dm + 1 x m = 0 and all higher
differential coefficients of x m are zero. The whole process will be clear from the
following examples.
The following binomial expansions should be remembered well.
(i) (1 − x)−1 = 1 + x + x2 + x3 + x4 + …
(ii) (1 + x)−1 = 1 − x + x2 − x3 + x4 − …
(iii) (1 − x)−2 = 1 + 2 x + 3 x2 + 4 x3 + …
(iv) (1 + x)−2 = 1 − 2 x + 3 x2 − 4 x3 + …
Examples on Case III:

Example 14: Solve (d2 y / dx2 ) − 4 y = x2 . (Lucknow 2010)
2
Solution: The auxiliary equation is m − 4 = 0 or m= ±2.
∴ C. F. = c1e2 x + c2 e −2 x .
−1
1 1 1  1 2
And P. I. = x2 = x2 = − 1− D x2
2
D −4 − 4 [1 − 1 2
D ] 4  4 
4
1 1 2
=− 1 + D + … x2 ,
4  4 
expanding by binomial theorem upto the terms containing D2
D-109
1
 x2 + 1 D2 ( x2 ) , because all the remaining terms vanish
=−

4 4 
1 1 1 1
= −  x2 + ⋅ 2 = −  x2 +  .
4  4  4  2 
1 1
or y = c1e2 x + c2 e −2 x − ( x2 + ) .
4 2
d3 y d2 y dy
Example 15: Solve 3
− −6 = 1 + x2 .
dx dx2 dx (Gorakhpur 2006; Avadh 06)

m3 − m2 − 6 m = 0 or m (m + 2) (m − 3) = 0 .
∴ m = 0 , − 2, 3.
∴ C. F. = c1e0 x + c2 e −2 x + c3 e3 x = c1 + c2 e −2 x + c3 e3 x .
1 1
And P. I. = 4 2
(1 + x2 ) = (1 + x2 )
D − D − 6D − 6 D [1 + 1
6
D− 1
6
D2 ]
−1
1  1
=− 1 − (− D + D2 ) (1 + x2 )
6 D  6 
1  1 1
=− 1 + (− D + D2 ) + (− D + D2 )2 + … (1 + x2 ) ,
6 D  6 36 
the terms in the expansion being needed only upto D2
1  1 1 1 2
=− 1 − D + D2 + D + … (1 + x2 )
6 D  6 6 36 
1  1 7
=− 1− D + D2 + … (1 + x2 )
6 D  6 36 
1  1 7
=− (1 + x2 ) − D (1 + x2 ) + D2 (1 + x2 ) ,
6 D  6 36 
because all other terms vanish
1  1 7 1  25 1
=− 2
1 + x − x +  = −  − x + x2 
6D  3 18  6 D  18 3 
1  25 1 2 ∵ 1/ D ≡
=−
6 ∫ 18 − 3 x + x  dx ,  ∫ dx
1 25 1 1
=− x − x2 + x3  .
6 18 6 3 
1 25 1 1
or y = c1 + c2 e −2 x + c3 e3 x −  x − x2 + x3  .

6 18 6 3 
Example 16: Solve ( D3 + 8) y = x4 + 2 x + 1 .

D-110
m3 + 8 = 0 or (m + 2) (m2 − 2 m + 4) = 0 .
∴ m = − 2, 1 ± i √ 3
Hence C.F. = c1e −2 x + e x { c2 cos ( x √ 3) + c3 sin ( x √ 3)} .
1 1
And P.I. = 3
( x4 + 2 x + 1) = ( x4 + 2 x + 1)
D +8  1 3
8 1+ D
 8 
1
= [1 + 1
8
D3 ]−1( x4 + 2 x + 1)
8
1
= [1 − 1
8
D3 + …] ( x4 + 2 x + 1),
8
the other terms in the expansion being of no need
1 1
= [( x4 + 2 x + 1)] − D3 ( x4 + 2 x + 1)
8 8
1 1
= [ x4 + 2 x + 1 − 3 x] = ( x4 − x + 1) .
8 8
1 4
or y = c1e −2 x + e x { c2 cos ( x √ 3) + c3 sin ( x √ 3)} + ( x − x + 1) .
8
Example 17: Solve ( D2 + D − 2) y = x + sin x.
Solution: The auxiliary equation is m2 + m − 2 = 0
or (m − 1) (m + 2) = 0 .
∴ m = 1, − 2 . ∴ C.F. = c1e x + c2 e −2 x .
1 1 1
And P.I. = 2 ( x + sin x) = x+ 2 sin x
D + D−2 ( D2 + D − 2) D + D−2
1 1
= x+ 2
sin x
1 1
− 2 1 − D − D2  −1 + D − 2
 2 2 
−1
1   1 D + 1 D2   ( D + 3)
=−
2 1 − 
2 2


x+
( D − 3) ( D + 3)
sin x
=−
1 1 + 1 D + … x + ( D + 3) sin x = − 1  x + 1 + D + 3 sin x
 2   
2 D2 − 9 2 2 − 1 − 9
1 1 1
=− (x + ) − { D (sin x) + 3 sin x} .
2 2 10
1 1 1
=− x− − {cos x + 3 sin x} .
2 4 10
1 1 1 3
or y = c1e x + c2 e −2 x − x − − cos x − sin x.
2 4 10 10
d2 y dy
Example 18: Solve −4 + 4 y = x2 + e x + cos 2 x .
dx2 dx (Meerut 2001)
D-111
Solution: The auxiliary equation is m2 − 4 m + 4 = 0 .
∴ m = 2, 2 .
Hence C. F. = (c1 + c2 x) e2 x ,
1
and P. I. = 2
( x2 + e x + cos 2 x) = P1 + P2 + P3 , (say),
( D − 4 D + 4)
−2
1 1 1 1 − 1 D
where P1 = x2 = x2 = x2
D2 − 4 D + 4 ( D − 2)2 4  2 
1 1 + D + 3 D2 + …  x2 = 1  x2 + 2 x + 3  ,
=
4  4  4  2 
1 ex ex
P2 = 2
ex = 2
= = e x,
( D − 4 D + 4) 1 − 4 ⋅1 + 4 1
1 1 1
and P3 = 2
cos 2 x = 2
cos 2 x = − cos 2 x
( D − 4 D + 4)] − 2 − 4D + 4 4D
1 1 1 1
=−
4 ∫ cos 2 x dx = − 4 ⋅ 2 sin 2 x = − 8 sin 2 x. (Rohilkhand 2008)
Hence the complete solution is y = C. F. + P. I.

or y = C. F. + P1 + P2 + P3
1 3 1
or y = (c1 + c2 x) e2 x + ( x2 + 2 x + ) + e x − sin 2 x.
4 2 8
1. ( D2 + D − 6) y = x . (Bundelkhand 2007)
3 3 2 2 2
2. (d y / dx ) + 3(d y / dx ) + 2(dy / dx) = x . (Bundelkhand 2001)
3 2
d y d y dy
3. 3
−4 2
+5 − 2 = 0.
dx dx dx
4. ( D + 2 D + 1) y = 2 x + x2 .
2
(Kumaun 2007)
5. ( D2 + 3 D + 2) y = x2 . (Bundelkhand 2001)
6. ( D4 + D2 + 16) y = 16 x2 + 256 . (Lucknow 2005)

2
7. ( D − 5 D + 6) y = x + sin 3 x . (Kanpur 2007)
8. ( D2 − 4 D + 3) y = e − x + 5 .
9. (d2 y / dx2 ) − 4 y = sin2 x. (Gorakhpur 2007)
10. ( D2 − 2 D + 3) y = cos x + x2 . (Meerut 2009B; Bundelkhand 11)

D-112
d3 y d2 y dy
11. 3
+2 2
+ = e2 x + x2 + x .
dx dx dx (Meerut 2009; Gorakhpur 10)
4 3 2 3
12. ( D − 2 D + D ) y = x . (Garhwal 2007)
A nswers 4
1
1. y = c1e2 x + c2 e −3 x − (6 x + 1)
36
1
2. y = c1 + c2 e − x + c3 e −2 x + x (2 x2 − 9 x + 21)
12
2
3. y = c1 + c2 e2 x cos ( x + c3 ) + x
5
4. y = (c1 + c2 x) e − x + ( x2 − 2 x + 2)
1 2 3 7
5. y = c1 e − x + c2 e − 2 x + x − x+ ⋅
2 2 4
3 3 127
6. y = c1 e −(1 /2)√7. x sin  x + c2  + c3 e(1 /2)√7 x sin  x + c4  + x2 +
2  2  8
1 5 1
7. y = c1e2 x + c2 e3 x + x +  − (3 sin 3 x − 15 cos 3 x)
6  6  234
1 −x 5
8. y = c1 e x + c2 e3 x + e +
8 3
1 1
9. y = c1e2 x + c2 e −2 x − + cos 2 x
8 16
1 2 4 2 1
10. y = e x [c1 cos √ 2 x + c2 sin √ 2 x] + x + x+ + (cos x − sin x)
3 9 27 4
1 2x 1 3 3 2
11. y = c1 + (c2 + c3 x) e − x + e + x − x + 4x
18 3 2
1 5 1 4
12. y = c1 + c2 x + (c3 + c4 x) e x + x + x + 3 x3 + 36 x2
20 2
5 P.I. when Q = eax V , where V is any Function of x

By successive differentiation we notice that
D (e axV ) = e ax D (V ) + Vae ax = e ax ( D + a) V ,
D2 (e ax V ) = D { e ax ( D + a) V }
= e ax D ( D + a) V + ae ax ( D + a) V = e ax ( D + a)2 V .
Similarly, D3 (e axV ) = e ax ( D + a)3 V , … ,
D-113
Dn (e ax V ) = e ax ( D + a)nV .
∴ F ( D) (e axV ) = e ax F ( D + a) V . …(1)
1
The result (1) is true for any function V of x. Taking V in place of V in (1), we
F ( D + a)
have
 1   1 
F ( D)  e ax V  = e ax F ( D + a)  V
 F ( D + a)   F ( D + a) 
 1 
i. e., F ( D)  e ax V  = e axV . …(2)
 F ( D + a) 
1
Operating by on both sides of (2), we get
F ( D)
1 1
e ax V = (e ax V ).
F ( D + a) F ( D)
1 1
Thus, (e axV ) = e ax V.
F ( D) F ( D + a)
Working Rule: Replace D by ( D + a) and take out e ax before the operator1 / F ( D). Then
determine {1 / F ( D + a)} V by the methods discussed in 4.
This method also enables us to find {1 / F ( D)} e ax when F (a) is zero. We shall discuss it
later on in article 6.
Example 19: Solve ( D2 − 2 D + 1) y = x2 e3 x .

(Agra 2005; Kumaun 06. 08; Avadh 09; Garhwal 13)

m2 − 2 m + 1 = 0 or (m − 1)2 = 0 .
∴ m = 1, 1 .
∴ C.F. = (c1 + c2 x) e x .
1 1
And P. I. = x2 e3 x = e3 x x2
D2 − 2 D + 1 ( D − 1)2
1
= e3 x x2 ,
{( D + 3) − 1}2
putting D + 3 for D and bringing e3 x before the operator
−2
1 1 1 3x  1
= e3 x x2 = e3 x x2 = e 1 + D x2
( D + 2)2 1
2 4  2 
4 1 + D
 2 
D-114
1 3x 1 1
= e 1 − 2 ⋅ D + 3 ⋅ D2 + … x2 ,
4 
 2 4 
expanding by binomial theorem
1 3
1 − D + D2 + … x2
= e3 x
4  4 
1 3x  x2 − 2 x + 3 ⋅ 2 = 1 e3 x  x2 − 2 x + 3  .
= e    
4  4  4  2
Hence the required solution is y = (C. F. ) + (P. I. )
1 3
or y = (c1 + c2 x) e x + e3 x  x2 − 2 x +  .
4  2
Example 20: Solve ( D2 − 2 D + 5) y = e2 x sin x.
∴ m = {2 ± √ (4 − 20 )} / 2 = 1 ± 2 i.
Hence C.F. = e x (c1 cos 2 x + c2 sin 2 x) .
1
And P. I. = 2
e2 x sin x
D − 2D + 5
1 1
= e2 x sin x = e2 x 2 sin x
( D + 2)2 − 2 ( D + 2) + 5 D + 2D + 5
1
= e2 x sin x, putting − 12 for D2
(− 12 + 2 D + 5)
1 2x 1 1 D−2
= e sin x = e2 x sin x
2 D+2 2 ( D − 2) ( D + 2)
1 2x D − 2 1 D−2
= e 2
sin x = e2 x sin x
2 ( D − 4) 2 − 1− 4
1 2x 1 2x
=− e ( D − 2) sin x = − e { D (sin x) − 2 sin x}
10 10
1 2x
=− e (cos x − 2 sin x) .
10
1 2x
or y = e x (c1 cos 2 x + c2 sin 2 x) − e (cos x − 2 sin x) .
10

1. ( D + 1)3 y = x2 e − x .
2. ( D2 − 2 D + 1) y = x2 e x . (Avadh 2007; Kumaun 09)
D-115
3. ( D2 + 2 D + 1) y = e − x /( x + 2).
4. ( D2 − 3 D + 2) y = xe x . (Bundelkhand 2008)
2 2 2x
5. ( D − 2 D + 1) y = x e . (Purvanchal 2007)
2 2x
6. ( D − 4 D + 4) y = e sin 3 x. (Purvanchal 2006)
4 x
7. ( D − 1) y = e cos x. (Lucknow 2010)
2 x
8. ( D − 2 D + 4) y = e cos x . (Garhwal 2001)
2 2x
9. ( D + 4 D − 12) y = ( x − 1) e .
10. ( D2 − 5 D + 6) y = e2 x sin 2 x.
11. ( D2 − 4 D + 3) y = e2 x sin 3 x. (Garhwal 2008)
2 x
12. ( D + 2 D + 2) y = xe .
13. ( D2 − 1) y = e x cos x.
A nswers 5
1 5 −x
1. y = (c1 + c2 x + c3 x2 ) e − x + x e
60
1 4 x
2. y = (c1 + c2 x) e x + x e
12
3. y = (c1 + c2 x) e − x + e − x { x log ( x + 2) − x + 2 log ( x + 2)}
1
4. y = c1e x + c2 e2 x − e x  x2 + x
2 
5. y = (c1 + c2 x) e x + e2 x ( x2 − 4 x + 6)
1 2x
6. y = (c1 + c2 x) e2 x − e sin 3 x
9
1 x
7. y = c1 e x + c2 e − x + c3 cos x + c4 sin x − e cos x
5
1 x
8. y = e x (c1 cos √ 3 x + c2 sin √ 3 x) + e cos x
2
1 2 9
9. y = e2 x (c1 + x − x) + c2 e −6 x
16 64
1 2x
10. y = c1e2 x + c2 e3 x + e (cos 2 x − 2 sin 2 x)
10
1 2x
11. y = c1e x + c2 e3 x − e sin 3 x
10
1 4
12. y = e − x (c1 cos x + c2 sin x) + e x  x − 
5  5
1
13. y = c1e x + c2 e − x − e x (cos x − 2 sin x)
5
D-116
6 P.I. when Q = eax and F ( a ) = 0

In this case we use the method explained in article 5 to find the P.I.
1 1
We have P.I. = e ax = e ax . 1, if F (a) = 0 .
F( D) F( D)
Now this is of the form e ax V . Hence by the method of article 5, i. e., by putting ( D + a)
for D in the operator and by making e ax free from the operator, we have
1
P. I. = e ax 1.
F ( D + a)
Now this can be evaluated by using the method for finding P.I. in the case of x m .
d2 y dy
Example 21: Solve 2
−3 + 2 y = e x.
dx dx (Garhwal 2003, 05)
2
Solution: The auxiliary equation is m − 3 m + 2 = 0
or (m − 1) (m − 2) = 0 . ∴ m = 1, 2 .
x 2x
Hence C. F. = c1e + c2 e .
1 1 1
And P. I. = ex = ex = ex
D2 − 3 D + 2 ( D − 2) ( D − 1) (1 − 2) ( D − 1)
[putting 1 for D in the factor D − 2 because
it does not vanish by doing so]
1
=− e x ⋅1
D −1
[Note that D − 1 becomes zero by putting 1 for D ; so
here we shall apply the method for e axV by taking 1 for V.]
1 1
= − ex 1= − ex 1= − ex x .
( D + 1) − 1 D
or y = c1e x + c2 e2 x − x e x .
Note: While finding P.I. in the case of e ax if F ( D) becomes zero by putting a for D, we
factorise F ( D) . Then we put D = a in the factors which do not vanish by doing so. The
remaining operator is then dealt with by using the method for e axV on taking 1 for V.
Example 22: Solve ( D3 − 7 D + 6) y = e2 x .

m3 − 7 m + 6 = 0 or (m − 1) (m − 2) (m + 3) = 0 .
∴ m = 1, 2, − 3 .
D-117
∴ C. F. = c1e x + c2 e2 x + c3 e −3 x .
1 1
And P. I. = 3
e2 x = e2 x
( D − 7 D + 6) ( D − 1) ( D − 2) ( D + 3)
1
= e2 x ,
(2 − 1) ( D − 2) (2 + 3)
putting 2 for D in all the factors except D − 2
1 1 1 1 1 1 1
= e2 x ⋅ 1 = e2 x 1 = e2 x 1 = xe2 x .
5 D−2 5 ( D + 2) − 2 5 D 5
Hence the complete solution is
1 2x
y = (C. F. ) + (P. I. ) or y = c1e x + c2 e2 x + c3 e −3 x + xe .
5
Example 23: Solve ( D2 + 4 D + 4) y = e2 x − e −2 x
or ( D2 + 4 D + 4) y = 2 sinh 2 x. (Garhwal 2010)

2
Solution: The auxiliary equation is (m + 2) = 0 .
∴ m = − 2, − 2.
Hence C. F. = (c1 + c2 x) e −2 x .
1 1 1
And P. I. = 2
(e2 x − e −2 x ) = 2
e2 x − e −2 x .
( D + 2) ( D + 2) ( D + 2)2
1 e2 x 1 2x
Now 2
e2 x = 2
= e . [∵ Here, F (a) ≠ 0]
( D + 2) (2 + 2) 16
1 1
Also e −2 x = e −2 x ⋅ 1, [∵ Here F (a) = 0]
( D + 2)2 ( D + 2)2
1 1 1 1
= e −2 x 2
1 = e −2 x 2
1 = e −2 x x = x2 e −2 x .
{( D − 2) + 2} D D 2

1 2 x 1 2 −2 x
or y = (c1 + c2 x) e −2 x + e − x e .
16 2
Example 24: Solve D2 y − 3 Dy + 2 y = cosh x . (Rohilkhand 2010)

m2 − 3 m + 2 = 0 or (m − 1) (m − 2) = 0 ∴ m = 1, 2 .
x 2x
∴ C. F. = c1e + c2 e .
1 1  e x + e− x 
And P. I. = cosh x =  
D2 − 3 D + 2 ( D2 − 3 D + 2)  2 
1 1 1 1
= ex + e− x .
2 D2 − 3 D + 2 2 D2 − 3 D + 2
D-118
1 1 1 1
Now 2
ex = ex
2 D − 3D + 2 2 ( D − 1) ( D − 2)
1 1 1 1
= ex = − e x ⋅1
2 ( D − 1) (1 − 2) 2 D −1
1 x 1 1 1 1
=− e 1= − ex 1 = − xe x .
2 D + 1− 1 2 D 2
1 1 1 1
Also e− x = e − x , putting −1 for D
2 D2 − 3 D + 2 2 (− 1)2 − 3 (− 1) + 2
1 −x
= e .
12
1 x 1 −x
∴ P. I. = − xe + e .
2 12
1 1 −x
or y = c1e x + c2 e2 x − xe x + e .
2 12
Example 25: Solve ( D2 − 4 D + 4) y = e2 x + sin 2 x .

m2 − 4 m + 4 = 0 or (m − 2)2 = 0 .
∴ m = 2, 2 .
Hence C. F. = (c1 + c2 x) e2 x .
1 1
And P.I. = 2
e2 x + 2
sin 2 x .
( D − 2) D − 4D + 4
1 1 1
Now 2
e2 x = 2
e2 x ⋅ 1 = e2 x 1
( D − 2) ( D − 2) ( D + 2 − 2)2
1 1 x2 1 2 2 x
= e2 x 2
(1) = e2 x ( x) = e2 x ⋅ = x e ,
D D 2 2
1 1
and 2
sin 2 x = 2
sin 2 x, putting − 22 for D2
D − 4D + 4 − 2 − 4D + 4
1 1 1 1 1
=
− 4D
sin 2 x = −
4 ∫ sin 2 x dx = − 4 ⋅ 2 (− cos 2 x) = 8 cos 2 x .
1 1
or y = (c1 + c2 x) e2 x + x2 e2 x + cos 2 x .
2 8

1. ( D2 − 2 D + 1) y = e x . (Garhwal 2006)
2. ( D2 + D − 6) y = e2 x .
D-119
3. ( D2 − 4 D + 3) y = 2 e3 x .
4. ( D3 + 3 D2 + 3 D + 1) y = e − x .
5. ( D2 − a2 ) = cosh ax .
6. ( D3 − 5 D2 + 7 D − 3) y = e2 x cosh x . (Rohilkhand 2005)
2 2 2x
7. ( D − 4 D + 4) y = 8 ( x + e + sin 2 x) .
A nswers 6
1 2 x 1 2x
1. y = (c1 + c2 x) e x + x e 2. y = c1e2 x + c2 e −3 x + xe
2 5
1 3 −x
3. y = c1e x + c2 e3 x + xe3 x 4. y = (c1 + c2 x + c3 x2 ) e − x + x e
6
5. y = c1e ax + c2 e − ax + ( x / 2 a)sinh ax
1 1
6. y = (c1 + c2 x) e x + c3 e3 x + xe3 x − x2 e x
8 8
3
7. y = (c1 + c2 x) e2 x + 2 ( x2 + 2 x + ) + 4 x2 e2 x + cos 2 x
2
7 P.I. when Q = sin ax or cos ax and F ( − a 2 ) = 0

To find out these types of particular integrals, it is convenient to replace sin ax or cos ax
by the exponential value and then apply 6. Thus the particular integral of cos ax
1
i. e., cos ax, when F (− a2 ) = 0 , can be written as
F ( D2 )
1
= (Real part of e iax ), [∵ e iax = cos ax + i sin ax]
F ( D2 )
1
= Real Part of e iax ⋅ 1, which is now of the form of article 5.
F ( D2 )
Similarly, the particular integral of sin ax
1
i. e., 2
sin ax, when F (− a2 ) = 0
F (D )
1
= (Imaginary part of e iax ), [∵ e iax = cos ax + i sin ax]
F ( D2 )
1
= Imaginary part of e iax ⋅ 1,.
F( D2 )
1
which is now of the form e ax V of article 5.
F ( D)
Now the method of article 5 is to be followed.

D-120
d2 y
Example 26: Solve + a2 y = cos ax .
dx2 (Meerut 2003)

m2 + a2 = 0 , i. e., m = 0 ± ai .
∴ C. F. = e0 x (c1 cos ax + c2 sin ax)

= c1 cos ax + c2 sin ax .
1
And P. I. = cos ax
D2 + a2
1
= the real part in (cos ax + i sin ax)
D + a2
2
1
= the real part in e iax ,
D + a2
2
by Euler’s theorem of trigonometry.

1 iax 1
Now e = e iax
D2 + a2 ( D + ia) ( D − ia)
1
= e iax , putting ia for D in the factor D + ia
(ia + ia) ( D − ia)
1
= e iax ⋅ 1
2 ia ( D − ia)
1 iax 1 1 iax 1
= e 1= e 1
2 ia ( D + ia) − ia 2 ia D
1 iax x
= e .x = (cos ax + i sin ax), [∵ e iax = cos ax + i sin ax]
2 ia 2 ai
ix  1 i i 
=− (cos ax + i sin ax), ∵ = 2 = = − i
2a  i i −1 
x x
=−i cos ax + sin ax, [∵ i2 = − 1]
2a 2a
1 x x
∴ 2 2
cos ax = the real part in  sin ax − i cos ax
D +a  2 a 2 a 
x
= sin ax .
2a (Agra 2006)

or y = c1 cos ax + c2 sin ax + ( x / 2 a)sin ax.
Remember: While solving the differential equation ( D2 + a2 ) y = cos ax , the P.I.
1 x x
2
D +a 2
cos ax =
2a
sin ax =
2 ∫ cos ax dx .
D-121
Example 27: Solve ( D2 + a2 ) y = sin ax . (Bundelkhand 2008; Rohilkhand 10)
Solution: Here as in Example 26, C.F. = c1 cos ax + c2 sin ax.

1
Also P. I. = sin ax
D + a2
2
1
= the coefficient of i in (cos ax + i sin ax)
D2 + a2
1
= the coefficient of i in e iax .
D + a2
2
1 x x
Now 2 2
e iax = − i cos ax + sin ax . [Proceed as in Example 26]
D +a 2a 2a
1
∴ sin ax
D + a2
2
x x x
= the coefficient of i in  − i cos ax + sin ax = − cos ax .
 2a 2a  2a
or y = c1 cos ax + c2 sin ax − ( x / 2 a)cos ax.
1 x x
Remember : 2
D +a 2
sin ax = −
2a
cos ax =
2
sin ax dx .
∫
Example 28: Solve ( D2 + 4) y = sin2 x .
(Rohilkhand 2007; Gorakhpur 10; Avadh 11)
2
Solution: The auxiliary equation is m + 2 = 0 .
∴ m = 0 ± 2i .
Hence C. F. = c1 cos 2 x + c2 sin 2 x. [∵ e0 x = 1]
1 1 1 1 1
And P. I. = sin2 x = (2 sin2 x) = (1 − cos 2 x)
D2 + 4 D2 + 4 2 2 ( D2 + 4)
1 1 1 1
= 2
1− 2
cos 2 x .
2 D +4 2 D +4
−1
1 1 1 1 1 1 2 1
Now 1 = 1 + D2  1= 
1 − D + …  1 = ⋅
2 D2 + 4 8 4  8 4  8
1 1 1 1
Again cos 2 x = ⋅ x sin 2 x. [Refer Example 26]
2 D2 + 4 2 4
1 1
or y = c1 cos 2 x + c2 sin 2 x + − x sin 2 x .
8 8
Example 29: Solve ( D2 + 1) y = sin x sin 2 x .
Solution: The auxiliary equation is m2 + 1 = 0 or m = ± i .
∴ C. F. = c1 cos x + c2 sin x .
D-122
1 11
And P. I. = 2
(sin x sin 2 x) = 2
(2 sin x sin 2 x)
D +1 D +12
1 1 1 1 1 1
= (cos x − cos 3 x) = cos x − cos 3 x .
2 D2 + 1 2 D2 + 1 2 D2 + 1
1 1
Now 2
cos x = the real part in 2
e ix
D +1 D +1
1
= the real part in e ix
( D + i) ( D − i)
1
= the real part in e ix
(i + i) ( D − i)
1 1 1
= the real part in e ix ⋅ 1 = the real part in e ix 1
2 i ( D − i) 2i D + i− i
1 ix 1 1
= the real part in e 1 = the real part in e ix x
2i D 2i
x
= the real part in (cos x + i sin x)
2i
1
= the real part in − xi (cos x + i sin x)
2
1 1 1
= the real part in − i x cos x + x sin x = x sin x.
2 2 2
1 1 1
Again 2
cos 3 x = 2
cos 3 x = − cos 3 x .
D +1 − 3 +1 8
1 1 1 1 1 1
∴ the P.I. = ⋅ x sin x − ⋅ (− cos 3 x) = x sin x + cos 3 x .
2 2 2 8 4 16
1 1
or y = c1 cos x + c2 sin x + x sin x + cos 3 x .
4 16

1. (i) ( D2 + 4) y = sin 2 x. (Agra 2005; Gorakhpur 09; Kumaun 11)
(ii) ( D2 + 4) y = cos 2 x. (Kumaun 2012)
2. ( D3 + a2 D) y = sin ax. (Avadh 2005, Gorakhpur 05)
2 2
3. ( D + 9) ( D + 1) y = cos 3 x .
4. ( D4 − m4 ) y = sin mx.
5. ( D2 + 9) y = 2 sin 3 x + cos 3 x.
6. ( D4 + 2 a2 D2 + a4 ) y = cos ax.
D-123
d2 y
7. Find the solution of the equation + 4 y = 8 cos 2 x, given that y = 0 and
dx2
dy
= 2 when x = 0.
dx
A nswers 7
1
1. (i) y = c1 cos 2 x + c2 sin 2 x − x cos 2 x
4
x
(ii) y = c1 cos 2 x + c2 sin 2 x + sin 2 x
4
2. y = c1 + c2 cos ax + c3 sin ax − (1/ 2 a2 ) x sin ax
1
3. y = c1 cos 3 x + c2 sin 3 x + c3 cos x + c4 sin x − x sin 3 x
48
x
4. y = c1e mx + c2 e − mx + c3 cos mx + c4 sin mx + cos mx
4 m3
1 1
5. y = c1 cos 3 x + c2 sin 3 x − x cos 3 x + x sin 3 x.
3 6
x2
6. y = (c1 + c2 x)cos ax + (c3 + c4 x)sin ax − 2 cos ax
8a
7. y = sin 2 x + 2 x sin 2 x
8 P.I. when Q = xV , where V is any Function of x

1
Here P.I. = ( xV ).
F ( D)
By Leibnitz’s theorem, we have
 d 
Dn ( xV ) = x ⋅ DnV + nDn −1V = x DnV +  Dn V ,
 d D 
showing that F ( D) ( xV ) = xF ( D) V + F ′ ( D) V . …(1)
1 1
Taking V in place of V in (1) and then operating on both sides by , we get
F ( D) F ( D)
after transposition
1 1 F ′ ( D)
( xV ) = x V − V [Remember]
F( D) F ( D) { F( D)}2
1 1  d 1 
or ( xV ) = x V +  V.
F ( D) F ( D)  d D F ( D)
Proceeding by repeated application of the above formula {1 / F ( D)} x mV can be
determined.
D-124
Note: The method of article 8 cannot be used when V is of the form cos ax or sin ax and
F (− a2 ) = 0 i. e., F ( D2 ) vanishes by putting − a2 for D2 . In that case P.I. is found by
the method discussed in article 7.
d2 y dy
Example 30: Solve −2 + y = x sin x .
dx2 dx (Rohilkhand 2010)
∴ m = 1, 1 . Hence C.F. = (c1 + c2 x) e x .

1 1 2
And P. I. = 2
x sin x = x 2
sin x − sin x, by article 8
( D − 1) ( D − 1) ( D − 1)3
1 2
=x sin x − sin x
D2 − 2 D + 1 D3 − 3 D2 + 3 D − 1
1 2
= x⋅ sin x − sin x , putting − 12 i. e., −1 for D2
−2 D − D + 3 + 3D − 1
x 1 2 x 1
=− ⋅ sin x − sin x = cos x − sin x
2 D 2 ( D + 1) 2 D +1
x ( D − 1) x cos x D −1
= cos x − sin x = − 2 sin x
2 ( D + 1) ( D − 1) 2 D −1
1 1 1 1
= x cos x + ( D − 1) sin x = x cos x + (cos x − sin x).
2 2 2 2
1 1
or y = (c1 + c2 x) c x + x cos x + (cos x − sin x).
2 2
d2 y
Example 31: Solve + 4 y = x sin x .
dx2
Solution: The auxiliary equation is m2 + 4 = 0 .
Therefore m = 0 ± 2 i .
∴ C. F. = c1 cos 2 x + c2 sin 2 x . [∵ e0 x = 1]
1 1 2D
And P. I. = x sin x = x sin x − sin x, by article 8
D2 + 4 D2 + 4 ( D2 + 4)2
x sin x 2D
= 2
− sin x, putting − 12 for D2
−1 + 4 (− 1 + 4)2
2
1 2 1 2
= x sin x − D (sin x) = x sin x − cos x .
3 9 3 9
D-125
1 2
or y = c1 cos 2 x + c2 sin 2 x + x sin x − cos x .
3 9
Example 32: Solve ( D4 − 1) y = x sin x . (Lucknow 2006)
4 2 2
Solution: The auxiliary equation is m − 1 = 0 or (m − 1) (m + 1) = 0 .
∴ m = 1, − 1, 0 ± i .
∴ C. F. = c1e x + c2 e − x + e0 x (c3 cos x + c4 sin x) .
1
And P. I. = 4
x sin x .
D −1
Here we cannot use the method given in 8 because D4 − 1vanishes by putting − 12 i. e.,
−1 for D2 . So here we shall proceed by the method given in article 7.
1 [∵ e ix = cos x + i sin x]
∴ P.I. = Imaginary part of 4 x e ix ,
D −1
1
= I. P. of e ix x, by using the method for e ax V
{( D + i)4 − 1}
1
= I. P. of e ix x
( D + 4 iD + 6 i D2 + 4 i3 D + i4 − 1)
4 3 2
1
= I. P. of e ix 4 3 2
x, [∵i2 = − 1]
D + 4 iD − 6 D − 4 iD
1
= I. P. of e ix x
− 4 iD [1 + (3 D / 2 i) − D2 − ( D3 / 4 i)]
−1
e ix 1   3 D − .... 
= I. P. of ⋅ 1 +   x
−4i D  2i 
e ix 1 3D
= I. P. of ⋅ 1 − + ... x
−4i D 2i 
e ix 1  3 ∵ D ≡ d 
= I. P. of x −  ,
− 4i D  2 i  dx 
e ix  x2 3 x 
= I. P. of  − , [∵ (1/ D) stands for integration w.r.t. x]
− 4i  2 2i 
1 ix  1 2 3 
= I. P. of i e  x + i x ; [∵ i2 = − 1 ⇒ 1/ i = − i]
4 2 2 
1 1 3
= I. P. of i (cos x + i sin x)  x2 + ix
4 2 2 
1 1 2 3 
=  x cos x − x sin x
4 2 2 
1 2
= ( x cos x − 3 x sin x) .
8
D-126
1 2
or y = c1e x + c2 e − x + c3 cos x + c4 sin x + ( x cos x − 3 x sin x).
8
Example 33: Solve ( D4 + 2 D2 + 1) y = x2 cos x .
(Gorakhpur 2010; Purvanchal 07, 08; Garhwal 09, 11)
Solution: The auxiliary equation is m4 + 2 m2 + 1 = 0
or (m2 + 1)2 = 0 giving m = ± i, ± i .

∴ C. F. = (c1 + c2 x) cos x + (c3 + c4 x) sin x . [∵ e0 x = 1]
1
And P. I. = x2 cos x
D + 2 D2 + 1
4
1
= Real part of x2 e ix , [∵ e ix = cos x + i sin x]
{( D2 + 1)2
1
= R. P. of e ix x2 , by article 5
{( D + i)2 + 1}2
1
= R. P. of e ix 2 2
x2 , [∵ i2 = − 1]
( D + 2 iD)
1
= R. P. of e ix x2
4 i2 D2 [1 + ( D / 2 i)]2
−2
1 ix 1  D
= R. P. of − e 2 
1+  x2 , [∵ i2 = − 1]
4 D  2 i
1 ix 1 1 ∵ 1 = − i
= R. P. of − e [1 − iD]−2 x2 ,
4 D2 2  i 
1 ix 1  1 1
= R. P. of − e 1 + 2 ⋅ iD + 3 ⋅ i2 D2 + … x2 ,
4 2
D  2 4 
expanding by binomial theorem

1 ix 1  3
= R. P. of − e 1 + iD − D2 + … x2
4 D2  4 
1 ix  1 i 3
= R. P. of − e + − + terms in D, D2 , and so on  x2
4  D2 D 4 
1 ix  1 x4 1 3 
= R. P. of − e  + i x3 − x2 + terms in x1 , x 0 ,
4  3 4 3 4 
(∵ 1/ D stands for integration w.r.t. x)
1 1 4 1 3 3 2
= R. P. of − (cos x + i sin x)  x + ix − x + terms in x1, x0 
4  12 3 4 
1  1 4 3 2 1  1 x3  sin x
=−  x − x  cos x +  
4 12 4  4 3 
+ terms already included in the C.F.
D-127
1 4 1 3
=− ( x − 9 x2 ) cos x + x ⋅ sin x ,
48 12
neglecting the terms already included in the C.F.
1 4 1 3
or y = (c1 + c2 x) cos x + (c3 + c4 x) sin x − ( x − 9 x2 ) cos x + x sin x .
48 12
Example 34: Solve ( D2 − 2 D + 1) y = xe x sin x ..
(Purvanchal 2010; Kanpur 12; Avadh 13; Kumaun 13)

m2 − 2 m + 1 = 0 or (m − 1)2 = 0 . ∴ m = 1, 1 .
x
∴ C. F. = (c1 + c2 x) e .
1 1
And P.I.= 2
xe x sin x = e x x sin x
D − 2D + 1 ( D − 1)2
1
= ex x sin x , by 4.5
{( D + 1) − 1}2
1 1
= ex ( x sin x) = e x ⋅
D 2 D ∫ x sin x dx,
[∵ (1 / D) stands for integration w.r.t. x]
1
= ex (− x cos x + sin x) , integrating by parts
D
= ex x
∫ (− x cos x + sin x) dx = e [∫ − x cos x dx + ∫ sin x dx]
= e x [− x sin x + sin x dx + sin x dx] = e x (− x sin x − 2 cos x) .
∫ ∫
or y = (c1 + c2 x) e x − e x ( x sin x + 2 cos x) .

1. d2 y / dx2 + 9 y = x sin x .
2. ( D2 + 2 D + 1) y = x cos x .
3. ( D2 − 4 D + 4) y = 8 x2 e2 x sin 2 x . (Gorakhpur 2011; Meerut 05B; Garhwal 12)
2
4. ( D + 4) y = x sin 2 x .
5. ( D2 + D) y = x cos x .
6. ( D2 − 1) y = x2 cos x .
7. ( D2 + 1) y = x2 sin 2 x . (Meerut 2004, 07B; Lucknow 11)
D-128
8. ( D4 − 1) y = x sin 2 x . (Meerut 2001, 06B)

9. ( D2 + m2 ) y = x cos mx .
10. (d2 y / dx2 ) − y = x sin x + (1 + x2 ) e x . (Bundelkhand 2009)
A nswers 8
1 1
1. y = c1 cos 3 x + c2 sin 3 x + x sin x − cos x
8 32
1 1 1
2. y = (c1 + c2 x) e − x + x sin x − sin x + cos x
2 2 2
3. y = (c1 + c2 x) e2 x + e2 x (− 2 x2 sin 2 x + 3 sin 2 x − 4 x cos x)
1 2 1
4. y = c1 cos 2 x + c2 sin 2 x − x cos 2 x + x sin 2 x
8 16
1 1
5. y= c1 + c2 e − x − x (cos x − sin x) + cos x + sin x
2 2
x −x 1 2
6. y= c1e + c2 e − ( x − 1) cos x + x sin x
2
1
7. y= c1 cos x + c2 sin x − [24 x cos 2 x − (9 x2 − 26) sin 2 x]
27
1  32
8. y= c1 e x + c2 e − x + c3 cos x + c4 sin x +  cos 2 x + x sin 2 x
15  15 
9. y = c1 cos mx + c2 sin mx + ( x / 4 m2 ) cos mx + ( x2 / 4 m)sin mx
1 1 x
10. y = c1e x + c2 e − x − (cos x + x sin x) + e (2 x3 − 3 x2 + 9 x)
2 12
1 1
9 The Operator and , α being a Constant
D −α D+α
If Q is any function of x, then

1 1
Q = eα x −α x
Q = e − α x e α x Q dx
D−α ∫e Q dx and
D+α ∫
(Meerut 20, 03; Rohilkhand 09)
1 1
Proof. We have Q= [e α x (e − α x Q)]
D−α D−α
1
= eα x e − α x Q , by article 5
( D + α) − α
1 −α x
= eα x Q = e α x e − α x Q dx .
D
e
∫
Similarly proceed for second operator proof.
D-129
d2 y
Example 35: Solve + a2 y = sec ax .
dx2
(Garhwal 2002; Agra 06; Bundelkhand 07; Purvanchal 07, 08; Lucknow 09)

m2 + a2 = 0 or m = ± ia .
∴ C. F. = c1 cos ax + c2 sin ax .
1 1
And P. I. = 2 2
sec ax = sec ax
D +a ( D + ia) ( D − ia)
1  1 1 
=  −  sec ax [by resolving into partial fractions]
2 ia  D − ia D + ia 
1  1 1 
=  sec ax − sec ax
2 ia  D − ia D + ia 
1  iax − iax
sec ax dx − e − iax e iaxsec ax dx , by article 9
=
2 ia 
e
∫e ∫ 
1  iax cos ax − i sin ax cos ax + i sin ax 

dx − e − iax
=
2 ia 
e
∫cos ax ∫
cos ax
dx ,

(∵ e − iax = cos ax − i sin ax and e iax = cos ax + i sin ax)
1  iax  i  x − i log cos ax 

= e  x + log cos ax − e − iax  
2 ia   a   a  
x  e iax − e − i ax  1  e iax + e − iax 

=   + 2 (log cos ax)  
a 2i  a  2 
x 1
= sin ax + 2 (log cos ax) cos ax .
a a
y = (C. F. ) + (P. I. )
x 1
or y = c1 cos ax + c2 sin ax + sin ax + 2 cos ax ⋅ log (cos ax).
a a
d2 y
Example 36: Solve + 4 y = 4 tan2 x. (Garhwal 2015)
dx2
m2 + 4 = 0
or m = ± 2 i.
∴ C.F. = c1 cos 2 x + c2 sin 2 x.
D-130
1 1
And P.I. = 4 tan 2 x = 4 tan 2 x
D2 + 22 ( D + 2 i)( D − 2 i)
1 1 1  1 1 1
=  − tan 2 x =  tan 2 x − tan 2 x

i  D − 2 i D + 2 i  
i  D − 2i D + 2i 
1 sin 2 x
tan 2 x = e2 ix e −2 ix tan 2 x dx = e2 ix (cos 2 x − isin 2 x)
Now
D − 2i ∫ ∫ cos 2 x
dx
 isin2 2 x   i(1 − cos2 2 x)

= e2 ix sin 2 x −  dx = e2 ix sin 2 x −
 ∫ cos 2 x  
∫ cos 2 x
 dx

= e2 ix 2 ix −2 ix
∫ {sin 2 x − isec 2 x + icos 2 x} dx = e i ∫ (e − sec 2 x) dx
 e −2 ix 1 
= ie2 ix  − log (sec 2 x + tan 2 x)
 − 2 i 2 
1 ie2 ix
=− − log (sec 2 x + tan 2 x).
2 2
1 −1 ie2 ix
Similarly, tan 2 x = + log (sec 2 x + tan 2 x)
D + 2i 2 2
1   e2 ix + e −2 ix  
P.I. = − i   log (sec 2 x + tan 2 x)
i   2  
= − cos 2 x log (sec 2 x + tan 2 x).
y = C.F. + P.I.
= c1 cos 2 x + c2 sin 2 x − cos 2 x log (sec 2 x + tan 2 x).

1. (d2 y / dx2 ) + 9 y = sec 3 x. (Meerut 2007)
2 2
2. ( D + a ) y = cosec ax. (Lucknow 2008)
2 2
3. ( D + a ) y = tan ax.
x
4. ( D2 + 3 D + 2) y = e e . (Kumaun 2014)
2
5. ( D + 1) y = x − cot x.
A nswers 9
1 1
1. y = c1 cos 3 x + c2 sin 3 x + x sin 3 x + cos 3 x log (cos 3 x)
3 9
D-131
2. y = c1 cos ax + c2 sin ax + (1/ a2 )sin ax log(sin ax) − ( x / a)cos ax

1 1
3. y = c1 cos ax + c2 sin ax − (1/ a2 )cos ax log tan  π + ax
4 2 
x
4. y = c1e − x + c2 e −2 x + e −2 x e e
5. y = c1 cos x + c2 sin x + x − sin x log (cosec x − cot x)

(b), (c) and (d).
d2 y
1. The solution of differential equation + y = 0 is
dx2
(a) c1 e − x + c2 e x (b) c1 cos x + c2 sin x
(c) (c1 + c2 x) cos x + (c3 + c4 x) sin x (d) none of these.
(Bundelkhand 2001; Kumaun 07)
2. P.I. of the differential equation ( D + D + 1) y = e x is
2
1 x
(a) e (b) 3e x
3
(c) e x (d) none of these.
(Avadh 2005; Garhwal 11)
1
3. For the differential equation F ( D) y = e ax , if F (a) = 0, then P.I. = e ax is
F ( D)
given by
1 1
(a) e ax (b) e ax
F (a) F (− a)
1
(c) e ax 1 (d) none of these
F ( D + a) (Rohilkhand 2005; Kumaun 11)
1
4. For the particular integral sin ax when f (− a2 ) = 0 , which one of the
f ( D2 )
following is correct
1 1 1 x
(a) sin ax = sin ax (b) sin ax = − cos ax
f ( D2 ) f (− a2 ) D2 + a2 2a
1 x
(c) 2 2
sin ax = cos ax (d) none of these
D +a 2a
(Bundelkhand 2001; Agra 08; Kumaun 12)
D-132
d2 y dy
5. The solution of the differential equation −3 + 2 y = e x is
dx2 dx
(a) y = c1e x + c2 e3 x + x (b) y = (c1 + c2 ) e x − xe x
(c) y = c1 e x + c2 e2 x − xe x (d) y = c1 x + c2 e x − e2 x .
6. The P.I. of the differential equation ( D2 + 1) y = cos x is

1 x
(a) sin x (b) − sin x
2 2
x x
(c) sin x (d) cos x
2 2 (Rohilkhand 2007)
2
d y
7. The solution of differential equation 2
+ a2 y = 0 is
dx
(a) y = c1e ax + c2 e − ax (b) y = c1 sin ax + c2 cos ax
(c) y = sin ax e ax (d) y = cos ax e ax
(Garhwal 2005, 08; Kumaun 10)
8. The solution of the differential equation ( D2 + 9) y = 0 is

(a) y = c1e3 x + c2 e −3 x (b) y = (c1 + c2 x) e3 x
(c) y = c1 cos 3 x + c2 sin 3 x (d) y = (c1 + c2 x)sin 3 x
(Garhwal 2006)
2 2
9. The general solution of ( D − m ) y = 0 is
(a) y = (c1 + c2 x) e mx (b) y = c1 sin mx + c2 cos mx
mx − mx
(c) y = c1e + c2 e (d) none of these (Garhwal 2007)
10. The P.I. of the differential equation ( D2 − 3 D + 2) y = e5 x is

1 5x 1 −5 x
(a) e (b) e
2 5
1 5x
(c) e (d) 0 (Garhwal 2010B)
12
11. C.F. of the differential equation ( D2 − 3 D + 2) y = e x is

(a) c1e − x + c2 e −2 x (b) c1e x + c2 e −2 x
(c) c1e − x + c2 e2 x (d) c1e x + c2 e2 x (Garhwal 2013)
d2 y
12. A part of the C.F. of the differential equation + a2 y = cos ax is
dx2
(a) c1 sin ax − c2 cos ax (b) − (c1 sin ax + c2 cos ax)
(c) c1 cos ax + c2 sin ax (d) − c1 sin−1 ax + c2 cos −1 ax
(Garhwal 2015)
D-133
d3 y d2 y dy
13. The solution of the differential equation 3
−2 −4 + 8 y = 0 is
dx dx2 dx
(a) y = c1e2 x + c2 e −2 x (b) y = c1e x + c2 e2 x
(c) y = c1 x + c2 e −2 x (d) y = (c1 + c2 x) e2 x + c3 e −2 x
14. The particular integral of the differential equation y ′′ − 5 y ′ + 4 y = 0 is
(a) 0 (b) y = 4 e x + e4 x
(c) y = c1 e 4 x + c2 e x (d) y = 2 x (Kumaun 2014)
ax
15. For the differential equation F ( D) y = e if F (a) ≠ 0 , then the value of the
1
particular integral e ax is
F ( D)
e ax e − ax
(a) − (b)
F (a) F (a)
e ax
(c) (d) none of these
F (a) (Kumaun 2013)
1
16. The value of particular integral sin ax of the differential equation
D2 + a2
( D2 + a2 ) y = sin ax is
x
∫
(a) x sin ax dx (b)
2 ∫
sin ax dx
x
2∫
(c) cos ax dx (d) none of these
(Kumaun 2013)
17. The general solution of the differential equation y ′′ − 2 y ′ + 5 y = 0 will be
(a) y = e x (c1 cos 2 x + c2 sin 3 x) (b) y = c1 e x + c2 e3 x
(c) y = e x (c1 cos 2 x + c2 sin 2 x) (d) y = c1 cos 2 x + c2 sin 3 x
(Kumaun 2014)
18. The particular integral of ( D2 + 4) y = cos 2 x is
x x
(a) cos 2 x (b) sin 2 x
4 4
1
(c) x cos 2 x (d) sin 2 x
4 (Kanpur 2016)
2
d y
19. The solution of differential equation + y = 0 satisfying the initial
dx2
condition y (0 ) = 1, y (π / 2) = 2 is
(a) y = 2 cos x + sin x (b) y = cos x + 2 sin x
(c) y = cos x + sin x (d) y = 2 cos x + 2 sin x
(Kanpur 2016)
d2 y dy
20. Particular integral of + + y = x2 is
dx2 dx
2
(a) x − 2 x (b) x2 + 2 x
(c) x2 + 2 (d) x2 − 2 (Kanpur 2016)
D-134
1
21. x n cos ax is equal to
f ( D)
1
(a) real part of e iax xn
f ( D + ia)
1
(b) imaginary part of e iax xn
f ( D + ia)
1
(c) e iax xn
f ( D + ia)
(d) none of these (Kanpur 2016)
2
22. The solution of ( D + 2 D + 2) y = 0 , y (0 ) = 0 , y ′ (0 ) = 1 is
(a) e x sin x (b) e − x cos x
−x
(c) e sin x (d) e x cos x (Kanpur 2016)
23. The roots of the auxiliary equation of the differential equation
d2 y dy
2
−6 + 9 y = 4 e3 t are
dt dt
(a) 3, − 3 (b) 3, 3
(c) −3, − 3 (d) none of these (Kanpur 2016)
24. Complete solution of the differential equation y ′′ + y = sec ax is
(a) y = c1 cos x + c2 sin x + cos x log cos x + x sin x
(b) y = c1 cos x + c2 sin x + x sin x
(c) y = c1 cos x − c2 sin x
x2
(d) y = c1 cos x + c2 sin x +
2 (Kanpur 2016)
2
d y
25. Particular integral of + 9 y = sin 3 x is
dx2
x x
(a) − cos 3 x (b) cos 3 x
6 6
x x
(c) sin 3 x (d) − sin 3 x
6 6 (Kanpur 2016)
3 2
26. Solution of ( D + 2 D + D) y = 0 is
(a) c1 e − x + c2 (b) (c1 + c2 x) e − x
(c) (c1 + c2 x + c3 x2 ) e − x (d) c1 + (c2 + c3 x) e − x
27. Complementary function of ( D4 + 2 D3 − 3 D) y = x3 is

(a) (c1 + c2 x) + c3 e x + c4 e − x (b) (c1 + c2 x) + c3 e x + c4 e − 3 x
(c) (c1 + c2 x + c3 x) e − x + c4 e − 3 x (d) (c1 + c2 x) e − 3 x + (c3 + c4 x) e − x
28. Complementary function of ( D2 − 4) y = 0 is

(a) (c1 + c2 x) e 2 x (b) (c1 + c2 x) e − 2 x
(c) c1 e2 x + c2 e − 2 x (d) c1 sin 2 x + c2 cos 2 x

D-135
29. Complementary function of ( D2 + 1) y = 0 is

(a) c1 + c2 x (b) (c1 + c2 x) e − x
(c) (c1 + c2 x) e x (d) c1 cos x + c2 sin x
30. The roots of the auxiliary equation of ( D4 + 2 D2 + 1) y = 0 are

(a) m = ±1, ± 1 (b) m = ±1, ± i
(c) m = ± i, ± i (d) m = 1, 1, i, i
31. The roots of the auxiliary equation of ( D4 − 1) y = 0 are
(a) m = 1, − 1, i, − i (b) m = ±1, ± 1
(c) m = ± i, ± i (d) m = 1, 1, ± i
32. The roots of the auxiliary equation of ( D2 − 2 D + 5) y = 0 are
(a) 1 + 2 i (b) 1 − 2 i
(c) 1 ± 2 i (d) 2 ± i
33. Complementary function of ( D + 2 a D + a4 ) y = cos ax is
4 2 2
(a) (c1 + c2 ) x cos ax + (c 3 + c4 ) x sin ax

(b) (c1 + c2 x)cos ax + (c 3 + c4 x)sin ax
(c) (c1 x + c2 x2 )cos ax + (c 3 x + c4 x2 )sin ax
(d) none of these

1. The general solution of the equation f ( D) y = 0 is called the …… of the
equation f ( D) y = Q .
d2 y
2. The solution of the differential equation + a2 y = 0 is …… .
dx2
3. The complementary function (C.F.) of the differential equation
d2 y dy
2
−5 + 4 y = 0 is …… .
dx dx
4. The particular integral (P.I.) of the differential equation
d3 y d2 y dy
3
−4 2
+5 − 2 = 0 is …… .
dx dx dx
5. The particular integral of the differential equation f ( D) y = Q will be …… .
6. The particular integral of the differential equation
d2 y
+ a2 y = sin ax is …… .
dx2 (Kumaun 2009)
1
7. For the differential equation F ( D) y = e − ax if F (− a) = 0 then P.I. = e − ax
F ( D)
is …… . (Kumaun 2009)
D-136
True or False
1. A linear differential equation is an equation in which the dependent variable
and its derivatives appear only in the first degree.
d4 y d3 y dy
2. 4
−5 3
−2 + 4 xy = 0 is a linear differential equation with constant
dx dx dx
coefficients. (Meerut 2003)
3. e − x (c1 cos √ 3 x + c2 sin √ 3 x) + c3 e2 x is the solution of the differential equation

d3 y
− 8y = 0 .
dx3
A nswers
1. (b) 2. (a) 3. (c) 4. (b) 5. (c)
6. (c) 7. (b) 8. (c) 9. (c) 10. (c)
11. (d) 12. (c) 13. (d) 14. (a) 15. (c)
16. (b) 17. (c) 18. (b) 19. (b) 20. (a)
21. (a) 22. (c) 23. (b) 24. (a) 25. (a)
26. (d) 27. (b) 28. (c) 29. (d) 30. (c)
31. (a) 32. (c) 33. (b)

1. C.F. 2. y = c1 cos ax + c2 sin ax 3. c1 e x + c2 e4 x
2 1 −x e − ax
4. x 5. Q 6. cos ax 7. ⋅1
5 F ( D) 2a F ( D − a)
True or False
1. T 2. F 3. T
¨
D-137
5
H omogeneous L inear
( C auchy- E uler E quations)
1 Homogeneous Linear Differential Equations

(Meerut 2009B)
efinition: A differential equation of the form

D xn
dn y
+ a1 x n−1
dn −1
+ … + an−1 x
dy
+ an y = Q, …(1)
n n−1 dx
dx dx
where a1, a2 , … , an are constants and Q is either a constant or a function of x is called a
homogeneous linear differential equation of order n.
A linear homogeneous differential equation with variable coefficients is called
Cauchy-Euler equation or Euler-Cauchy equation or simply Euler's equation.
2 Method of Solution
To solve such equations, we introduce a new independent variable z such that
x = e z or log x = z , so that 1/ x = dz / dx.
dy dy dz 1 dy
We have = ⋅ = ⋅
dx dz dx x dz
D-138
dy dy d d
∴ x = so that x ≡ ≡ D , say.
dx dz dx dz
d  n−1 d n−1 y  n
nd y n−1 d
n−1
y
Now x x


n−1 
= x n
+ (n − 1) x n−1
dx  dx  dx dx
dn y d d n−1 y
or xn =  x − n + 1 x n−1 n−1
dx n  dx  dx
d n−1 y
= ( D − n + 1) x n−1 ⋅ …(2)
dx n−1
Putting n = 2, 3, 4,… etc. in (2), we have
d2 y dy ∵ x d ≡ D
x2 = ( D − 1) x = ( D − 1) Dy
dx 2 dx  dx 
= D ( D − 1) y, because the operators can be interchanged

d3 y d2 y
x3 3
= ( D − 2) x2
dx dx2
= ( D − 2) ( D − 1) Dy
= D ( D − 1) ( D − 2) y .
Whence, generalizing, we have
dn y
xn = D ( D − 1) ( D − 2) … ( D − n + 1) y .
dx n
dy 2 d2 y dn y
Substituting these values of x ,x 2
, … , x n n in (1) and thus changing the
dx dx dx
independent variable from x to z, we have
[{ D ( D − 1) … ( D − n + 1)} + a1 { D ( D − 1) … ( D − n + 2)}]
+ … + an−1 D + an] y = Q ,
or f ( D) y = Q, …(3)
where Q has now become a function of z.
In the differential equation (3) the independent variable is z and the operator D stands
for d / dz . Obviously (3) is a linear differential equation with constant coefficients and
so it can be solved by the methods given in the previous chapter i. e., chapter 3. Thus as
in the case of the linear equations with constant coefficients, the general solution of
(3) is the sum of any particular integral of (3) and the complementary function i. e.,
the general solution of
f ( D) y = 0 . …(4)
To find the Complementary function (C.F.):
(i) If m1, m2 , …, mn are the roots of the auxiliary equation of (4), and no two of
them are equal, the general solution of (4) i. e., the C.F. of the solution of (3) is
easily seen to be
D-139
y = c1e m1z + c2 e m2 z + ... + c ne mnz ,
or y = c1 x m1 + c2 x m2 + … + c n x mn . [∵ e z = x]
(ii) In case there are r roots alike, each equal to m, and the rest all different, then the
C.F. = (c1 + c2 z + … + c r z r −1) e mz + c r +1e mr +1z + ... + c ne mn z
= [c1 + c2 log x + … + c r (log x)r −1] x m + c r +1 x mr +1 + ... + c n x mn .
(iii) In case the roots are imaginary, say of the form α ± iβ, then the
C. F. = e α x (c1 cos βz + c2 sin βz ) = x α [c1 cos (β log x) + c2 sin (β log x)] .
In this case we can also write the
C. F. = c1e α z cos (βz + c2 ) = c1 x α cos (β log x + c2 ) .
(iv) In case the roots α ± iβ occur r times, the C.F. corresponding to these roots will
be
e α z [(c1 + c2 z + … + c r z r −1)cos βz + (c1 ′ + c2 ′ z + … + c r ′ z r −1)sin βz ]
= x α [{ c1 + c2 log x + … + c r (log x)r −1} cos (β log x)
+ { c1 ′ + c2 ′ log x + … + c r ′ (log x)r −1} sin (β log x)].
To find the Particular integral. (P.I.):
The particular integral (P.I.) of (3) is given by
1
Q.
f ( D)
If α is a constant, we have
1 1
Q= { e α z (e − αz Q)}
D−α D−α
1
= eα z e−α z Q
( D + α) − α
1 − αz
= eα z e Q = e α z e − α z Q dz .
D ∫
Methods to find the P.I:
General Methods:
(i) We resolve the operator f ( D) into linear factors. Thus we write
f ( D) = ( D − m1) ( D − m2 ) … ( D − mn) . Then the
1 1
P. I. = Q= Q
f ( D) ( D − m1) ( D − m2 ) .. ( D − mn−1)( D − mn)
1
e mn z − mn z
=
( D − m1) ( D − m2 ) … ( D − mn−1) ∫e Q dz ,
1
by operating upon Q as explained above.
D − mn
Similarly we operate with other remaining factors in succession and thus we find the
P.I.
D-140
(ii) First we resolve f ( D) into linear factors as in (i) and then we break up
{ f ( D)} −1 into partial fractions. Then the
1  A1 A2 An 
P. I. = Q= + + ... + Q
f ( D)  D − m1 D − m2 D − mn 
= A1e m1z − m1z

Q dz + … + An e mn z − mn z
∫e ∫e Q dz .
Special short Methods.

(i) When Q is of the form e a z , then
1 1 az
P. I. = ea z = e , provided f (a) ≠ 0 .
f ( D) f (a)
(ii) When Q is of the form cos az or sin az, then the P.I. is given by
1 1
cos az = cos az ,
f ( D2 ) f (− a2 )
1 1
and sin az = sin az , [Provided f (− a2 ) ≠ 0]
f ( D2 ) f (− a2 )
1
(iii) If Q is of the form z m , we have P.I. = z m.
f ( D)
We expand { f ( D)} −1 in ascending powers of D retaining terms as far as Dm and
operate each term on z m .
(iv) If Q is of the form e a zV , where V is any function of z, we have the
1 1
P.I. = e a zV = e a z V.
f ( D) f ( D + a)
(v) If Q is of the form zV, where V is any function of z, we have the
1 1  d 1 
P.I. = (zV ) = z V +  V.
f ( D) f ( D)  dD f ( D)
Example 1: Solve x2 D2 y + 5 x Dy + 4 y = 0 , where D ≡ d / dx. (Kumaun 2008)

d2 y dy
x2 2
+ 5x +4y =0. …(1)
dx dx
Now (1) is a homogeneous linear differential equation of order 2. So putting x = e z and
denoting d / dz by D′, we have
dy d2 y
x = D′ y, x2 = D′ ( D′ − 1) y .
dx dx2
D-141
∴ the differential equation (1) transforms to

{ D′ ( D′ − 1) + 5 D′ + 4} y = 0
or ( D′2 + 4 D′ + 4) y = 0
or ( D′ + 2)2 y = 0 ,
which is a linear differential equation with constant coefficients, the independent

variable being z.
∴ auxiliary equation is (m + 2)2 = 0 , giving m = − 2, − 2 .
∴ the complete solution is

y = (c1 + c2 z ) e −2 z
or y = (c1 + c2 z ) (e z )−2
or y = (c1 + c2 log x) x −2 .
Example 2: Solve ( x3 D3 + 3 x2 D2 − 2 xD + 2) y = 0 , where D ≡ d / dx.
Solution: Putting x = e z and denoting d / dz by D ′, we have
dy d2 y
x Dy = x = D ′ y, x2 D2 y = x2 = D ′ ( D ′ − 1) y
dx dx2
d3 y
and x3 D3 y = x3 = D ′ ( D ′ − 1) ( D ′ − 2) y.
dx3
∴ the given differential equation transforms to
[ D ′ ( D ′ − 1) ( D ′ − 2) + 3 D ′ ( D ′ − 1) − 2 D ′ + 2] y = 0
or [ D ′ ( D ′ − 1) ( D ′ − 2) + 3 D ′ ( D ′ − 1) − 2 ( D ′ − 1)] y = 0
or ( D ′ − 1) [ D ′ ( D ′ − 2) + 3 D ′ − 2] y = 0
or ( D ′ − 1) ( D ′2 + D ′ − 2) y = 0
or ( D ′ − 1) ( D ′ + 2) ( D ′ − 1) y = 0
or ( D ′ − 1)2 ( D ′ + 2) y = 0 .
∴ the auxiliary equation is (m − 1)2 (m + 2) = 0 , giving

m = 1, 1, − 2 .
∴ the general solution is
y = (c1 + c2 z ) e z + c3 e −2 z = (c1 + c2 z ) e z + c3 (e z )−2
or y = (c1 + c2 log x) x + c3 x −2 , [∵ x = e z and z = log x]
or y = (c1 + c2 log x) x + c3 ⋅ (1/ x2 ).
d3 y d2 y dy
Example 3: Solve x4 3
+ 2 x3 2
− x2 + xy = 1 .
dx dx dx (Garhwal 2011)
D-142
Solution: Dividing out by x, the given equation may be written as

d3 y d2 y dy 1
x3 3
+ 2 x2 2
−x + y= , …(1)
dx dx dx x
which is a linear homogeneous equation of order 3.
Putting x = e z in (1) and denoting d / dz by D, we have
[ D ( D − 1) ( D − 2) + 2 D ( D − 1) − D + 1] y = e − z
or ( D − 1)2 ( D + 1) y = e − z .
∴ auxiliary equation is
(m − 1)2 (m + 1) = 0 , giving m = 1, 1, − 1 .
∴ C. F. = (c1 + c2 z ) e z + c3 e − z
= (c1 + c2 log x) x + c3 x −1. [∵ e z = x and z = log x]
1
And P. I. = 2
e−z
( D − 1) ( D + 1)
1  1 
=  2
e−z 
( D + 1) ( D − 1) 
1 1
= ⋅ e−z
( D + 1) 4
1 1
= e−z ⋅ 1
4 D +1
1 −z 1
= e 1
4 D − 1+ 1
1 −z 1
= e 1
4 D
1 1
= ze − z = log x.
4 4x
c3 1
y = (c1 + c2 log x) x + + log x .
x 4x
d3 y 2
2 d y dy
+ 2 x 2
+ 3x − 3 y = x2 + x.
dx dx dx
(Meerut 2001, 07; Gorakhpur 05, 08; Kanpur 15)
z
Solution: Putting x = e and denoting d / dz by D, the given differential equation
becomes
[ D ( D − 1) ( D − 2) + 2 D ( D − 1) + 3 D − 3] y = e2 z + e z
or [ D ( D − 1) ( D − 2) + 2 D ( D − 1) + 3 ( D − 1)] y = e2 z + e z
or ( D − 1) { D ( D − 2) + 2 D + 3} y = e2 z + e z
D-143
or ( D − 1) ( D2 + 3) y = e2 z + e z .
Hence the auxiliary equation is

(m − 1) (m2 + 3) = 0 , giving m = 1, 0 ± √ 3 i.
∴ C. F. = c1e z + e0 z (c2 cos √ 3 z + c3 sin √ 3 z )
= c1e z + c2 cos √ 3 z + c3 sin √ 3 z .
1 1
Also P. I. = e2 z + ez
( D − 1) ( D2 + 3) ( D − 1) ( D2 + 3)
e2 z 1
= + ez ⋅ 1
(2 − 1) ( D2 + 3) (12 + 3) ( D − 1)
1 2z 1 z 1
= e + e 1
7 4 D + 1− 1
1 2z 1 z 1
= e + e 1
7 4 D
1 1
= e2 z + e z . z .
7 4
∴ the complete solution is
1 2z 1 z
y = c1 e z + c2 cos √ 3 z + c3 sin √ 3 z + e + ze
7 4
1 1
or y = c1 x + c2 cos (√ 3 log x) + c3 sin (√ 3 log x) + x2 + x log x .
7 4
[∵ e z = x and z = log x]
d2 y dy
−x − 3 y = x2 log x .
dx dx
(Meerut 2004B, 10; Gorakhpur 07; Kanpur 07, 14; Lucknow 09, 11;
Avadh 08; Bundelkhand 04; Garhwal 07, 10; Kumaun 09)
Solution: Putting x = e z or z = log x and denoting d / dz by D the equation becomes
[ D ( D − 1) − D − 3] y = e2 z . z
or ( D2 − 2 D − 3) y = e2 z . z .
∴ auxiliary equation is
(m2 − 2 m − 3) = 0
or (m − 3) (m + 1) = 0 i. e., m = − 1, 3 .
∴ C.F. = c1e3 z + c2 e − z .
1
And P. I. = 2
e2 z . z
( D − 2 D − 3)
1
= e2 z . 2
z
[( D + 2) − 2( D + 2) − 3]
D-144
1
= e2 z 2
z
( D + 2 D − 3)
1
= e2 z z
− 3 [1 − 2
3
D− 1
3
D2 ]
−1
1  2 1 
= e2 z − 1 −  D + D2   z
3   3 3 
1 2z  2 1
=− e 1 + D + D2 + … z
3  3 3 
1 2z 2
=− e (z + ).
3 3
∴ complete solution is
1 2z 2
y = c1 e3 z + c2 e − z − e (z + )
3 3
1 2 2
or y = c1 x3 + c2 x −1 − x (log x + ).
3 3

1. x2 (d2 y / dx2 ) − 4 x (dy / dx) + 6 y = x . (Garhwal 2005; Bundelkhand 10)
2 2 2 4
2. x (d y / dx ) − 4 x (dy / dx) + 6 y = x . (Garhwal 2006; Gorakhpur 09; Kumaun 11)
3. x2 (d2 y / dx2 ) − 2 x (dy / dx) + 2 y = 1/ x .
4. x2 (d2 y / dx2 ) + x (dy / dx) − 4 y = x2 . (Rohilkhand 2009)

2 2 2 2
5. x (d y / dx ) + 2 x (dy / dx) − 20 y = ( x + 1) .
6. x2 (d2 y / dx2 ) + 4 x (dy / dx) + 2 y = e x .

(Meerut 2005; Avadh 08, 11; Rohilkhand 07, 11)
2 2 2
7. x (d y / dx ) + 7 x (dy / dx) + 13 y = log x. (Bundelkhand 2006)
8. (i) x2 (d2 y / dx2 ) − x (dy / dx) + 2 y = x log x.
(Rohilkhand 2010; Gorakhpur 06, 10, 11; Purvanchal 09;

Avadh 06; Bundelkhand 13; Garhwal 13)
3 3 3 2 2 2
(ii) x (d y / dx ) + 3 x (d y / dx ) + x (dy / dx) + y = x log x.
(Kanpur 2009; Rohilkhand 08; Purvanchal 08)
3 3 3 2 2 2
9. x (d y / dx ) + 3 x (d y / dx ) + x (dy / dx) + y = x + log x. (Avadh 2006, 14)
10. x2 (d2 y / dx2 ) − x (dy / dx) + 4 y = cos (log x) + x sin (log x). (Garhwal 2009)
D-145
11. x3 (d3 y / dx3 ) + 2 x2 (d2 y / dx2 ) + 2 y = 10 [ x + (1/ x)].

(Meerut 2007B; Bundelkhand 09; Agra 2006; Lucknow 06; Garhwal 2010B)
2
d y dy log x sin (log x) + 1
12. x2 2
− 3x
+ y= ⋅
dx dx x
1
13. ( x2 D2 + 3 xD + 1) y = ⋅
(1 − x)2 (Rohilkhand 2008; Purvanchal 08; Bundelkhand 01)
2
d2 y  dy   dy 
14. 2 x2 y + 4 y2 = x2   + 2 xy   ⋅
dx2  dx   dx 
A nswers 1
1
1. y = c1 x2 + c2 x3 + x
2
1
2. y = c1 x2 + c2 x3 + x4
2
1
3. y = c1 x + c2 x2 + ⋅ (1/ x)
6
1 2
4. y = c1 x2 + (c2 / x2 ) + x log x
4
1 2 1 1
5. y = c1 x4 + c2 x −5 − x − x−
14 9 20
6. y = (c1 / x) + (c2 / x2 ) + (e x / x2 )
1 6
7. y = c1 {cos (log x2 + c2 )} / x3 + log x −
13 169
8. (i) y = c1 x cos (log x + c2 ) + x log x
1 3
(ii) y = c1 x −1 + c2 √ x cos {(√ 3 / 2) log x + c3 } + x log x − x
2 4
1
9. y = c1 x − 1 + c2 √ x cos {(√ 3 / 2) log x + c3 } + x + log x
2
10. y = x [c1 cos (√ 3 log x) + c2 sin (√ 3 log x)]
1 1
+ [3 cos (log x) − 2 sin (log x)] + x sin (log x)
13 2
11. y = c1 x −1 + c2 x cos (log x + c3 ) + 5 x + 2 x −1 log x
1
12. y = x2 (c1 x √3 + c2 x −√3 ) +
6x
1 (5 sin log x + 6 cos log x) 382 cos log x + 54 sin log x 
+ log x ⋅ +
x  61 3721 
D-146
1 1 x
13. y = (c1 + c2 log x) + log
x x 1− x
14. y = x2 (c1 + c2 log x)2
3 Equations Reducible to Homogeneous Form

dn y dn −1 y dy
(a + bx)n n
+ P1 (a + bx)n − 1 n −1
+ … + Pn − 1 (a + bx) + Pn y = Q,
dx dx dx
where P1, P2 , …, Pn are constants and Q is a function of x, can be reduced to the
homogeneous linear form by putting a + bx = t and then can be solved by the method
explained in article 2. We can also solve this differential equation directly by making
the substitution.
e z = a + bx
or z = log (a + bx).
If we denote the operator d / dz by D, we can easily see that
dy
(a + bx) = b Dy,
dx
d2 y
(a + bx)2 2
= b2 D (d − 1) y,and so on.
dx
d2 y dy
Example 6: Solve ( x + a)2 2
− 4 ( x + a) +6y = x.
dx dx (Meerut 2001, 06, 11; Kanpur 11)
Solution: Putting ( x + a) = e z and denoting d / dz by D, the given equation becomes
[ D ( D − 1) − 4 D + 6] y = e z − a
or ( D − 3) ( D − 2) y = e z − a .
∴ auxiliary equation is (m − 3) (m − 2) = 0 , giving m = 3, 2 .
∴ C. F. = c1e3 z + c2 e2 z = c1 (e z )3 + c2 (e z )2
= c1 ( x + a)3 + c2 ( x + a)2
1 1
and P. I. = ez − ae0 z , [∵ e0 z = 1]
( D − 3) ( D − 2) ( D − 3) ( D − 2)
1 1
= ez − a e0 z ,
( D − 3) ( D − 2) ( D − 3) ( D − 2)
D-147
1 1
= ez − a e0 z
(1 − 3) (1 − 2) (0 − 3) (0 − 2)
1 z 1
= e − a
2 6
1 1
= ( x + a) − a .
2 6
Hence the complete solution is y = C. F. + P. I.
1 1
or y = c1 ( x + a)3 + c2 ( x + a)2 + ( x + a) − a .
2 6

d2 y dy
1. (3 x + 2)2 + 3 (3 x + 2) − 36 y = 3 x2 + 4 x + 1.
dx2 dx
(Bundelkhand 2009; Agra 06; Lucknow 06)
2
d y dy
2. (1 + x)2 2
+ (1 + x) + y = 4 cos log (1 + x).
dx dx
d2 y dy
3. (1 + 2 x)2 2
= 6 (1 + 2 x) + 16 y = 8 (1 + 2 x)2 .
dx dx
d2 y dy
4. (3 x + 2)2 − (3 x + 2) − 12 y = 6 x.
dx2 dx
d2 y dy
5. ( x + 1)2 2
+ ( x + 1) = (2 x + 3)(2 x + 4).
dx dx
A nswers 2
1
1. y = c1 (3 x + 2)2 + c2 (3 x + 2)−2 + [(3 x + 2)2 log (3 x + 2) + 1]
108
2. y = c1 cos log (1 + x) + c2 sin log (1 + x) + 2 log (1 + x) sin log (1 + x)
3. y = (1 + 2 x)2 log(1 + 2 x)[1 + log(1 + 2 x)]

2 1
4. y = c1(3 x + 2)2 + c2 (3 x + 2)−2 /3 − (3 x + 2) +
15 3
5. y = c1 + c2 log( x + 1) + ( x + 1)2 + 6 ( x + 1) + [log ( x + 1)]2
D-148

(b), (c) and (d).
1. The complementary function (C.F.) of the differential equation
d2 y dy
x2 2
−x − 3 y = x2 log x is
dx dx
(a) c1 cos x + c2 sin x (b) c1 e3 z + c2 e − z
(c) c1 e z + c2 e −3 z (d) none of these
d2 y dy
2. C.F. of the differential equation x2 + 4x + 2 y = e x is
dx2 dx
(a) c1 + c2 x (b) c1 x −1 + c2 x −2
(c) c1 e − x + c2 e x (d) none of these
d3 y d2 y dy
3. C.F. of the differential equation x4 3
+ 2 x3 2
− x2 + xy = 1 is
dx dx dx
x −x
(a) (c1 + c2 x) e + c3 e (b) c1e x + c2 e − x + c3 e2 x
c3 1
(c) (c1 + c2 log x) x + + log x (d) none of these
x 4x
d2 y dy
4. C.F. of the differential equation ( x + a)2 − 4 ( x + a) + 6 y = x is
dx2 dx
(a) c1 x2 + c2 x3 (b) c1 ( x + a)3 + c2 ( x + a)2
(c) c1 e3 x + c2 e2 x (d) none of these
5. To solve homogeneous linear differential equation, we put :

1
(a) x = X + h, y = Y + k (b) y =
x
(c) z = log x (d) none of these (Garhwal 2008)
6. The substitution z = log x transforms the differential equation into the following :
d2 y dy
x2 2
− 3x +4y =0
dx dx
d2 y dy d2 y
dy
(a) −4 +4y =0 (b) 4 +4y =0
−
dz 2 dz dz 2
dz
d2 y dy d2 y dy
(c) −3 +4y =0 (d) z 2 2 − 3 z +4y =0
dz 2 dz dz dz
(Garhwal 2009)
D-149
d2 y dy
7. The general solution of the differential equation x2 −x + y =0
dx2 dx
(a) Ax + Bx2 (b) Ax + B log x
(c) Ax + Bx2 log x (d) Ax + Bx log x
where A and B are constants. (Kumaun 2015)
d2 y dy
8. On putting x = e z , the transformal differential equation of x2 2
+x + y=x
dx dx
is
d2 y d2 y
(a) 2
+ y = ez (b) 2
− y = ez
dz dx
2
d y d2 y
(c) + y = ex (d) − y = e−z
dx2 dx2 (Kumaun 2016)
9. The substitutions in solving the homogeneous linear equation are

dy d2 y d
(a) z = e x , x = Dy, x2 2 = D ( D − 1) y, where D ≡
dx dx dz
dy d2 y d
(b) x = e − z , x = Dy, x2 2 = D ( D − 1) y, where D ≡
dx dx dz
dy d2 y d
(c) x = e z , x = Dy, x2 2 = D ( D − 1) y, where D ≡
dx dx dz
(d) none of these
10. The auxiliary roots of the equation ( x3 D3 + 2 x2 D2 + 3 xD − 3) y = x2 + x are

(a) m = −1, ± 3 i (b) m = ±1, 3 i
(c) m = 1, ± 3 i (d) m = ±1, − 3 i

d2 y dy
1. Homogeneous linear differential equation x2 + 5x +4y =0
dx2 dx
will reduce to a linear differential equation with constant coefficients by putting
x = ……
2. To reduce the homogeneous linear differential equation

d3 y d2 y dy
( x + a)3 3
− 2 ( x + a)2
− 5 ( x + a) + 3 y = 2x
dx dx2 dx
to a linear differential equation with constant coefficients, we have to put ……
D-150
True or False
d2 y dy
1. x2 −x − 3 y = x2 log x is a linear homogeneous differential equation of
dx2 dx
order two. (Meerut 2003)
d3 y d2 y dy
2. x4 3
+ 2 x3 2
− x2 + xy = 1 is not a linear homogeneous differential
dx dx dx
equation of order 3.
d3 y d2 y dy
3. x3 3
+ 4 x2 2
+ 2x + xy = x2 is a linear homogeneous differential
dx dx dx
equation of order 3.
A nswers
1. (b) 2. (b) 3. (c) 4. (b) 5. (c)
6. (a) 7. (d) 8. (a) 9. (c) 10. (c)

1. ez 2. ( x + a) = e z
True or False
1. T 2. F 3. F
¨
D-151
6
O rdinary S imultaneous
1 Introduction
n the present chapter, we shall discuss differential equations containing one
I independent variable and two or more than two dependent variables. To
completely solve such equations we need as many simultaneous equations as there are
dependent variables.
2 Methods of Solving Simultaneous Linear Differential

Equations with Constant Coefficients
Let x and y be the two dependent variables and t be the independent variable. Thus the
equations will contain differential coefficients of x, y with respect to t. Let D ≡ d / dt.
Then such equations can be put in the form
f1( D) x + f2 ( D) y = T1 …(1)
and φ1( D) x + φ2 ( D) y = T2 , …(2)
D-152
where T1 and T2 are some functions of the independent variable t. Here

f1( D), f2 ( D), φ1( D) and φ 2( D) are all rational integral functions of D with constant
coefficients.
There are two methods to solve such equations.
First method: Method of Elimination (Symbolic method):
To eliminate y between (1) and (2), operating both sides of (1) by φ 2( D) and of (2) by
by f2 ( D), we get
φ 2 ( D) f1 ( D) x + φ 2 ( D) f2 ( D) y = φ 2 ( D) T1 …(3)
and f2 ( D) φ 1 ( D) x + f2 ( D) φ2 ( D) y = f2 ( D) T2 . …(4)
Subtracting (4) from (3), we get
[φ2 ( D) f1 ( D) − f2 ( D) φ1 ( D)] x = φ2 ( D) T1 − f2 ( D) T2 ,
which is of the form F ( D) x = T . …(5)
The equation (5) is a linear differential equation with constant coefficients in x and t.
Solving it we can find the value of x in terms of t.
Putting this value of x in either (1) or (2), we get the value of y.
Note 1: The equations can also be solved by first finding the value of y by eliminating x
between (1) and (2) and then solving the resulting linear differential equation in y and
t. The value of x can now be obtained from (1) or (2) after putting the value of y.
Note 2: Since f2 ( D) and φ2 ( D) are functions of D with constant coefficients, so
f2 ( D) φ2 ( D) = φ2 ( D) f2 ( D) .
Second method: Method of differentiation:
Sometimes x or y can be conveniently eliminated if we differentiate the equations (1)
and (2). For example suppose the two given equations (1) and (2) connect four
quantities x, y, dx / dt and dy / dt. Differentiating (1) and (2) w.r.t. t, we get in all four
dx dy d2 x d2 y dy
equations containing x, y, , , , ⋅ Eliminating three quantities y, and
dt dt dt2 dt2 dt
d2 y
from these four equations, we get an equation of the second order with x as the
dt2
dependent and t as the independent variable. Solving this equation we find the value
of x in terms of t. Then the value of the other variable can be found.
3 Number of Arbitrary Constants

In the general solution of (1) and (2) the number of arbitrary constants is equal to the
degree of D in the determinant
f1( D) f2 ( D)
∆= , if ∆ ≠ 0 .
φ 1( D) φ 2( D)
If ∆ vanishes then the system is dependent. Here we shall not consider such cases.
D-153
Example 1: Solve the simultaneous equations

dx dy
− 7x + y = 0 , − 2x − 5 y = 0 .
dt dt
(Purvanchal 2008; Rohilkhand 09; Lucknow 10; Kumaun 11; Kanpur 15)
Solution: Writing D for d / dt, the given equations are

( D − 7) x + y = 0 …(1)
and − 2 x + ( D − 5) y = 0 . …(2)
Let us eliminate y between (1) and (2). Operating on both sides of (1) by D − 5 and
then subtracting (2) from it, we get
{( D − 5) ( D − 7) + 2} x = 0
or ( D2 − 12 D + 37) x = 0 , …(3)
which is a linear differential equation of second order in x and t with constant

coefficients.
A.E. is m2 − 12 m + 37 = 0 ⇒ m = 6 ± i .
∴ the solution of (3) is x = e6 t (c1 cos t + c2 sin t) …(4)
dx
∴ = 6 e6 t (c1 cos t + c2 sin t) + e6 t (− c1 sin t + c2 cos t) .
dt
Putting the values of x and dx / dt in (1), we get
y = 7 x − Dx = 7 x − (dx / dt)
= 7 e6 t (c1 cos t + c2 sin t) − 6 e6 t (c1 cos t + c2 sin t)
− e6 t (− c1 sin t + c2 cos t)
or y = e6 t [(c1 − c2 ) cos t + (c1 + c2 ) sin t] . …(5)
The required general solution of the given equations consists of the equations (4) and
(5).
d2 y d2 y
− 3x − 4 y = 0 , + x + y =0.
dt2 dt2 (Kanpur 2003; Meerut 09)
Solution: Writing D for (d / dt), the given equations are

( D2 − 3) x − 4 y = 0 …(1)
and x + ( D2 + 1) y = 0 . …(2)
Let us eliminate y from (1) and (2). Multiplying both sides of (2) by 4 and operating
on both sides of (1) by D2 + 1 and adding, we get
{( D2 + 1) ( D2 − 3) + 4} x = 0
or ( D4 − 2 D2 + 1) x = 0 . …(3)
D-154
A.E. is m4 − 2 m2 + 1 = 0 or (m2 − 1)2 = 0 .

∴ m = ± 1, ± 1.
∴ the solution of (3) is x = (c1 + c2 t) e t + (c3 + c4 t) e − t …(4)
dx
∴ = (c1 + c2 t) e t − (c3 + c4 t) e − t + c2 e t + c4 e − t ,
dt
d2 x
and = (c1 + c2 t) e t + (c3 + c4 t) e − t + 2 c2 e t − 2 c4 e − t .
dt2
Putting these values in (1), we get
4 y = D2 x − 3 x = (d2 x / dt2 ) − 3 x
= (c1 + c2 t) e t + (c3 + c4 t) e − t
+ 2 c2 e t − 2 c4 e − t − 3 (c1 + c2 t) e t − 3 (c3 + c4 t) e − t
= − 2 (c1 + c2 t) e t − 2 (c3 + c4 t) e − t + 2 c2 e t − 2 c4 e − t
1 t 1
or y= e (c2 − c1 − c2 t) − e −1 (c4 + c3 + c4 t). …(5)
2 2
The required general solution of the given equations consists of the equations (4) and
(5).
d2 x d2 y
+ 4 x + y = te3 t , + y − 2 x = cos2 t .
dt2 dt2
( D2 + 4) x + y = te3 t …(1)
2 2
and − 2 x + ( D + 1) y = cos t …(2)
Eliminating y from (1) and (2), we get
{( D2 + 1) ( D2 + 4) + 2} x = ( D2 + 1)(te3 t ) − cos2 t
or ( D4 + 5 D2 + 6) x = 10 te3 t + 6 e3 t − cos2 t. …(3)
4 2 2 2
A.E. is m + 5m + 6 = 0 or (m + 3) (m + 2) = 0 .
∴ m = ± i √ 3, ± i √ 2 .
∴ C.F. = (c1 cos √ 3 t + c2 sin √ 3 t) + (c3 cos √ 2 t + c4 sin √ 2 t) .
10 6 1
P. I. = 4 2
te3 t + 4 2
e3 t − 4 2
cos2 t
D + 5D + 6 D + 5D + 6 D + 5D + 6
1 1
= 10 e3 t 4 2
t+6 ⋅ e3 t
( D + 3) + 5 ( D + 3) + 6 3 + 5 . 32 + 6
4
1 1
− 4
(1 + cos 2 t)
2
D + 5D + 6 2
1 1 3t
= 10 e3 t t+ e
132 + 138 D + 52 D2 + … 22
1
1 1  1 cos 2 t
− 2
− 4 42  
6 + 5D + D 2 D + 5 D + 6 2 
D-155
−1
1  23 1 3t
= 10 e3 t ⋅ 1 + D + … t + e
132  22  22
1 1 1 cos 2 t
− ⋅ e0 t −
2 6 + 5 D2 + D4 2 (− 4)2 + 5 (− 4) + 6
5 e3 t  23  e3 t 1 1
= t −  + − − cos 2 t .
66  22  22 12 4
∴ x = (c1 cos √ 3 t + c2 sin √ 3 t) + (c3 cos √ 2 t + c4 sin √ 2 t)
5 3t 49 3 t 1 1
+ te − e − cos 2 t − ⋅ …(4)
66 1452 4 12
dx d2 x
Now from (4), we find and 2 .
dt dt
Putting the values of x and d2 x / dt2 in (1), we get
d2 x
y=− 2
− 4 x + te3 t
dt
or y = − (c1 cos √ 3 t + c2 sin √ 3 t) − 2 (c3 cos √ 2 t + c4 sin √ 2 t)
1 33 3 t 1
+ t . e3 t − e + ⋅ …(5)
66 2452 3
The required solution consists of the equations (4) and (5).
Example 4: Solve the simultaneous differential equations

dy dz
x + z =0 …(1) x + y =0. …(2)
dx dx
(Meerut 2001, 09B; Kumaun13)
Solution: Differentiating (1) with respect to x, we get

d2 ydy dz 2
2 d y dy dz
x 2
+
+ = 0 or x 2
+x +x =0.
dx dx dx dx dx dx
dz
Putting the value of x from (2), we get
dx
d2 y dy
x2 2
+x − y =0, …(3)
dx dx
which is a linear homogeneous equation.
To solve (3), putting x = e t and denoting d / dt by D, we have
dy d2 y
x = Dy, x2 = D ( D − 1) y .
dx dx2
∴ the equation (3) becomes
{ D ( D − 1) + D − 1} y = 0 or ( D2 − 1) y = 0 .
∴ y = c1e t + c2 e − t = c1 x + c2 x −1. …(4)
dy c dy c
∴ = c1 − 22 ⇒ x = c1 x − 2 ⋅
dx x dx x
D-156
dy
∴ from (1), we get z = − x = − c1 x + c2 x −1. …(5)
dx
Example 5: Solve the simultaneous differential equations

d2 x d2 y
+ m2 y = 0 , − m2 x = 0 .
dt2 dt2
D2 x + m2 y = 0 …(1)
2 2
and D y − m x =0 . …(2)
( D4 + m4 ) x = 0 .
A.E. is M 4 + m4 = 0 ⇒ ( M 2 + m2 )2 − 2 M 2 m2 = 0 .
or ( M 2 − √ 2 Mm + m2 ) ( M 2 + √ 2 Mm + m2 ) = 0 .
∴ M 2 − √ 2 Mm + m2 = 0 , or M 2 + √ 2 Mm + m2 = 0 .
Solving these, we get
m m m m
M= ±i and M = − ±i ⋅
√2 √2 √2 √2
m m
∴ x = c1e(m / √2)t cos  t + c2  + c3 e −(m / √2)t cos  t + c4  …(3)
√2  √2 
dx m   m t + c  − sin  m t + c  
∴ = c1e(m / √2)t . cos  2  2 
dt √2  √2  √2 
m   m t + c  + sin  m t + c  
− c3 e −(m / √2)t . cos  4  4 
√2  √2  √2 
d2 x m2   m t + c  − 2 sin  m t + c 
and = c1e(m / √2)t . cos  2  2
dt2 2  √2  √2 
m  m2   m t+c 
− cos  t + c2   + c3 e −(m / √2)t . cos  4
√2  2  √2 
m m 
+ 2 sin  t + c4  − cos  t + c4  
√2  √2 
 m m 
= m2  c3 e −(m / √2)t sin  t + c4  − c1e(m / √2)t sin  t + c2   .
 √2  √2 
1 d2 x
From the equation (1), we get y = − ⋅
m2 dt2
Putting the value of d2 x / dt2 , we have
m m
y = c1e(m / √2)t sin  t + c2  − c3 e −(m / √2)t .sin  t + c4  . …(4)
√2  √2 
The equations (3) and (4) constitute the required general solution of the given
equations.
D-157
Example 6: Solve t dx = (t − 2 x) dt, t dy = (tx + ty + 2 x − t) dt . (Rohilkhand 2001)
Solution: The given equations are

t dx = (t − 2 x) dt …(1)
and t dy = (tx + ty + 2 x − t) dt . …(2)
dx 2
From (1), we have + x = 1, which is linear.
dt t
I.F. = e ∫ (2 / t)dt = e2 log t = t2 .
1 3
xt2 = 2
∴
∫t ⋅ 1 dt + c1 =
3
t + c1
1
or t + c1 t −2 .
x= …(3)
3
Now adding (1) and (2), we get
dx + dy
t (dx + dy) = t ( x + y) dt or = dt .
x+ y
Integrating, log ( x + y) = t + log c2 .
∴ x + y = c2 e t or y = c2 e t − x . …(4)
Putting in (4) the value of x found in (3), we get
1
y = c2 e t − t − c1t −2 . …(5)
3
The equations (3) and (5) give the required general solution of the given equations.

d2 x dx d2 y dy
t2 +t + 2 y = 0 , t2 2 + t − 2x = 0 .
dt2 dt dt dt
Solution: Here both the given equations are linear homogeneous equations. First we
shall change them into linear differential equations with constant coefficients.
For this, we put t = e z .
Then denoting d / dz by D, we have
dx d2 x dy d2 y
t = Dx, t2 2 = D ( D − 1) x, t = Dy, t2 2 = D ( D − 1) y .
dt dt dt dt
∴ the given equations transform to
{ D ( D − 1) + D} x + 2 y = 0
i. e., D2 x + 2 y = 0 …(1)
and { D ( D − 1) + D} y − 2 x = 0
i. e., D2 y − 2 x = 0 . …(2)
( D4 + 4) x = 0 .
A.E. is m4 + 4 = 0 , or (m2 + 2)2 − 4 m2 = 0
or (m2 − 2 m + 2) (m2 + 2 m + 2) = 0 .
D-158
∴ m2 − 2 m + 2 = 0 or m2 + 2 m + 2 = 0 .
2 ± √ (4 − 8) − 2 ± √ (4 − 8)
∴ m= and m =
2 2
i. e., m = 1 ± i, − 1 ± i .
∴ x = e z (c1 cos z + c2 sin z ) + e − z (c3 cos z + c4 sin z ) …(3)
dx
∴ = e z (c1 cos z + c2 sin z ) + e z (− c1 sin z + c2 cos z )
dz
− e − z (c3 cos z + c4 sin z ) + e − z (− c3 sin z + c4 cos z )
d2 x
and = e z (c1 cos z + c2 sin z ) + 2 e z (− c1 sin z + c2 cos z )
dz 2
− e z (c1 cos z + c2 sin z ) + e − z (c3 cos z + c4 sin z )
− 2 e − z (− c3 sin z + c4 cos z ) − e − z (c3 cos z + c4 sin z )
= 2 e z (− c1 sin z + c2 cos z ) − 2 e − z (− c3 sin z + c4 cos z ) .
From (1), we have
1 2 1 d2 x
y=− D x=− ⋅
2 2 dz 2
∴ y = e z (c1 sin z − c2 cos z ) + e − z (− c3 sin z + c4 cos z ) . …(4)
Now changing the variable z into t in the equations (3) and (4) by the relation t = e z or
z = log t, we get
x = t (c1 cos log t + c2 sin log t) + t −1 (c3 cos log t + c4 sin log t)
and y = t (c1 sin log t − c2 cos log t) + t −1 (− c3 sin log t + c4 cos log t) .
These equations give the required general solution of the given equations.
dx dy dz
Example 8: Solve = ny − mz , = lz − nx , = mx − ly.
dt dt dt (Kanpur 2012)
Solution: Multiplying the given equations by x, y, z respectively and then adding, we

get
dx dy dz
x + y +z =0.
dt dt dt
Integrating, we get
1 2 1 2 1 2
x + y + z =k
2 2 2
or x2 + y2 + z 2 = c1. …(1)
Now multiplying the given equations by l, m, n respectively and then adding, we get
dx dy dz
l +m +n =0 .
dt dt dt
Integrating, we get lx + my + nz = c2 . …(2)
The equations (1) and (2) constitute the complete solution of the given differential
equations.
D-159

dx dy dx dy
3 +2 − 4 x + 3 y = 8 e −3 t , 4 + + 3 x + 4 y = 8 e −3 t
dt dt dt dt
1
given that x = , y = 0 , when t = 0 .
5
Solution: Writing D for (d / dt), the given equations can be written as
(3 D − 4) x + (2 D + 3) y = 8 e −3 t …(1)
and (4 D + 3) x + ( D + 4) y = 8 e −3 t …(2)
[( D + 4)(3 D − 4) − (2 D + 3) (4 D + 3)] x
= 8 [( D + 4) e −3 t − (2 D + 3) e −3 t ]
or − 5 ( D2 + 2 D + 5) x = 8 (− 3 e −3 t + 4 e −3 t + 6 e −3 t − 3 e −3 t )
32 −3 t
or ( D2 + 2 D + 5) x = − e .
5
A.E. is m2 + 2 m + 5 = 0 , giving m = − 1 ± 2 i .
∴ C.F. = e −1 (c1 cos 2 t + c2 sin 2 t) .
1  − 32 e −3 t  = − 4 e −3 t .
Also P.I. = 2  
D + 2D + 5  5  5
4 −3 t
∴ x = e − t (c1 cos 2 t + c2 sin 2 t) − e .
5
Given that x = 1/ 5, when t = 0, we have
1 4
= c1 − or c1 = 1 .
5 5
4 −3 t
∴ x = e − t (cos 2 t + c2 sin 2 t) − e
5
dx 12 −3 t
⇒ = − e − t (cos 2 t + c2 sin 2 t) + e − t (− 2 sin 2 t + 2 c2 cos 2 t) + e .
dt 5
Now multiplying (2) by 2 and then subtracting (1) from it, we get
dx
5y = −5 − 10 x + 8 e −3 t .
dt
Putting the values of x and dx / dt, we get
4
y = e − t (− cos 2 t − 2 c2 cos 2 t + 2 sin 2 t − c2 sin 2 t) + e − 3 t .
5
Given that y = 0, when t = 0, we have
4 1
0 = − 1 − 2 c2 + , or c2 = − ⋅
5 10
1 −t 4
∴ x= e (10 cos 2 t − sin 2 t) − e −3 t …(3)
10 5
1 −t 4
and y= e (21 sin 2 t − 8 sin 2 t) + e −3 t . …(4)
10 5
The required solution consists of equations (3) and (4).
D-160
Solve the following simultaneous differential equations :

dx dy
1. + 5 x + y = et , − x + 3 y = e2 t .
dt dt (Meerut 2004B)
dx dy t dx dy 2t
2. +2 − 2 x + 2 y = 3e , 3 + + 2 x + y = 4e .
dt dt dt dt (Meerut 2007)
dx dy 2t
3. + 2 x − 3 y = t, − 3x + 2 y = e .
dt dt (Lucknow 2011)
d d
 + 2 x + 3 y = 0 , 3 x +  + 2 y = 2 e3 t .
4.    
 dt   dt 
dx dy
5. = 3x + 2 y , + 5x + 3 y = 0 .
dt dt (Gorakhpur 2009; Kumaun 12)
dx dy
6. = ax + by , = a ′ x + b ′ y.
dt dt
dx dy
7. + ωy = 0 , − ωx = 0 .
dt dt (Kanpur 2003, 05; Kumaun 06)
dx dy t
8. + 4 x + 3 y = t, + 2x + 5 y = e .
dt dt
dx dy dx dy
9. + − 2 y = 2 cos t − 7 sin t, − + 2 x = 4 cos t − 3 sin t .
dt dt dt dt
(Lucknow 2005, 09; Avadh 10)
10. ( D − 17) y + (2 D − 8) z = 0 , (13 D − 53) y − 2 z = 0 .
dx dy dx dy
11. 4 +9 + 11x + 31 y = e t , 3 +7 + 8 x + 24 y = e2 t .
dt dt dt dt
dx dy dy
12. +2 + x + 7 y = et − 3 , − 2 x + 3 y = 12 − 3 e t .
dt dt dt
d2 x d2 y
13. 2
+ 4 x + y = te t , 2
+ y − 2 x = sin2 t.
dt dt
d2 xdy dx dy
14. 2
− = 2 x + 2t , +4 =3y.
dt dt dt dt
dx 2 dy 1
15. + ( x − y) = 1, + ( x + 5 y) = t .
dt t dt t
A nswers 1
4 t 1 2t 7 2t 1 t
1. x = (c1 + c2 t) e −4 t + e − e , y = − (c1 + c2 + c2 t) e −4 t + e + e
25 36 36 25
1 2t 3 t 15 t
2. x = e − e + c1 e −6 t /5 , y = e − 8 c1 e −6 t /5 + c2 e − t
2 11 22
3 2t 2 13 4 3 12
3. x = c1 e −5 t + c2 e t + e − t− , y = − c1e −5 t + c2 e t + e2 t − t −
7 5 25 7 5 25
D-161
3 3t 5
4. x = c1 e t + c2 e −5 t − e , y = − c1e t + c2 e −5 t + e3 t
8 8
1 1
5. x = c1 cos t + c2 sin t, y = (c2 − 3 c1) cos t − (c1 + 3 c2 ) sin t
2 2
1
6. x = c1e m1t + c2 e m2 t , y = (m1 − a) c1 e m1t + (m2 − a) c2 e m2 t , where
{ }
b
(a + b ′ ) + √ [(a − b ′ )2 + 4 a ′ b] (a + b ′ ) + √ [(a − b ′ )2 + 4 a ′ b]
m1 = and m2 =
2 2
7. x = c1 cos ωt + c2 sin ωt, y = c1 sin ωt − c2 cos ωt
5 1 31
8. x = c1e −2 t + c2 e −7t + t − et − ,
14 8 196
1 3 −2 t 5 27 
y=  − t − 2 c1 e + 3 c2 e −7t + e t + 
3 7 8 98 
9. x = c1e √2 t + c2 e −√2 t + 3 cos t,
y = (√ 2 + 1) c1e √2 t + (1 − √ 2) c2 e −√2 t + 2 sin t
10. y = c1e3 x + c2 e5 x , z = − 7 c1e3 x + 6 c2 e5 x

31 t 49 2 t 19 2 t 11 t
11. y = (c1 + c2 t) e −4 t + e − e , y = − (c2 + c1 + c2 t) e −4 t + e − e .
25 36 36 25
1 31 t 93
12. x = e −4 t [(c2 − c1) cos t − (c1 + c2 ) sin t] + e − ⋅
2 26 17
6 2 t
y = e −4 t (c1 cos t + c2 sin t) + − e
17 13
et  1 1 cos 2 t
13. x = c1 cos √ 2 t + c2 sin √ 2 t + c3 cos √ 3 t + c4 sin √ 3 t + t −  − + ⋅
6  6  12 4
1 7 1
y = − 2 (c1 cos √ 2 t + c2 sin √ 2 t ) − (c3 cos √ 3 t + c4 sin √ 3 t) + e t  t −  + ⋅
6  6 3
1 1
14. x = (c1 + c2 t) e t + c3 e −3 /2 t − t, y = (3 c2 − c1 − c2 t) e t − c3 e −(3 /2)t − ⋅
6 3
c1 c2 t2 3t c c 2 t2 1
15. x = + + + , y = − 13 − 42 + − t⋅
t3 t4 15 10 2t t 15 20
4 Simultaneous Equations of the Form

P1 dx + Q1 dy + R1 dz = 0 …(1)
P2 dx + Q2 dy + R2 dz = 0 , …(2)
where the coefficients P1, P2 , Q1, Q2, R1 and R2 are functions of x, y, z .
Above is the general form of a system of simultaneous equations of the first order in
three variables.
D-162
Taking z as the independent variable, the equations (1) and (2) can be written in the
form
dx dy
P1 + Q1 + R1 = 0
dz dz
dx dy
P2 + Q2 + R2 = 0 .
dz dz
Solving these equations for dx / dz and dy / dz by the method of cross- multiplication, we
get
dx / dz dy / dz 1
= =
Q1 R2 − Q2 R1 R1 P2 − R2 P1 P1Q2 − P2 Q1
dx dy dz
or = = ,
Q1 R2 − Q2 P1 R1 P2 − R2 P1 P1Q2 − P2 Q1
which is of the form

dx dy dz
= = , …(3)
P Q R
where P, Q, R are functions of x, y and z.
Thus the equations (1) and (2) can always be expressed in the form (3). To solve the
simultaneous equations in the form (3) we can use the following methods.
First Method: Suppose one variable is absent from two members of (3), say, z is
absent from P and Q. Then the solution of
dx dy
=
P Q
gives a relation between x and y and we thus get one equation constituting the
complete solution of the given equations. If necessary, we can use this relation to
eliminate x or y from one of the remaining equations in (3) and we can find the other
integral relation which gives us the second equation of the complete solution.
Second Method: By a property of ratio and proportion in algebra, we know that each
dx dy dz
of the equal fractions , , in (3), is also equal to
P Q R
l dx + m dy + n dz L dx + M dy + N dz
, , etc.
lP + mQ + nR LP + MQ + NR
dx dy dz l dx + m dy + n dz L dx + M dy + N dz
∴ = = = = ⋅ …(4)
P Q R lP + mQ + nR LP + MQ + NR
The relations between x, y, z that satisfy (3) also satisfy (4) .
By a proper choice of multipliers l, m, n ; L, M , N etc., it is possible to find equations
which can be easily solved. Particularly l, m, n can be so chosen that
lP + mQ + nR = 0.
which makes l dx + m dy + n dz = 0 . …(5)
Let l dx + m dy + n dz be an exact differential, say du.
D-163
Then (5) ⇒ du = 0 ⇒ u = a , which gives one equation constituting the complete

solution.
Again L, M , N may be found such that
LP + MQ + NR = 0 ⇒ L dx + M dy + N dz = 0 .
If L dx + M dy + N dz be an exact differential, say dv , then dv = 0 ⇒ v = b, which gives
the second equation constituting the complete solution.
Hence the complete solution consists of the equations u = a and v = b which must of
course be independent of each other.
Note 1: Sometimes only one set of multipliers may serve the purpose.
Note 2: We can find one relation, say u = a, by the first method and the second relation,
say v = b, by the second method.
5 Geometrical Interpretation of the Differential

Equations dx = dy = dz ⋅
P Q R
(Gorakhpur 2005)
From solid geometry, we know that the direction cosines of the tangent to a curve in
three dimensional Euclidean space at any point ( x, y, z ) are
dx dy dz
, ,
ds ds ds
i. e., the direction cosines are in the ratio of dx : dy : dz . Hence the given differential
equations represent a system of curves in space, the direction cosines of the tangent to
any member of this system at any point ( x, y, z ) are proportional to P, Q and R. Let
u = a and v = b be two component solutions of the given equations. Then the curves are
obtained by the intersection of the surfaces u = a and v = b. Since the arbitrary
constants a and b can take infinite values, the curves are doubly infinite in number.
x dx dy dz
Example 10: Solve 2 = = ⋅
z − 2 yz − y2 y+z y−z
(Meerut 2001, 04; Avadh 08; Garhwal 08; Kumaun 13)
Solution: Taking the last two members of the given equations, we get
( y − z ) dy = ( y + z ) dz or y dy − z dz − ( y dz + z dy) = 0
or 2 y dy − 2 z dz − 2 ( y dz + z dy) = 0 .
Integrating, we get
y2 − z 2 − 2 yz = c1 . …(1)
D-164
Again choosing 1, y, z as multipliers, we get

x dx dy dz
2 2
= =
z − 2 yz − y y + z y −z
x dx + y dy + z dz x dx + y dy + z dz
= 2 2
= ⋅
1 . (z − 2 yz − y ) + y ( y + z ) + z ( y − z ) 0
∴ x dx + y dy + z dz = 0 .
Integrating, we get x2 + y2 + z 2 = c2 . …(2)
dx dy dz
Example 11: Solve = = ⋅
y2 + z 2 − x2 − 2 xy − 2 xz (Garhwal 2002; Kumaun 10)
Solution: Taking the last two members of the given equations, we get
dy dz
= ⋅
y z
Integrating, we get
log y = log z + log c1
or log y = log (c1 z ) or y = c1z . …(1)
Again choosing x, y, z as multipliers, we get
dx dy dz x dx + y dy + z dz
2 2 2
= = = ⋅
y +z −x − 2 xy − 2 xz − x ( x2 + y2 + z 2 )
Taking the last two fractions, we get
dz 2 x dx + 2 y dy + 2 z dz
= ⋅
z x2 + y2 + z 2
Integrating, log z + log c2 = log ( x2 + y2 + z 2 )
or x2 + y2 + z 2 = zc2 . …(2)
dx dy dz
1 − 2 3 x2 sin ( y + 2 x)
(Meerut 2001, 10B; Gorakhpur 07; Garhwal 06; Kumaun 06)
Solution: From the first two members, we get

2 x + y = c1.
Taking the first and the last members, we get
dx dz dz
= 2 = 2 , using (1).
1 3 x sin ( y + 2 x) dx sin c1
∴ dz = 3 x2 sin c1 dx .
Integrating, z = x3 sin c1 + c2 or z − x3 sin ( y + 2 x) = c2 . …(2)

D-165
dx dy dz
x ( y2 − z 2 ) − y (z 2 + x2 ) z ( x2 + y2 ) (Meerut 2005, 11)
1 1 1
Solution: Choosing , − , − as multipliers, we have
x y z
dx dy dz
− −
x y z
each fraction = ⋅
0
dx dy dz dx dy dz
∴ − − =0 ⇒ = + ⋅
x y z x y z
Integrating, log x + log c1 = log y + log z
or yz = c1 x. …(1)
Again choosing x, y, z as multipliers, we get
x dx + y dy + z dz
each fraction = ⋅
0
∴ x dx + y dy + z dz = 0 .
Integrating, x2 + y2 + z 2 = c2 . …(2)
The equations (1) and (2) give the required solution of the given equations.
dx dy dz
cos ( x + y) sin ( x + y) z
(Meerut 2006, 07B; Gorakhpur 08, 11; Garhwal 11)
Solution: From the given equations, we have

dx + dy dx − dy dz
= = ⋅
cos ( x + y) + sin ( x + y) cos ( x + y) − sin ( x + y) z
Taking the first two members, we get
cos ( x + y) − sin ( x + y)
(dx + dy) = dx − dy .
cos ( x + y) + sin ( x + y)
Integrating, log [cos ( x + y) + sin ( x + y)] = x − y + log c1
or [cos ( x + y) + sin ( x + y)] e y− x = c1. …(1)
Again taking the first and the third members, we get
dx + dy dz dz (1/ √ 2)(dx + dy)
= or =
cos ( x + y) + sin ( x + y) z z 1
sin ( x + y + π)
4
dz 1
or √2 = cosec ( x + y + π)(dx + dy) .
z 4
1 1
Integrating, √ 2 log z = log tan ( x + y + π) + log c2
2 4
x y π
or z √2 cot  + +  = c2 . …(2)
2 2 8
D-166
dx dy dz
x y z − a √ ( x2 + y2 + z 2 ) (Meerut 2005B, 10B)
Solution: From the first two members of the given differential equations, we have
x = c1 y. …(1)
Again choosing x, y, z as multipliers, we have
dx dy dz x dx + y dy + z dz
= = =
x y z − a √ ( x2 + y2 + z 2 ) x2 + y2 + z 2 − az √ ( x2 + y2 + z 2 )
dy dz u du
or = = , putting x2 + y2 + z 2 = u2
y z − au u2 − azu
dy dz du du + dz
or = = = ⋅
y z − au u − az (u + z ) (1 − a)
Taking the first and the last members and integrating, we get
(1 − a) log y = log (u + z ) + log c2 or y(1− a) = (u + z ) c2
or y(1− a) = c2 { √ ( x2 + y2 + z 2 ) + z } . …(2)
dx dy dz
Example 16. Solve = = ⋅
y3 x − 2 x4 2 y4 − x3 y 9 z ( x3 − y3 )
Solution: From the first two members of the given differential equations, we get
(2 y4 − x3 y) dx = ( y3 x − 2 x4 ) dy
2 y 1  1 2x 
or 
3
− 2  dx =  2 − 3  dy , dividing by x3 y3
x y  x y 
 1 2y   1 2x 
or  2 dy − 3 dx +  2 dx − 3 dy = 0 .
x x   y y 
y x
Integrating, + 2 = c1 . …(1)
x2 y
1 1 1
Again taking , , as multipliers, we get
x y 3z
1 1 1 1 1 1
dx + dy + dz dx + dy + dz
x y 3z x y 3z
each fraction = 3 = ⋅
( y − 2 x3 ) + (2 y3 − x3 ) + 3 ( x3 − y3 ) 0
1 1 1
∴ dx + dy + dz = 0 .
x y 3z
1
Integrating, log x + log y + log z = log c2
3
or xyz1 /3 = c2 . …(2)

D-167
dx dy dz
Example 17: Solve = 2 = ⋅
xy y zxy − 2 x2 (Purvanchal 2006; Garhwal 09)
Solution: From the first two members of the given differential equations, we get
dx dy
= ⋅
x y
Integrating, log x = log y + log c1
or x = c1 y. …(1)
Again taking the last two members, we get
dy dz dy dz
2
= 2
or 2
= [∵ x = c1 y]
y xyz − 2 x y c1 y z − 2 c12 y2
2
dz dz
or dy = or c1 dy = ⋅
c1z − 2 c12 z − 2 c1
Integrating, c1 y = log (z − 2 c1) + c2

 x
or x = log  z − 2  + c2 [∵ x = c1 y]
 y
or x = log ( yz − 2 x) − log y + c2 . …(2)

dx dy dz
Example 18: Solve 2 = = ⋅
x + y2 2 xy ( x + y) z (Kumaun 2009)
Solution: From the given equations, we have

dx + dy dx − dy dz
2 2
= 2 2
= ⋅
x + y + 2 xy x + y − 2 xy ( x + y) z
Taking the first two members, we have

dx + dy dx − dy
= ⋅
( x + y)2 ( x − y)2
Integrating, − ( x + y)−1 = − ( x − y)−1 + c1
1 1 2y
or − = c1, or = c1
x− y x+ y x − y2
2
or 2 y = c1 ( x2 − y2 ) …(1)
Taking the first and the last members, we have
dx + dy dz dx + dy dz
2
= or = ⋅
( x + y) ( x + y) z x+ y z
Integrating, log ( x + y) = log z + log c2

or ( x + y) = c2 z . …(2)
D-168
dx dy dz
Example 19: Solve 2 = 2 = ⋅
x y nxy
dx dy
Solution: We have 2 = 2 .
x y
1 1
Integrating, = + c1 or y − x = c1 xy. …(1)
x y
Also using 1/ x, − 1/ y and c1 / n as multipliers, we have
(1/ x) dx − (1/ y) dy + (c1 / n) dz
each fraction =
x − y + c1 xy
(1/ x) dx − (1/ y) dy + (c1 / n) dz
= , using (1).
0
∴ (1/ x) dx − (1/ y) dy + (c1 / n) dz = 0 .
Integrating, log x − log y + (c1 / n) z = c2
c1  y nxy  y
or z = log   + c2 or z = log   + c2 . …(2)
n  x y−x  x
Solve the following simultaneous differential equations :

dx dy dz
1. = = ⋅
yz zx xy (Kumaun 2008, 12)
dx dy dz
2. = = .
mz − ny nx − lz ly − mx
Also give geometrical interpretation of the solution of these equations.
dx dy dz
3. = = 2 ⋅
z − z z + ( x + y)2
a dx b dy c dz
4. = = ⋅
(b − c ) yz (c − a) zx (a − b) xy (Lucknow 2006; Kumaun 11)
dx dy dz
5. = = ⋅
xz (z 2 + xy) − yz (z 2 + xy) x4
dx dy dz
6. = = 2 ⋅
z ( x + y) z ( x − y) x + y2
− dx dy dz
7. = = ⋅
x ( x + y) y ( x + y) ( x − y) (2 x + 2 y + z )
D-169
dx dy dz
8. = = ⋅
1 3 5 z + tan ( y − 3 x) (Meerut 2006B; Gorakhpur 08, 11)
dx dy dz
9. = = ⋅
x ( y − z ) y (z − x) z ( x − y)
dx dy dz
10. 2 2
= 2 2
= ⋅
x(y − z ) y (z − x ) z ( x − y2 )
2
(Gorakhpur 2005, 09)
dx dy dz
11. = = ⋅
1+ y 1+ x z
dx dy dz
12. 2 2
= 2 2
= ⋅
y + yz + z z + zx + x x + xy + y2
2
(Avadh 2007)
 dy − dx dz − dy dz − dx 
Hint. = = 
 ( x − y ) ( x + y + z ) ( y − z ) ( x + y + z ) ( x − z ) ( x + y + z )
dx dy dz
13. = = ⋅
x2 − yz y2 − zx z 2 − xy (Gorakhpur 2006; Garhwal 01, 10B; Kumaun 15)
 dx − dy dy − dz dz − dx 
Hint. = = ⋅
 ( x − y) ( x + y + z ) ( y − z ) ( x + y + z ) (z − x) ( x + y + z )
dx dy dz
14. 2 2
= 2 2
= ⋅ (Kanpur 2009)
x + y + yz x + y − xz z ( x + y)
 dx − dy dz x dx + y dy 
Hint. = = 
 z ( x + y) z ( x + y) ( x + y) ( x2 + y2 )
dx dy dz
15. = = ⋅
y+z z+x x+ y
 dx − dy dy − dz dx + dy + dz 
Hint. = = .
 y−x z−x 2 ( x + y + z )
A nswers 2
1. x2 − y2 = c1, x2 − z 2 = c2
2. lx + my + nz = c1, x2 + y2 + z 2 = c2
These equations constituting the general solution of the given differential
equations represent a family of circles.
3. x + y = c1, log { z 2 + ( x + y)2 } − 2 x = c2
4. a2 x2 + b2 y2 + c 2 z 2 = c1, ax2 + by2 + cz 2 = c2
5. xy = c1, (z 2 + xy)2 − x4 = c2
D-170
6. x2 − y2 − z 2 = c1, 2 xy − z 2 = c2
7. xy = c1, x2 + y2 + ( x + y) z = c2
8. y − 3 x = c1, 5 z + tan ( y − 3 x) = c2 e5 x
9. x + y + z = c1, xyz = c2
10. x2 + y2 + z 2 = c1, xyz = c2
11. z ( x − y) = c1, z = c2 ( x + y + 2)
12. ( y − x) = c1 (z − y), ( y − x) = c2 (z − x)
13. ( x − y) = c1 ( y − z ), ( y − z ) = c2 (z − x)
14. x − y − z = c1, x2 + y2 = z 2 c2
15. ( y − x) = c1 (z − y), ( x − y)2 ( x + y + z ) = c2

(b), (c) and (d).
dx dy
1. Solving the equations + 5 x + y = et , − x + 3 y = e2 t , for x, we get x =
dt dt
4 t 1 2t 2
(a) (c1 + c2 t) e −4 t + e − e (b) c1 e −4 t + c2 e t −
25 36 5
(c) c1e t + c2 e −4 t + c3 e −2 t (d) none of these.
dx dy dy
2. Solving the equations + + 2x + y = 0 , + 5 x + 3 y = 0 , we get x =
dt dt dt
(a) c1 e t + c2 e − t (b) c1 cos t + c2 sin t
(c) (c1 + c2 t) cos t (d) c1 e t + c2 e − t . (Garhwal 2008)
dx dy
3. Solving the equations + ωy = 0 , − ωx = 0 for x, we get x =
dt dt
(a) c1e ωt + c2 e − ωt (b) (c1 + c2 t) e ωt
(c) c1 cos ωt + c2 sin ωt (d) none of these. (Kumaun 2008)
dx dy dz
4. Solution set of the equations = = is
y+z z+x x+ y
(a) ( y − x) = c1 (z − y) (b) x + y = c1 ( y + z )
2
( x − y) ( x + y + z ) = c2 ( x − y)2 ( x + y + z ) = c2 .
(c) ( y − z ) = c1 ( x − y) (d) none of these.
2
( x + y) ( x − y + z ) = c2
D-171
dx dy dz
x y z
(a) xy = c1, yz = c2 (b) x = c1 z , y = c2 z

(c) x = c1 y, y = c2 z (d) none of these.
dx dy dz
yz zx xy
(a) x2 + y2 = c1, y2 − z 2 = c2 (b) x2 − z 2 = c1, y2 + z 2 = c2
(c) x2 − y2 = c1, x2 − z 2 = c2 (d) none of these. (Kumaun 2007)

dx dy dz
7. Solution set of the simultaneous equations = = is
y x z
(a) x2 − y2 = c1, x + y = c2 z (b) x2 + y2 = c1, x + y = c2 z
(c) x2 − y2 = c1, x − y = c2 z (d) x2 − y2 = c1, y2 − z 2 = c2

dx dy
8. In equation − 7 x + y = 0, − 2 x − 5 y = 0, value of x is
dt dt
(a) e6 t (c1 cos t + c2 sin t) (b) e −6 t (c1 cos t + c2 sin t)
(c) e t (c1 cos 6 t + c2 sin 6 t) (d) none of these
(Garhwal 2007, 10B)
9. Solution of the differential equation dx = dy = dz is
(a) x − y = c1 and x − z = c2 (b) x + y = c1 and x − z = c2
(c) x − y = c1 and x + z = c2 (d) x + y = c1 and x + z = c2
dx dy
10. If = 3 x + 2 y, = 5 x + 3 y then x is equal to
dt dt
(a) c1 cosh 10 t + c2 sinh 10 t (b) e 3 t (c1 cosh 10 t + c2 sinh 10 t)
(c) e − 3 t (c1 cosh 10 t + c2 sinh 10 t) (d) e 3 t (c1 cosh 10 t − c2 sinh 10 t)

(Kanpur 2016)
dx dy
11. If + 7 x − y = 0, + 2 x + 5 y = 0 then y is equal to
dt dt
(a) e − 6 t (c 1 cos t + c2 sin t) (b) e 6 t (c 1 cos t + c2 sin t)
e−6 t e 6t
(c) (c 1 cos t + c2 sin t) (d) (c 1 cos t + c2 sin t)
2 2
x dx dy dz
12. Solution set of the equations 2
= = 2 is
y z xz y
(a) x3 + y3 = c 1, x2 − z 2 = c2 (b) x3 − y3 = c 1, x2 + z 2 = c2
(c) x3 − y3 = c 1, x2 − z 2 = c2 (d) x3 + y3 = c 1, x2 + z 2 = c2
D-172
dx dy dz
1 − 2 3 x2 sin ( y + 2 x)
(a) 2 x + y = c 1, z − x3 sin ( y + 2 x) = c2
(b) 2 x − y = c 1, z − x3 sin ( y + 2 x) = c2
(c) 2 x + y = c 1, z + x3 sin ( y + 2 x) = c2
(d) 2 x − y = c 1, z + x3 sin ( y + 2 x) = c2
dx dy
14. Solving the equations − y = t, + x = 1 for x, we get x =
dt dt
(a) c 1 e − 2 t + c2 e 2 t (b) (c 1 + c2 x) e t
(c) c 1 cos t + c2 sin t + 2 (d) c 1 cos 2 t + c2 sin 2 t + 2

1. Eliminating y between the equations
dx dy
− 7x + y = 0 , − 2x − 5 y = 0
dt dt
we obtain the differential equation…… .
2. Eliminating y between the simultaneous equations
d2 x d2 y
+ m2 y = 0 ,− m2 x = 0
dt2 dt2
we obtain the differential equation …… . (Meerut 2003)
3. The complete solution of the simultaneous equations

dx dy dz
= ny − mz , = lz − nx, = mx − ly is …… .
dt dt dt
A nswers
1. (a) 2. (b) 3. (c) 4. (a) 5. (b)
6. (c) 7. (a) 8. (a) 9. (a) 10. (b)
11. (a) 12. (c) 13. (a) 14. (c)

1. ( D2 − 12 D + 37) x = 0 2. ( D4 + m4 ) x = 0
3. x2 + y2 + z 2 = c1, lx + my + nz = c2
¨
D-173
7
L inear E quations of S econd
O rder with V ariable C oefficients
1 Linear Equation of Second Order

n equation of the form
A d2 y
dx2
+P
dy
dx
+ Qy = R ,
where P, Q, R are functions of x only, is called the ‘linear equation of second order’.
There is no general method for solving such equations. Here we shall discuss certain
methods by which the solutions of such equations can be found.
2 The Complete Solution in Terms of a Known Integral

The following theorem relating to the linear differential equation of the second order
is of great importance :
If an integral included in the complementary function of such an equation be known,
then the complete primitive (or the general solution) can be found in terms of the known
integral.
Let y = u be a known integral in the complementary function of
D-174
d2 y dy
+P + Qy = R …(1)
dx2 dx
i. e., it is a solution of
d2 y dy
2
+P + Qy = 0 .
dx dx
d2 u du
∴ +P + Qu = 0 . …(2)
dx2 dx
Let y = uv be the solution of (1).
dy du dv
Putting y = uv, we get =v +u
dx dx dx
d2 y d2 u
du dv d2 v
and 2
=v 2
+2
+u 2 ⋅
dx dx dx dx dx
Substituting these values the equation (1) becomes
 d2 u du dv d2 v  du dv
v 2 + 2 + u 2  + P  v + u  + Q vu = R
 dx dx dx dx   dx dx 
d2 v dv 2
2 du + Pu + v  d u + P du + Qu = R
or u 2
+    2 
dx dx  dx   dx dx 
d2 vdv  du
or u + 2 + Pu + v. 0 = R , using (2)
dx2 dx  dx 
d2 v  2 du dv R
or + P +  = ⋅ …(3)
dx2  u dx  dx u
dv d2 v dp
Putting = p, 2 = , (3) becomes
dx dx dx
dp  2 du R
+ P +  p= , …(4)
dx  u dx  u
which is linear with p as dependent variable.
 2 du  2 
∫  P +  dx ∫  P dx + du
I. F. = e u dx  =e u = e2 log u + ∫ P dx
= u2 e ∫ P dx .
Hence solution of (4) is
pu2 e ∫ P dx = R 2
e ∫ P dx  dx + c1.
∫  u u 
…(5)
dv c1e − ∫ P dx e − ∫ P dx P dx
∴ p=
dx
=
u2
+
u2 ∫ u Re ∫ dx.
Integrating this, we get

e − ∫ P dx  e − ∫ P dx P dx 
y = c2 + c1
∫ u 2
dx +
∫ 
 u
2 ∫ u Re ∫ dx dx.

Hence the solution of (1) is
e − ∫ P dx  e − ∫ P dx P dx 
y = uv = c2 u + c1 u
∫ u2
dx + u
∫ 
 u
2
⋅
∫ u Re ∫ dx dx.

…(6)
D-175
It contains the given solution y = u and since it contains two arbitrary constants so it is
the complete primitive or the general solution of (1). It is evident from (6) that the
second part of the complementary function is
e − ∫ P dx
u
∫ u2
dx and the particular integral is
 e − ∫ P dx P dx 
u
∫  u2
⋅
∫ u Re ∫ dx dx.

3 To Find one Integral In C.F. by Inspection

d2 y dy
Consider 2
+P + Q y = 0. …(1)
dx dx
1. y = e mx is a solution of (1) if
m2 + Pm + Q = 0 .
dy d2 y
We have y = e mx ⇒ = me mx , 2 = m2 e mx .
dx dx
If y = e mx is a solution of (1), then
(m2 + Pm + Q) e mx = 0 ⇒ m2 + Pm + Q = 0 .
Deductions. (i) y = e x is a solution of (1) if 1 + P + Q = 0 .
(ii) y = e − x is a solution of (1) if 1 − P + Q = 0 .
2. y = x m is a solution of (1) if m (m − 1) + Pmx + Qx2 = 0 .
dy d2 y
We have y = x m ⇒ = mx m −1, 2 = m (m − 1) x m −2 .
dx dx
If y = x m is a solution of (1), then
m (m − 1) x m −2 + Pmx m −1 + Qx m = 0
or m (m − 1) + Pmx + Qx2 = 0 .
Deductions. (i) y = x is a solution of (1) if P + Qx = 0.
(ii) y = x2 is a solution of (1) if 2 + 2 Px + Qx2 = 0 .
All the above results are summarised below :

(i) y = e x is a part of C.F. if 1 + P + Q = 0 .
(ii) y = e − x is a part of C.F. if 1 − P + Q = 0 .
(iii) y = e mx is a part of C.F. if m2 + m P + Q = 0 .
(iv) y = x is a part of C.F. if P + Qx = 0 .
(v) y = x2 is a part of C.F. if 2 + 2 P x + Qx2 = 0 .
(vi) y = x m is a part of C.F. if m (m − 1) + Pmx + Qx2 = 0 .
D-176
d2 y dy
Example 1: Solve ( x + 2) 2
− (2 x + 5) + 2 y = ( x + 1) e x .
dx dx
(Gorakhpur 2009; Lucknow 07;
Rohilkhand 09; Kanpur 07, 09; Garhwal 09; Kumaun 11)
d2 y (2 x + 5) dy 2 ( x + 1) x
2
− + y= e . …(1)
dx x + 2 dx x + 2 x +2
Comparing (1) with the standard form
d2 y dy
2
+ Qy = R, we have
+P
dx dx
2x + 5 2 x +1 x
P=− ,Q = ,R= e .
x +2 x +2 x +2
Here 22 + 2 P + Q = 0 ,
∴ y = e2 x is a part of the C.F. of the solution of the equation (1).
dy dv d2 y d2 v dv
Putting y = ve2 x , = e2 x + 2 v e2 x , and 2 = e2 x 2 + 4 e2 x + 4 v e2 x in (1), we
dx dx dx dx dx
get
d2 v 2 x + 3 dv x + 1 − x
2
+ = e
dx x + 2 dx x + 2
dp 2 x + 3 x + 1 −x  where p = dv
or + p= e ,  
dx x +2 x +2  dx 
which is a linear equation with p as dependent variable.
2 x +3  1 
∫ dx ∫ 2 −  dx
e2 x
x +2 x + 2
I. F. = e =e 
= e2 x − log ( x +2) =
x +2
e2 x x + 1 − x e2 x
∴ p⋅
x +2
=
∫ x +2
e ⋅
x +2
dx + c1
x +1  1 1  x
e x dx + c1 =
=
∫ ( x + 2) 2 ∫  −
 x + 2 ( x + 2) 
2
e dx + c1
1 1 1
ex + ex ⋅ ex
=
x +2 ∫ ( x + 2)2
dx −
∫ ( x + 2)2
dx + c1 ,
integrating the first integral by parts.

dv −x −2 x
∴ p= = e + c1 e ( x + 2) .
dx
1 1
v = − e − x − c1e −2 x ( x + 2) − c1 e −2 x + c2
2 4
1
= − e − x − c1 (2 x + 5) e −2 x + c2 .
4
D-177
∴ The general solution of (1) is

1
y = v e2 x = − e x − c1 (2 x + 5) + c2 e2 x .
4
d2 y dy
Example 2: Solve x 2
− (2 x − 1) + ( x − 1) y = 0 .
dx dx
(Garhwal 2003; Gorakhpur 08; Meerut 10)
Solution: The given equation can be written in the standard form as
d2 y 1 dy  1
− 2 −  + 1 −  y = 0 . …(1)
dx2  x  dx  x
Here 1 + P + Q = 0,
∴ y = e x is a part of the C.F. of the solution of (1).
dy dv x d2 y d2 v dv x
Putting y = v e x , = e + v e x , and 2 = 2 e x + 2 e + v e x in (1), we get
dx dx dx dx dx
d2 v 1 dv dp p dv
2
+ =0 ⇒ + = 0 , where p =
dx x dx dx x dx
dp dx
⇒ =− ⇒ log p = − log x + log c1
p x
dv c1
⇒ p= = ⇒ v = c1 log x + c2 .
dx x
∴ The complete solution of (1) is
y = v e x = (c1 log x + c2 ) e x .
d2 y dy
− ( x2 + 2 x) + ( x + 2) y = x3 e x .
dx dx
(Bundelkhand 2001; Garhwal 02;
Gorakhpur 06; Purvanchal 07; Kumaun 09; Meerut 2010B)
Solution: The standard form of the given equation is
d2 y 2  dy  1 2 
2
− 1 +  + + x
 y = xe . …(1)
dx  x  dx  x x2 
2 1 2
Here P + Qx = − 1 +  +  + 2  x = 0 ,
 x x x 
∴ y = x is a part of the C.F. of the solution of (1).
dy dv d2 y d2 v dv
Putting y = vx, =x + v, 2
= x 2
+2 in (1), we get
dx dx dx dx dx
d2 v dv dp  where p = dv
2
− = e x or − p = e x,  
dx dx dx  dx 
which is linear in p.
I. F. = e ∫
− dx
∴ = e− x .
pe − x = x
⋅ e − x dx + c1 = x + c1.
∴
∫e
dv
∴ p= = x e x + c1 e x .
dx
D-178

v = x e x − e x + c1e x + c2 .
y = v x = x2 e x − x e x + c1 x e x + c2 x .
d2 y dy
Example 4: Solve 2
− cot x − (1 − cot x) y = e x sin x .
dx dx (Meerut 2007B)

d2 y dy
− cot x − (1 − cot x) y = e x sin x .
dx2 dx
Here 1 + P + Q = 0 ,
∴ y = e x is a part of the C.F. of (1).
dy dv x d2 y d2 v dv x
Putting y = ve x , = e + ve x , and 2 = 2 e x + 2 e + ve x in (1), we get
dx dx dx dx dx
d2 v dv
2
+ (2 − cot x) = sin x
dx dx
dp  where p = dv
or + (2 − cot x) p = sin x,  
dx  dx 
e2 x
I. F. = e ∫ (2 − cot x) dx
= e2 x − log sin x = ⋅
sin x
e2 x e2 x 1
sin x dx + c1 = e2 x + c1 .
∴ p
sin x
=
∫ sin x 2
dv 1
∴ p= = sin x + c1 e −2 x sin x .
dx 2
1 c
v = − cos x + 1 e −2 x (− 2 sin x − cos x) + c2 .
2 5
1 c
y = v e x = − e x cos x − 1 e − x (2 sin x + cos x) + c2 e x .
2 5
d2 y dy
− 2 x (1 + x) + 2 (1 + x) y = x3 .
dx dx (Rohilkhand 2001;
Garhwal 02, 07, 14; Kumaun 08, 10; Gorakhpur 10; Lucknow 05, 06)
d2 y 1 dy 1 1
2
− 2  + 1 + 2  2 +  y = x . …(1)
dx  x  dx x x
Here P + Q x = 0.
∴ y = x is a part of the C.F. of (1).
D-179
dy dv d2 y d2 v dv
Putting y = v x, =x + v and 2 = x 2 + 2 in (1), we have
dx dx dx dx dx
d2 v dv dp  where p = dv
2
−2 = 1 or − 2 p = 1,  
dx dx dx  dx 
− 2 dx
I. F. = e ∫ = e −2 x .
1 −2 x
pe −2 x = 1 ⋅ e −2 x dx + c1 = −
∴
∫ 2
e + c1.
dv 1
∴ p= = − + c1 e2 x .
dx 2
1 c
Integrating, we get v = − x + 1 e2 x + c2 .
2 2
1 c
y = v x = − x2 + 1 x e2 x + c2 x.
2 2
d2 y
Example 6: Solve sin2 x = 2 y, given y = cot x is a solution.
dx2
(Purvanchal 2010; Lucknow 08; Kanpur 07, 08)
dy dv
Solution: Putting y = v cot x, = cot x − v cosec2 x,
dx dx
d2 y d2 v dv
2
=cot x − 2 cosec2 x + 2 v cosec2 x cot x,
dx dx2 dx
the given differential equation becomes
d2 v dv d2 v 2 dv
sin2 x cot x 2 − 2 =0 or 2
− =0
dx dx dx sin x cos x dx
dp 2 dv
or − p = 0 , where p =
dx sin x cos x dx
dp 2
or = dx = 4 cosec 2 x dx .
p sin x cos x
Integrating, we get
1
log p = 4 ⋅ log tan x + log c1
2
dv
or p= = c1 tan2 x = c1 (sec2 x − 1) .
dx
2
∴ v = c1
∫ (sec x − 1) dx + c2 = c1 (tan x − x) + c2 .
∴ the complete solution of the given equation is

y = v cot x = c1 (1 − x cot x) + c2 cot x .
d2 y dy
Example 7: Solve ( x sin x + cos x) 2
− x cos x + y cos x = sin x ( x sin x + cos x)2 .
dx dx
Solution: The given equation in the standard form is
D-180
d2 y x cos x dy cos x
2
− + y = sin x ( x sin x + cos x) .
dx x sin x + cos x dx x sin x + cos x
…(1)
Here P + Qx = 0,
dy d2 y
Putting y = vx and the corresponding values of , in (1), we get
dx dx2
d2 v 2 x cos x  dv sin x ( x sin x + cos x)
+ −  =
dx2  x x sin x + cos x  dx x
dp  2 x cos x  sin x ( x sin x + cos x)
or + −  p= ⋅
dx  x x sin x + cos x  x
dv
which is linear in p where p = ⋅
dx
2 x cos x 
∫ −  dx
 x x sin x + cos x 
∴ I. F. = e
x2
= e2 log x − log ( x sin x + cos x)
= ⋅
x sin x + cos x
x2
∴ p⋅
x sin x + cos x
=
∫ x sin x dx + c1 = − x cos x + sin x + c1.
dv 1 c
∴ p= = (− x cos x + sin x) ( x sin x + cos x) + 12 ( x sin x + cos x)
dx x2 x
dv 1 1 1 1
or = − sin x cos x − cos 2 x + 2 sin x cos x + c1  sin x + 2 cos x
dx x x x x 
1 1 2 1 
or v=−
2 ∫ sin 2 x dx − 2 ∫  x cos 2 x − x2 sin 2 x dx
 sin x cos x 
+ c1
∫  x + x2  dx + c2
1 1 d  sin 2 x  d  cos x 
=
4
cos 2 x −
2 ∫ dx  x 
 dx − c1 ∫ dx  x 
 dx + c2
1 1 sin 2 x cos x
= cos 2 x − − c1 + c2 .
4 2 x x
1 1
y = vx = x cos 2 x − sin 2 x − c1 cos x + c2 x .
4 2
dy  d2 y 
Example 8: Solve x − y = ( x − 1)  2 − x + 1 .
dx  dx  (Garhwal 2001)
Solution: The given equation in the standard form is

d2 y x dy y
− + = x − 1. …(1)
dx2 x − 1 dx x − 1
D-181
Here P + Qx = 0.
dy d2 y
Putting y = vx and the corresponding values of and 2 in (1), we get
dx dx
d2 v 2 x  dv x − 1
+  −
2  =
dx  x x − 1 dx x
dp  2 x  x −1  where p = dv
or + −  p= ,  
dx  x x − 1 x  dx 
2 x 
∫ −  dx 2 1 
x x − 1 ∫  −1 −  dx x2 − x
∴ I. F. = e 
= e x x −1 = e2 log x − x − log ( x −1)
= e .
x −1
x2 x − 1 x2 − x
e− x =
∴ p⋅
( x − 1) ∫ x
⋅
x −1
e dx + c1
xe − x dx + c1 = − xe − x − e − x + c1 .
=
∫
dv x − 1 ( x − 1) c1 ( x − 1) x
∴ p= =− − + e
dx x x2 x2
dv 1 1 1
or = − 1 + 2 + c1  − 2  e x .
dx x  x x 
Integrating, we get
1 1
v=− x− + c1 e x + c2 .
x x
y = vx = − x2 − 1 + c1 e x + c2 x
or y = c1e x + c2 x − (1 + x2 ) .

d2 y dy
1. ( x + 1) 2
− 2 ( x + 3) + ( x + 5) y = e x .
dx dx (Kumaun 2012)
d2 y dy
2. x − 2 ( x + 1) + ( x + 2) y = ( x − 2) e x .
dx2 dx
d2 y dy
3. x 2
− (2 x + 1) + ( x + 1) y = ( x2 + x − 1) e2 x .
dx dx
2
d y dy
4. 2
+ (1 − cot x) − y cot x = sin2 x .
dx dx
D-182
d2 y dy
5. ( x − x2 ) 2
− (1 − 2 x) + (1 − 3 x + x2 ) y = (1 − x)3 .
dx dx
2
d y dy
6. (3 − x) 2
− (9 − 4 x) + (6 − 3 x) y = 0 .
dx dx
2
d y dy
7. x 2
+ ( x − 2) − 2 y = x3 .
dx dx (Kanpur 2011)
2
d y dy
8. (2 x − 1) 2
−2 + (3 − 2 x) y = 2 e x .
dx dx
2
d y dy
9. x 2
+ (1 − x) − y = e x.
dx dx (Kumaun 2009)
2 3
10. Solve x y2 + xy1 − 9 y = 0 , given that y = x is a solution.
d2 y 2 2
11. Solve 2
+ 1 + cot x − 2  y = x cos x ,
dx  x x  (Gorakhpur 2008, 11)
sin x
given that is an integral included in the C.F.
x
d2 y dy −1
12. Solve (1 − x2 ) 2
−x − a2 y = 0 , given that y = ce a sin x is an integral.
dx dx
A nswers 1
1 1
1. y=− x e x + c1e x ( x + 1)5 + c2 e x
4 5
1 2 x 1
2. y = − x e + xe x + c1 x3 e x + c2 e x
2 3
3. y = x e2 x + c1 x2 e x + c2 e x
1
4. y=− (sin 2 x − 2 cos 2 x) + c1 (sin x − cos x) + c2 e − x
10
1
5. y = − x + c1 x2 e − x + c2 e x
2
6. y = c1e3 x (4 x3 − 42 x2 + 150 x − 183) + c2 e x
7. y = x3 + (c1 − 3) x2 − 2 (c1 − 3) ( x − 1) + c2 e − x
y = − c1 xe − x − xe x + e x
8.
∫ (2 x − 1) log (2 x − 1) dx
− 2 e x [e −2 x (2 x − 1) e2 x log (2 x − 1) dx] dx + c2 e x
∫ ∫
e− x
y = e x log x + c1 e x dx + c2 e x
9.
∫ x
−3 3
10. y = c1 x + c2 x
D-183
x2 sin x  2 sin x  sin x

11. y =
6
+ c1 − x cos x + 2 sin x log sin x −
 x ∫ log sin x dx + c2 x
c − 1 − 1
12. y = − 1 e −2 a sin x + c2  ⋅ e a sin x
 2 a 
4 Removal of the First Derivative

(Reduction to Normal Form)
(Agra 2007; Gorakhpur 05)
If we are unable to obtain a part of the C.F. of the solution of the differential equation
d2 y dy
2
+P + Qy = R, …(1)
dx dx
then we cannot solve (1) by the method given in article 2. In such cases the equation
(1) can sometimes be solved by reducing it into the form in which the term containing
the first derivative is absent. For this we first change the dependent variable from y to v
in the equation (1) by putting y = uv, where u is some function of x.
dy du dv d2 y d2 u dv du d2 v
Then =v +u , = v + 2 ⋅ + u ⋅
dx dx dx dx2 dx2 dx dx dx2
dy d2 y
Putting the values of y , , in (1), we get
dx dx2
 d2 u dv du d2 v   du + u dv + Q vu = R
 v 2
+ 2 ⋅ + u 2  + P v
 dx dx dx dx   dx dx 
d2 v 2
 P + 2 du dv + v  d u + P du + Qu = R.
or u + u   …(2)
dx2  u dx  dx  dx
2 dx 
dv
To remove the term of the first derivative we choose u such that
dx
2 du du 1
P+ =0 or = − P dx
u dx u 2
1 − 1 ∫ P dx
or log u = −
2 ∫ P dx or u=e 2 …(3)
Now the equation (2) becomes

d2 v  d2 u v
du  R
2 2 +P
+ + Qu = ⋅ …(4)
dx  dx u
dx  u
du 1
But from (3), = − Pu,
dx 2
d2 u 1  du dP  1  − 1 Pu + u dP 
=− P +u = − P  
dx 2 2  dx dx  2  2  dx 
D-184
1 2 1 dP
= P u− u ⋅
4 2 dx
d2 v 1 1 dP 1 R
+ v  P2 −
2
− P ⋅ P + Q =
dx 
4 2 dx 2  − 1 ∫ P dx
e 2
d2 v  P2 1 dP  1 P dx
∫ d2 v
or 2
+ v  Q − −  = R e 2 or + Xv = Y ,
dx  4 2 dx  dx2
1 P dx
1 dP 1 2 ∫
where X =Q− − P and Y = Re2 .
2 dx 4
The reduced equation (5) may easily be integrated. The equation (5) is called the
normal form of the equation (1).
Note: The students should remember the values of u, X and Y . They can write the
reduced equation (5) directly.
d2 y dy 2
Example 9: Solve 2
− 4x + (4 x2 − 1) y = − 3 e x sin 2 x .
dx dx (Rohilkhand 2001;
Meerut 05; Agra 07; Gorakhpur 09; Garhwal 10B, 15; Kumaun 06, 13)
2
Solution: Here P = − 4 x, Q = 4 x2 − 1, R = − 3 e x sin 2 x .
− 1 ∫ P dx − 1 ∫ −4 x dx 2
We choose u = e 2 =e 2 = ex .
Putting y = uv in the given equation, it reduces to its normal form
d2 v
+ Xv = Y , …(1)
dx2
1 dP 1 2 1 1
where X =Q− − P = 4 x2 − 1 − (− 4) − ⋅ 16 x2 = 1 ,
2 dx 4 2 4
1 P dx 2 2
∫
and Y = Re2 = −3 e x sin 2 x ⋅ e − x = −3 sin 2 x.
Hence the normal form (1) of the given equation is
d2 v
2
+ v = − 3 sin 2 x or ( D2 + 1) v = − 3 sin 2 x. …(2)
dx
Now (2) is a linear differential equation with constant coefficients.
A.E. is m2 + 1 = 0 ⇒ m2 = − 1 ⇒ m = ± i .
∴ C.F. of the solution of (2) = c1 cos x + c2 sin x .
1 −3
P. I. = 2 (− 3 sin 2 x) = sin 2 x = sin 2 x.
D +1 − 22 + 1
∴ the solution of (2) is v = c1 cos x + c2 sin x + sin 2 x.
D-185
Hence the general solution of the given equation is

2
y = uv = e x (c1 cos x + c2 sin x + sin 2 x) .
d2 y dy
Example 10: Solve 2
− 2 tan x + 5 y = sec x . e x .
dx dx
(Agra 2006; Gorakhpur 06; Kanpur 09; Garhwal 10)
Solution: Here P = − 2 tan x, Q = 5, R = e x sec x .
− 1 ∫ P dx
We choose u=e 2 = e ∫ tan x dx = e log sec x
= sec x.
Substituting y = uv in the given equation, it reduces to its normal form
d2 v
+ Xv = Y , …(1)
dx2
1 dP 1 2 1
where X =Q− − P = 5 − (− 2 sec2 x) − tan2 x = 6
2 dx 4 2
1 P dx
∫
and Y = Re2 = e x sec x (sec x)−1 = e x .
Hence the equation (1) is
d2 v
2
+ 6v = e x or ( D2 + 6) v = e x . …(2)
dx
A.E. is m2 + 6 = 0 ⇒ m = ± √ 6 i .
∴ C.F. of the solution of (2) = c1 cos √ 6 x + c2 sin √ 6 x.
1 1
P. I. = 2 e x = e x.
D +6 7
1 x
∴ the solution of (2) is v = c1 cos √ 6 x + c2 sin √ 6 x + e .
7
Hence the complete solution of the given equation is
1 x
y = uv = sec x (c1 cos √ 6 x + c2 sin √ 6 x + e ).
7
d2 y 1 dy  1 1 6
Example 11: Solve 2
+ 1 /3
+  2 /3 − 4 /3 − 2  y = 0 .
dx x dx  4x 6x x 
1 1 6
Solution: Here P = x −1 /3 , Q = − − ,R=0.
4 x2 /3 6 x4 /3 x2
− 1 ∫ P dx − 1 ∫ x −1 /3 dx − 3 x2 /3
We choose u = e 2 =e 2 =e 4 .
d2 v
+ Xv = Y , …(1)
dx2
1 dP 1 2
where X =Q− − P
2 dx 4
1 1 6 1 1 1 6
= 2 /3 − 4 /3 − 2 −  − x −4 /3  − x −2 /3 = − 2 ,
4x 6x x 2 3  4 x
D-186
1
∫ P dx
and Y = Re2 =0.
d2 v 6 d2 v
2
− 2
v = 0 or x2 − 6v = 0 , …(2)
dx x dx2
which is a homogeneous linear equation.
In order to solve it, putting x = e z , the differential equation (2) becomes
[ D ( D − 1) − 6] v = 0 , where D ≡ d / dz
or ( D2 − D − 6) v = 0 .
A.E. is m2 − m − 6 = 0 , giving m = − 2, 3 .
∴ the solution of (2) is v = c1e −2 z + c2 e3 z = c1 x −2 + c2 x3 .
2 /3
y = uv = e(−3 /4) x (c1 x −2 + c2 x3 ).
d2 y dy 2
Example 12: Solve 2
− 4x + (4 x2 − 3) y = e x .
dx dx
(Gorakhpur 2005, 07; Lucknow 09; Garhwal 11; Kumaun 12)
2
Solution: Here P = − 4 x, Q = 4 x2 − 3, R = e x .
− 1 ∫ P dx 2
We choose u1 = e 2 = e ∫ 2 x dx = e x .
d2 v
+ Xv = Y , …(1)
dx2
1 dP 1 2
where X =Q− − P = −1
2 dx 4
1 P dx
∫
and Y = R e2 = 1.
d2 v
− v = 1 or ( D2 − 1) v = 1 . …(2)
dx2
A.E. is m2 − 1 = 0 , giving m = ± 1.
∴ C.F. of the solution of (2) = c1 e x + c2 e − x .
1 1 1
P. I. = 2
(1) = 2
e0 x = 2
e0 x = − 1 .
D −1 D −1 0 −1
∴ the solution of (2) is v = c1 e x + c2 e − x − 1.
2
y = uv = e x (c1e x + c2 e − x − 1) .
D-187
d2 y dy
Example 13: Solve 2
+ 2x + ( x2 + 1) y = x3 + 3 x .
dx dx
Solution: Here P = 2 x, Q = x2 + 1, R = x ( x2 + 3) .
− 1 ∫ P dx 2
We choose u = e 2 = e− x /2
.

d2 v
+ Xv = Y . …(1)
dx2
1 dP 1 2
where X =Q− − P =0
2 dx 4
1 P dx 2
∫
and Y = R e2 = x( x2 + 3) e x /2
.
d2 v 2
2
= x ( x2 + 3) e x /2
. …(2)
dx
Integrating (2), we get
dv 2
x2 / 2
= x3 e x /2 dx + 3
dx ∫ ∫xe dx + c1
2  x e x2 /2  dx + 3 x2 / 2
=
∫x 


 ∫ xe dx + c1
2 2 2
= x2 e x /2
x ex /2
x ex /2
−2
∫ dx + 3
∫ dx + c1,
integrating the first integral by parts

2 2
= x2 e x /2
+ ex /2
+ c1 .
Integrating again, we get
2 x2 / 2 x2 / 2 2
dx + c1 x + c2 = x e x /2
v=
∫x e dx +
∫e + c1 x + c2 .

2
y = uv = x + (c1 x + c2 ) e − x /2
.
d2 y1 dy 1
Example 14: Solve 2
− + 2 ( x + x1 /2 − 8) y = 0 .
1 /2
dx x dx 4x
1 1
Solution: Here P = − 1 /2 , Q = ( x + x1 /2 − 8), R = 0 .
x 4 x2
−1 /2
− 1 ∫ P dx 1
∫x dx 1 /2
We choose u = e 2 = e2 = ex = e√ x .
d2 v
+ Xv = Y , …(1)
dx2
1 dP 1 2 2
where X =Q− − P =− 2
2 dx 4 x
D-188
1 P dx
∫
and Y = R e2 = 0.
d2 v 2 d2 v
+ v  − 2  = 0
2
or x2 − 2v = 0. …(2)
dx  x  dx2
This is a homogeneous linear equation.
Putting x = e z , the differential equation (2) becomes
{ D ( D − 1) − 2} v = 0 or ( D2 − D − 2) v = 0 ,
where D ≡ d / dz .
A.E. is m2 − m − 2 = 0 or (m − 2) (m + 1) = 0 .
∴ m = 2, − 1 .
Thus the solution of (2) is v = c1 e2 z + c2 e − z = c1 x2 + c2 x −1.
y = uv = e √ x (c1 x2 + c2 x −1).
Solve the following equations by removing the first derivative :

d2 y dy
1. x2 2
− 2 ( x2 + x) + ( x2 + 2 x + 2) y = 0 .
dx dx
(Garhwal 2000; Bundelkhand 01; Kanpur 08)
2
d y dy 2
2. 2
− 4x + (4 x2 − 1) y = − 3 e x (sin 2 x + 5 e −2 x + 6) .
dx dx
d2 y 2 dy  2
3. − + 1 + 2  y = x e x .
dx2 x dx  x  (Kanpur 2009; Garhwal 14)
d  2 dy  2
4. cos x  + y cos x = 0 .
dx  dx  (Rohilkhand 2007)
2
d y 2 dy
5. 2
+ + n2 y = 0 .
dx x dx (Agra 2008)
d  dy  dy
6. x x − y − 2 x + 2 y + x2 y = 0 .
dx  dx  dx (Rohilkhand 2008)
2
d y 2 dy  2 2
7. 2
− +  a + 2  y = 0 .
dx x dx  x 
 d2 y   dy 
8.  2 + y cot x + 2  + y tan x = sec x .
 dx   dx 
(Garhwal 2008)
2
d y dy
9. x2 (log x)2 2
− 2 x log x + [2 + log x − 2 (log x)2 ] y = x2 (log x)3 .
dx dx
D-189
d2 y dy
10. 2
− 2 tan x +5y =0.
dx dx (Agra 2008)
A nswers 2
1. y = x e x (c1 x + c2 )
2
2. y = e x (c1 cos x + c2 sin x + sin 2 x − 3 e −2 x − 18)
1 x
3. y = x (c1 cos x + c2 sin x + e )
2
4. y = (c1 cos √ 2 x + c2 sin √ 2 x) sec x
1 1
5. y = c1 cos (nx + c2 ) or y = (c1 e nx + c2 e − nx )
x x
6. y = x (c1 cos x + c2 sin x)
7. y = x c1 cos (ax + c2 ) or y = x (c1 cos ax + c2 sin ax)
1
8. y = sin x + (c1 x + c2 ) cos x
2
9. y = { c1 x2 + c2 x −1 + 1
3
x2 log x} . log x
10. y = (sec x) . { c1 cos √ 6 x + c2 sin √ 6 x}
5 Transformation of the Equation by Changing

the Independent Variable
Sometimes the equation may become easily integrable by changing the independent
variable.
Let the linear equation of second order be
d2 y dy
+P + Q y = R. …(1)
dx2 dx
Let the independent variable be changed from x to z with the help of a relation of the
form z = f ( x).
2
dy dy dz d2 y d2 y 2
 dz  + dy ⋅ d z ⋅
Then = ⋅ , = 2  
dx dz dx dx2 dz  dx  dz dx2
Substituting these values in (1), we get
d2 y  dz  2 dy d2 z  dy dz 
2  dx 
+ ⋅ 2 +P ⋅ + Qy = R
dz dz dx  dz dx 
d2 y  dz  2 dy  d2 z dz 
or   +  2 + P  + Qy = R
dz 2  dx  dz  dx dx 
D-190
d2 z dz
2 +P
d y 2 dx ⋅ dy + Q y = R
or + dx
dz 2  dz 
2 dz  dz  2  dz 
2
     
 dx   dx   dx 
d2 y dy
or + P1 + Q1 y = R1, …(2)
dz 2 dz
 d2 z 2 2 2
dz   dz  dz dz
where P1 =  2 + P    , Q1 = Q   and R1 = R   .
 dx dx   dx   dx   dx 
Here P1, Q 1 and R1 are functions of x and may be expressed as functions of z with the
help of the relation z = f ( x) .
How to choose z ? We would like to choose z in such a way that the equation (2) can
be easily integrated.
Case I: Let us choose z in such a way that P1 vanishes i. e.,
d2 z dz d  dz  dz
2
+P =0 ⇒   +P =0
dx dx dx  dx  dx
dz
= e − ∫ P dx ⇒ z = − ∫ P dx
⇒
dx ∫ [e ] dx .
d2 y
Then the equation (2) is reduced to + Q 1 y = R1 .
dz 2
Now this equation can be easily solved, if
(i) Q 1 is a constant because then it is a linear equation with constant coefficients,
(ii) Q1 is of the form (constant)/z 2 because then it is a linear homogeneous equation
with variable coefficients.
Case II: Suppose we choose z such that
Q
Q1 = 2
= ± a2 (i. e., a constant)
 dz 
 
 dx 
dz
or a
dx
= √ (± Q) or az = √ (± Q) dx ,
∫
(+ ive or – ive sign is taken to make the expression under the radical sign + ive).
With this choice of z, the equation (2) is reduced to
d2 y dy
2
+ P1 + a2 y = R1.
dz dz
This equation can be easily solved if P1 comes out to be a constant because then it
is a linear equation with constant coefficients.
Note 1: Students should remember the values of P1, Q1 and R1 in equation (2).
Note 2: There are only two choices for z, either P1 = 0 or Q 1 = a2 . Sometimes it is
possible to make both the choices to get the solution of the given equation.
D-191
d2 y dy
− − 4 x3 y = 8 x3 sin x2 .
dx dx
(Meerut 2004, 13B; Garhwal 09; Agra 05; Gorakhpur 09; Kumaun 08, 11)
Solution: The given equation written in the standard form is

d2 y 1 dy
2
− − 4 x2 y = 8 x2 sin x2 . …(1)
dx x dx
Here P = − 1/ x, Q = − 4 x2 , R = 8 x2 sin x2 .
Changing the independent variable from x to z by a relation of the form z = f ( x), the
given equation is transformed into
d2 y dy
2
+ P1
+ Q 1 y = R1, …(2)
dz dz
d2 z dz
2
+P
where P1 = dx dx , Q = Q , R = R ⋅
2 1 2 1 2
 dz   dz   dz 
     
 dx   dx   dx 
− 4 x2
Choosing z such that Q 1 = = constant = −1 (say), we have
(dz / dx)2
(dz / dx)2 = 4 x2 . ∴ dz / dx = 2 x ⇒ z = x2 .
1
2− ⋅2x
Now P1 = x = 0 , R1 = 2 sin x2 = 2 sin z .
4 x2
∴ The transformed equation (2) is
d2 y
− y = 2 sin z or ( D2 − 1) y = 2 sin z . …(3)
dz 2
A.E. is m2 − 1 = 0 ⇒ m2 = 1 ⇒ m = ± 1 .
∴ C.F. of the solution of (3) = c1 e z + c2 e − z .
1 2
P. I. = 2
2 sin z = sin z = − sin z .
D −1 − 1− 1
∴ Solution of the equation (3) is
y = c1e z + c2 e − z − sin z .
Hence the complete solution of the equation (1) is
2 2
y = c1e x + c2 e − x − sin x2 .
d2 y dy
Example 16: Solve + cot x + 4 y cosec2 x = 0 .
dx2 dx
(Meerut 2001; Agra 06; Gorakhpur 05)
D-192
Solution: Here P = cot x, Q = 4 cosec2 x, R = 0 .
Changing the independent variable from x to z by a relation of the form z = f ( x), the
given equation is transformed into
d2 y dy
2
+ P1 + Q 1 y = R1, …(1)
dz dz
d2 z dz
2
+P
2 1 2 1 2
 dz   dz   dz 
     
 dx   dx   dx 
4 cosec2 x
Choosing z such that Q1 = = constant =1 (say), we have
(dz / dx)2
(dz / dx)2 = 4 cosec2 x
dz x
or = 2 cosec x or z = 2 log tan ⋅
dx 2
− 2 cosec x cot x + 2 cosec x cot x
Now P1 = = 0 , R1 = 0 .
4 cosec2 x
∴ the transformed equation (1) is
d2 y
+ y =0 or ( D2 + 1) y = 0 . …(2)
dz 2
A.E. is m2 + 1 = 0 ⇒ m2 = − 1 ⇒ m = ± i .
∴ Solution of the equation (2) is
y = c1 cos z + c2 sin z .
x x
y = c1 cos 2 log tan  + c2 sin 2 log tan  .
 2  2
d2 y dy
Example 17: Solve cos x 2
+ sin x − 2 y cos3 x = 2 cos5 x .
dx dx
(Meerut 2006; Rohilkhand 07; Agra 07; Avadh 11; Purvanchal 06, 10)

d2 y dy
+ tan x − 2 y cos2 x = 2 cos4 x. …(1)
dx2 dx
Here P = tan x, Q = − 2 cos2 x, R = 2 cos4 x.
Changing the independent variable from x to z by a relation of the form z = f ( x),
the given equation is transformed into
d2 y dy
2
+ P1 + Q1 y = R1, …(2)
dz dz
D-193
d2 z dz
2
+P
2 1 2 1 2
dz
   dz   dz 
     
 dx   dx   dx 
− 2 cos2 x
Let us choose z such that Q1 = = constant = − 2 (say).
(dz / dx)2
2
Then  dz  = cos2 x or
dz
= cos x or z = sin x .
 
 dx  dx
Now P1 = 0 , R1 = 2 cos2 x = 2 (1 − sin2 x) = 2 (1 − z 2 ) .
∴ The transformed equation (2) is

d2 y
− 2 y = 2 (1 − z 2 ) . …(3)
dz 2
A.E. is m2 − 2 = 0 , giving m = ± √ 2.
∴ C.F. of the solution of (3) = c1 e √2 z + c2 e −√2 z .

−1
1  D2 
P. I. = 2
2 (1 − z 2 ) = − 1 −  (1 − z 2 )
D −2  2 
1 2 1
= − (1 + D + … ) (1 − z 2 ) = − (1 − z 2 ) + ⋅ 2 = z 2 .
2 2
∴ The solution of the equation (3) is
y = c1 e √2 z + c2 e −√2 z + z 2.
y = c1e √2 sin x + c2 e −√2 sin x + sin2 x.
d2 y dy 1
Example 18: Solve x6 + 3 x5 + a2 y = 2 ⋅
dx2 dx x (Garhwal 2001)

d2 y 3 dy a2 1
2
+ + y= 8 ⋅
dx x dx x6 x
3 a2 1
Here P= ,Q = 6 , R= 8 ⋅
x x x
To solve this differential equation we change the independent variable from x to z by
choosing z such that
Q a2 / x6
2
= = constant = a2 (say).
(dz / dx) (dz / dx)2
2
Then  dz  = 1 or
dz
=
1
or z = − 1/(2 x2 ).
 
 dx  x6 dx x3
D-194
Now by the substitution z = − 1/(2 x2 ), the given differential equation is transformed

into
d2 y dy
2
+ P1 + Q1 y = R1, where
dz dz
d2 z
dz
+P 4 3
P1 = dx2 dx = (−3 / x ) + (3 / x)(1/ x ) = 0 ,
(dz / dx)2 (dz / dx)2
Q R 1/ x8 1
Q1 = 2
= a2 , R1 = 2
= = = −2 z .
(dz / dx) (dz / dx) 1/ x6 x2
∴ the transformed equation is
d2 y
+ a2 y = − 2 z or ( D2 + a2 ) y = − 2 z . …(1)
dz 2
The C.F. of the solution of (1) = c1 cos az + c2 sin az .
−1
1 1  D2 
P. I. = (− 2 z ) = 1 +  (−2 z )
D2 + a2 a2  a2 
1  1 1
= 2 
1 − 2 D2 + … (−2 z ) = 2 (−2 z ).
a a  a
∴ the solution of the equation (1) is
1
y = c1 cos az + c2 sin az − ⋅ 2z .
a2
a a 1
y = c1 cos 2 − c2 sin 2 + 2 2
2x 2x a x
a a 1
or y = c1 cos 2 + c2 sin 2 + 2 2 ⋅
2x 2x a x
d2 y dy
Example 19: Solve 2
+ (tan x − 1)2 − n (n − 1) y sec 4 x = 0 .
dx dx
Solution: Comparing the given differential equation with the standard form
d2 y dy
+P + Qy = R, we have
dx2 dx
P = (tan x − 1)2 , Q = − n (n − 1) sec4 x, R = 0 .
To solve the given differential equation we change the independent variable from x to
z by choosing z such that
Q − n (n − 1) sec4 x
= = constant = − n (n − 1), say.
(dz / dx)2 (dz / dx)2
2
Then  dz  = sec4 x or
dz
= sec2 x or z = tan x.
 
 dx  dx
D-195
Now by the substitution z = tan x,the given differential equation is transformed into
d2 y dy
2
+ P1 + Q 1 y = R1, where
dz dz
d2 z dz
+P 2 2 2
P1 = dx
2dx = 2 sec x tan x + (tan x − 1) sec x = 1,
(dz / dx)2 sec4 x
Q R
Q1 = = − n (n − 1), R1 = =0.
(dz / dx)2 (dz / dx)2
∴ the transformed equation is
d2 y dy
+ − n (n − 1) y = 0 or { D2 + D − n (n − 1)} y = 0 . …(1)
dz 2 dz
A.E. is m2 + m − n (n − 1) = 0 or (m + n) { m − (n − 1)} = 0 .
∴ m = − n, n − 1 .
∴ the solution of the equation (1) is
y = c1e − nz + c2 e(n − 1)z .
y = c1e − n tan x + c2 e(n − 1) tan x .
Solve the following equations :

d2 y dy
1. 2
+ (3 sin x − cot x) + 2 y sin2 x = e − cos x sin2 x.
dx dx
d2 y dy x
2. 2
− (1 + 4 e x ) + 3 e2 x y = e2( x + e ).
dx dx (Gorakhpur 2008, 11)
2
d y 1 dy
3. 2
+ 1 −  + 4 x2 ye − 2 x = 4( x2 + x3 ) e −3 x .
dx  x  dx (Rohilkhand 2006)
2 2 2
d y a dy x
4. (a2 − x2 ) 2
− + y = 0.
dx x dx a (Bundelkhand 2001)
d2 y 1 dy
5. − + 4 x2 y = x4 .
dx2 x dx
d2 y dy
6. x 2
+ (4 x2 − 1) + 4 x3 y = 2 x3 .
dx dx
d2 y dy
7. (1 + x)2 + (1 + x) + y = 4 cos log (1 + x) .
dx2 dx (Gorakhpur 2010)
d2 y dy
8. − cot x − y sin2 x = 0 .
dx2 dx
D-196
d2 y dy
9. x4 2
+ 2 x3 + n2 y = 0 .
dx dx (Kumaun 2010)
2
d y dy
10. (1 + x2 )2 + 2 x (1 + x2 ) +4y =0.
dx2 dx
A nswers 3
1 − cos x
1. y = c1e cos x + c2 e2 cos x + e
6
x x x
2. y = c1e3 e + c2 e e − e2 e
3. y = c1 cos {2 e − x (1 + x)} − c2 sin {2 e − x (1 + x)} + e − x (1 + x)
 √ (a2 − x2 )  √ (a2 − x2 )
4. y = c1 cos   + c2 sin  
 √a   √a 
1
5. y = c1 cos ( x2 + c2 ) + x2
4
2 1
6. y = (c1 + c2 x2 ) e − x +
2
7. y = c1 cos {log (1 + x) + c2 } + 2 log (1 + x) ⋅ sin log (1 + x)
8. y = c1e − cos x
+ c2 e cos x − cos x
n
9. y = c1 cos  + c2 
x 
10. y = c1 cos (2 tan−1 x) + c2 sin (2 tan−1 x), or y (1 + x2 ) = c1 (1 − x2 ) + 2 c2 x.
6 Method of Variation of Parameters

Now we shall discuss an elegant but somewhat artificial method for finding the
complete primitive of a linear equation whose complementary function is known.
Let the linear equation of second order be
d2 y dy
+P + Q y = R. …(1)
dx2 dx
Suppose y = Au + Bv is the complementary function of (1), where A and B are
arbitrary constants and u and v are functions of x. Then y = Au + Bv is the solution of
d2 y dy
2
+P + Qy = 0 .
dx dx
∴ ( Au2 + Bv2 ) + P ( Au1 + Bv1) + Q ( Au + Bv) = 0
or A (u2 + Pu1 + Qu) + B (v2 + Pv1 + Qv) = 0
⇒ u2 + Pu1 + Qu = 0 and v2 + Pv1 + Qv = 0 . …(2)
D-197
Now, suppose that

y = Au + Bv …(3)
is the complete primitive of (1) where A and B are not constants but functions of x, so
chosen that (1) will be satisfied.
Differentiating (3),
dy
= A1u + B1v + Au1 + Bv1.
dx
Let us choose A and B such that
A1u + B1v = 0 . …(4)
2
dy d y
Then = Au1 + Bv1 and = Au2 + Bv2 + A1 u1 + B1 v1.
dx dx2
( Au2 + Bv2 + A1u1 + B1v1) + P ( Au1 + Bv1) + Q ( Au + Bv) = R
or A (u2 + Pu1 + Qu) + B (v2 + Pv1 + Qv) + A1u1 + B1 v1 = R
or A1 u1 + B1v1 = R1, [using 2] …(5)
Solving (4) and (5), we get
vR uR
A1 = and B1 = , which on integration give A and B.
u1v − uv1 v1u − vu1
v R dx u R dx
∴ A=
∫ u1v − uv1 + c1 and B=
∫ v1u − vu1 + c2 ,
where c1 and c2 are parameters.
Substituting these values of A and B in (3), we get the complete primitive of (1).
Note : Since the general solution is obtained by varying the arbitrary constants of the
complementary function, therefore the method is known as method of variation of
parameters.
7 Wronskian Method
Alternate method to find C.S:
Consider the linear equation of second order
d2 y dy
2
+P + Qy = R . …(1)
dx dx
The P.I. of (1) is given by
vR uR vR uR
P.I. = − u
∫
W
dx + v
∫ W dx = − u ∫ uv1 − vu1 dx + v ∫ uv1 − vu1 dx , …(2)
where u and v are defined by the C.F. of
d2 y dy
2
+P + Qy = 0 …(3)
dx dx
i. e., by y = c1 u + c2 v, where c1 and c2 are arbitrary constants
D-198
u v
and W = is called the Wronskian of u and v, also written as W (u v).
u1 v1
∴ C. S. = C. F. + P. I.
Note : The method of variation of parameters must be used, if instructed to do so in the
question.
Example 20: Solve by the method of variation of parameters

d2 y
+ 4 y = 4 tan 2 x.
dx2 (Rohilkhand 2001, 11; Lucknow 09;
Garhwal 05; Kumaun 06, 13; Avadh 05; Meerut 04B)
Solution: The C.F. of the given equation i. e., the solution of the equation
d2 y
+ 4 y = 0 is
dx2
y = c1 cos 2 x + c2 sin 2 x, where c1 and c2 are constants.
Let y = A cos 2 x + B sin 2 x …(1)
be the complete primitive of the given equation where A and B are functions of x, so
chosen that the given equation will be satisfied.
dy dA dB
Then = − 2 A sin 2 x + 2 B cos 2 x + cos 2 x + sin 2 x .
dx dx dx
dA dB
cos 2 x + sin 2 x = 0 . …(2)
dx dx
dy
Then = − 2 A sin 2 x + 2 B cos 2 x
dx
d2 y dA dB
and 2
= −2 sin 2 x + 2 cos 2 x − 4 A cos 2 x − 4 B sin 2 x.
dx dx dx
Putting these values in the given equation, we get
dA dB
− 2 sin 2 x + 2 cos 2 x = 4 tan 2 x
dx dx
dA dB
or − sin 2 x + cos 2 x = 2 tan 2 x . …(3)
dx dx
dA 2 sin2 2 x dB
=− , = 2 sin 2 x.
dx cos 2 x dx
Integrating these, we get
(1 − cos2 2 x)
A= −2
∫ cos 2 x ∫
dx + c1 = − 2 (sec 2 x − cos 2 x) dx + c1
= − log (sec 2 x + tan 2 x) + sin 2 x + c1

D-199
and B = − cos 2 x + c2 .
Putting the values of A and B in (1), the complete primitive of the given equation
is
y = c1 cos 2 x + c2 sin 2 x − [ log (sec 2 x + tan 2 x)] ⋅ cos 2 x.
Alternate method to find P.I.:
Here u = cos 2 x, v = sin 2 x, R = 4 tan 2 x
u v cos 2 x sin 2 x
and W= = =2.
u1 v1 − 2 sin 2 x 2 cos 2 x
vR uR
Thus, P.I. = − u
∫ W dx + v ∫ W dx
∫ ∫
= − cos 2 x 2 sin 2 x tan 2 x dx + sin 2 x 2 cos 2 x tan 2 x dx
∫ ∫
= − 2 cos 2 x (sec 2 x − cos 2 x) dx + 2 sin 2 x sin 2 x dx
= − cos 2 x [log (sec 2 x + tan 2 x) − sin 2 x] − sin 2 x cos 2 x .

= − cos 2 x log (sec 2 x + tan 2 x) .
Hence, the C.S. is y = c1 cos 2 x + c2 sin 2 x − (cos 2 x) log (sec 2 x + tan 2 x) .
Example 21: (i) Solve by the method of variation of parameters

d2 y dy
2
−2 + y = e x log x .
dx dx
Solution: First we shall find the C.F. of the given equation i. e., the solution of the
equation
d2 y dy
2
−2 + y =0 or ( D2 − 2 D + 1) y = 0 .
dx dx
A.E. is m2 − 2 m + 1 = 0 or (m − 1)2 = 0 , giving m = 1, 1 .
∴ C.F. is y = (c1 + c2 x) e x .
Now assume A and B as the functions of x in such a way that the given equation is
satisfied by
y = ( A + Bx) e x = Ae x + Bxe x . …(1)
dy dA dB
∴ = Ae x + B (e x + xe x ) + e x + xe x
dx dx dx
dA dB
= Ae x + B (e x + xe x ), assuming e x + xe x =0. …(2)
dx dx
d2 y dA dB
∴ 2
= Ae x + B (2 e x + xe x ) + e x + (e x + xe x ) ⋅
dx dx dx
Now putting these values in the given equation, we get
dA dB
ex + e x (1 + x) = e x log x . …(3)
dx dx
D-200

dA dB
= − x log x and = log x .
dx dx
 x2 1 x2 
A= −
∫ x log x dx + c1 = − 
2
log x −
∫ ⋅
x 2
dx + c1

 x2 x2 
=− log x −  + c1,
2 4
1
B=
∫ log x dx + c2 = x log x − ∫ x ⋅ x dx + c2 = x log x − x + c2 .
Putting these values of A and B in (1), the general solution of the given equation is
 x2 x2 
y = − ex  log x − + c1 + x e x ( x log x − x + c2 )
2 4 
1 2 x
= (c1 + c2 x) e x + x e (2 log x − 3) .
4
Alternate method to find P.I.
Here u = e x , v = x e x , R = e x log x
u v ex xex
and W= = = e2 x .
u1 v1 x x
e (1 + x) e
uR uR
Thus P.I. = − u
∫ W dx + v ∫ W dx
xe x ⋅ e x log x e x ⋅ e x log x
= − ex dx + xe x
∫ e2 x ∫ e2 x
dx
= − e x x log x dx + xe x log x dx
∫ ∫
 x2 1 x2  1
= − ex  dx + x ⋅ e x  x log x − 
2
log x −
∫ ⋅
x 2   ∫ x ⋅ x dx
 x2 x2  1
= − ex  log x −  + x e x ( x log x − x) = x2 e x (2 log x − 3) .
 2 4  4
1
Hence, the C.S. is y = (c1 + c2 x) e x + x2 e x (2 log x − 3).
4
Example 22: (ii) Solve by the method of variation of parameters
d2 y
2
+ n2 y = sec nx .
dx (Meerut 2003, 07; Gorakhpur 11;
Garhwal 07, 11; Kumaun 09, 12; Lucknow 10; Avadh 09, 10, 13)
Solution: The C.F. of the given equation i. e. the solution of the equation
d2 y
+ n2 y = 0 is
dx2
y = c1 cos nx + c2 sin nx, where c1 and c2 are constants.
D-201
Let y = A cos nx + B sin nx …(1)

be the general solution of the given equation where A and B are functions of x, so
dy dA dB
Then = − An sin nx + Bn cos nx + cos nx + sin nx .
dx dx dx
dA dB
cos nx + sin nx = 0 …(2)
dx dx
dy
Then = − An sin nx + Bn cos nx
dx
d2 y dA dB
and = − An2 cos nx − Bn2 sin nx − n sin nx + n cos nx .
dx2 dx dx
dA dB
−n sin nx + n cos nx = sec nx . …(3)
dx dx
dA 1 dB 1
= − tan nx and = ⋅
dx n dx n
1 x
A = 2 log cos nx + c1, B = + c2 .
n n
Putting these values of A and B in (1), the general solution of the given equation is
1 x
y = c1 cos nx + c2 sin nx + 2 cos nx log cos nx + sin nx .
n n
d2 y
Example 23: Apply the method of variation of parameters to solve + y = tan x.
dx2
d2 y
2
+ y = tan x or ( D2 + 1) y = tan x …(1)
dx
A.E. is m2 = − 1 or m=±i
∴ C.F. is y = A cos x + B sin x
Here u = cos x, v = sin x, R = tan x
and W = uv1 − u1 v = cos2 x + sin2 x = 1
vR uR
∴ P.I. = −
∫ W
dx + v
∫ W dx .
− vR sin x tan x sin2 x 1 − cos2 x
We have
∫ W
dx = −
∫ 1
dx = −
∫ cos x
dx = −
∫ cos x
dx
=
∫ (cos x − sec x) dx = sin x − log (sec x + tan x)
D-202
uR cos x ⋅ tan x
and
∫ W dx = ∫ 1
=
∫ sin x dx = − cos x
∴ P. I. = [sin x − log (sec x + tan x)] cos x − cos x sin x
= − cos x log (sec x + tan x)
Hence the complete solution is y = A cos x + B sin x − cos x log (sec x + tan x).

d2 y dy
x2 2
+x − y = x2 e x .
dx dx
(Rohilkhand 2006, 11; Purvanchal 06; Meerut 07;
Garhwal 01, 10; Gorakhpur 06, 10; Lucknow 08, 11; Avadh 14; Kanpur 15)
Solution: First we shall find the C.F. of the given equation i.e., the solution of the
equation
d2 y dy
x2 2
+x − y =0. …(1)
dx dx
This is a homogeneous linear equation. Putting x = e z , the differential equation (1)
becomes
d
{ D ( D − 1) + D − 1} y = 0 , where D ≡
dz
or ( D2 − 1) y = 0
whose solution is y = c1 e z + c2 e − z .
1
∴ solution of (1) is y = c1 x + c2 ⋅
x
B
Let y = Ax + …(2)
x
dy B dA 1 dB
Then = A− 2 + x+ ⋅ ⋅
dx x dx x dx
dA 1 dB
x + ⋅ =0. …(3)
dx x dx
dy B d2 y dA 1 dB 2
Then = A− 2 and 2
= − 2 + 3 B.
dx x dx dx x dx x
dA 1 dB
− 2 = ex . …(4)
dx x dx
dA 1 x dB x2 x
= e and =− e .
dx 2 dx 2
D-203

1 1 2 x
A = e x + c1 and B=− x e + xe x − e x + c2 .
2 2
Putting the values of A and B in (2), the complete primitive of the given equation
is
c 1
y = c1 x + 2 + e x − e x .
x x

d2 y dy
(1 − x) 2
+x − y = (1 − x)2 .
dx dx (Kumaun 2011)
Solution: First we shall find the C.F. of the given equation i. e., the solution of the
equation
d2 y dy
(1 − x) 2
+x − y =0
dx dx
d2 y x dy y
or 2
+ − =0. …(1)
dx 1 − x dx 1 − x
Here P + Qx = 0.
∴ y = x is a part of the C.F.
Putting y = vx in (1), the reduced equation is
d2 v  x 2  dv
2
+ +  =0
dx 1 − x x  dx
dp  x 2 dv
or + +  p = 0, where p =
dx 1 − x x  dx
dp  1 2 dp  1 2
or + − 1 − +  p=0 or = 1 + −  dx .
dx  x − 1 x p  x − 1 x
Integrating, log p = x + log ( x − 1) − 2 log x + log c1
dv c1 ( x − 1) e x 1 1
or p= = 2
= c1  − 2  e x .
dx x  x x 
Integrating again, we get
1 1 c1 x
v = c1  − 2  e x dx + c2
∫
x x 
or v=
x
e + c2 .
∴ solution of (1) is y = vx = c1 e x + c2 x .
Let y = Ae x + Bx …(2)
dy dA dB
Then = Ae x + B + e x +x ⋅
dx dx dx
D-204
dA dB
ex +x =0. …(3)
dx dx
dy d2 y dA dB
Then = Ae x + B and 2
= ex + + ex A .
dx dx dx dx
dA dB
ex + = 1 − x. …(4)
dx dx
dA dB
= − xe − x and = 1.
dx dx
A = (1 + x) e − x + c1 and B = x + c2 .
Putting the values of A and B in (2) the general solution of the given equation is
y = c1 e x + c2 x + 1 + x + x2 .

d2 y dy
+ (1 − cot x) − y cot x = sin2 x.
dx2 dx (Kumaun 2008; Garhwal 12)
Solution: First we shall find the C.F. of the given equation i.e., the solution of the
equation
d2 y dy
2
+ (1 − cot x) − y cot x = 0 . …(1)
dx dx
Here 1 − P + Q = 0 , ∴ y = e − x is a part of the C.F.
Putting y = ve − x , the equation (1) reduces to
d2 v dv dp dv
2
− (1 + cot x) =0 or − (1 + cot x) p = 0 , where p =
dx dx dx dx
dp
or = (1 + cot x) dx .
p
Integrating, log p = x + log sin x + log c1.
dv
∴ p= = c1e x sin x
dx
x 1 x
or v = c1
∫e sin x dx + c2 = c1 ⋅
2
e (sin x − cos x) + c2 .
∴ the solution of (1) i.e., the C.F. of the given equation is

1
y = ve − x = c1 ⋅ (sin x − cos x) + c2 e − x .
2
Let y = A (sin A − cos x) + Be − x …(2)
be the complete solution of the given equation where A and B are functions of x, so
dy dA dB − x
∴ = A (cos x + sin x) − Be − x + (sin x − cos x) + e .
dx dx dx
D-205

dA dB − x
(sin x − cos x) + e =0. …(3)
dx dx
dy
∴ = A (cos x + sin x) − Be − x
dx
d2 y dA dB − x
and 2
= (cos x + sin x) − e + A (− sin x + cos x) + Be − x .
dx dx dx
dA dB − x
(cos x + sin x) − e = sin2 x. …(4)
dx dx
dA 1 dB 1 x
= sin x and = e (sin x cos x − sin2 x) .
dx 2 dx 2
1
A = − cos x + c1
2
1
e x (sin 2 x − 1 + cos 2 x) dx + c2
and B=
4 ∫
1 ex ex 1 ex
= ⋅ (sin 2 x − 2 cos 2 x) − + ⋅ (cos 2 x + 2 sin 2 x) + c2
4 5 4 4 5
ex ex
= (3 sin 2 x − cos 2 x) − + c2 .
20 4
Putting the values of A and B in (2), the general solution of the given equation is
1
y = c1 (sin x − cos x) + c2 e − x − (sin 2 x − 2 cos 2 x) .
10
Solve the following differential equations by the method of variation of

parameters:
d2 y
1. + y = x.
dx2 (Agra 2007)
2
d y
2. 2
+ a2 y = cosec ax.
dx (Kumaun 2010)
d2 y dy
3. x2 − 2 x (1 + x) + 2 ( x + 1) y = x3 .
dx2 dx (Gorakhpur 2007; Garhwal 13)
d2 y dy
4. (1 − x2 ) − 4x − (1 + x2 ) y = x .
dx2 dx (Garhwal 2002, 08)
2
d y dy
5. (1 − x2 ) +x − y = x (1 − x2 )3 /2 .
dx2 dx
D-206
d2 y dy
6. ( x + 2) 2
− (2 x + 5) + 2 y = ( x + 1) e x .
dx dx (Garhwal 2009)
2
d y
7. + y = cosec x .
dx2 (Purvanchal 2007, 10; Kumaun 14)
2
d y 2
8. − y= ⋅
dx2 1+ ex (Agra 2008; Gorakhpur 05; Purvanchal 09; Kanpur 07, 14;
Kumaun 07, 15)
9. xy1 − y = ( x − 1) ( y2 − x + 1) .
A nswers 4
1. y = c1 cos x + c2 sin x + x
x 1
2. y = c1 cos ax + c2 sin ax − cos ax + 2 sin ax log sin ax
a a
1 2 1
3. y = c1 x + c2 x e2 x − x − x
2 4
1
4. y= ⋅ (c1 cos x + c2 sin x + x)
(1 − x2 )
5. y = c1 [√ (1 − x2 ) + x sin−1 x] + c2 x − ( x / 9) (1 − x2 )3 /2
6. y = c1 (2 x + 5) + c2 e2 x − e x
7. y = c1 cos x + c2 sin x − x cos x + sin x log sin x
1 + e x 
8. y = c1 e x + c2 e − x + e x log  x  − 1 − e − x log (1 + e x )
 e 
9. y = c1 e x + c2 x − (1 + x2 )
8 Method of Operational Factors

In some cases the method of operational factors easily solves the equation.
Let the linear equation of second order
d2 y dy
2
+P + Qy = R ,
dx dx
be expressed in the form f ( D) y = R
or f1 ( D) f2 ( D) y = R
i. e., f ( D) can be resolved into a product of two factors f1( D) and f2 ( D), such that, if
f2 ( D) operates upon y, and then f1( D) operates upon the result of the operation, the
same result is obtained as if f ( D) operates upon y.
Note: With the exception of the classes of equations, the factors of f ( D) are generally
not commutative. Hence a great care is to be taken in writ ing them in the right order.
D-207
d2 y dy
+ (1 − x) − y = e x.
dx dx
Solution: In the symbolic form, the given equation is
[ xD2 + (1 − x) D − 1] y = e x , where D ≡ d / dx
or ( xD + 1) ( D − 1) y = e x . …(1)
[Note that here the factors are not commutative because on expansion
{( D − 1)( xD + 1)} y gives { xD2 + (2 − x) D − 1} y] .
Let ( D − 1) y = v . …(2)
dv dv 1 ex
Then (1) gives ( xD + 1) v = e x or x + v = ex or + v= ,
dx dx x x
which is a linear differential equation of the first order.
Now I.F. = e ∫ (1 / x) dx = e log x
= x.
ex
dx + c1 = e x + c1 v = e x x −1 + c1 x −1.
∴ vx =
∫x x
or
Putting the value of v in (2), we get

( D − 1) y = e x x −1 + c1 x −1
dy
or − y = e x x −1 + c1 x −1, which is again linear.
dx
Here I.F. = e ∫ − dx = e − x .
ye − x = x −1
+ c1 x −1) e − x dx + c2
∴
∫ (e x
1 e− x e− x
=
∫ x
dx + c1
∫ x
dx + c2 = log x + c1
∫ x
dx + c2 .
e− x
y = e x log x + c1e x dx + c2 e x ,
∴
∫ x
which is the required general solution of the given differential equation.
d2 y dy
Example 28: Solve 3 x2 2
+ (2 + 6 x − 6 x2 ) −4y =0.
dx dx
[3 x2 D2 + (2 + 6 x − 6 x2 ) D − 4] y = 0
or [(3 x2 D2 + 6 x D + 2 D) − (6 x2 D + 4)] y = 0
or [ D (3 x2 D + 2) − 2 (3 x2 D + 2)] y = 0
or ( D − 2) (3 x2 D + 2) y = 0 . …(1)
D-208
[Here factors are not commutative because on expansion

(3 x2 D + 2) ( D − 2) y gives {3 x2 D2 + (2 − 6 x2 ) D − 4} y].
Let (3 x2 D + 2) y = v . …(2)
dv
Then (1) gives ( D − 2) v = 0 ⇒ − 2v = 0
dx
dv
= 2v ⇒ v = c1e2 x .
dx
(3 x2 D + 2) y = c1 e2 x
dy dy 2 c
or 3 x2 + 2 y = c1e2 x ⇒ + y = 12 e2 x .
dx dx 3 x2 3x
2 dx
∴ I. F. = e
∫
3 x2 = e3
2 −1
x( ) = e−32x .
−2 c1 −2 2 x −2
∴ ye 3x =
3 ∫x e ⋅e 3x dx + c2
2
c1 32x 2
−2 2 x − 3 x
or y = c2 e3 x +
3
e
∫x e dx,
which is the required solution.
d2 y dy
− ( x + 2) + 2 y = x3 .
dx dx
[ xD2 − ( x + 2) D + 2] y = x3
or ( xD − 2) ( D − 1) y = x3 …(1)
{( D − 1) ( xD − 2)} y gives { xD2 − ( x + 1) D + 2} y].
Let ( D − 1) y = v . …(2)
Then (1) gives
dv
( xD − 2) v = x3 or x − 2 v = x3
dx
dv 2
or − v = x2 , which is linear.
dx x
1
∴ I. F. = e ∫ (−2 / x) dx = e −2 log x = 2
x ⋅
1  2 1
∴ v⋅
x 2
=
∫  x ⋅  dx + c1 = x + c1.
x2 
∴ v = x3 + c1 x2 .
D-209
dy
( D − 1) y = x3 + c1 x2 or − y = x3 + c1 x2 , which is again linear.
dx
∴ I. F. = e ∫ − dx = e − x .
ye − x = −x
( x3 + c1 x2 ) dx + c2
∴
∫e
= − ( x3 + c1 x2 ) e − x − (3 x2 + 2 c1 x) e − x − (6 x + 2 c1) e − x − 6 e − x + c2
or y = − ( x3 + c1 x2 ) − (3 x2 + 2 c1 x) − (6 x + 2 c1) − 6 + c2 e x
= − x3 − (c1 + 3) x2 − 2 (c1 + 3) x − 2 (c1 + 3) + c2 e x
or y = − x3 − (c1 + 3) { x2 + 2 x + 2} + c2 e x .
Example 30: Solve [( x + 3) D2 − (2 x + 7) D + 2] y = ( x + 3)2 e x .

{( x + 3) D − 1} ( D − 2) y = ( x + 3)2 e x . …(1)

( D − 2) {( x + 3) D − 1} y gives {( x + 3) D2 − (2 x + 5) D + 2} y]
Let ( D − 2) y = v . …(2)
dv
Then (1) gives {( x + 3) D − 1} v = ( x + 3)2 e x or ( x + 3) − v = ( x + 3)2 e x
dx
dv 1
or − v = ( x + 3) e x , which is linear.
dx x + 3
dx
−∫ 1
∴ I. F. = e x +3 = e − log ( x +3) = ⋅
x +3
1 1 x
dx + c1 = e x + c1.
∴ v⋅
x +3
=
∫ x + 3 ⋅ ( x + 3) e
∴ v = e x ( x + 3) + c1 ( x + 3) .
( D − 2) y = e x ( x + 3) + c1 ( x + 3)
dy
or − 2 y = e x ( x + 3) + c1 ( x + 3), which is again linear.
dx
∴ I. F. = e ∫ −2 dx = e −2 x .
ye −2 x = −2 x −2 x
∴
∫e ( x + 3) dx + c1
∫e ( x + 3) dx + c2
1 1
= − ( x + 3) e − x − e − x − c1 ( x + 3) e −2 x − c1e −2 x + c2
2 4
1 1
or y = − ( x + 3) e x − e x − c1 ( x + 3) − c1 + c2 e2 x
2 4
D-210
1
or y = − xe x − 4 e x − c1 (2 x + 7) + c2 e2 x
4
or y = − xe x − 4 e x + A (2 x + 7) + Be2 x ,
Solve the following equations by the method of factorization of the

operator:
d2 y dy
1. x + ( x − 2) − 2 y = x3 .
dx2 dx
d2 y dy
2. x 2
+ ( x − 1) − y = x2 .
dx dx (Kumaun 2010, 12)
2 2 2
3. 3 x D y + (2 − 6 x ) Dy − 4 y = 0 .
d2 y dy
4. x2 2
+ − (1 + x2 ) y = e − x .
dx dx
5. [ xD2 + (1 − x) D − 2(1 + x)] y = e − x (1 − 6 x) .
A nswers 5
1. y = x3 + a ( x2 − 2 x + 2) + be − x 2. y = c1 ( x − 1) + c2 e − x + x2
1 −x
y = c1e2 x (−2 x) + 2 /(3 x)
dx + c2 e2 x y = c1e x − 2 x + (1 / x)
dx + c2 e x −
3.
∫e 4.
∫e 2
e
1
y = xe − x + c1e2 x −3 x
dx + c2 e2 x
5.
∫xe
9 Guidelines of the Procedure for the Solution of Linear

Differential Equations of Second Order
Before closing this chapter we give the guidelines of the procedure to be applied for the
solution of linear differential equation of the second order.
1. First of all put the differential equation in the standard form
dy d2 y
+P + Qy = R .
dx2 dx
2. Then try to find by inspection an integral belonging to the complementary
function of the given differential equation. If it is found, then proceed as in
article 2.
D-211
3. If a part of the C.F. is not found by inspection, then find the value of
1 dP 1 2
Q− − P . If it is a constant or a constant divided by x2 , then the normal
2 dx 4
form of the equation is easily integrable and so proceed as in article 4.
4. In case the methods given in (2) and (3) do not succeed, the method of change
of independent variable given in article 5 may be tried.
Q
In this method we often choose z such that = some suitable constant.
(dz / dx)2
5. In some cases the method of operational factors solves the equation easily.
6. If the students are instructed in the question to solve the equation by the
method of variation of parameters, then only the method of variation of
parameters must be applied.
Objective Type Questions

(b), (c) and (d).
d2 y dy
1. y = x m is a part of C.F. of equation 2
+P + Q y = R if
dx dx
(a) m (m − 1) + Pmx + Qx2 = 0 (b) m (m + 1) + Pmx + Qx2 = 0
(c) m (m − 1) + Pm2 x + Qx2 = 0 (d) none of these
(Kumaun 2008, 13)
2
d y dy
2. y = e x is a solution of 2
+P + Q y = R if
dx dx
(a) 1 − P + Q = 0 (b) 1 + P − Q = 0
(c) 1 + P + Q = 0 (d) 1 − P − Q = 0
(Kumaun 2007)
2
d y dy
3. y = e − x is a part of C.F. of 2
+P + Q y = R if
dx dx
(a) 1 + P + Q = 0 (b) 1 − P + Q = 0
(c) 1 − P − Q = 0 (d) 1 + P − Q = 0
(Kumaun 2010)
 d2 y   dy 
4. y = e mx is a part of C.F. of  2  + P   + Q y = R if
 dx   dx 
(a) m2 + mP + Q = 0 (b) m (m − 1) + Pmx + Qx2 = 0

(c) m2 + mQ + P = 0 (d) none of these
D-212
 d2 y   dy 
5. y = x is a solution of  2  + P   + Q y = R if
 dx   dx 
(a) P + Q = 0 (b) P − Qx = 0
Q
(c) P + Qx = 0 (d) P + =0
x
(Kumaun 2011; Kanpur 16)
 d y 2
 dy 
6. y = x2 is a part of C.F. of equation  2  + P   + Q y = R if
 dx   dx 
(a) P + Qx = 0 (b) 1 + P + Q = 0
(c) 1 − P + Q = 0 (d) 2 + 2 Px + Qx2 = 0
(Kumaun 2009)
d2 y dy
7. y = e2 x is a part of C.F. of the differential equation +P + Q y = 0 if
dx2 dx
(a) 2 + 2 P + Q = 0 (b) 4 + 2 P + Q = 0
(c) 4 − 2 P + Q = 0 (d) 4 + 2 P − Q = 0
(Garhwal 2008, 12, 13)
8. A part of the C.F. of the differential equation
d2 y dy
2
− cot x − (1 − cot x) y = e x sin x
dx dx
(a) y = x (b) y = x2
(c) y = e x (d) y = e − x (Garhwal 2014)
9. Normal form of second order differential equation

d2 y dy d2V
+P + Q = R is given by 2 + IV = S, where I and S are
dx2 dx dx
1
1 dP P2 ∫ P dx
(a) I = Q − − , S = Re 2
2 dx 4
1
1 dP P2 R ∫ P dx
(b) I = Q − − , S = e2
3 dx 4 2
1
1 dP P2 ∫ P dx
(c) I = Q + − , S = Re 2
2 dx 4
1
2 ∫ P dx
1 dP  − P 
(d) I = Q − +  , S = Re 2
2 dx  4  (Kanpur 2016)
ex
10. If y1( x) and y2 ( x) are solutions of y ′′ + 3 y ′ + 2 y = e , then Wronskian
W ( y1 y2 ) is
(a) e −3 x (b) e 3 x
−3 x
(c) − e (d) 2 e − 3 x
(Kanpur 2016)
D-213
11. On solving by changing the independent variable the differential equation

d2 y dy d2 y dy
(1 + x2 )2 2
+ 2 x(1 + x2 ) + 4 y = 0 reduces into + P1 + Q1 y = R1
dx dx dx2 dz
where P1 , Q1 , R1 are
(a) P1 = 0 , Q1 = 4, R1 = 0 (b) P1 = 2, Q1 = 4, R1 = 0
(c) P1 = 0 , Q1 = 2, R1 = 0 (d) P1 = 0 , Q1 = 1, R1 = 0
(Kanpur 2016)
d2 y dy d2 v
− 2 tan x + 5 y = (sec x) e x is given by 2 + X v = Y , where X and Y are
dx2 dx dx
(a) X = 6, Y = e x (b) X = − 6, Y = e x
(c) X = 6, Y = e − x (d) X = − 6, Y = e − x
d dy d2 v
(cos2 x ) + y cos2 x = 0 is given by 2 + X v = Y , where X and Y are
dx dx dx
(a) X = − 2, Y = 1 (b) X = − 2, Y = −1
(c) X = 2, Y = 0 (d) X = 0 , Y = 2
14. On solving by changing the independent variable the differential equation
d2 y dy 2 ( x +e x) d2 y dy
2
− (1 + 4 e x )
+ 3 e 2x
y = e , reduces into 2
+ P1 + Q1 y = R1
dx dx dz dz
where P1 , Q1 , R1 are
(a) P1 = − 4, Q1 = 3, R1 = e 2 z (b) P1 = 4, Q1 = − 3, R1 = e 2 z
(c) P1 = − 4, Q1 = 3, R1 = e − 2 z (d) P1 = − 4, Q1 = − 3, R1 = e 2 z
d2 y
15. If u and v are solutions of + 4 y = 4 tan 2 x, then Wronskian W ( u v) is
dx2
(a) − 2 (b) 2
(c) 0 (d) 1
Fill in The Blank(s)

d2 y dy
1. y = x m is a part of C.F. of equation 2
+P + Qy = R if ……
dx dx (Meerut 2003)
2
d y dy
2. y = e x is a solution of 2
+P + Qy = R if ……
dx dx
d2 y dy
3. y = e − x is a part of C.F. of 2
+P + Qy = R if ……
dx dx
 d2 y   dy 
4. y = e mx is a part of C.F. of  2  + P   + Qy = R if ……
 dx   dx 
D-214
 d2 y   dy 
5. y = x is a solution of  2  + P   + Qy = R if ……
 dx   dx 
 d2 y   dy 
6. y = x2 is a part of C.F. of equation  2  + P   + Qy = R if ……
 dx   dx 
True or False
1. y = x2 is a part of C.F. of the solution of equation
d2 y dy
(1 − x) +x − y = (1 − x)2 .
dx2 dx
2. y = e x is a part of C.F. of the solution of equation
d2 y dy
x 2
− (2 x − 1) + ( x − 1) y = 0 .
dx dx
3. y = x is a part of C.F. of the solution of equation
dy  d2 y 
x − y = ( x − 1)  2 − x + 1 .
dx  dx 
d2 y 2 2
4. 2
+ 1 + cot x − 2  y = x cos x is a linear differential equation of second
dx  x x 
order.
A nswers
1. (a) 2. (c) 3. (b) 4. (a) 5. (c)
6. (d) 7. (b) 8. (d) 9. (a) 10. (c)
11. (d) 12. (a) 13. (c) 14. (a) 15. (b)

1. m (m − 1) + Pmx + Qx2 = 0 2. 1 + P + Q = 0
3. 1− P + Q = 0 4. m2 + Pm + Q = 0
5. P + Qx = 0 6. 2 + 2 Px + Qx2 = 0
True or False
1. F 2. T 3. T 4. T
¨
D-215
D ifferentiation of V e ctors
1 Vector Function
e know that a scalar quantity possesses only magnitude and has no concern
W with direction. A single real number gives us a complete representation of a
scalar quantity. Thus a scalar quantity is nothing but a real number.
Let D be any subset of the set of all real numbers. If to each element t of D , we
associate by some rule a unique real number f (t), then this rule defines a scalar
function of the scalar variable t . Here f (t) is a scalar quantity and thus f is a scalar
function.
In a similar manner we define a vector function.
Let D be any subset of the set of all real numbers. If to each element t of D, we associate by some
rule a unique vector f (t), then this rule defines a vector function of the scalar variable t.
Here f (t) is a vector quantity and thus f is a vector function.
We know that every vector can be uniquely expressed as a linear combination of
three fixed non-coplanar vectors. Therefore we may write
f (t) = f1 (t) i + f 2 (t) j + f 3 (t) k
D-216
where i , j , k denote a fixed right handed triad of three mutually perpendicular

non-coplanar unit vectors.
2 Scalar Fields and Vector Fields

If to each point P( x , y , z ) of a region R in space there corresponds a unique scalar
f ( P), then f is called a scalar point function and we say that a scalar field f has
been defined in R .
Examples: (1) The temperature at any point within or on the surface of earth at a
certain time defines a scalar field.
(2) f ( x , y , z ) = x 2 − y 3 − 3z 2 defines a scalar field.
If to each point P ( x , y , z ) of a region R in space there corresponds a unique vector
f ( P), then f is called a vector point function and we say that a vector field f has
been defined in R.
Examples: (1) If the velocity at any point ( x, y, z ) of a particle moving in a curve

is known at a certain time, then a vector field is defined.
(2) f ( x , y , z ) = xy 2 i + 3 yz 3 j − 2 x 2 zk defines a vector field.
3 Limit and Continuity of a Vector Function

Definition 1: A vector function f (t) is said to tend to a limit l, when t tends to t0 , if for any
given positive number ε , however small, there corresponds a positive number δ such that
|f (t) − l | < ε whenever 0 < |t − t0 | < δ.
If f (t) tends to a limit l as t tends to t0 , we write lim f (t) = l .

t → t0
Definition 2: A vector function f (t) is said to be continuous for a value t0 of t if
(i) f (t0 ) is defined and
(ii) for any given positive number ε, however small, there corresponds a positive number δ
such that |f (t) − f (t0 )| < ε , whenever |t − t0| < δ.
Further a vector function f (t) is said to be continuous if it is continuous for every value of t for
which it has been defined.
We shall give here (without proof ) some important results about the limits and
continuity of a vector function.
Theorem 1: The necessary and sufficient condition for a vector function f (t) to be
continuous at t = t0 is that lim f (t) = f (t0 ).

t → t0 (Purvanchal 2014)
D-217
Theorem 2: If f (t) = f1 (t)i + f 2 (t) j + f 3 (t)k , then f (t) is continuous if and only if
f1 (t), f 2 (t), f 3 (t) are continuous.
Theorem 3: Let f (t) = f1 (t)i + f 2 (t) j + f 3 (t)k and l = l1 i + l2 j + l3 k .
Then the necessary and sufficient conditions that lim f (t) = l are
t → t0
lim f (t) = l , lim f 2 (t) = l2 and lim f 3 (t) = l3 .
t → t0 1 1
t → t0 t → t0
Theorem 4: If f (t), g (t) are vector functions of scalar variable t and φ (t) is a scalar
function of scalar variable t , then
(i) lim [f (t) ± g(t)] = lim f (t) ± lim g(t)
t → t0 t → t0 t → t0
(ii) lim [f (t) • g(t)] =  lim f (t) •  lim g(t)

t → t0  t→ t   t→ t 
 0   0 
(iii) lim [f (t) × g(t)] =  lim f (t) ×  lim 

t → t0  t→ t   t → t g(t)
 0   0 
(iv) lim [φ (t) f (t)] =  lim φ (t)  lim f (t)

t → t0  t→ t   t→ t 
 0  0 
(v) lim |f (t) | = 

 lim f (t) 
⋅
t → t0  t → t0 
4 Derivative of a Vector Function with Respect to a Scalar

Definition: Let r = f (t) be a vector function of the scalar variable t. We define
r + δr = f (t + δt).
∴ δr = f (t + δt) − f (t).
δr f (t + δt) − f (t)
Consider the vector = ⋅
δt δt
δr f (t + δt) − f (t)
If lim = lim exists, then the value of this limit, which we shall
δt → 0 δt δt → 0 δt
dr
denote by , is called the derivative of the vector function r with respect to the scalar t.
dt
Symbolically
dr (r + δr) − r f (t + δt) − f (t)
= lim = lim ⋅
dt δt → 0 δt δt → 0 δt
dr δr
If exists, then r is said to be differentiable. Since is a vector quantity,
dt δt
dr
therefore is also a vector quantity.
dt
D-218
dr d2 r
Successive Derivatives: If we differentiate again, we get 2 which is called
dt dt
the second derivative of r w.r.t. t , and so on.
Thus, we continue differentiating successively upto n times and get
dr d2 r d3 r d nr
, , , …… , ,
dt dt 2 dt 3 dt n
d nr
where n
is called the n th diff. coeff. (or derivative) of r w.r.t. t. (Here diff. coeff.
dt
means differential coefficient.)
dr d2 r . ..
We often represent , , …… by r , r, …… respectively.
dt dt 2
A scalar or vector function of t is called differentiable of order n if its n th order
derivative exists.
5 Differentiation Formulae
Theorem: If a, b and c are differentiable vector functions of a scalar t and φ is a
differentiable scalar function of the same variable t, then
d da db
1. (a + b) = +
dt dt dt
d db da
2. (a • b) = a • + •b
dt dt dt
d db da
3. (a × b) = a × + × b
dt dt dt
d da dφ
4. (φa) = φ + a
dt dt dt
d da db   dc
5. [a b c] =  b c + a c + a b 
dt  dt   dt   dt  (Meerut 2013B; Purvanchal 08)
d da db dc
6. {a × (b × c)} = × (b × c) + a ×  × c + a ×  b ×  ⋅
dt dt  dt   dt 
(Meerut 2000; Purvanchal 08)
d {( a + δa) + ( b + δb)} − (a + b)
Proof: 1. (a + b) = lim
dt δt → 0 δt
δa + δb  δa δb
= lim = lim  + 
δt → 0 δt δt → 0  δ t dt 
δa δb d a d b
= lim + lim = + ⋅
δt → 0 δ t δt → 0 dt dt dt
D-219
Thus the derivative of the sum of two vectors is equal to the sum of their derivatives, as it is also
in Scalar Calculus.
d da d b
Similarly we can prove that (a – b) = − ⋅
dt dt dt
In general if r1 , r2 ,……, r n are vector functions of a scalar t, then
d dr dr dr
(r1 + r2 + …… + r n ) = 1 + 2 + …… + n ⋅
dt dt dt dt
d (a + δa) • ( b + δb) − a • b
2. (a • b) = lim
dt δt → 0 δt
= lim a • b + a • δb + δa • b + δa • δb − a • b
δt → 0 δt
= lim a • δb + δa • b + δa • δb
δt → 0 δt
lim  δb δa δa 
= a • + • b+ • δb
δt → 0  δt δt δt 
lim a • δb + lim δa • b + lim δa • δb
=
δt → 0 δt δt → 0 δt δt → 0 δt
db da da
= a• + • b+ • 0, since δb → zero vector as δt → 0
dt dt dt
db da
= a• + •b
dt dt
d
Note: We know that a • b = b • a. Therefore while evaluating (a • b), we should
dt
not bother about the order of the factors.
d (a + δa) × ( b + δb) − a × b
3. (a × b) = lim
dt δ t → 0 δt
= lim a × b + a × δb + δa × b + δa × δb − a × b
δt → 0 δt
= lim a × δb + δa × b + δa × δb
δt → 0 δt
= lim  a × δb + δa × b + δa × δb
 
= lim a × δb + lim δa × b + lim δa × δb

db da da
=a× + × b+ × 0 , since δb → zero vector as δt → 0
dt dt dt
db da db da
=a× + × b+ 0 = a × + × b.
dt dt dt dt
D-220
Note: We know that cross product of two vectors is not commutative because
d
a × b = – b × a . Therefore while evaluating (a × b), we must maintain the order
dt
of the factors a and b .
d (φ + δφ)(a + δa) − φa
4. (φa) = lim
dt δt → 0 δt
= lim φa + φδa + δφa + δφδa − φa

δt → 0 δt
= lim φδa + δφa + δφδa

δt → 0 δt
= lim  φ δa + δφ a + δφ δa
 
= lim φ δa + lim δφ a + lim δφ δa

da dφ dφ
=φ + a+ 0 , since δa → zero vector as δt → 0
dt dt dt
da dφ da dφ
=φ + a+0=φ + a.
dt dt dt dt
Note: φa is the multiplication of a vector by a scalar. In the case of such
multiplication we usually write the scalar in the first position and the vector in the
second position.
d d d da
5. [a b c] = {a • (b × c)} = a • ( b × c) + • ( b × c) [by rule (2)]
dt dt dt dt
dc db da
= a •  b × + × c + • ( b × c) [by rule (3)]
 dt dt  dt
dc db da
= a •  b ×  + a •  × c + • ( b × c)
 dt   dt  dt
dc db   da
= a b  + a c + bc 
 dt   dt   dt 
da db   dc
= b c  + a c + a b ⋅
 dt   dt   dt 
Note: Here [abc] is the scalar triple product of three vectors a , band c. Therefore
d
while evaluating [abc] we must maintain the cyclic order of each factor.
dt
d d da
6. {a × (b × c)} = a × (b × c) + × (b × c) [by rule (3)]
dt dt dt
db dc da
= a ×  × c + b ×  + × (b × c)
 dt dt  dt
D-221
db dc da
= a ×  × c + a ×  b ×  + × (b × c)
 dt   dt  dt
da db dc
= × (b × c) + a ×  × c + a ×  b ×  ⋅
dt  dt   dt 
6 Derivative of a Function of a Function

Suppose r is a differentiable vector function of a scalar variable s and s is a
differentiable scalar function of another scalar variable t.Then r is a function of t.
An increment δt in t produces an increment δr in r and an increment δs in s. When
δt → 0, δr → 0 and δs → 0.
dr δr  δs δr 
We have = lim = lim  
dt δt → 0 δt δt → 0  δt δs 
 δs  lim δr  ds dr
=  lim    = ⋅
δt → 0 δt  δt → 0 δs  dt ds
7 Derivative of a Constant Vector

A vector is said to be constant only if both its magnitude and direction are fixed. If
either of these changes then the vector will change and thus it will not be constant.
Let r be a constant vector function of the scalar variable t. Let r = c , where c is a
constant vector. Then r + δr = c .
δr 0
∴ δr = 0 (zero vector). ∴ = = 0.
δt δt
lim δr = lim 0 = 0 . dr
∴ ∴ = 0 (zero vector).
δt → 0 δt δt → 0 dt
Thus the derivative of a constant vector is equal to the null vector.
8 Derivative of a Vector Function in Terms

of its Components
Let r be a vector function of the scalar variable t . (Avadh 2014)
Let r = xi + yj + zk where the components x , y , z are scalar functions of the
scalar variable t and i, j, k are fixed unit vectors.
We have r + δr = ( x + δx) i + ( y + δy) j + (z + δz ) k .
∴ δr = (r + δr) − r = δx i + δyj + δzk .
δr δx δy δz
∴ = i+ j+ k.
δt δt δt δt
δy
∴ lim δr = lim  δx i + j+
δz 
k ⋅

δt → 0 δt δt → 0  δt δt δt 
D-222
dr dx dy dz
∴ = i+ j+ k.
dt dt dt dt
Thus in order to differentiate a vector we should differentiate its components.
Note: If r = xi + yj + zk , then sometimes we also write it as r = ( x, y, z ). In this
2
dr  dx dy dz  d 2 r  d2 x d y d2 z 
notation = , , , 2 = , , 2 , and so on.
dt  dt dt dt  dt  dt 2 dt 2 dt 

Alternative Method:
We have r = xi + yj + zk , where i , j, k are constant vectors and so their
derivatives will be zero.
dr d d d d
Now, = ( xi + yj + zk ) = ( xi ) + ( yj) + (zk )
dt dt dt dt dt
dx di dy dj dz dk
= i+ x + j+ y + k +z
dt dt dt dt dt dt
dx dy dz di
= i+ j+ k , since , etc. vanish.
dt dt dt dt
9 Some Important Results

Theorem 1: The necessary and sufficient condition for the vector function a (t) to be
da
constant is that = 0.
dt (Purvanchal 2014)
Proof: The condition is necessary. Let a (t) be a constant vector function of
the scalar variable t . Then a (t + δt) = a (t). We have
da a (t + δt) − a (t) 0
= lim = lim = 0.
dt δ t → 0 δt δ t → 0 δt
Therefore the condition is necessary.
da
The condition is sufficient. Let = 0. Then to prove that a is a constant
dt
vector. Let
a (t) = a1 (t) i + a2 (t) j + a3 (t) k .
da da1 da da
Then = i + 2 j+ 3 k.
dt dt dt dt
da da1 da2 da
Therefore = 0 gives, i+ j + 3 k = 0.
dt dt dt dt
Equating to zero the coefficients of i, j and k , we get
da1 da da
= 0, 2 = 0, 3 = 0.
dt dt dt
Hence a1 , a2 , a3 are constant scalars i. e., they are independent of t . Therefore a (t)
is a constant vector function.
D-223
Theorem 2: If a is a differentiable vector function of the scalar variable t and if | a | = a,

then
d 2 da da da
(i) (a ) = 2a ; (ii) a • =a ⋅
dt dt dt dt
Proof: (i) We have a2 = a • a = (a) (a) cos 0 = a2 .
d 2 d 2 da
Therefore (a ) = (a ) = 2a ⋅
dt dt dt
d 2 d da da da
(ii) We have (a ) = (a • a) = •a+a• = 2a • ⋅
dt dt dt dt dt
d 2 d 2 da
Also (a ) = (a ) = 2a ⋅
dt dt dt
da da da da
∴ 2a • = 2a or a• =a ⋅
dt dt dt dt
da
Theorem 3: If a has constant length (fixed magnitude), then a and are
dt
da
perpendicular provided  ≠ 0.
 dt 
Proof: Let |a | = a = constant. Then a • a = a2 = constant.
d da da
∴ (a • a) = 0 or •a+a• =0
dt dt dt
da da
or 2a • =0 or a• = 0.
dt dt
da
Thus the scalar product of two vectors a and is zero.
dt
da da
Therefore a is perpendicular to provided is not null vector i. e., provided
dt dt
da ≠ 0.
dt
Thus the derivative of a vector of constant length is perpendicular to the vector provided the
vector itself is not constant.
Theorem 4: The necessary and sufficient condition for the vector a (t) to have constant
da
magnitude is a • = 0. (Meerut 2001, 04B, 06, 07, 11, 13; Avadh 13;
dt
Kashi 10, 12; Purvanchal 11)
Proof: Let a be a vector function of the scalar variable t . Let|a| = a = constant.
Then a • a = a2 = constant.
d da da
∴ (a • a) = 0 or a• + •a=0
dt dt dt
da da
or 2a • =0 or a• = 0.
dt dt
D-224
da
Condition is sufficient. If a • = 0, then
dt
da da d
a• + •a=0 or (a • a) = 0
dt dt dt
or a • a = constant or a2 = constant
or a2 = constant or |a | = constant
Theorem 5: If a is a differentiable vector function of the scalar variable t, then
d  da  d2 a
a ×  = a × ⋅
dt  dt  dt 2
Proof: We have
2 2
d  a × da  = da × da + a × d a = 0 + a × d a ,
 
dt  dt  dt dt dt 2 dt 2
da d2 a
since the cross product of two equal vectors is zero = a × ⋅
dt dt 2
Theorem 6: The necessary and sufficient condition for the vector a (t) to have constant
da
direction is a × = 0.
dt (Meerut 2004B, 06, 07, 10, 11, 13B; Avadh 14)
Proof: Let a be a vector function of the scalar variable t . Let A be a unit vector in
the direction of a . If a be the magnitude of a , then a = aA .
da dA da
∴ =a + A.
dt dt dt
da dA da 
Hence a× = (a A) ×  a + A
dt  dt dt 
dA da
= a2 A × +a A×A
dt dt
dA
= a2 A × [ ∵ A × A = 0] …(1)
dt
The condition is necessary. Suppose a has a constant direction. Then A is a
constant vector because it has constant direction as well as constant magnitude.
dA
Therefore = 0.
dt
da
∴ From (1), we get a × = a2 A × 0 = 0.
dt
da
The condition is sufficient. Suppose that a × = 0.
dt
Then from (1), we get
dA dA
a2 A × =0 or A× = 0. …(2)
dt dt
D-225
dA
Since A is of constant length, therefore A • = 0. …(3)
dt
dA
From (2) and (3), we get = 0.
dt
Hence A is a constant vector i. e., the direction of a is constant.
10 Curves in Space
A curve in a three dimensional Euclidean space may be regarded as the intersection
of two surfaces represented by two equations of the form
F1 ( x, y, z ) = 0, F2 ( x, y, z ) = 0.
It can be easily seen that the parametric equations of the form
x = f1 (t), y = f 2 (t), z = f 3 (t)
where x , y , z are scalar functions of the scalar t, also represent a curve in three
dimensional space. Here ( x, y, z ) are the coordinates of a current point on the
curve. The scalar variable t may range over a set of values a ≤ t ≤ b .
In vector notation an equation of the form r = f (t), represents a curve in three
dimensional space if r is the position vector of a current point on the curve. As t
changes, r will give position vectors of different points on the curve. The vector f (t)
can be expressed as
f1 (t)i + f 2 (t) j + f 3 (t)k .
Also if ( x, y, z ) are the coordinates of a current point on the curve whose position
vector is r, then r = xi + yj + zk .
Therefore the single vector equation r = f (t)
i. e., xi + yj + zk = f1 (t)i + f 2 (t) j + f 3 (t)k
is equivalent to the three parametric equations
x = f1 (t), y = f 2 (t), z = f 3 (t).
Thus a curve in a space may be defined as the locus of a point whose coordinates may be
expressed as a function of a single parameter.
For example, the two equations
x2 y2 z
2
− 2
= 1, x = a cosh
a b a
specify a curve in three dimensional space. The parametric equations of this curve
are x = a cosh u, y = b sinh u, z = au.
And its vectorial equation is
r = a cosh u i + b sinh u j + au k .
The vector equation r = a cos ti + b sin tj + 0 k represents an ellipse, as for different
values of t, the end point of r describes an ellipse.
Similarly r = at 2 i + 2atj + 0 k is the vector equation of a parabola.
The terms skew, twisted or tortuous are often used for curves in a space.
D-226
11 Geometrical Significance of dr / dt and Unit Tangent

Vector to a Curve
Geometrical Significance of dr / dt :
Let r = f (t) be the vector equation of a curve
in space. Let r and r + δr be the position
vectors of two neighbouring points P and Q
on this curve. Thus we have
→
OP = r = f (t)
→
and OQ = r + δr = f (t + δt).
→ → →
∴ PQ = OQ − OP
= (r + δr) − r= δr.
δr
Thus is a vector parallel to the chord PQ .
dt
As Q → P i. e., as δt → 0, chord PQ → tangent at P to the curve.
∴ lim δr = dr is a vector parallel to the tangent at P to the curve r = f (t).

δt → 0 δt dt
Unit tangent vector to a curve:
Suppose in place of the scalar parameter t , we take the parameter as s where s
denotes the arc length measured along the curve from any convenient fixed point C
on the curve. Thus arc CP = s and arc CQ = s + δs .
dr
In this case will be a vector along the tangent at P to the curve and in the
ds
direction of s increasing. Also we have
|δr |
d r 
  = lim  δr
   = lim = lim
chord PQ
= 1.
 ds  δt → 0  δs  δt → 0 arc PQ Q → P arc PQ
dr
Thus is a unit vector along the tangent at P in the direction of s increasing. We
ds
denote it by t. Note that t always points in the direction of motion along the curve.
12 Application to Velocity and Acceleration

Velocity: If the scalar variable t be the time and r be the position vector of a
moving particle P with respect to the origin O, then δr is the displacement of the
particle in time δt .
δr
The vector is the average velocity of the particle during the interval δt . If v
δt
δr dr
represents the velocity vector of the particle at P, then v = lim = ⋅
δt → 0 δ t dt
D-227
dr
Since is a vector along the tangent at P to the curve in which the particle is
dt
moving, therefore the direction of velocity is along the tangent.
δv
Acceleration: If δv be the change in the velocity v during the time δt , then is
δt
the average acceleration during that interval. If a represents the acceleration of the
particle at time t , then
δv dv d  dr  d2 r
a = lim = =   = ⋅
δt → 0 δt dt dt  dt  dt 2
Example 1: If r = sin t i + cos t j + tk , find

dr d2 r
(i) , (Kashi 2013) (ii) ,
dt dt 2 (Meerut 2010B)
2
dr d r
(iii)  , (Kashi 2013) (iv)  2  .
dt dt  (Meerut 2009)
di
Solution: Since i , j , k are constant vectors, therefore = 0 etc. Therefore
dt
dr d d d
(i) = (sin t) i + (cos t) j + (t) k = cos t i − sin t j + k .
dt dt dt dt
d2 r d  dr  = d (cos t) i − d (sin t) j + dk
(ii) =  
dt 2 dt  dt  dt dt dt
= − sin t i − cos t j + 0 = − sin t i − cos t j.
(iii) dr = √ [(cos t)2 + (− sin t)2 + (1)2 ] = √ 2 .

dt
d2 r
(iv)   = √ [(− sin t)2 + (− cos t)2 ] = 1.
2
dt 
Example 2: If a, b are constant vectors, ω is a constant, and r is a vector function of the
scalar variable t given by r = cos ωt a + sin ωt b , show that
d2 r
(i) 2
+ ω2 r = 0
dt (Rohilkhand 2006)
dr
(ii) r× = ωa × b .
dt (Bundelkhand 2004; Kanpur 05)
Solution: Since a, b are constant vectors, therefore
da db
= 0, = 0.
dt dt
D-228
dr d d
(i) = (cos ωt) a + (sin ωt) b
dt dt dt
= − ω sin ωt a + ω cos ωt b .
2
d r
∴ 2
= − ω 2 cos ωt a − ω 2 sin ωt b
dt
= − ω 2 (cos ωta + sin ωtb) = − ω 2 r.
d2 r
∴ + ω 2 r = 0.
dt 2
dr
(ii) r× = (cos ωta + sin ωt b) × (− ω sin ωta + ω cos ωtb)
dt
= ω cos 2 ωt a × b − ω sin2 ωt b × a [ ∵ a × a = 0, b × b = 0]
= ω cos 2 ωt a × b + ω sin2 ωt a × b
= ω (cos 2 ωt + sin2 ωt) a × b = ωa × b .
Example 3: If r = a cos t i + a sin t j + at tan α k , find
dr d2 r  dr d2 r d3 r 
 ×  and  , , 3⋅
dt dt 2   dt dt
2
dt  (Purvanchal 07, 10; Agra 14)
dr
Solution: We have = − a sin t i + a cos t j + a tan α k
dt
d2 r  dk 
2
= − a cos t i − a sin t j ,  ∵ dt = 0
dt  
d3 r
= a sin t i − a cos t j .
dt 3
 i j k 
dr d2 r 
∴ × 2 = − a sin t a cos t a tan α 
dt dt  
 − a cos t − a sin t 0 
= a2 sin t tan α i − a2 cos t tan α j + a2 k .

dr d2 r
∴  ×  = √ (a4 sin2 t tan2 α + a4 cos 2 t tan2 α + a4 )
dt 2
 dt 
= a2 sec α.
 dr d2 r d3 r   dr d2 r  d3 r
Also  , 2
, = × 2 • 3
 dt dt dt 3   dt dt  dt
= (a2 sin t tan α i − a2 cos t tan α j + a2 k ) • (a sin t i − a cos t j)
= a3 sin2 t tan α i • i + a3 cos 2 t tan α j • j [∵ i • j = 0 etc.]
= a3 tan α (sin2 t + cos 2 t) [ ∵ i • i = 1 = j • j]
3
= a tan α.
D-229
du dv d
Example 4: If = w × u, = w × v, show that (u × v) = w × (u × v).
dt dt dt
(Garhwal 2003; Bundelkhand 09)
Solution: We have
d du dv
(u × v) = × v+ u × = ( w × u) × v + u × ( w × v)
dt dt dt
= ( v • w) u − ( v • u) w + (u • v) w − (u • w) v
= ( v • w) u − (u • w) v [ ∵ u • v = v • u]
= (w • v) u – (w • u) v = w × (u × v).
Example 5: If R be a unit vector in the direction of r, prove that
dR 1 dr
R× = 2 r× , where r = | r |.
dt r dt (Bundelkhand 2008, 11)
1
Solution: We have r = rR ; so that R = r.
r
dR 1 dr 1 dr
∴ = − r.
dt r dt r 2 dt
dR 1 1 d r 1 dr 
Hence R× = r× − 2 r
dt r  r dt r dt 
1 dr 1 dr
= 2
r× − r×r
r dt r 3 dt
1 dr
= r× ⋅ [ ∵ r × r = 0]
r2 dt
Example 6: If r is a vector function of a scalar t and a is a constant vector, m a constant,
differentiate the following with respect to t :
(i) r • a , (ii) r × a ,
dr dr
(iii) r × , (iv) r • ,
dt dt
1 dr 2
(v) r 2 + 2 , (vi) m   ,
r  dt 
r+a r×a
(vii) 2 2
, (viii) ⋅
r +a r •a
Solution: (i) Let R = r • a. [Note that r • a is a scalar]

dR dr da
Then = •a+r •
dt dt dt
dr ∵ da
= •a+r •0 = 0, as a is constant 
dt  dt 
dr
= •a+0
dt
dr
= • a.
dt
D-230
(ii) Let R = r × a .
dR dr da
Then = ×a+r×
dt dt dt
dr ∵ da
= ×a+r×0 = 0
dt  dt 
dr dr
= ×a+0= × a.
dt dt
dr
(iii) Let R = r × ⋅
dt
dR dr dr d2 r
Then = × +r× 2
dt dt dt dt
d2 r ∵ dr dr
=0+r× × = 0
dt 2  dt dt 
2
d r
=r× ⋅
dt 2
dr
(iv) Let R = r • ⋅
dt
dR dr dr d2 r
Then = • +r• 2
dt dt dt dt
2
dr d2 r
=   +r• ⋅
 dt  dt 2
1
(v) Let R = r2 + ⋅
r2
dR d 2 d  1
Then = (r ) +  
dt dt dt  r 2 
d 2 d  1
= (r ) +   , where r = | r |
dt dt  r 2 
dr 2 dr
= 2r − ⋅
dt r 3 dt
2
dr
(vi) Let R = m   ⋅
 dt 
dR d  dr  2
Then =m  
dt dt  dt 
dr d2 r  dr 2 dr 
=2m • Note : =2r • 
dt dt 2  dt dt 
dr d2 r
=2m • ⋅
dt dt 2
D-231
r+a
(vii) Let R = ⋅
r 2 + a2
dR 1 d  d  1  
Then (r + a) +  
dt
= 2 2 dt   r 2 + a2   (r + a)
(r + a )  dt   
[Note that r 2 + a2 is a scalar]
1  dr + da  −  1 d 2
(

2 
)
= 
 dt
  2 r + a  (r + a)
r 2 + a2 dt   (r + a2 )2 dt 
dr
2r•
1 dr dt (r + a)
= 2
−
r + a2 dt (r 2 + a2 )2
 ∵ da = 0, d r 2 = 2r • dr , d a2 = 0
 dt dt dt dt 
r×a
(viii) Let R = ⋅
r •a
dR 1 d d  1 
Then = (r × a) +     (r × a)
dt r • a dt  dt  r • a 
[Note that r • a is a scalar quantity]

1 d r d a   1 d 
=  ×a+r×  − (r • a)  (r × a)
r • a  dt dt   (r • a)2 dt 
dr
×a 
= dt −
 1  dr • a + r • da   (r × a)
 
r •a  (r • a)  dt
2 dt  
dr dr
×a •a
dt da
= − dt 2 (r × a). ∵ = 0
r •a (r • a)  dt 
Example 7: Find
d  dr d2 r  d2  dr d2 r 
(i) r, , ; (ii) r, , ;
dt  dt dt 2  dt 2  dt dt 2 
d   dr d2 r  
(iii) r ×  × 2  ⋅
dt   dt dt   (Kumaun 2010)
 dr d r  2
Solution: (i) Let R = r, , 2  ⋅ Then R is the scalar triple product of three
 dt dt 
dr d2 r
vectors r, and 2 ⋅ Therefore using the rule for finding the derivative of a scalar
dt dt
triple product, we have
D-232
dR  dr dr d2 r   d2 r d2 r   dr d3 r 
= , ,  + r, ,  + r, , 
dt  dt dt dt 2   dt 2 dt 2   dt dt 3 
 dr d3 r 
= r, , 3 ,
 dt dt 
since scalar triple products having two equal vectors vanish.
 dr d2 r 
(ii) Let R = r, , 2  ⋅ Then as in part (i)
 dt dt 
dR  dr d3 r 
= r, , ⋅
dt  dt dt 3 
Differentiating again, we get
d2 R  dr dr d3 r   d2 r d3 r   dr d4 r 
= , , 3  + r, 2 , 3  + r, , 
dt 2  dt dt dt   dt dt   dt dt 4 
 d2 r d3 r   dr d4 r 
= r, 2 , 3  + r, , ⋅
 dt dt   dt dt 4 
 dr d2 r 
(iii) Let R = r ×  × 2⋅
 dt dt 
Then R is the vector triple product of three vectors. Therefore using the rule for
finding the derivative of a vector triple product, we have
dR dr  dr d2 r   d2 r d2 r   dr d3 r 
= × × 2 +r × 2 × 2 +r × × 3
dt dt  dt dt   dt dt   dt dt 
dr  dr d2 r   dr d3 r 
= × × 2 +r × × 3 ,
dt  dt dt   dt dt 
d2 r d2 r
since × = 0 , being vector product of two equal vectors.
dt 2 dt 2
Example 8: If a = sin θ i + cos θ j + θk , b = cos θ i − sin θ j − 3k , and

d π
c = 2i + 3 j − 3k , find {a × ( b × c)} at θ = ⋅
dθ 2 (Kumaun 2010)
 i j k 
Solution: 
We have b × c = cos θ − sin θ − 3
 
 2 3 − 3
= (3 sin θ + 9) i + (3 cos θ − 6) j + (3 cos θ + 2 sin θ) k .
 i j k 
∴ a × (b × c) =  sin θ cos θ θ 
 
3 sin θ + 9 3 cos θ − 6 3 cos θ + 2 sin θ 
D-233
= (3 cos 2 θ + 2 sin θ cos θ − 3θ cos θ + 6θ) i

+ (3θ sin θ + 9θ − 3 sin θ cos θ − 2 sin2 θ) j
+ (− 6 sin θ − 9 cos θ) k .
d
∴ {a × (b × c)} = (−6 cos θ sin θ + 2 cos θ − 2 sin2 θ
2
dθ
− 3 cos θ + 3θ sin θ + 6) i + (3 sin θ + 3θ cos θ + 9
− 3 cos 2 θ + 3 sin2 θ − 4 sin θ cos θ) j
+ (−6 cos θ + 9 sin θ) k .
Putting θ = π / 2 , we get the required derivative
3
= 4 + π i + 15 j + 9k .
 2 
Example 9: A particle moves along the curve x = 4 cos t , y = 4 sin t , z = 6t . Find the
1
velocity and acceleration at time t = 0 and t = π . Find also the magnitudes of the velocity
2
and acceleration at any time t. (Bundelkhand 2007, 14)
Solution: Let r be the position vector of the particle at time t .
Then r = xi + yj + zk = 4 cos t i + 4 sin t j + 6t k . If v is the velocity of the
particle at time t and a its acceleration at that time, then
dr
v= = − 4 sin t i + 4 cos t j + 6k
dt
d2 r
and a = 2 = − 4 cos t i − 4 sin t j.
dt
Magnitude of the velocity at time t = | v|
= √ (16 sin2 t + 16 cos 2 t + 36) = √ (52) = 2 √ (13).
Magnitude of the acceleration = |a | = √ (16 cos 2 t + 16 sin2 t) = 4.
At t = 0, v = 4 j + 6 k , a = − 4 i.
1
At t = π , v = − 4 i + 6 k , a = − 4 j.
2
Ex am ple 10: A par ti cle moves along the curve x = t 3 + 1, y = t 2 , z = 2t + 5, where t

is the time. Find the com po nents of its ve loc ity and ac cel er a tion at t = 1in the di rec tion
i + j + 3k . (Purvanchal 2012)
Solution: If r is the position vector of any point ( x , y , z ) on the given curve, then
r = xi + yj + zk = (t 3 + 1) i + t 2 j + (2t + 5) k .
dr
Velocity = v = = 3t 2 i + 2tj + 2k = 3i + 2 j + 2k at t = 1.
dt
d2 r d  dr 
Acceleration = a = =   = 6ti + 2 j = 6i + 2 j at t = 1.
dt 2 dt  dt 
Now the unit vector in the given direction i + j + 3k
D-234
i + j + 3k i + j + 3k
= = = b , say.
|i + j + 3k| √ (11)
∴ the component of velocity in the given direction
(3i + 2 j + 2k ) • (i + j + 3k ) 11
= v• b= = = √ (11);
√ (11) √ (11)
and the component of acceleration in the given direction
(6i + 2 j) • (i + j + 3k ) 8
= a • b= = ⋅
√ (11) √ (11)
1. (i) If r is the position vector of a moving point and r is the modulus of r ,

dr dr dr dr
show that r • =r ⋅ Interpret the relations r • = 0 and r × = 0.
dt dt dt dt
(ii) If r × dr = 0, show that r$ = constant. (Garhwal 2001; Meerut 05B;
Purvanchal 12; Bundelkhand 12; Avadh 13)
dr d2 r
2. (i) If r = (t + 1) i + (t 2 + t + 1) j + (t 3 + t 2 + t + 1) k , find
and 2 ⋅
dt dt
2
(ii) If r = t i − t j + (2t + 1) k , find at t = 0, the values of
dr d2 r dr d2 r
, , ,  ⋅
dt dt 2 dt dt 2  (Kanpur 2002; Bundelkhand 2010, 13)
3. Show that ^
r × d^
r = (r × d r) / r 2 , where r = r ^
r.
4. (i) The position vector of a moving particle at time t is given by
r = (3t − 4) i + (t 2 − 2) j + 4t 3 k. Find its velocity and acceleration at
time t = 2. (Kanpur 2005; Bundelkhand
13)
 1  dr
(ii) If r = t 3 i + 2t 3 − 2
j , show that r × = k.
 5t  dt (Agra 2007)
(iii) If r = (cos nt) i + (sin nt) j , where n is a constant and t varies, show that
dr
r× = nk . (Garhwal 2002)
dt
5. (i) If r = (sinh t) a + (cosh t) b, where a and b are constant vectors, then
d2 r
show that = r.
dt 2
(ii) If u = t 2 i − tj + (2t + 1) k and v = (2t − 3) i + j − tk , find
d
(u • v), when t = 1.
dt (Bundelkhand 2007)
D-235
6. (i) If r = e n t a + e − n t b , where a , b are constant vectors, show that

d2 r
2
− n2 r = 0 . (Kumaun 2000; Rohilkhand 07; Agra 07; Kashi 14)
dt
(ii) Show that r = a e mt + b e nt , where a and b are the constant vectors, is the
d2 r dr
solution of the differential equation 2
− (m + n) + mn r = 0.
dt dt
Hence solve the equation
d2 r
dr dr
− 2r = 0, where r = i and
2
− = j for t = 0.
dt dt dt
ct
(iii) If r = a sin ωt + b cos ωt + 2 sin ωt, prove that
ω
d2 r 2c
2
+ ω2 r = cos ωt,
dt ω
where a , b, c are constant vectors and ω is a constant scalar.
7. (i) A particle moves along the curve x = e − t , y = 2 cos 3t, z = 2 sin 3t.
Determine the velocity and acceleration at any time t and their
magnitudes at t = 0. (Bundelkhand 2005; Rohilkhand 08)
(ii) Show that if a , b, c are constant vectors, then r = at 2 + bt + c is the

path of a particle moving with constant acceleration.
8. A particle moves so that its position vector is given by r = cos ωt i + sin ωt j

where ω is a constant; show that
(i) the velocity of the particle is perpendicular to r,
(ii) the acceleration is directed towards the origin and has magnitude
proportional to the distance from the origin.
9. If A = 5t 2 i + t j − t 3 k and B = sin t i − cos t j , find

d d d
(i) (A • B) ; (ii) (A × B) ; (iii) (A • A). (Agra 2005)
dt dt dt
10. Prove the following :
2 2
d a • db − da • b = a • d b − d a • b
(i)
dt  dt dt  dt 2 dt 2
2 2
d a × db − da × b = a × d b − d a × b
(ii)
dt  dt dt  dt 2 dt 2
11. If r is a unit vector, then prove that

 r × dr  = dr ⋅
 dt  dt (Purvanchal 2006)
D-236
12. If r is a vector function of a scalar t, r its module, and a , b are constant vectors,
differentiate the following with respect to t :
dr
(i) r 3 r + a × , (ii) r 2 r + (a • r) b ,
dt
(iii) r n r , (iv) (ar + r b)2 .
13. Find the unit tangent vector to any point on the curve
x = a cos t , y = a sin t , z = bt .
14. If the direction of a differentiable vector function r (t) is constant, show that
 d r
r×  = 0. Or
 dt 
If r (t) is a vector of constant direction, show that its derivative is collinear
with it.
15. If e is the unit vector making an angle θ with x-axis, show that de / dθ is a unit
vector obtained by rotating e through a right angle in the direction of θ
increasing.
A nswers 1
2. (i) i + (2t + 1) j + (3t 2 + 2t + 1) k ; 2 j + (6t + 2) k
(ii) − j + 2k ; 2i; √ 5 ; 2
4. (i) v = 3 i + 4 j + 48 k , a = 2 j + 48 k
5. (i) r (ii) 6t 2 − 10 t − 2 ; − 6
7. (i) √ 37; √ (325)
9. (i) (5t 2 − 1) cos t + 11t sin t
(ii) t 2 (t sin t − 3 cos t) i − t 2 (t cos t + 3 sin t) j
− (11t cos t − 5t 2 sin t + sin t) k
3 5
(iii) 100 t + 2 t + 6t
dr dr d2 r
12. (i) 3r 2 r+r3 +a ×
dt dt dt 2
dr dr  d r
(ii) 2 r r +r2 + a •  b
dt dt  dt 
dr  dr
(iii)  nr n−1
 r+r
n
 dt  dt
 d r dr 
(iv) 2 (a r + r b) •  a + b
 dt dt 
1
13. (− a sin t i + a cos t j + b k)
√ (a2 + b 2 )
D-237

Indicate the correct answer for each question by writing the corresponding letter from
(a), (b), (c) and (d).
d2 r
1. If r = a e ω t + b e − ω t , where a, b are constant vectors then 2 − ω 2 r is
dt
equal to
(a) 1 (b) 0
(c) 2 (d) none of these
(Bundelkhand 2001; Agra 06)
2. A particle moves along the curve r = e − t cos t i + e − t sin t j + e − t k . The

magnitude of its velocity at t = 0 is
(a) 2 3 (b) 3 /2
2
d r
3. If r = x i + y j + z k , then = …… .
dt 2
dx dy dz
(a) i x + j y + k z (b) i + j +k
dt dt dt
d2 x d2 y d2 z
(c) i + j +k (d) none of these
dt 2 dt 2 dt 2
(Kumaun 2011)
4. If r = (cos nt) i + (sin nt) j , then
dr dr
(a) r × = ni (b) r × = nj
dt dt
dr
(c) r × = nk (d) none of these
dt (Kumaun 2006)
d
5. If a and b are two vectors then the value of (a • b) is
dt
da db
(a) •b (b) a •
dt dt
db da db da
(c) a • + •b (d) a × + ×b
dt dt dt dt
d
6. If a and b are two vectors then the value of (a × b) is
dt
db da
(a) a × (b) ×b
dt dt
db da db da
(c) a × + b× (d) a × + ×b
dt dt dt dt
D-238
dr
7. If r = 3 i − 6t 2 j + 4t k , then the value of is
dt
(a) 3 t i − 6 t 2 j + 4 t k (b) 3 i − 12 t j + 4 t k
(c) − 6t i + 4 k (d) −12 t j + 4 k
da
8. Unit vector a and its derivative are
dt
(a) parallel (b) perpendicular
(c) along the same direction (d) none of these
2
d r
9. If r = 3i − t 2 j + 4t k , then is equal to
dt 2
(a) 0 (b) 1
(c) 2 (d) 3 (Garhwal 2004, 13)
10. The necessary and sufficient condition for the vector a(t) to have
constant direction is
da da
(a) a • =0 (b) a × =0
dt dt
da
(c) =0 (d) none of these
dt (Garhwal 2006)
2
d r
11. If r = sin t i + cos t j + t k , then the value of is
dt 2
(a) 1 (b) 2
(c) 2 (d) 0 (Garhwal 2010, 14)
−t
12. A particle moves along the curve x = e , y = 2 cos 3t, z = 2 sin 3t, then
magnitude of velocity at t = 0 is
(a) 325 (b) 37
(Garhwal 2011)
d2 r
13. If r = (sinh t) a + (cosh t) b, where a and b are constant vectors then =
dt 2
(a) r (b) − r
2 3 d
14. If A = 5 t i + t j − t k , then (A • A) is
dt
(a) (5 t 3 − 1) + 11t (b) 10 t 3 + 2 t + 6 t 5
(c) 100 t 3 + 2 t + 6 t 5 (d) 100 t 5 + 2 t + 6 t 3
15. The necessary and sufficient condition for the vector function a(t) to be
constant is
D-239
da da
(a) =0 (b) =0
dt dt
da da
(c) a • =0 (d) ≠0
dt dt
16. If a is a differential vector function of the scalar variable t and if a = a,
d
then (a2 ) is
dt
da da
(a) a (b)
dt dt
da da
(c) 2a (d) a •
dt dt
Fill in the blanks “……”, so that the following statements are complete and correct.
dr d2 r
1. If r = 3i − 6t 2 j + 4t k , then = …… ; = …… .
dt dt 2 (Bundelkhand 2008)
d
2. If u = t 2 i − t j + (2t + 1) k , v = (2t − 3) i + j − t k , then (u • v) = …… .
dt
d2 r
3. If r = (cos ωt) i + (sin ωt) j , then r × = …… .
dt 2
4. The necessary and sufficient condition for the vector a (t) to have constant
direction is ……
dr d2 r
5. If r = 5i + 3t 2 j + 2t k , then = …… ; = …… . (Bundelkhand 2010)
dt dt 2
d
6. (a • b) = …… .
dt (Kumaun 2009)
True or False
1. A vector is said to be constant only if its magnitude is fixed and direction
changes.
2. The necessary and sufficient condition for the vector a (t) to have constant
da
magnitude is a • = 0.
dt
A nswers

1. (b) 2. (c) 3. (c) 4. (c) 5. (c)
6. (d) 7. (d) 8. (b) 9. (c) 10. (a)
11. (a) 12. (b) 13. (a) 14. (c) 15. (a)
16. (c)
D-240

1. − 12 t j + 4 k ; − 12 j 2. 6t 2 − 10 t − 2 3. 0
da
4. a × =0 5. 6 t j + 2 k ;6 j
dt
db da
6. a • + •b
dt dt
True or False
1. F 2. T
¨
D-241
9
G radient, D ivergence
and C url
1 Partial Derivatives of Vectors

uppose r is a vector depending on more than one scalar variable. Let
S r = f ( x, y, z ) i. e., let r be a function of three scalar variables x , y and z . The
partial derivative of r with respect to x is defined as
∂r f ( x + δx, y, z ) − f ( x, y, z )
= lim
∂x δx → 0 δx
if this limit exists. Thus ∂r / ∂x is nothing but the ordinary derivative of r with
respect to x provided the other variables y and z are regarded as constants.
∂r ∂r
Similarly we may define the partial derivatives and ⋅
∂y ∂z
Higher partial derivatives can also be defined as in Scalar Calculus. Thus, for
example,
∂2 r ∂  ∂r  ∂2 r ∂  ∂r  ∂2 r ∂  ∂r 
=   , =   , =  ,
∂x 2 ∂x  ∂x  ∂y 2 ∂y  ∂y  ∂z 2 ∂z  ∂z 
∂2 r ∂  ∂r  ∂2 r ∂  ∂r 
=  , =  ⋅
∂x ∂y ∂x  ∂y  ∂y ∂x ∂y  ∂x 
D-242
If r has continuous partial derivatives of the second order at least, then,

∂2 r ∂2 r
= i. e., the order of differentiation is immaterial. If r = f ( x, y, z ), the
∂x ∂y ∂y ∂x
∂r ∂r ∂r
total differential dr of r is given by dr = dx + dy + dz .
∂x ∂y ∂z
2 The Vector Differential Operator Del (∇ )

The vector differential operator ∇ (read as del or nabla) is defined as
∂ ∂ ∂ ∂ ∂ ∂
∇≡ i+ j+ k ≡i + j +k
∂x ∂y ∂z ∂x ∂y ∂z
and operates distributively.
The vector operator ∇ can generally be treated to behave as an ordinary vector. It
∂ ∂ ∂
possesses properties like ordinary vectors. The symbols , , can be treated
∂x ∂y ∂z
as its components along i , j , k .
3 Gradient of a Scalar Field

(Agra 2005)
Definition: Let f ( x, y, z ) be defined and differentiable at each point ( x, y, z ) in a
certain region of space (i.e., defines a differentiable scalar field). Then the gradient of f, written
as ∇f or grad f, is defined as
∂ ∂ ∂  ∂f ∂f ∂f
∇f =  i + j+ k f = i+ j+ k.
 ∂x ∂y ∂z  ∂x ∂y ∂z
∂f ∂f
It should be noted that ∇f is a vector whose three successive components are ,
∂x ∂y
∂f
and ⋅ Thus the gradient of a scalar field defines a vector field. If f is a scalar point
∂z
function, then ∇f is a vector point function.
4 Formulas Involving Gradient

Theorem 1: Gradient of the sum of two scalar point functions. If f and g are
two scalar point functions, then
grad ( f + g) = grad f + grad g or ∇ ( f + g) = ∇f + ∇g .
Proof: We have
 ∂ ∂ ∂
grad ( f + g) = ∇ ( f + g) =  i + j +k  ( f + g)
 ∂x ∂y ∂z 
D-243
∂ ∂ ∂
=i ( f + g) + j ( f + g) + k ( f + g)
∂x ∂y ∂z
∂f ∂g ∂f ∂g ∂f ∂g
=i +i + j + j +k +k
∂x ∂x ∂y ∂y ∂z ∂z
 ∂f ∂f ∂f   ∂g ∂g ∂g 
= i + j +k  + i + j +k 
 ∂x ∂y ∂z   ∂x ∂y ∂z 
 ∂ ∂ ∂  ∂ ∂ ∂
= i + j +k  f + i + j +k  g
 ∂x ∂y ∂z   ∂x ∂y ∂z 
= ∇f + ∇g = grad f + grad g .
Similarly, we can prove that ∇ ( f − g) = ∇f − ∇g.
Theorem 2: Gradient of a constant. The necessary and sufficient condition for a
scalar point function to be constant is that ∇f = 0.
∂f ∂f ∂f
Proof: If f ( x, y, z ) is constant, then = 0, = 0, = 0.
∂x ∂y ∂z
∂f ∂f ∂f
Therefore, grad f = i + j +k = 0 i + 0 j + 0 k = 0.
∂x ∂y ∂z
Hence the condition is necessary.
∂f ∂f ∂f
Conversely, let grad f = 0. Then i + j +k = 0.
∂x ∂y ∂z
∂f ∂f ∂f
Therefore, = 0, = 0, = 0.
∂x ∂y ∂z
∴ f must be independent of x, y and z .
∴ f must be a constant. Hence the condition is sufficient.
Theorem 3: Gradient of the product of two scalar point functions.

If f and g are scalar point functions, then grad ( fg) = f grad g + g grad f
or ∇ ( fg) = f ∇g + g ∇f . (Kumaun 2014)
 ∂ ∂ ∂
Proof: We have ∇ ( fg) =  i + j +k  ( fg)
 ∂x ∂y ∂z 
∂ ∂ ∂
=i ( fg) + j ( fg) + k ( fg)
∂x ∂y ∂z
 ∂g ∂f   ∂g ∂f   ∂g ∂f 
=i f + g  + jf + g  +k f + g 
 ∂x ∂x   ∂y ∂y   ∂z ∂z 
 ∂g ∂g ∂g   ∂f ∂f ∂f 
= f i + j +k  + g i + j +k 
 ∂x ∂y ∂z   ∂x ∂y ∂z 
= f ∇g + g ∇f = f grad g + g grad f .
In particular, if c is a constant, then
∇ (c f ) = c ∇f + f ∇c = c ∇f + 0 = c ∇f .
D-244
Theorem 4: Gradient of the quotient of two scalar functions. If f and g are two
f  g ∇f − f ∇g
scalar point functions, then ∇   = ⋅
 g g2 (Kumaun 2013)
Proof: We have
f   ∂ ∂ ∂ f 
∇   = i + j +k   
 g   ∂x ∂y ∂z   g 
∂ f  ∂ f  ∂ f 
=i   + j   +k  ⋅
∂x  g  ∂y  g  ∂z  g 
∂f ∂g ∂f ∂g
g − f g − f
∂  f  ∂ f  ∂y ∂y
But   = ∂x 2 ∂x ,   = ,
∂x  g  g ∂y  g  g2
∂f ∂g
g − f
∂ f  ∂ z ∂ z ⋅
and   = 2
∂z  g  g
f  1   ∂f ∂g   ∂f ∂g   ∂f ∂g  
∴ ∇   = 2  i g − f  + jg − f  + k g − f 
 g g   ∂x ∂x   ∂y ∂y   ∂z ∂z  
1   ∂f ∂f ∂f   ∂g ∂g ∂g 
= 2 
g i + j +k  − f i + j +k 
g   ∂x ∂y ∂z   ∂x ∂y ∂z 
1
= { g ∇f − f ∇g }.
g2
Example 1: If A = x 2 yz i − 2 xz 3 j + xz 2 k , B = 2zi + yj − x 2 k , find the value of

∂2
(A × B) at (1 , 0, − 2).
∂x ∂y
 i j k 
 
Solution: We have A × B = x 2 yz −2 xz 3 xz 2

 

2z y − x2 

= (2 x 3 z 3 − xyz 2 ) i + (2 xz 3 + x 4 yz ) j
+ ( x 2 y 2 z + 4 xz 4 ) k .
∂
∴ (A × B) = − xz 2 i + x 4 z j + 2 x 2 yzk .
∂y
∂2 ∂ ∂ 
Again (A × B) =  (A × B) = − z 2 i + 4 x 3 zj + 4 xyz k . …(1)
∂x ∂y ∂x  ∂y 
D-245
Putting x = 1, y = 0 and z = − 2 in (1), we get the required derivative at the point

(1, 0, − 2) = − 4i − 8 j .
Example 2: If f ( x , y , z ) = 3 x 2 y − y 3 z 2 , find grad f at the point (1, − 2 , − 1).

(Rohilkhand 2009)
Solution: We have
 ∂ ∂ ∂
grad f = ∇f =  i + j +k  (3 x 2 y − y 3 z 2 )
 ∂x ∂y ∂z 
∂ ∂
=i (3 x 2 y − y 3 z 2 ) + j (3 x 2 y − y 3 z 2 )
∂x ∂y
∂
+k (3 x 2 y − y 3 z 2 )
∂z
= i (6 xy) + j (3 x 2 − 3 y 2 z 2 ) + k (−2 y 3 z )
= 6 x y i + (3 x 2 − 3 y 2 z 2 ) j − 2 y 3 z k .
Putting x = 1, y = − 2 , z = − 1, we get
∇f = 6 (1) (−2) i + {3 (1)2 − 3 (−2)2 (−1)2 } j − 2 (−2)3 (−1) k
= − 12 i − 9 j − 16k .
Example 3: If r = | r | where r = xi + yj + zk , prove that

(i) ∇ f (r) = f ′ (r) ∇r , (Rohilkhand 2008; Meerut 11)
1
(ii) ∇r = r ,
r
(iii) ∇ f (r) × r = 0,
1 r
(iv) ∇   = − 3 , (Rohilkhand 2008, 09B)
 r r
r
(v) ∇ log | r | = 2 ,
r (Garhwal 2000; Kumaun 07; Rohilkhand 08;
Bundelkhand 08; Meerut 11)
n n−2
(vi) ∇r = nr r. (Rohilkhand 2008, 11; Bundelkhand 11)
Solution: If r = x i + y j + z k , then r = | r | = √ ( x 2 + y 2 + z 2 ).
∴ r 2 = x 2 + y 2 + z 2.
 ∂ ∂ ∂ ∂ ∂ ∂
(i) ∇f (r) =  i + j +k  f (r) = i f (r) + j f (r) + k f (r)
 ∂x ∂y ∂z  ∂x ∂y ∂z
∂r ∂r ∂r  ∂r ∂r ∂r 
= i f ′ (r) + j f ′ (r) + k f ′ (r) = f ′ (r)  i + j +k 
∂x ∂y ∂z  ∂ x ∂ y ∂ z
= f ′ (r) ∇r .
∂r ∂r ∂r
(ii) We have ∇r = i + j +k ⋅
∂x ∂y ∂z
D-246
Now r 2 = x2 + y2 + z 2 .
∂r ∂r x
∴ 2r = 2 x i. e., = ⋅
∂x ∂x r
∂r y ∂r z
Similarly, = and = ⋅
∂y r ∂z r
x y z 1 1 ∧
∴ ∇r = i+ j + k = ( xi + yj + zk ) = r = r .
r r r r r
(iii) We have as in part (i), ∇f (r) = f ′ (r) ∇r.

1
But as in part (ii), ∇r = r .
r
1
∴ ∇f (r) = f ′ (r) r .
r
1 1
∴ ∇f (r) × r =  f ′ (r) r × r =  f ′ (r) (r × r) = 0 .
 r  r 
[ ∵ r × r = 0]
1 ∂  1 + j ∂  1 + k ∂  1
(iv) We have ∇   = i      
 r ∂x  r ∂y  r ∂z  r
 1 ∂r   1 ∂r   1 ∂r 
= i − 2  + j − 2  + k − 2 
 r ∂x   r ∂y   r ∂z 
1  ∂r ∂r ∂r 
=− 2
 i+ j+ k
r  ∂x ∂y ∂z 
1 x y z  1 1
=− 2
 i + j + k  = − 3 ( xi + yj + zk ) = − 3 r .
r  r r r  r r
[see part (ii)]
(v) We have ∇ log |r | = ∇ log r
∂ ∂ ∂
=i log r + j log r + k log r
∂x ∂y ∂z
1 ∂r 1 ∂r 1 ∂r 1 x y z 
= i+ j+ k =  i+ j + k
r ∂x r ∂y r ∂z r r r r 
1 1
= 2
( xi + yj + zk ) = r.
r r2
∂ n ∂ n ∂ n
(vi) We have ∇r n = i r + j r +k r
∂x ∂y ∂z
∂r ∂r ∂r
= i nr n − 1 + j nr n − 1 + k nr n − 1
∂x ∂y ∂z
 ∂r ∂r ∂r 
= nr n − 1  i + j +k  = nr n − 1 ∇r
 ∂x ∂y ∂z 
D-247
1 ∵ ∇r = r as in part (ii)
= nr n − 1 r
r  r 
= nr n − 2 r .
Example 4: (i) Interpret the symbol a • ∇ (ii) Show that (a • ∇) φ = a • ∇ φ
(iii) Show that (a • ∇) r = a . (Kumaun 2008; Purvanchal 14)
Solution: (i) Let a = a1 i + a2 j + a3 k . Then
 ∂ ∂ ∂
a • ∇ = (a1 i + a2 j + a3 k ) •  i + j +k 
 ∂ x ∂ y ∂ z
∂ ∂ ∂
= a1 + a2 + a3 ⋅
∂x ∂y ∂z
 ∂ ∂ ∂
(ii) (a • ∇) φ =  a1 + a2 + a3  φ.
 ∂x ∂y ∂z 
 ∂φ ∂φ ∂φ 
Also a • ∇φ = (a1 i + a2 j + a3 k ) •  i+ j+ k
 ∂ x ∂ y ∂z 
∂φ ∂φ ∂φ
= a1 + a2 + a3 ⋅
∂x ∂y ∂z
Hence (a • ∇) φ = a • ∇φ.
 ∂ ∂ ∂ ∂r ∂r ∂r
(iii) (a • ∇) r =  a1 + a2 + a3  r = a1 + a2 + a3 ⋅
 ∂x ∂y ∂z  ∂x ∂y ∂z
But r = xi + yj + zk .
∂r ∂r ∂r
∴ = i, = j, = k.
∂x ∂y ∂z
∴ (a • ∇) r = a1 i + a2 j + a3 k = a .
x y
1. If F = e i + ( x − 2 y) j + x sin y k , calculate
∂F ∂F ∂2 F
(i) , (ii) , (iii) ,
∂x ∂y ∂x 2
∂2 F ∂2 F
(iv) , (v) ⋅
∂x ∂y ∂y 2
2. If f = (2 x 2 y − x 4 ) i + (e xy − y sin x) j + x 2 cos y k , verify that
∂2 f ∂2 f
= ⋅
∂y ∂x ∂x ∂y
D-248
2
3. If u = x y z i + x z j − y 3 k and v = x 3 i − x y z j + x 2 z k , calculate
∂2 u ∂2 v
× at the point (1, 1, 0).
∂y 2 ∂x 2
4. If φ ( x , y , z ) = x 2 y + y 2 x + z 2 , find ∇φ at the point (1, 1, 1).
5. Find grad f , where f is given by f = x 3 − y 3 + xz 2 , at the point (1, − 1, 2).
6. If φ ( x, y, z ) = xy 2 z and f = xzi − xyj + yz 2 k , show that
∂3
(φf ) at (2 , − 1, 1) is 4i + 2 j .
∂x 2 ∂z
7. If u = x + y + z , v = x 2 + y 2 + z 2 , w = yz + zx + xy, prove that
(grad u) • [(grad v) × (grad w)] = 0. (Meerut 2007B)
 ∂f ∂f   ∂f ∂f   ∂f ∂f 
8. If F =  y −z  i + z − x  j + x − y  k , prove that
 ∂z ∂y   ∂x ∂z   ∂y ∂x 
(i) F = r × ∇f , (ii) F • r = 0, (iii) F • ∇f = 0 .
2 1 /2 −1 /3 −3 /2 −7 /3
9. If φ = (3r − 4r + 6r ), show that ∇φ = 2 (3 − r −r )r.
10. Prove that ∇φ • d r = dφ . (Meerut 2005, 06, 09B; Kumaun 08)
11. Prove that f (u) ∇u = ∇ ∫ f (u) du .
(Kumaun 2012, 13)
12. ρ and p are two scalar point functions such that ρ is a function of p; show
dρ
that ∇ρ = ∇p .
dp
dφ dr
13. Show that = ∇φ • , where r = x i + y j + z k and φ is a function of
ds ds
x , y and z .
1 A•r
14. Prove that A • ∇  = − 3 ⋅
 r r (Meerut 2010)
−3 −5
15. Prove that ∇r = − 3r r. (Meerut 2009, 12)
16. Show that
(i) grad (r • a) = a , (Avadh 2010)
(ii) grad [r , a , b] = a × b ,
where a and b are constant vectors.
A nswers 1
x y x y
1. (i) ye i + j + sin y k (ii) xe i − 2 j + x cos y k
2 x y x y
(iii) y e i (iv) e ( x y + 1) i + cos y k
D-249
x y
(v) x2 e i − x sin y k
3. − 36 j 4. 3i + 3 j + 2 k
5. 7i − 3 j + 4k
5 Equipotential Surfaces or Level Surfaces

Let f ( x , y , z ) be a scalar field over a region R .The points satisfying an equation of
the type
f ( x , y , z ) = c , (arbitrary constant )
constitute a family of surfaces in three dimensional space. The surfaces of this
family are called level surfaces. Any surface of this family is such that the value of
the function f at any point of it is the same. Therefore these surfaces are also called
iso-f-surfaces.
Theorem 1: Let f ( x , y , z ) be a scalar field over a region R . Then through any point of
R there passes one and only one level surface.
Proof: Let ( x1 , y1 , z1 ) be any point of the region R .
Then the level surface f ( x, y, z ) = f ( x1 , y1 , z1 ) passes through this point.
Now suppose the level surfaces f ( x , y , z ) = c1 and f ( x , y , z ) = c 2 pass through
the point ( x1 , y1 , z1 ). Then
f ( x1 , y1 , z1 ) = c1 and f ( x1 , y1 , z1 ) = c 2 .
Since f ( x, y, z ) has a unique value at ( x1 , y1 , z1 ), therefore we have c1 = c 2 .
Hence only one level surface passes through the point ( x1 , y1 , z1 ).
Theorem 2: ∇f is a vector normal to the surface f ( x , y , z ) = c where c is a constant.
Proof: Let r = xi + yj + zk be the position vector of any point P ( x , y , z ) on the

level surface f ( x, y, z ) = c . Let Q ( x + δx , y + δy , z + δz ) be a neighbouring point
on this surface.
Then the position vector of Q = r + δr = ( x + δx) i + ( y + δy) j + (z + δz ) k .
→
∴ PQ = (r + δr) − r = δr = δx i + δy j + δz k .
As Q → P , the line PQ tends to tangent at P to the level surface.
Therefore d r = dx i + dy j + dz k lies in the tangent plane to the surface at P .
From the differential calculus, we have
∂f ∂f ∂f
df = dx + dy + dz
∂x ∂y ∂z
 ∂f ∂f ∂f 
= i + j +k  • (dx i + dy j + dz k ) = ∇f • d r.
 ∂x ∂y ∂z 
D-250
Since f ( x , y , z ) = constant, therefore df = 0.

∴ ∇f • dr = 0 so that ∇f is a vector perpendicular to d r and therefore to the
tangent plane at P to the surface f ( x, y, z ) = c .
Hence ∇ f is a vector normal to the surface f ( x, y, z ) = c .
Thus if f ( x, y, z ) is a scalar field defined over a region R , then ∇f at any point
( x , y , z ) is a vector in the direction of normal at that point to the level surface
f ( x, y, z ) = c passing through that point.
6 Directional Derivative of a Scalar Point Function

Definition: Let f ( x, y, z ) define a scalar field in a region R and let P be any point in
this region. Suppose Q is a point in this region in the neighbourhood of P in the direction of a
∧
given unit vector a .
lim f (Q) − f ( P )
Then Q → P , if it exists, is called the directional derivative of f at P in the
PQ
∧
direction of a .
Interpretation of Directional Derivative: Let P be the point ( x , y , z ) and

let Q be the point ( x + δx, y + δy, z + δz ). Suppose PQ = δs . Then δs is a small
∧
element at P in the direction of a. If
δf = f ( x + δx , y + δy , z + δz ) − f ( x , y , z ) = f (Q) − f ( P ),
δf
then represents the average rate of change of f per unit distance in the direction
δs
∧ ∧
of a . Now the directional derivative of f at P in the direction of a is
lim f (Q) − f ( P ) = lim δf = df ⋅ It represents the rate of change of f with
Q→ P
PQ δ s → 0 δs ds
∧
respect to distance s at the point P in the direction of unit vector a .
Theorem 1: The directional derivative of a scalar field f at a point P ( x, y, z ) in the
∧ df ∧
direction of a unit vector a is given by = ∇ f • a.
ds
Proof: Let f ( x, y, z ) define a scalar field in the region R . Let r = xi + yj + zk
denote the position vector of any point P ( x, y, z ) in this region. If s denotes the
∧
distance of P from some fixed point A in the direction of a, then δs denotes small
∧ dr
element at P in the direction of a . Therefore is a unit vector at P in this
ds
dr ∧
direction i. e., = a.
ds
But r = x i + y j+ z k.
D-251
dr dx dy dz ∧
∴ = i+ j+ k = a.
ds ds ds ds
∧  ∂f ∂f ∂f   dx dy dz 
Now ∇f • a =  i + j +k  • i + j+ k 
 ∂x ∂y ∂z   ds ds ds 
∂f dx ∂f dy ∂f dz d f
= + + =
dx ds dy ds dz ds ds
∧
= directional derivative of f at P in the direction of a .
Alternative Proof: Let Q be a point in the neighbourhood of P in the direction of

∧
the given unit vector a . If l, m, n are the direction cosines of the line PQ , then
∧
l i + m j + n k = the unit vector in the direction of PQ = a . Further if PQ = δs, then
the co-ordinates of Q are ( x + lδs, y + mδs, z + nδs). Now the directional derivative
∧
of f at P in the direction of a is
f (Q) − f ( P )
= lim
Q→ P PQ
f ( x + lδs , y + mδs , z + nδs) − f ( x , y , z )
= lim
δs → 0 δs
 ∂f ∂f ∂f 
f ( x , y , z ) +  lδs + mδ s + nδs  +… − f ( x , y , z )
 ∂x ∂y ∂z 
= lim ,
δs → 0 δs
on expanding by Taylor’s theorem

∂f ∂f ∂f
=l +m +n
∂x ∂y ∂z
 ∂f ∂f ∂f  ∧
= i+ j+ k  • (l i + m j + n k ) = ∇f • a .
 ∂x ∂y ∂z 
∧
Theorem 2: If n be a unit vector normal to the level surface f ( x , y , z ) = c at a point
∧
P ( x , y , z ) and n be the distance of P from some fixed point A in the direction of n so that δn
∧ df ∧
represents element of normal at P in the direction of n , then grad f = n.
dn
∂f ∂f ∂f
Proof: We have grad f = ∇f = i+ j+ k.
∂x ∂y ∂z
∧
Also grad f is a vector normal to the surface f ( x , y , z ) = c . Since n is a unit vector
∧
normal to the surface f ( x , y , z ) = c , therefore let grad f = A n, where A is some
scalar to be determined.
D-252
df ∧
Now = directional derivative of f in the direction of n
dn
∧ ∧ ∧ ∧
= ∇f • n = A n • n = A. [∵ ∇ f = grad f = A n ]
df ∧
∴ grad f = ∇ f = n.
dn
∧ df
Note: If the vector n is in the direction of f increasing, then is positive.
dn
Therefore ∇f is a vector normal to the surface f ( x , y , z ) = c in the direction of f
increasing.
Theorem 3: Grad f is a vector in the direction of which the maximum value of the
df
directional derivative of f i.e., occurs.
ds
∧
Proof: The directional derivative of f in the direction of a is given by
df ∧  df ∧  ∧  df ∧ 
=∇ f •a=  n  •a ∵ ∇f = n
ds  dn   dn 
df ∧ ∧ df ∧ ∧
= (n • a ) = cos θ, where θ is the angle between a and n .
dn dn
df df
Now is fixed. Therefore cos θ is maximum when cos θ is maximum i. e., when
dn dn
∧ ∧ ∧
cos θ = 1 . But cos θ will be 1 when the angle between a and n is 0 i. e., when a is along
∧
the unit normal vector n .
Therefore the directional derivative is maximum along the normal to the surface.
df
Its maximum value is = = |grad f |.
dn
7 Tangent Plane and Normal to a Level Surface

Tangent plane to a surface: The tangent to any curve drawn on a surface is
called a tangent line to the surface. All the tangent lines to a surface at the point P lie
in a plane. This plane is called the tangent plane to the surface at the point P on it.
Normal to a surface: The normal to a surface at the point P is a straight line
passing through P and perpendicular to the tangent plane at P.
Angle between two surfaces: The angle between the two surfaces at a point P is
the angle between the normals to the two surfaces at that point.
Equations of the Tangent Plane and Normal to a Surface:
To find the equations of the tangent plane and normal to the surface
f ( x , y , z) = c .
Let f ( x , y , z ) = c be the equation of a level surface. Let r = x i + y j + z k be the
position vector of any point P ( x , y , z ) on this surface.
D-253
∂f ∂f ∂f
Then ∇f = i+ j+ k is a vector along the normal to the surface at P i. e.,∇f
∂x ∂y ∂z
is perpendicular to the tangent plane at P .
Tangent plane at P: Let R = X i + Y j + Z k be the position vector of any
current point Q ( X , Y , Z ) on the tangent plane at P to the surface. The vector
→
PQ = R − r = ( X − x) i + (Y − y) j + (Z − z ) k
lies in the tangent plane at P . Therefore it is perpendicular to the vector ∇f .
∴ (R − r) • ∇f = 0
 ∂f ∂f ∂f 
or [( X − x) i + (Y − y) j + (Z − z ) k ] •  i+ j+ k = 0
 ∂x ∂y ∂z 
∂f ∂f ∂f
or ( X − x) + (Y − y) + (Z − z ) = 0, …(1)
∂x ∂y ∂z
is the equation of the tangent plane at P.
Normal at P: Let R = X i + Y j + Z k
be the position vector of any current point Q ( X , Y , Z ) on the normal at P to the
→
surface. The vector PQ = R − r = ( X − x) i + (Y − y) j + (Z − z ) k lies along the
normal at P to the surface. Therefore it is parallel to the vector ∇f .
∴ (R − r) × ∇f = 0 …(2)
is the vector equation of the normal at P to the given surface.
Cartesian form: The vectors
∂f ∂f ∂f
( X − x) i + (Y − y) j + (Z − z ) k and ∇ f = i+ j+ k,
∂x ∂y ∂z
will be parallel if
 ∂f ∂f ∂f 
( X − x) i + (Y − y) j + (Z − z ) k = p  i+ j+ k ,
 ∂ x ∂ y ∂z 
where p is some scalar.
Equating the coefficients of i , j, k , we get
∂f ∂f ∂f
X − x= p ,Y − y = p ,Z − z = p
∂x ∂y ∂z
X − x Y − y Z −z
or = = …(3)
∂f ∂f ∂f
∂x ∂y ∂z
which are the equations of the normal at P.
 ∂F ∂F ∂F 
Note: The vector  , ,  is along the normal to the surface
 ∂x ∂y ∂z 
F ( x, y, z ) = 0 at the point ( x, y, z ).
D-254
Example 5: Find a unit normal vector to the level surface x 2 y + 2 xz = 4 at the point
(2 , − 2 , 3). (Kashi 2014)
2
Solution: The equation of the level surface is f ( x, y, z ) ≡ x y + 2 xz = 4.
The vector grad f is along the normal to the surface at the point ( x , y , z ).
We have grad f = ∇ ( x 2 y + 2 xz ) = (2 x y + 2z ) i + x 2 j + 2 x k .
∴ at the point (2 , − 2 , 3), grad f = − 2 i + 4 j + 4k .
∴ −2 i + 4 j + 4 k is a vector along the normal to the given surface at the point
(2 , − 2 , 3).
Hence a unit normal vector to the surface at this point
−2 i + 4 j + 4 k −2 i + 4 j + 4 k 1 2 2
= = = − i + j+ k.
|−2 i + 4 j + 4 k| √ (4 + 16 + 16) 3 3 3
 1 2 2  1 2 2
The vector −  − i + j + k  i. e., i − j − k is also a unit normal vector to
 3 3 3  3 3 3
the given surface at the point (2 , − 2 , 3).
Alternate Solution: The given surface is
φ ( x, y, z ) ≡ x 2 y + 2 xz − 4 = 0.
∂φ ∂φ ∂φ
We have = 2 x y + 2z , = x2 , = 2z .
∂x ∂y ∂z
∴ At the point (2, − 2, 3), we have
∂φ ∂φ ∂φ
= − 2, = 4, = 4.
∂x ∂y ∂z
∴ A vector along the normal to the given surface at the point (2, − 2, 3)
 ∂φ ∂φ ∂φ
= , ,  = (− 2, 4, 4) or (− 1, 2, 2).
 ∂x ∂y ∂z 
We have |(− 1, 2, 2)| = (− 1)2 + 22 + 22 = 3.

Hence, a unit normal vector to the given surface at the point (2, − 2, 3)
1 1 2 2
= (− 1, 2, 2) =  − , ,  ⋅
3  3 3 3
Example 6: Find the directional derivative of f ( x, y, z ) = x 2 yz + 4 xz 2 at the point
(1, − 2 , − 1) in the direction of the vector 2 i − j − 2 k .
(Bundelkhand 2007; Rohilkhand 11; Kashi 13; Agra 14)
Solution: We have f ( x, y, z ) = x 2 yz + 4 xz 2 .
∴ grad f = (2 xyz + 4z 2 ) i + x 2 zj + ( x 2 y + 8 xz ) k
= 8i − j − 10 k at the point (1, − 2 , − 1).
D-255
∧
If a be the unit vector in the direction of the vector 2i − j − 2k , then
∧ 2i − j−2k 2 1 2
a= = i − j − k.
√ (4 + 1 + 4) 3 3 3
Therefore the required directional derivative is
df ∧  2 1 2 
= grad f • a = (8i − j − 10 k ) •  i − j − k 
ds  3 3 3 
16 1 20 37
= + + = ⋅
3 3 3 3
Since this is positive, f is increasing in this direction.
Example 7: Find the directional derivative of the function f = x 2 − y 2 + 2z 2 at the
point P (1, 2 , 3) in the direction of the line PQ where Q is the point (5, 0, 4).
(Purvanchal 2007; Kashi 14)
Solution: Here
∂f ∂f ∂f
grad f = i+ j+ k
∂x ∂y ∂z
= 2 xi − 2 yj + 4zk = 2i − 4 j + 12k at the point (1, 2 , 3).
→
Also PQ = position vector of Q − position vector of P
= (5i + 0 j + 4k ) − (i + 2 j + 3k ) = 4i − 2 j + k .
∧ →
If a be the unit vector in the direction of the vector PQ , then
∧ 4i −2 j+ k 4i −2 j+ k
a= = ⋅
√ (16 + 4 + 1) √ (21)
∴ the required directional derivative
∧  4i − 2 j + k 
= (grad f ) • a = (2i − 4 j + 12 k ) •  
 √ (21) 
28 28 4
= = √ (21) = √ (21).
√ (21) 21 3
Example 8: In what direction from the point (1, 1, − 1) is the directional derivative of
2 2
f = x −2y + 4z 2 a maximum ? Also find the value of this maximum directional
derivative. (Kanpur 2008)
Solution: We have
grad f = 2 x i − 4 y j + 8z k
= 2 i − 4 j − 8 k at the point (1, 1, − 1).
The directional derivative of f is maximum in the direction of grad f
=2i −4 j−8k.
The maximum value of this directional derivative
= |grad f | = |2 i − 4 j − 8 k | = √ (4 + 16 + 64) = √ (84) = 2 √ (21).
D-256
Example 9: What is the greatest rate of increase of u = xyz 2 at the point (1, 0 , 3) ?
Solution: We have ∇u = yz 2 i + xz 2 j + 2 xyz k .

∴ at the point (1, 0, 3), we have ∇u = 0 i + 9 j + 0 k = 9 j.
The greatest rate of increase of u at the point (1, 0, 3)
du
= the maximum value of at the point (1, 0, 3)
ds
= |∇u |, at the point (1, 0, 3) = | 9 j | = 9.
Example 10: Find the equations of the tangent plane and normal to the surface
2
2 xz − 3 xy − 4 x = 7 at the point (1, − 1, 2).
Solution: The equation of the surface is

f ( x, y, z ) ≡ 2 xz 2 − 3 xy − 4 x = 7.
We have
grad f = (2z 2 − 3 y − 4) i − 3 x j + 4 x z k
= 7i − 3 j + 8k , at the point (1, − 1, 2).
∴ 7 i − 3 j + 8 k is a vector along the normal to the surface at the point (1, − 1, 2).
The position vector of the point (1, − 1, 2) is = r = i – j + 2k .
If R = X i + Y j + Z k is the position vector of any current point ( X , Y , Z ) on the
tangent plane at (1, − 1, 2), then the vector R − r is perpendicular to the vector
grad f .
∴ the equation of the tangent plane is (R − r) • grad f = 0
i. e., {( X i + Y j + Z k ) − (i − j + 2k )} • (7i − 3 j + 8k ) = 0
i. e., {( X − 1) i + (Y + 1) j + (Z − 2) k } • (7i − 3 j + 8k ) = 0
i. e., 7 ( X − 1) − 3 (Y + 1) + 8 (Z − 2) = 0.
The equations of the normal to the surface at the point (1, − 1, 2) are
X −1 Y +1 Z − 2 X −1 Y +1 Z − 2
= = i. e., = = ⋅
∂f ∂f ∂f 7 −3 8
∂x ∂y ∂z
Example 11: Find the equations of the tangent plane and normal to the surface xyz = 4 at
the point (1, 2 , 2). (Meerut 2000)
Solution: The equation of the surface is f ( x , y , z ) ≡ x y z − 4 = 0.

We have grad f = yz i + x z j + x y k = 4i + 2 j + 2k , at the point (1, 2 , 2).
∴ 4i + 2 j + 2k is a vector along the normal to the surface at the point (1, 2 , 2).
The position vector of the point (1, 2 , 2) is = r = i + 2 j + 2k .
If R = X i + Y j + Z k is the position vector of any current point ( X , Y , Z ) on the
tangent plane at (1, 2 , 2), the equation of the tangent plane is
(R − r) • grad f = 0,
D-257
i. e., { ( X i + Y j + Z k ) − (i + 2 j + 2k )} • (4i + 2 j + 2k ) = 0
i. e., { ( X − 1) i + (Y − 2) j + (Z − 2) k} • (4i + 2 j + 2k ) = 0
i. e., 4 ( X − 1) + 2 (Y − 2) + 2 (Z − 2) = 0
i. e., 4 X + 2Y + 2 Z = 12 , i. e., 2 X + Y + Z = 6 .
The equations of the normal to the surface at the point (1, 2 , 2) are
X −1 Y − 2 Z − 2
= =
∂f ∂f ∂f
∂x ∂y ∂z
X −1 Y −2 Z −2 X −1 Y −2 Z −2
i. e., = = , i. e., = = ⋅
4 2 2 2 1 1
Example 12: Find the angle between the surfaces x 2 + y 2 + z 2 = 9, and
z = x 2 + y 2 − 3 at the point (2 , − 1, 2). (Meerut 2001)
Solution: Angle between two surfaces at a point is the angle between the normals
to the surfaces at that point. Let f1 = x 2 + y 2 + z 2 and f 2 = x 2 + y 2 − z .
Then grad f1 = 2 x i + 2 y j + 2 z k and grad f 2 = 2 x i + 2 y j − k .
Let n1 = grad f1 at the point (2 , − 1, 2) and n 2 = grad f 2 at the point (2 , − 1, 2).
Then n1 = 4i − 2 j + 4k and n 2 = 4i − 2 j − k .
The vectors n1 and n 2 are along normals to the two surfaces at the point (2 , − 1, 2).
If θ is the angle between these vectors, then
n1 • n 2 = |n1||n 2| cos θ
or 16 + 4 − 4 = √ (16 + 4 + 16) √ (16 + 4 + 1) cos θ.
16 8
∴ cos θ = or θ = cos −1 ⋅
6 √ (21) 3 √ (21)
1. (i) Find the gradient and the unit normal to the level surface
x 2 + y − z = 4 at the point (2 , 0, 0).
(ii) Find the unit normal to the surface z = x 2 + y 2 at the point
(−1, − 2 , 5).
2. (i) Find the unit vector normal to the surface x 2 − y 2 + z = 2 at the point
(1, − 1, 2).
(ii) Find the unit normal to the surface x 4 − 3 xyz + z 2 + 1 = 0 at the point
(1, 1, 1).
(iii) Find a unit normal vector to the surface x 2 y + 2 xz = 4 at the point
(2 , − 2 , 3).
D-258
3. (i) Find the directional derivatives of a scalar point function f in the

direction of coordinate axes.
(ii) Find the directional derivative of φ = xy + yz + zx in the direction of
the vector i + 2 j + 2k at (1, 2 , 0). (Kumaun 2015)
(iii) Find the directional derivative of φ ( x , y , z ) = x 2 yz + 4 xz 2 at the
point (1, − 2 , 1) in the direction of 2 i − j − 2k .
(iv) Find the directional derivative of f ( x , y , z ) = x 2 − 2 y 2 + 4z 2 at the
point (1, 1, − 1) in the direction of 2 i + j − k .
(v) Find the directional derivative of the function f = xy + yz + zx in the
direction of the vector 2i + 3 j + 6k at the point (3 , 1, 2).
(vi) Find the directional derivative of the function xy 2 + yz 2 + zx 2 along
the tangent to the curve x = t , y = t 2 , z = t 3 at the point (1, 1, 1.
)
4. Find the directional derivatives of φ = x y z at the point (2 , 2 , 2), in the
directions (i) i , (ii) j , (iii) i + j + k .
5. For the function f = y / ( x 2 + y 2 ), find the value of the directional
derivative making an angle 30° with the positive x-axis at the point (0, 1).
6. (i) Find the greatest value of the directional derivative of the function
2 x 2 − y − z 4 at the point (2 , − 1, 1).
(ii) In what direction the directional derivative of φ = x 2 y 2 z from (1, 1, 2)
will be maximum and what is its magnitude ? Also find a unit normal
vector to the surface x 2 y 2 z = 2 at the point (1, 1, 2).
(iii) Find the maximum value of the directional derivative of φ = x 2 yz at
the point (1, 4 , 1).
(iv) Calculate the maximum rate of change and the corresponding
direction for the function φ = x 2 y 3 z 4 at the point 2i + 3 j − k .
(v) Find the values of the constants a, b, c so that the directional derivative
of φ = ax 2 + by 2 + c z 2 at (1, 1, 2) has a maximum magnitude 4 in
the direction parallel to y-axis.
7. Find the equation of the tangent plane to the surface yz − zx + x y + 5 = 0,
at the point (1, − 1, 2).
8. (i) Find the equations of the tangent plane and normal to the surface
x 2 + y 2 + z 2 = 25 at the point (4, 0, 3).
(ii) Given the curve x 2 + y 2 + z 2
= 1, x + y + z = 1 (intersection of two
surfaces), find the equations of the tangent line at the point (1, 0, 0).
(iii) Find the equations of the tangent plane and normal to the surface
z = x 2 + y 2 at the point (2 , − 1, 5).
D-259
9. (i) Find the equations of the tangent plane and the normal to the surface
x 2 + 2 y 2 + 3z 2 = 12 at the point (1, 2 , − 1).
(ii) Find the equations of the tangent plane and the normal to the surface
xy + yz + z x = 1, at the point (2, 3, − 1).
(iii) Find the equations of the tangent plane and the normal to the surface
z = x 2 − 2 xy − y 2 at the point (1, 2 , − 7).
(iv) Find the equation of the tangent plane to the surface x 2 + y 2 + z 2 = 9
at (2 , − 1, 2).
(v) Find the equation of the tangent plane to the surface z = x 2 + y 2 at
the point (1, − 1, 2).
10. Show that the directional derivative of a scalar point function at any point
along any tangent line to the level surface at the point is zero.
11. If F and f are point functions, show that the components of the former,
tangential and normal to the level surface f = 0 are
∇f × (F × ∇f ) (F • ∇f ) ∇f
2
and ⋅
(∇f ) (∇f )2
12. Find the angle of intersection at (4, − 3, 2) of spheres x 2 + y 2 + z 2

= 29 and
2 2 2
x + y +z + 4 x − 6 y − 8z − 47 = 0.
13. Find the constants a and b so that the surface ax 2 − byz = (a + 2) x will be
orthogonal to the surface 4 x 2 y + z 3 = 4 at the point (1, − 1, 2).
14. Show that the sum of the squares of the intercepts on the coordinate axes
made by the tangent plane to the surface x 2 /3 + y 2 /3 + z 2 /3 = a2 /3 is
constant.
A nswers 2
1 − (2i + 4 j + k)
1. (i) 4i + j − k ; (4 i + j − k ) (ii)
3 √2 √ (21)
1 i − 3j − k 1 2 2
2. (i) (2 i + 2 j + k ) (ii) (iii)  − , , 
3 √ (11)  3 3 3
∂f ∂f ∂f
3. (i) , and in the directions of i , j and k
∂x ∂y ∂z
10 13 8
(ii) (iii) − (iv)
3 3 6
45 18
(v) (vi)
7 √ (14)
D-260
4. (i) 4 (ii) 4 (iii) 4 3

5. −1 / 2
6. (i) 9
4i +4 j+ k
(ii) √ (33) in the direction of the vector 4 i + 4 j + k ;
√ (33)
(iii) 9
(iv) 324 √ 2 in the direction of the vector 108 (i + j − 4 k)
(v) a = 0, b = 2, c = 0
7. 3 X − 3Y + 2 Z = 10
x−4 y z −3
8. (i) 4 x + 3z = 25 ; = =
4 0 3
Y Z x−2 y +1 z −5
(ii) X = 1, = (iii) 4 x − 2 y − z = 5 ; = =
−1 1 4 −2 −1
x −1 y − 2 z +1
9. (i) x + 4 y − 3z = 12 ; = =
1 4 −3
x−2 y − 3 z +1
(ii) 2 x + y + 5z = 2 ; = =
2 1 5
x −1 y − 2 z + 7
(iii) 2 x + 6 y + z = 7; = =
2 6 1
(iv) 2 x − y + 2 z = 9 (v) 2 x − 2 y − z = 2
12. cos −1 √ (19 / 29) 13. a = 5 / 2, b = 1
8 Divergence of a Vector Point Function

(Agra 2005)
Definition: Let V be any given differentiable vector point function. Then the divergence of
V, written as, ∇ • V or div V,
 ∂ ∂ ∂
is defined as div V = ∇ • V =  i + j +k  •V
 ∂x ∂y ∂z 
∂V ∂V ∂V ∂V
=i• + j• +k • = ∑i • ⋅
∂x ∂y ∂z ∂x
It should be noted that div V is a scalar quantity. Thus the divergence of a vector point
function is a scalar point function.
Theorem: If V = V1 i + V2 j + V3 k is a differentiable vector point function, then

∂V1 ∂V2 ∂V3
div V = + + ⋅
∂x ∂y ∂z
Proof: We have by definition

D-261
∂V ∂V ∂V
div V = ∇ • V = i • + j• +k • ⋅
∂x ∂y ∂z
Now V = V1 i + V2 j + V3 k .
∂V ∂V1 ∂V2 ∂V3
∴ = i+ j+ k.
∂x ∂x ∂x ∂x
∂V  ∂V ∂V2 ∂V3  ∂V1
∴ i• =i • 1 i + j+ k = ⋅
∂x  ∂x ∂x ∂x  ∂x
∂V ∂V2 ∂V ∂V3
Similarly, j• = and k • = ⋅
∂y ∂y ∂z ∂z
∂V1 ∂V2 ∂V3
Hence, div V = + + ⋅
∂x ∂y ∂z
Solenoidal Vector: A vector V is said to be solenoidal if div V = 0.
9 Curl of a Vector Point Function

(Agra 2005)
Definition: Let f be any given differentiable vector point function. Then the curl or
rotation of f , written as ∇ × f , curl f or rot f is defined as
 ∂ ∂ ∂
curl f = ∇ × f =  i + j +k  ×f
 ∂x ∂y ∂z 
∂f ∂f ∂f ∂f
=i× + j× +k × = ∑i × ⋅
∂x ∂y ∂z ∂x
It should be noted that curl f is a vector quantity. Thus the curl of a vector point function
is a vector point function.
Theorem: If f = f1 i + f 2 j + f 3 k is a differentiable vector point function, then
 ∂f ∂f   ∂f ∂f   ∂f ∂f 
curl f =  3 − 2  i +  1 − 3  j +  2 − 1  k .
 ∂y ∂z   ∂z ∂x   ∂x ∂y 
Proof: We have by definition
∂f ∂f ∂f
curl f = ∇ × f = i × + j× +k ×
∂x ∂y ∂z
∂ ∂
=i× ( f1 i + f 2 j + f 3 k ) + j × ( f1 i + f 2 j + f 3 k )
∂x ∂y
∂
+k× ( f1 i + f 2 j + f 3 k )
∂z
 ∂f ∂f ∂f   ∂f ∂f ∂f 
= i ×  1 i + 2 j + 3 k + j ×  1 i + 2 j + 3 k
 ∂x ∂x ∂x   ∂y ∂y ∂y 
 ∂f ∂f ∂f 
+ k ×  1 i + 2 j + 3 k
 ∂z ∂z ∂z 
D-262
 ∂f ∂f   ∂f ∂f   ∂f ∂f 
= 2 k − 3 j +  − 1 k + 3 i +  1 j − 2 i
 ∂x ∂x   ∂y ∂y   ∂ z ∂z 
 ∂f ∂f   ∂f ∂f   ∂f 2 ∂f 
=  3 − 2 i +  1 − 3 j+  − 1 k.
 ∂ y ∂ z   ∂ z ∂x   ∂x ∂y 
Note: It should be noted that the expression for curl f can be written immediately
if we treat the operator ∇ as a vector quantity. Thus
 ∂ ∂ ∂
Curl f = ∇ × f =  i + j +k  × ( f1 i + f 2 j + f 3 k )
 ∂x ∂y ∂z 
i j k
∂ ∂ ∂ 
= 
∂x ∂y ∂z
f f2 f3
 1 
∂ ∂ ∂ ∂ ∂ ∂
= ∂y ∂z i −
∂x ∂z 
 j + ∂x ∂y k
  
 f2 f 3  f1 f 3  f1 f 2
 ∂f ∂f   ∂f ∂f   ∂f ∂f 
=  3 − 2  i +  1 − 3  j +  2 − 1 k .
 ∂ y ∂ z   ∂ z ∂ x   ∂ x ∂y 
But we must take care that in the expansion of the determinant the operators
∂ ∂ ∂
, , must precede the functions f1 , f 2 , f 3 .
∂x ∂y ∂z
Irrotational vector: A vector f is said to be irrotational if ∇ × f = 0.
10 The Laplacian Operator ∇ 2

∂2 ∂2 ∂2
The Laplacian operator ∇ 2 is defined as ∇ 2 ≡ + + ⋅
∂x 2 ∂y 2 ∂z 2
∂2 f ∂2 f ∂2 f
If f is a scalar point function, then ∇ 2 f = + + ⋅
∂x 2 ∂y 2 ∂z 2
It should be noted that ∇ 2 f is also a scalar quantity.
∂2 f ∂2 f ∂2 f
If f is a vector point function, then ∇ 2 f = + + ⋅
∂x 2 ∂y 2 ∂z 2
It should be noted that ∇ 2 f is also a vector quantity.
Laplace’s equation: The equation ∇ 2 f = 0 is called Laplace’s equation. A function

which satisfies Laplace’s equation is called a harmonic function.
D-263
Example 13: Prove that div r = 3. (Kumaun 2000; Garhwal 01;

Bundelkhand 01, 04; Meerut 03, 04, 07B, 08;
Agra 08; Lucknow 05; Kanpur 13; Avadh 09; Purvanchal 11)
Solution: We have r = x i + y j + z k .
 ∂ ∂ ∂ ∂r ∂r ∂r
By definition, div r = ∇ • r =  i + j +k  •r = i • + j• +k•
 ∂x ∂y ∂z  ∂x ∂y ∂z
 ∂r ∂r ∂r 
= i •i + j• j + k •k ∵ =i, = j, = k
 ∂x ∂y ∂z 
= 1 + 1 + 1 = 3.
Example 14: Prove that curl r = 0.

(Kumaun 2000; Garhwal 02; Meerut 11; Bundelkhand 06; Agra 08;
Avadh 09; Kashi 13; Purvanchal 11; Rohilkhand 14)
Solution: We have by definition
 ∂ ∂ ∂
curl r = ∇ × r =  i + j +k  ×r
 ∂x ∂y ∂z 
∂r ∂r ∂r
=i× + j× +k × ⋅
∂x ∂y ∂z
Now r = x i + y j + z k.
∂r ∂r ∂r
∴ =i, = j, = k.
∂x ∂y ∂z
∴ curl r = i × i + j × j + k × k
= 0 + 0 + 0 = 0.
Example 15: If f = x 2 y i − 2 xz j + 2 yz k , find

(i) div f , (Kumaun 2008)
(ii) curl f , (Meerut 2012)
(iii) curl curl f . (Meerut 2001; Kanpur 14)
Solution: (i) We have
 ∂ ∂ ∂
div f = ∇ • f =  i + j +k  • ( x 2 y i − 2 xz j + 2 yz k )
 ∂x ∂y ∂z 
∂ 2 ∂ ∂
= ( x y) + (−2 xz ) + (2 yz )
∂x ∂y ∂z
= 2 xy + 0 + 2 y
= 2 y ( x + 1).
D-264
 i j k
 ∂ ∂ ∂
(ii) We have curl f = ∇ × f =  
 ∂x ∂y ∂z 
2 2 yz
x y −2 xz 
∂ ∂  ∂ ∂ 2 
=  (2 yz ) − (−2 xz ) i −  (2 yz ) − ( x y) j
 ∂ y ∂ z   ∂ x ∂ z 
∂ ∂ 2 
+  (− 2 xz ) − ( x y) k
 ∂ x ∂ y 
2
= (2z + 2 x) i − 0 j + (−2z − x ) k
= (2 x + 2z ) i − ( x 2 + 2z ) k .
(iii) We have curl curl f = ∇ × ( ∇ × f ) = ∇ × [(2 x + 2z ) i − ( x 2 + 2z ) k ]
 i j k 
 ∂ ∂ ∂ 
= 
 ∂x ∂y ∂z 
2 x + 2z 0 − x 2 − 2z

∂  ∂ ∂
=  (− x 2 − 2z ) i −  (− x 2 − 2z ) − (2 x + 2z ) j
 ∂ y   ∂ x ∂ z 
 ∂ 
+ 0 − (2 x + 2z ) k
 ∂y 
= 0 i − (−2 x − 2) j + (0 − 0) k = (2 x + 2) j.
Example 16: Determine the constant a so that the vector

V = ( x + 3 y) i + ( y − 2 z ) j + ( x + az ) k is solenoidal.
(Rohilkhand 2009B)
Solution: A vector V is said to be solenoidal if div V = 0.
∂ ∂ ∂
We have div V = ∇ • V = ( x + 3 y) + ( y − 2z ) + ( x + az )
∂x ∂y ∂z
= 1 + 1 + a = 2 + a.
Now div V = 0 if 2 + a = 0 i. e., if a = − 2.
Example 17: Show that the vector V = (sin y + z ) i + ( x cos y − z ) j + ( x − y) k is
irrotational.
Solution: A vector V is said to be irrotational if curl V = 0. We have
curl V = ∇ × V
 i j k 
 ∂ ∂ ∂ 
= 
 ∂ x ∂y ∂z 
sin y + z x cos y − z x− y
 
D-265
∂ ∂  ∂ ∂
=  ( x − y) − ( x cos y − z ) i −  ( x − y) − (sin y + z ) j
 ∂ y ∂ z   ∂ x ∂ z 
∂ ∂ 
+  ( x cos y − z ) − (sin y + z ) k
 ∂x ∂y 
= (−1 + 1) i − (1 − 1) j + (cos y − cos y) k = 0.
∴ V is irrotational.
Example 18: Prove that ∇ • (r 3 r) = 6r 3 . (Purvanchal 2006, 10)

∴ r 3 r = r 3 ( x i + y j + z k ) = r 3 x i + r 3 y j + r 3 z k.
∴ ∇ • (r 3 r) = div (r 3 r)
∂ 3 ∂ 3 ∂ 3
= (r x) + (r y) + (r z )
∂x ∂y ∂z
∂r ∂r ∂r
= r 3 + 3r 2 x + r 3 + 3r 2 y + r 3 + 3r 2 z
∂x ∂y ∂z
 ∂r ∂r ∂r 
= 3r 3 + 3r 2  x + y +z  …(1)
 ∂x ∂y ∂z 
Now r 2 = x2 + y2 + z 2 .
∂r ∂r x
∴ 2r = 2 x or = ⋅
∂x ∂x r
∂r y ∂r z
Similarly = and = ⋅
∂y r ∂z r
 x y z
∴ from (1), ∇ • (r 3 r) = 3r 3 + 3r 2  x ⋅ + y ⋅ + z ⋅ 
 r r r
 x2 + y2 + z 2 
= 3r 3 + 3r 2  

 r 
r2
= 3r 3 + 3r 2 ⋅ = 3r 3 + 3r 3 = 6r 3 .
r
Example 19: If V is a constant vector, show that
(i) div V = 0, (Rohilkhand 2005; Purvanchal 14)
(ii) curl V = 0. (Purvanchal 2014; Kumaun 15)
Solution: (i) We have
∂V ∂V ∂V
div V = i • + j• +k •
∂x ∂y ∂z
= i • 0 + j • 0 + k • 0 = 0.
(ii) We have
D-266
∂V ∂V ∂V
curl V = i × + j× +k ×
∂x ∂y ∂z
= i × 0 + j × 0 + k × 0 = 0.
Example 20: If a is a constant vector, find

(i) div (r × a) , (Bundelkhand 2006; Kumaun 07, 11)
(ii) curl (r × a). (Bundelkhand 2006; Kanpur 06; Kumaun 07;
Agra 08; Rohilkhand 14)
Let a = a1 i + a2 j + a3 k . Then the scalars a1 , a2 , a3 are all constants.

We have
i j k
r × a = x y z
 
a1 a2 a3
= (a3 y − a2 z ) i + (a1 z − a3 x) j + (a2 x − a1 y) k .

∂ ∂ ∂
(i) div (r × a) = (a3 y − a2 z ) + (a1 z − a3 x) + (a3 x − a1 y)
∂x ∂y ∂z
= 0 + 0 + 0 = 0.
(ii) curl (r × a) = ∇ × (r × a)
 i j k 
 ∂ ∂ ∂ 
= 
 ∂x ∂y ∂z 
a3 y − a2 z a1 z − a3 x a2 x − a1 y
 
∂ ∂ 
=  (a2 x − a1 y) − (a1 z − a3 x) i
 ∂ y ∂ z 
∂ ∂
−  (a2 x − a1 y) − (a3 y − a2 z ) j
 ∂x ∂z 
∂ ∂ 
+  (a1 z − a3 x) − (a3 y − a2 z ) k
 ∂ x ∂ y 
= − 2a1 i − 2a2 j − 2a3 k = − 2 (a1 i + a2 j + a3 k ) = − 2a .
Example 21: If V = e xyz (i + j + k), find curl V.

 i j k 
 
∂ ∂ ∂
Solution: We have curl V = 
 x
∂ ∂y ∂z 
 xyz 
e e xyz e xyz
 
D-267
∂ ∂ xyz  ∂ ∂ xyz 
=  (e xyz ) − (e ) i −  (e xyz ) − (e ) j
 ∂ y ∂ z   ∂ x ∂ z 
∂ ∂ xyz 
+  (e xyz ) − (e ) k
 ∂x ∂y 
= e xyz ( xz − x y) i + e xyz ( x y − yz ) j + e xyz ( yz − xz ) k .
Example 22: Evaluate div f where f = 2 x 2 z i − xy 2 z j + 3 y 2 x k .
Solution: We have
 ∂ ∂ ∂
div f = ∇ • f =  i + j +k  • (2 x 2 z i − xy 2 z j + 3 y 2 x k )
 ∂x ∂y ∂z 
∂ ∂ ∂
= (2 x 2 z ) + (− xy 2 z ) + (3 y 2 x)
∂x ∂y ∂z
= 4 xz − 2 xyz + 0 = 2 xz (2 − y).
Example 23: Show that ∇ 2 ( x / r 3 ) = 0. (Kumaun 2008)
2 2 2
 x ∂ ∂ ∂   x
Solution: ∇ 2  3  =  2 + 2 + 2   3  ⋅
 r   ∂x ∂y ∂z   r 
∂2  x  ∂ ∂  x  ∂ 1 3 x ∂r 
Now 2  3
=   3  = − 4
∂x  r  ∂x  ∂x  r   ∂x  r 3 r ∂x 

∂ 1 3 x x ∵ r 2 = x 2 + y 2 + z 2 gives ∂r = x 
=  3 − 4  
∂x  r r r ∂x r 
∂ 1 3 x2  3 ∂r 6 x 15 x 2 ∂r
=  3 − 5  =− 4 − + 6
∂x  r r  r ∂x r 5 r ∂x
3 x 6 x 15 x 2 x 9 x 15 x 3
=− − + = − + 7 ⋅
r4 r r5 r6 r r5 r
∂2  x  ∂ ∂  x  ∂  3 x ∂r 
Again 2  3
=   3  = −
∂y  r  ∂ y  ∂y  r   ∂y  r 4 ∂y 
∂  3x y   ∂r y 
= − 4  ∵ = 
∂y  r r   ∂y r 
2
∂  3 xy  3 x 15 xy ∂r 3 x 15 x y
= − 5  = − 5 + = − + ⋅
∂y  r  r r 6 ∂y r5 r7
∂2  x  3 x 15 xz 2
Similarly   = − + ⋅
∂z 2  r 3  r5 r7
Therefore, adding we get
 x   ∂2 ∂2 ∂2   x 
∇ 2  3  =  2 + 2 + 2   3 
 r   ∂x ∂y ∂z   r 
D-268
9x 15 x 3 3x 15 xy 2 3x 15 xz 2
=− + − + − +
r5 r7 r5 r7 r5 r7
15 x 15 x 15 x 15 x
=− + (x2 + y2 + z 2 ) = − 5 + 7 r 2 = 0.
r5 r7 r r
1. (i) If F = x 2 z i − 2 y 3 z 2 j + xy 2 z k , find div F, curl F at (1, − 1, 1).
(ii) If f = ( y 2 + z 2 − x 2 ) i + (z 2 + x 2 − y 2 ) j + ( x 2 + y 2 − z 2 ) k ,
find div f and curl f .
(iii) If f = xy 2 i + 2 x 2 yz j − 3 yz 2 k , find div f and curl f . What are
their values at the point (1, − 1, 1) ? (Rohilkhand 2005)
3 3 3
2. Find div F and curl F where F = grad ( x + y +z − 3 xyz ).
(Rohilkhand 2007)
3. Find the divergence and curl of the vector
f = ( x 2 − y 2 ) i + 2 xyj + ( y 2 − xy) k . (Bundelkhand 2004)
4. Given φ = 2 x 3 y 2 z 4 , find div (grad φ).
5. If u = x 2 − y 2 + 4z , show that ∇ 2 u = 0.
6. If u = 3 x 2 y and v = xz 2 − 2 y, then find grad [(grad u) • (grad v)].
7. If f = ( x + y + 1) i + j + (− x − y) k , prove that f • curl f = 0.

8. (i) If f = f1 i + f 2 j + f 3 k , show that
∇ • f = ∇f1 • i + ∇f 2 • j + ∇f 3 • k . (Bundelkhand 2001)
3 3
(ii) Prove that ∇ • (r r) = 6 r .
9. Find the constants a, b, c so that the vector
F = ( x + 2 y + az ) i + (bx − 3 y − z ) j + (4 x + c y + 2 z ) k is irrotational.
(Bundelkhand 2005; Rohilkhand 08B)
10. Show that the vector F = 3 y z i + 4 x 3 z 2 j − 3 x 2 y 2 k is solenoid but not
4 2
irrotational . (Kashi 2014)
A nswers 3
1. (i) − 3 ; − 6i
(ii) − 2 ( x + y + z ) ; 2 ( y − z ) i + 2 (z − x) j + 2 ( x − y) k
D- 269
(iii) y 2 + 2 x 2 z − 6 yz ; 9 ; − (3z 2 + 2 x 2 y) i + (4 xyz − 2 xy) k; − i − 2k

2. 6 ( x + y + z ); 0 3. 4 x ; (2 y − x) i + y j + 4 y k
2 4 3 4 3 2 2
4. 12 xy z + 4x z + 24 x y z
2
6. (6 yz − 12 x) i + (6 xz 2 ) j + (12 xyz ) k
9. a = 4 , b = 2 , c = −1
11 Vector Identities
1. Prove that div (A + B) = div A + div B or ∇ • (A + B) = ∇ • A + ∇ • B .
Proof: We have
 ∂ ∂ ∂
div (A + B) = ∇ • (A + B) =  i + j +k  • (A + B)
 ∂x ∂y ∂z 
∂ ∂ ∂
=i• (A + B) + j • (A + B) + k • (A + B)
∂x ∂y ∂z
 ∂A ∂B   ∂A ∂B   ∂A ∂B 
=i • +  + j• +  +k • + 
 ∂x ∂x   ∂y ∂y   ∂z ∂z 
 ∂A ∂A ∂A   ∂B ∂B ∂B 
= i • + j• +k •  + i • + j• +k • 
 ∂x ∂y ∂z   ∂x ∂y ∂z 
= ∇ • A + ∇ • B = div A + div B .
2. Prove that curl (A + B) = curl A + curl B or ∇ × (A + B) = ∇ × A + ∇ × B .
(Rohilkhand 2008)
Proof: We have curl (A + B) = ∇ × (A + B)

 ∂ ∂ ∂
= i + j +k  × (A + B)
 ∂x ∂y ∂z 
∂  ∂A ∂B 
=Σi× (A + B) = Σ i ×  + 
∂x  ∂x ∂x 
∂A ∂B
=Σi× +Σi× = curl A + curl B .
∂x ∂x
3. If A is a differentiable vector function and φ is a differentiable scalar function, then
div (φA) = ( grad φ) • A + φ div A or ∇ • (φA) = (∇φ) • A + φ (∇ • A).
(Garhwal 2001; Agra 05; Kashi 14)
Proof: We have
 ∂ ∂ ∂
div (φA) = ∇ • (φA) =  i + j +k  • (φA)
 ∂x ∂y ∂z 
D-270
∂ ∂ ∂
=i• (φA) + j • (φA) + k • (φA)
∂x ∂y ∂z
 ∂    ∂φ ∂A  
= Σ  i •  (φA)  = Σ  i •  A+φ 
  ∂x     ∂x ∂x  
  ∂φ     ∂A  
= Σ i •  A  + Σ  i •  φ 
  ∂x     ∂x  
  ∂φ     ∂A  
= Σ  i • A + Σ  φ  i • 
  ∂x     ∂x  
[ ∵ a • (mb) = (ma) • b = m (a • b)]
 ∂φ   ∂A 
= Σ i • A + φ Σ i •  = (∇φ) • A + φ (∇ • A).
 ∂x   ∂x 
4. Prove that curl (φA) = ( grad φ) × A + φ curl A
or ∇ × (φA) = (∇φ) × A + φ (∇ × A).
(Garhwal 2002, 03; Bundelkhand 06)
 ∂ ∂ ∂
Proof: We have curl (φA) = ∇ × (φA) =  i + j +k  × (φA)
 ∂x ∂y ∂z 
∂   ∂φ ∂A  
= Σ  i × (φA) = Σ  i ×  A+φ 
 ∂ x    ∂ x ∂x  
  ∂φ     ∂A  
= Σ i ×  A  + Σ  i ×  φ 
  ∂x     ∂x  
  ∂φ     ∂A  
= Σ  i × A + Σ  φ  i × 

 ∂x     ∂x  
[∵ a × (mb) = (ma) × b = m (a × b)]
  ∂φ    ∂A 
= Σ  i  × A + φ Σ  i ×  = (∇φ) × A + φ (∇ × A).
  ∂x    ∂x 
5. Prove that div (A × B) = B • curl A − A • curl B
or ∇ • (A × B) = B • (∇ × A) − A • (∇ × B).
(Garhwal 2003; Agra 05; Meerut 04, 05B, 08, 09;
Bundelkhand 05, 07; Kashi 13; Avadh 09)
Proof: We have
∂   ∂A ∂B 
div (A × B) = Σ  i • (A × B) = Σ  i •  ×B+A× 
 ∂x    ∂ x ∂x 
  ∂A    ∂B  
= Σ i •  × B  + Σ  i •  A × 
  ∂ x     ∂x  
 ∂A     ∂B 
= Σ  i ×  • B − Σ  i •  × A 
  ∂ x     ∂ x 
[∵ a • (b × c) = (a × b) • c and a • (b × c) = – a • (c × b)]
D-271
  ∂A    ∂B  
= Σ  i ×   • B − Σ  i ×  • A
  ∂ x     ∂ x  
  ∂B  
= (curl A) • B −  Σ i ×  • A
  ∂x  
= (curl A) • B − (curl B) • A = B • curl A − A • curl B.
6. Prove that curl (A × B) = (B • ∇) A − B div A − (A • ∇) B + A div B.
(Garhwal 2001; Meerut 06B; Kumaun 09, 11, 12)
Proof: We have curl (A × B) = ∇ × (A × B)
∂   ∂B ∂A 
= Σ  i × (A × B) = Σ  i ×  A × + × B 
 ∂x    ∂x ∂x 
  ∂B     ∂A 
= Σ i ×  A ×   + Σ i ×  × B 
  ∂x     ∂x 
 ∂B  ∂B   ∂A  ∂A  
= Σ  i •  A − (i • A)  + Σ (i • B) − i •  B
 ∂x  ∂x   ∂x  ∂x  
 ∂B    ∂B 
= Σ  i •  A  − Σ (A • i) 
  ∂ x    ∂x 
∂A   ∂A  
+ Σ (B • i)  − Σ   i •  B
 ∂x   ∂x  
  ∂B    ∂  ∂
= Σ  i •   A − A • Σ i  B + B • Σ i  A
  ∂x    ∂x   ∂x 
  ∂A  
− Σ  i •  B
  ∂x  
= (div B) A − (A • ∇) B + (B • ∇) A − (div A) B .
7. Prove that grad (A • B) = (B • ∇ ) A + (A • ∇) B + B × curl A + A × curl B .

(Agra 2001; Kumaun 15)
Proof: We have
∂  ∂B ∂A 
grad (A • B) = ∇ (A • B) = Σ i (A • B) = Σ i  A • + • B
∂x  ∂x ∂x 
 ∂B    ∂A  
= Σ  A •  i + Σ   B •  i ⋅ …(1)
 ∂x    ∂x  
Now we know that a × (b × c) = (a • c) b - (a • b) c .
∴ (a • b) c = (a • c) b - a × (b × c).
 ∂B  ∂B  ∂B 
∴ A •  i = (A • i) −A× × i
 ∂x  ∂x  ∂x 
∂B  ∂B 
= (A • i) + A × i × ⋅
∂x  ∂x 
D-272
 ∂B   ∂B    ∂B  
Thus Σ  A •  i = Σ (A • i)  + Σ  A ×  i × 
  ∂x    ∂ x   ∂x  
∂  ∂B 
=  A • Σ i  B + A × Σ  i × 
 ∂x   ∂x 
= (A • ∇) B + A × (∇ × B). …(2)
 ∂A  
Similarly Σ  B •  i = (B • ∇) A + B × (∇ × A). …(3)
  ∂x  
Putting the values from (2) and (3) in (1), we get
grad (A • B) = (A • ∇ ) B + A × (∇ × B) + (B • ∇ ) A + B × (∇ × A).
Note: If we put A in place of B , then
grad (A • A) = 2 (A • ∇) A + 2A × (∇ × A)
1
or grad A2 = (A • ∇) A + A × curl A.
2
8. Prove that div grad φ = ∇ 2 φ i. e., ∇ • (∇φ) = ∇ 2 φ .

Proof: We have
 ∂ ∂ ∂   ∂φ ∂φ ∂φ 
∇ • (∇φ) =  i + j +k  • i+ j+ k
 ∂x ∂y ∂z   ∂x ∂y ∂z 
∂  ∂φ ∂  ∂φ ∂  ∂φ

=   +   +  
∂x  ∂x  ∂y  ∂y  ∂z  ∂z 
∂2 φ ∂2 φ ∂2 φ  ∂2 ∂2 ∂2 
= + + =  2 + 2 + 2  φ = ∇ 2 φ.
∂x 2 ∂y 2 ∂z 2  ∂x ∂y ∂z 
9. Prove that curl of the gradient of φ is zero i. e., ∇ × (∇φ) = 0 , i. e., curl grad φ = 0 .
(Bundelkhand 2014)
∂φ ∂φ ∂φ
Proof: We have grad φ = i+ j+ k.
∂x ∂y ∂z
∴ curl grad φ = ∇ × grad φ
 ∂ ∂ ∂   ∂φ ∂φ ∂φ 
= i + j +k  × i+ j+ k
 ∂ x ∂ y ∂ z   ∂ x ∂ y ∂z 
 
i j k
 
∂ ∂ ∂
= ∂x ∂y ∂z
 
∂φ ∂φ ∂φ
 ∂z
∂x ∂y
 
 ∂2 φ ∂2 φ   ∂2 φ ∂2 φ   ∂2 φ ∂2 φ 
= −  i+ −  j+  −  k
 ∂y ∂z ∂z ∂y   ∂z ∂x ∂x ∂z   ∂x ∂y ∂y ∂x 
D-273
= 0 i + 0 j + 0 k = 0,
provided we suppose that φ has continuous second partial derivatives so that the
order of differentiation is immaterial.
10. Prove that div curl A = 0 , i. e., ∇ • (∇ × A) = 0.

(Garhwal 2000; Meerut 02, 06B, 09; Agra 2000;
Rohilkhand 05; Bundelkhand 08; Kumaun 14)
Proof: Let A = A1 i + A2 j + A3 k . Then
 i j k 
 ∂ ∂ ∂ 
curl A = ∇ × A =  
 ∂x ∂y ∂y 
 A1 A2 A3 
 ∂A3 ∂A2   ∂A1 ∂A3   ∂A2 ∂A1 
=  −  i +  −  j+  −  k.
 ∂y ∂z   ∂z ∂x   ∂x ∂y 
Now div curl A = ∇ • (∇ × A)

∂  ∂A3 ∂A2  ∂  ∂A1 ∂A3  ∂  ∂A2 ∂A1 
=  −  +  −  +  − 
∂x  ∂y ∂z  ∂y  ∂z ∂x  ∂z  ∂x ∂y 
∂ 2 A3 ∂ 2 A2 ∂ 2 A1 ∂ 2 A3 ∂ 2 A2 ∂ 2 A1
= − + − + −
∂x ∂y ∂x ∂z ∂y ∂z ∂y ∂x ∂z ∂x ∂z ∂y
= 0 , assuming that A has continuous second partial derivatives.
11. Prove that ∇ × (∇ × A) = ∇ (∇ • A) − ∇ 2 A . (Meerut 2000, 07B, 10B;

Rohilkhand 14; Kumaun 10)
Proof: Let A = A1 i + A2 j + A3 k .
 i j k 
 ∂ ∂ ∂ 
Then ∇ × A = 
 ∂x ∂y ∂z 
 A1 A2 A3 
 ∂A3 ∂A2   ∂A1 ∂A3   ∂A2 ∂A1 
=  −  i +  −  j+  −  k.
 ∂y ∂z   ∂z ∂x   ∂x ∂y 
 
 i j k 
 
∂ ∂ ∂
∴ ∇ × (∇ × A) =  
 ∂x ∂y ∂z 
∂A3 ∂A2 ∂A1 ∂A3 ∂A2 ∂A1 
 − − − 
 ∂y ∂z ∂z ∂x ∂x ∂y 
D-274
 ∂  ∂A2 ∂A1  ∂  ∂A1 ∂A3   

= Σ   −  −  −  i 
 ∂y  ∂x ∂y  ∂z  ∂z ∂x   

  ∂ 2 A ∂ 2 A3   ∂ 2 A1 ∂ 2 A1   
 2    i 
=Σ  + − + 
  ∂y ∂x ∂z ∂x   ∂y 2 ∂z 2   
  

∂  ∂A2 ∂A3   ∂ 2 A1 ∂ 2 A1   
= Σ   +  −  +  i 

 ∂x  ∂y ∂z   ∂y 2 ∂z 2   
  

∂  ∂A1 ∂A2 ∂A3   ∂ 2 A1 ∂ 2 A1 ∂ 2 A1   
= Σ   + +  −  + +  i

 ∂x  ∂x ∂y ∂z   ∂x 2 ∂y 2 ∂z 2   
  
 ∂ 
= Σ  (∇ • A) − (∇ 2 A1 ) i 
  ∂ x  
 ∂ 
= Σ  (∇ • A) i  − ∇ 2 Σ A1 i = ∇ (∇ • A) − ∇ 2 A.
 ∂ x  
Example 24: Taking F = x 2 yi + xzj + 2 yzk , verify that div curl F = 0.

(Garhwal 2002; Bundelkhand 04; Kanpur 10)
 i j k 
∂ ∂ ∂ 
Solution: We have curl F =  
 ∂x ∂y ∂z 
2 2 yz
x y xz 
∂ ∂  ∂ ∂ 2 
=  (2 yz ) − ( xz ) i −  (2 yz ) − ( x y) j
 ∂ y ∂ z   ∂ x ∂ z 
∂ ∂ 2 
+  ( xz ) − ( x y) k
 ∂x ∂y 
= (2z − x) i − 0 j + (z − x 2 ) k = (2z − x) i + (z − x 2 ) k .
Now div curl F = div [(2z − x) i + (z − x 2 ) k ]
∂ ∂
= (2z − x) + (z − x 2 ) = − 1 + 1 = 0.
∂x ∂z
Example 25: Prove that div (r n r) = (n + 3) r n . (Bundelkhand 2006)
Solution: We have
div (φA) = φ (div A) + A • grad φ .
Putting A = r and φ = r n in this identity, we get
D-275
div (r n r) = r n div r + r • grad r n

= 3r n + r • (n r n − 1 grad r)
[∵ div r = 3 and grad f (u) = f ′ (u) grad u]
1
= 3r n + r • n r n − 1 r  ∵ grad r = ∧r = 1 r
 r   r 
= 3r n + n r n − 2 (r • r)
= 3 r n + n r n − 2 r2
= (n + 3) r n .
 r
Example 26: Prove that div  3  = 0.
r  (Bundelkhand 2010)
Solution: We have
 1 
div  3 r = div (r −3 r) = r −3 div r + r • grad r −3
r 
1
= 3r −3 + r • (−3r −4 grad r) = 3r −3 + r •  −3r −4 r
 r 
= 3r −3 − 3r −5 (r • r) = 3r −3 − 3r −5 r 2
= 3r −3 − 3r −3 = 0.
∴ the vector r −3 r is solenoidal.
∧
Example 27: Prove that div r = 2 / r. (Bundelkhand 2007; Kanpur 05)
∧ 1
Solution: div ( r ) = div  r ⋅ Now proceed as in Example 25.
r 
Alternative Method:
∧ 1 1
div r = div  r = div  ( x i + y j + z k )
r   r 
x y z  ∂  x ∂ y  ∂ z
= div  i + j + k =   +   +  
 r r r  ∂ x  r  ∂ y r
  ∂ z  r
1 x ∂r   1 y ∂r   1 z ∂r 
= − 2  + − 2  + − 2 ⋅
 r r ∂x   r r ∂y   r r ∂z 
Now r 2 = x2 + y2 + z 2 .
∂r ∂r x
∴ 2r = 2 x i. e., = ⋅
∂x ∂x r
∂r y ∂r z
Similarly = and = ⋅
∂y r ∂z r
∧ 3  x x y y z z
∴ div r= − 2 + 2 + 2 
r r r r r r r
D-276
2 2 2
3 x + y +z 3 r2 3 1 2
= − = − 3 = − = ⋅
r r3 r r r r r
2
Example 28: Prove that ∇ 2 f (r) = f ′ ′ (r) + f ′ (r).
r (Meerut 2003, 05, 06B)
Solution: We know that if φ is a scalar function, then ∇ 2 φ = ∇ • (∇φ).
∴ ∇ 2 f (r) = ∇ • {∇f (r)} = div { grad f (r) }
1
= div { f ′ (r) grad r} = div  f ′ (r) r
r 
1 1
= f ′ (r) div r + r • grad  f ′ (r)
r r 
3 d  1 f ′ (r) grad r
= f ′ (r) + r •    
r  dr r 
3  1 1  1 
= f ′ (r) + r •  − 2 f ′ (r) + f ′ ′ (r) r
r  r r  r 
3 1  1 1 
= f ′ (r) + r  − 2 f ′ (r) + f ′ ′ (r)  (r • r)
r   r r 
3 1  1 1 
= f ′ (r) +   − 2 f ′ (r) + f ′ ′ (r)  r 2
r r  r r 
3 1 2
= f ′ (r) − f ′ (r) + f ′ ′ (r) = f ′ ′ (r) + f ′ (r).
r r r
1 1
Example 29: Prove that ∇ 2   = 0 or div  grad  = 0 .
 r  r
( Agra 2002; Meerut 07, 13)
Solution: We have
1 1 1
∇ 2   = ∇ • ∇ 
 = div  grad 
 r  r  r
 1   1 1   1 
= div  − 2 grad r = div  − 2 r = div  − 3 r
 r   r r   r 
 1  1 3 d  1 
=  − 3  div r + r • grad  − 3  = − 3 + r •   − 3  grad r
 r   r  r  dr  r  
3 3 1  3 3 3 3
=− 3
+r • 4 r = − 3 + 5 (r • r) = − 3 + 5 r 2 = 0 .
r r r  r r r r
∴ 1 / r is a solution of Laplace’s equation.
Example 30: Prove that div grad r n = n (n + 1) r n − 2 , i. e., ∇ 2 r n = n (n + 1) r n − 2 .
(Bundelkhand 2005, 10; Meerut 2000, 08; Avadh 10; Agra 14)
Solution: We have ∇ 2 r n = ∇ • (∇r n ) = div (grad r n )
D-277
1
= div (nr n − 1 grad r) = div  nr n − 1 r = div (nr n − 2 r)
 r 
= (nr n − 2 ) div r + r • (grad nr n − 2 )
= 3nr n − 2 + r • [n (n − 2) r n − 3 grad r]
1
= 3nr n − 2 + r • n (n − 2) r n − 3 r
 r 
= 3nr n − 2 + r • [n (n − 2) r n − 4 r] = 3nr n − 2 + n (n − 2) r n − 4 (r • r)
= 3nr n − 2 + n (n − 2) r n − 4 r 2 = nr n − 2 (3 + n − 2) = n (n + 1) r n − 2 .
Note: If n = − 1, then ∇ 2 (r −1
) = (−1)(−1 + 1) r −3 = 0.
  1  3
Example 31: Prove that ∇ •  r ∇  3   = 4 or, div [r grad r −3 ] = 3r −4 .
  r  r (Meerut 2009B)
 1
Solution: We have ∇  3  = grad r −3
r 
∂ −3 ∂ −3 ∂ −3
= (r ) i + (r ) j + (r ) k .
∂x ∂y ∂z
∂ −3 ∂r
Now (r ) = − 3r − 4 ⋅
∂x ∂x
But r 2 = x2 + y2 + z 2 .
∂r ∂r x
Therefore 2r = 2x or = ⋅
∂x ∂x r
∂ −3 x
So (r ) = − 3r − 4 = − 3r −5 x .
∂x r
∂ −3 ∂ −3
Similarly (r ) = − 3r −5 y and (r ) = − 3r −5 z .
∂y ∂z
 1
Therefore ∇  3  = − 3r −5 ( x i + yj + zk ).
r 
 1
∴ r ∇  3  = − 3r − 4 ( xi + yj + zk ).
r 
 1 ∂ ∂ ∂
∴ ∇ • r ∇ 3  = (−3r − 4 x) + (− 3r − 4 y) + (−3r − 4 z ).
 r  ∂x ∂y ∂z
∂ ∂r
Now (−3r − 4 x) = 12r −5 x − 3r − 4
∂x ∂x
−5 x
= 12r x − 3r − 4 = 12r −6 x 2 − 3r − 4 .
r
∂
Similarly (−3r − 4 y) = 12r − 6 y 2 − 3r − 4
∂y
∂
and (−3r − 4 z ) = 12r − 6 z 2 − 3r − 4 .
∂z
D-278
 1
Hence ∇ •  r ∇ 3  = 12 r − 6 ( x 2 + y 2 + z 2 ) − 9r − 4
 r 
= 12 r −6 r 2 − 9r − 4 = 12 r − 4 − 9r − 4 = 3r − 4 .
Example 32: If a is a constant vector, prove that div { r n (a × r)} = 0.(Kanpur 2008)
Solution: We have div (φA) = φ div A + A • grad φ .

∴ div { r n (a × r) } = r n
div (a × r) + (a × r) • grad r n
= r n div (a × r) + (a × r) • (nr n − 1 grad r)
1
= r n (r • curl a − a • curl r) + (a × r) •  nr n −1 r
 r 
= r n (r • 0 – a • 0) + nr n − 2 (a × r) • r
[∵ curl of a constant vector is zero and curl r = 0 ]
n−2
= nr [a , r , r]
= 0 , since a sca lar triple product having two equal vec tors is zero.
Example 33: If a and b are constant vectors, prove that
(i) div [(r × a) × b] = − 2 b • a . (Bundelkhand 2010)
(ii) curl [(r × a) × b] = b × a . (Bundelkhand 2010)
Solution: (i) We have (r × a) × b = ( b • r) a − ( b • a) r .
∴ div [(r × a) × b] = div [( b • r) a − ( b • a) r]
= div [( b • r) a] − div [( b • a) r] …(1)
But div (φA) = φ div A + A • grad φ .
Taking φ = b • r and A = a , we get
div [( b • r) a] = ( b • r) div a + a • grad ( b • r).
Since a is a constant vector, therefore div a = 0 .
Also let b = b1 i + b2 j + b3 k .
Then b • r = (b1 i + b2 j + b3 k ) • ( xi + yj + zk )
= b1 x + b2 y + b3 z , where b1 , b2 , b3 are constants.
∴ grad ( b • r) = b1 i + b2 j + b3 k = b .
∴ div [( b • r) a] = a • b . …(2)
Again div [( b • a)] r = ( b • a) div r + r • grad ( b • a).
But div r = 3.
Also grad ( b • a) = 0 because b • a is constant.
∴ div [( b • a) r] = 3 ( b • a). …(3)
Substituting the values from (2) and (3) in (1), we get
div [(r × a) × b] = (a • b) − 3 ( b • a) = − 2 b • a .
(ii) curl [(r × a) × b] = curl [( b • r) a − ( b • a) r]
= curl [( b • r) a] − curl [( b • a) r].
D-279
But curl (φA) = grad φ × A + φ curl A .

∴ curl [( b • r ) a] = [ grad ( b • r )] × a + ( b • r ) curl a = b × a
[∵ curl a = 0 and grad ( b • r ) = b]
Also curl [( b • a ) r ] = [ grad ( b • a )] × r + ( b • a ) curl r = 0
[∵ grad ( b • a ) = 0 and curl r = 0]
∴ curl [( r × a ) × b] = b × a – 0 = b × a .
Example 34: Prove that curl [r × (a × r)] = 3r × a , where a is a constant vector.
(Purvanchal 2009)
Solution: curl [r × (a × r)]
= ∇ × [(r • r) a − (r • a) r] [∵ a × ( b × c) = (a • c) b − (a • b) c]
2
= ∇ × [r a − (r • a) r] [∵ r • r = r 2 = r 2 ]
= ∇ × (r 2 a) − ∇ × [(r • a) r] [∵ ∇ × (A + B) = ∇ × A + ∇ × B]
2 2
= (∇r ) × a + r (∇ × a) − [∇ (r • a)] × r − (r • a) (∇ × r)
[∵ ∇ × (φA) = (∇φ) × A + φ (∇ × A)]
2
= (2r ∇r) × a + r 0 − [∇ (r • a)] × r − (r • a) 0
[∵ ∇f (r) = f ′ (r) ∇r ; ∇ × a = 0, a
being a constant vector ; and ∇ × r = 0]
1
= 2r r × a − [∇ (r • a)] × r
 r 
= 2r × a − a × r [∵ ∇ (r • a) = a, if a is a constant vector]
= 2r × a + r × a = 3r × a.
Example 35: If a is a constant vector, prove that

a×r a 3r
curl 3 = − 3 + 5 (a • r).
r r r (Meerut 2009B; Bundelkhand 09, 11)
Solution: We have
a×r  a × r  ∂  a × r
curl 3 = ∇ ×  3  = Σ  i ×  3  ⋅
r  r   ∂ x  r 
∂  a × r 3 ∂r 1  ∂r  1  ∂a 
Now   =− 4 (a × r) + 3  a ×  + 3  × r …(1)
∂x  r 3  r ∂ x r  ∂ x  r  ∂ x 
∂a
Now = 0 because a is a constant vector.
∂x
Also r = x i + y j+ z k.
∂r
∴ =i.
∂x
∂r x
Further = ⋅
∂x r
D-280
∂  a × r 3 x 1
∴ (1) becomes  3  =− 4 (a × r) + 3 (a × i)
∂x r  r r r
3x 1
= − 5 (a × r) + 3 (a × i).
r r
∂  a × r 3x 1
∴ i×   = − 5 i × (a × r) + 3 i × (a × i)
∂x  r 3  r r
3x 1
= − 5 [(i • r) a – (i • a) r] + 3 [(i • i) a − (i • a) i]
r r
3x 3x 1 1
= − 5 xa + 5 a1 r + 3 a − 3 a1 i
r r r r
[∵ i • r = x and i • a = a1 if a = a1 i + a2 j + a3 k ]
2
3x 3 1 1
=− 5
a+ 5
a1 xr + 3
a− a1 i .
r r r r3
 ∂  a × r
∴ Σ i ×  
 ∂x  r 3  
 3  3  3 1
=  − 5 Σ x 2  a +  5 Σ a1 x r + 3 a − 3 Σ a1 i
 r  r  r r
3 2 3 3 1
= − 5 r a + 5 (r • a) r + 3 a − 3 a
r r r r
[∵ Σ x 2 = r 2 , Σ a1 x = r • a , Σ a1 i = a]
a 3
=− 3
+ (a • r) r .
r r5
1. (i) Verify that curl grad f = 0 , where f = x 2 y + 2 xy + z 2 .

(ii) Prove that grad f (u) = f ′(u) grad u.
2
+ y 2 + z 2 )1 /2
(iii) Find ∇φ and |∇φ| when φ = ( x 2 + y 2 + z 2 ) e − ( x .
(Kumaun 2012)
2. (i) Prove that curl (ψ∇φ) = ∇ψ × ∇φ = − curl (φ∇ψ).
(ii) Prove that ∇ 2 (φψ) = φ∇ 2 ψ + 2∇φ • ∇ψ + ψ∇ 2 φ. (Kumaun 2014)
(iii) Prove that div (∇φ × ∇ψ) = 0.
(iv) If A and B are irrotational, prove that A × B is solenoidal.
(v) Prove that curl (φ grad φ) = 0.
3. (i) Prove that a • {∇( v • a) − ∇ × ( v × a)} = div v, where a is a constant
unit vector.
(ii) Prove that vector f (r) r is irrotational.
D-281
(iii) If f and g are two scalar point functions, prove that

div ( f ∇ g) = f ∇ 2 g + ∇f • ∇g.
(iv) Show that curl (a • r) a = 0 , where a is a constant vector.

(Kumaun 2007, 09)
[ Hint: Use identity 4. Note that ∇ (a • r) = a , if a is a constant vector.]
4. If a is a constant vector, then prove that
(i) ∇ (a • u) = (a • ∇) u + a × curl u,
(ii) ∇ • (a × u) = − a • curl u,
(iii) ∇ × (a × u) = a div u − (a • ∇) u ,
5. (i) Given that ρ F = ∇p, where ρ, p, F are point functions, prove that
F • curl F = 0.
(ii) A vector function f is the product of a scalar function and the gradient
of a scalar function. Show that f • curl f = 0.
1
(iii) Show that curl a φ (r) = φ′ (r) r × a , where a is a constant vector.
r
n n n−2
6. (i) Prove that curl [r (a × r)] = (n + 2) r a − nr (r • a) r,
where a is a constant vector. (Kumaun 2008)
(ii) Prove that ∇ 2 (r n r) = n (n + 3) r n − 2 r.
7. (i) Prove that curl grad r n = 0. (Avadh 2010; Kanpur 11)
2 c1
(ii) If ∇ f (r) = 0 , show that f (r) = + c2 ,
r
where r 2 = x 2 + y 2 + z 2
and c1 , c 2 are arbitrary constants.
8. If r is the position vector of the point ( x , y , z ) show that curl (r n r) = 0, where

r is the module of r.
9. Prove that r n r is an irrotational vector for any value of n but is solenoidal
only if n + 3 = 0.
10. (i) If u = (1 / r) r, show that ∇ × u = 0. (Avadh 2010)
(ii) Prove that div (A × r) = r • curl A .
11. (i) If ∇ 2 f (r) = 0 show that f (r) = c1 log r + c 2 where r 2 = x 2 + y 2 and
c1 , c 2 are arbitrary constants.
(ii) If a and b are constant vectors, then show that ∇ • (a • br) = 3a • b .
12. If u = (1 / r) r find grad (div u).
  r 
13. (i) Prove that ∇ 2 ∇ •  2   = 2r − 4 .
  r  (Kumaun 2011, 13)
1 a•r
(ii) Prove that a • ∇  = − 3 ⋅
 r r
D-282
(iii) Prove that ∇ • (U ∇V − V ∇U ) = U ∇ 2 V − V ∇ 2 U.

1 3 (a • r) (b • r) a • b
(iv) Prove that b • ∇ a • ∇  = − ,
 r r5 r3
where a and b are constant vectors. (Kumaun 2015)
14. Evaluate div {a × (r × a)}, where a is a constant vector. (Kanpur 2007)
 f (r) r  1 d 2
15. (i) Prove that div   = (r f ).
 r  r 2 dr (Kumaun 2007, 14)
1 2
(ii) Prove that ∇ a = (a • ∇) a + a × curl a .
2
16. Prove that curl [r × (a × r)] = 3r × a , where a is a constant vector.
17. Prove that ∇ × (F × r) = 2F − (∇ • F) r + (r • ∇) F. (Kumaun 2015)
18. If a and b are constant vectors, prove that
grad [(r × a) • (r × b)] = (b × r) × a + (a × r) × b. (Kumaun 2015)
A nswers 4
2
1. (iii) (2 − r) e − r r ; (2 − r) e − r r 12. − r
r3
14. 2 a2

(a), (b), (c) and (d).
1. If the vector V = ( x + 3 y) i + ( y − 2z ) j + ( x + az ) k is solenoidal, then the
constant a is
(a) 0 (b) 1
(c) –2 (d) 2 (Garhwal 2006)
2. The directional derivative of φ ( x, y, z ) = x 2 yz + 4 xz 2 at (1, − 2 , − 1) in the

direction of the vector 2 i − j − 2 k is
(a) 37/3 (b) 3/37
D-283
3. If r = |r | where r = x i + y j + z k , then ∇ 2 r n
=
n n −1
(a) n (n + 1) r (b) n (n + 1) r
n−2
(c) n (n + 1) r (d) none of these
(Agra 2007; Garhwal 15)
1
4. ∇ 2   =
 r
(a) −2 / r 3 (b) 0
3
(c) 2 / r (d) none of these
(Kumaun 2014)
5. If r = xi + yj + zk and a is a constant vector, then curl (r × a) is

(a) − a (b) − 3a
(c) − 2 a (d) none of these
(Garhwal2014)
6. The value of div ^

r is
(a) 2 / r (b) 0
(c) 1 / r (d) none of these (Agra 2014)
7. The V is a constant vector, then div V is :

(a) 3 (b) 3V
(Kumaun 2007, 11)
2
8. If f = x yi − 2 xzj + 2 yzk , then div f is equal to
(a) 2 x ( x + 1) (b) 2 y ( x + 1)
(c) y ( y + 1) (d) none of these
(Kumaun 2008)
9. ∇ × (∇ f ) is equal to
(a) ∇ 2 f (b) 0
(Kumaun 2010)
10. grad (1 / r) is equal to
(a) − r / r 3 (b) r / r 3
(c) r / r 2 (d) none of these
(Kumaun 2006)
r
11. If f = , then curl f is equal to
r
2
(a) f (b) f
(Kumaun 2006)
D-284
12. Value of (2 i × j ) × ( j − k ) is equal to

(a) i + 2 j + 2k (b) − i + 2 j + 2k
(c) 2i − 2 j + k (d) 2i + 2 j + k
(Kumaun 2014)
13. The value of curl ( i × j) =

(a) 0 (b) 1
(c) −1 (d) 0
(Kumaun 2014)
14. If r = x i + y j + z k , then curl r is

(a) 1 (b) 2
(c) 3 (d) 4
(Garhwal 2007)
2 2 2
15. If F = xy i + 2 x yz j − 3 yz k , then div F at (1, − 1, 1) is
(a) − 9 (b) 9
(c) 8 (d) 10
(Garhwal 2008)
16. For any vector field F, the value of div curl A =

(a) 1 (b) −1
(Garhwal 2008)
17. The equation of normal plane is
(a) (R − r) × ∇f = 0 (b) (R + r) × ∇f = 0
2
(c) (R − r) × ∇ f = 0 (d) (R + r) × ∇ 2 f = 0
(Garhwal 2009)
2 3 2
18. If φ = 3 x y − y z , then ∇φ at the point (1, − 2, − 1) is
(a) −12 i − 9 j − 24k (b) −25
(c) 12 i + 9 j − 2k (d) none of these
(Garhwal 2011)
19. If for any vector V, then div V = 0, then V is
(a) solenoidal (b) irrotational
(c) normal to the surface (d) parallel to line surface
(Garhwal 2011)
20. Value of curl grad φ is
(a) 0 (b) ∇ 2 φ
1
(c) ∇ 2 φ (d) none of these
2 (Garhwal 2013)
D-285

1. If F = ( x 2 + y 2 ) i − 2 xy j , then F • dr = …… .
∂P
2. If P = e xy i + ( x − 2 y) j + ( x sin y) k , then = …… .
∂x
3. If a is a constant vector then grad (a • r) = …… .
(Rohilkhand 2007; Kumaun 12)
4. If a is a constant vector, then ∇ • (a × r) = …… .
5. If r = xi + yj + zk , then the value of div r = …… .
(Agra 2008; Kumaun 15)
6. If A = x 2 zi − 2 y 3 z 2 j + xy 2 zk , then div A at (1, − 1, 1) = …… .

7. If r = xi + yj + zk , then the value of curl r = …… .
(Agra 2008; Rohilkhand 14)
8. For any vector A, div curl A = …… .
9. A vector V is said to be solenoidal if …… . (Rohilkhand 2005; Agra 06)
10. A vector F is said to be irrotational if …… .
11. If φ = x 2 y + 2 xy + z 2 , then curl grad φ = …… . (Bundelkhand 2007)
12. If f = x 2 yi + 2 xzj + 2 yzk , then div (curl f ) = …… . (Kumaun 2010, 12)
13. Value of div grad φ is …… . (Kumaun 2011)
14. If a is a constant vector, then curl (r × a) = ...... . (Kumaun 2013)
True or False
1. If r = xi + yj + zk , then r is solenoidal.
2. If V is a constant vector, then div V = 0. (Rohilkhand 2005)
3. If F = 2 xyz i + y 2 zj − 2 yz 2 k , then F is irrotational.
4. If φ is a differentiable scalar function, then curl grad φ = 0.
5. If φ is a differentiable scalar function then div grad φ = ∇ 2 φ .
6. ∇ • (A × B) = A • (∇ × B) − B • (∇ × A).
7. A function which satisfies Laplace’s equation is called a harmonic function.
A nswers
1. (c) 2. (a) 3. (c) 4. (b) 5. (c)
6. (a) 7. (c) 8. (b) 9. (b) 10. (a)
D-286
11. (c) 12. (b) 13. (a) 14. (d) 15. (b)
16. (d) 17. (a) 18. (d) 19. (a) 20. (a)

1. ( x 2 + y 2 ) dx − 2 xydy 2. y e xy i + j + sin y k 3. a
4. 0 5. 3 6. –3 7. 0 8. 0
9. div V = 0 10. curl F = 0 11. 0
12. 0 13. ∇ 2 φ 14. −2a
True or False
1. F 2. T 3. F 4. T 5. T
6. F 7. T
¨
D-287
10
I ntegration of V e ctors
1 Integration of Vector Functions

e shall define integration as the reverse process of differentiation. Let f (t) and F (t)
W be two vector functions of the scalar t such that
d
F (t) = f (t).
dt
Then F (t) is called the indefinite integral of f (t) with respect to t and symbolically
we write
∫ f (t) dt = F (t). …(1)
The function f (t) to be integrated is called the integrand.
If c is any arbitrary constant vector independent of t, then
d
{F (t) + c } = f (t).
dt
This is equivalent to
∫ f (t) dt = F (t) + c . …(2)
From (2) it is obvious that the integral F (t) of f (t) is indefinite to the extent of an
additive arbitrary constant c .Therefore F (t) is called the indefinite integral of f (t).
D-288
The constant vector c is called the constant of integration. It can be determined if we

are given some initial conditions.
d
If F (t) = f (t) for all t in the interval [a , b], then the definite integral between the
dt
limits t = a and t = b can, in such case, be written as
b b d  b
∫ a f (t) dt = ∫ a  dt F (t) dt = [F (t) + c] a = F (b) − F (a).
Theorem: If f (t) = f1 (t) i + f 2 (t) j + f 3 (t) k , then
∫ f (t) dt = i ∫ f1 (t) dt + j ∫ f 2 (t) dt + k ∫ f 3 (t) dt.

d
Proof: Let F (t) = f (t). …(1)
dt
Then ∫ f (t) dt = F (t). …(2)
Let F (t) = F1 (t) i + F2 (t) j + F3 (t) k .
Then from (1), we have
d
{ F1 (t) i + F2 (t) j + F3 (t) k } = f (t)
dt
or  d F (t) i +  d F (t) j +  d F (t) k

 1   2   3 
 dt   dt   dt 
= f1 (t) i + f 2 (t) j + f 3 (t) k .
Equating the coefficients of i , j, k , we get
d d d
F1 (t) = f1 (t), F2 (t) = f 2 (t), F3 (t) = f 3 (t).
dt dt dt
∴ F1 (t) = ∫ f1 (t) dt, F2 (t) = ∫ f 2 (t) dt, F3 (t) = ∫ f 3 (t) dt.
∴ F (t) = {∫ f1 (t) dt} i + {∫ f 2 (t) dt} j + {∫ f 3 (t) dt} k .
So from (2), we get
∫ f (t) dt = i ∫ f1 (t) dt + j ∫ f 2 (t) dt + k ∫ f 3 (t) dt .
Note: From this theorem we conclude that the definition of the integral of a
vector function implies the definition of integrals of three scalar functions which
are the components of that vector function. Thus in order to integrate a vector
function we should integrate its components.
2 Some Standard Results

We have already obtained some standard results for differentiation. With the help
of these results we can obtain some standard results for integration.
d dr ds
1. We have (r • s) = •s+r • ⋅
dt dt dt
D-289
Therefore  dr • s + r • ds  dt = r • s + c ,
∫ 
 dt

dt 
where c is the constant of integration. It should be noted that c is here a scalar
quantity since the integrand is also scalar.
d 2 dr 2 r • dr  dt = r 2 + c .
2. We have (r ) = 2r • ⋅ Therefore ∫  
dt dt  dt 
Here the constant of integration c is a scalar quantity.
d  dr  2 dr d2 r
3. We have   =2 • ⋅
dt  dt  dt dt 2
 dr d2 r  dr 2
Therefore we have ∫ 2 • 2  dt =   + c .
 dt dt   dt 
Here the constant of integration c is a scalar quantity.

2
Also  dr  dr dr
=
• ⋅
 
 dt  dt dt
d  dr  dr dr d2 r d2 r
4. We have r ×  = × +r× 2 =r× 2 ⋅
dt  dt  dt dt dt dt
 2 
d r dr
∴ ∫  r × dt 2  dt = r × dt + c .
Here the constant of integration c is a vector quantity since the integrand
d2 r
r × 2 is also a vector quantity.
dt
5. If a is a constant vector, we have
d da dr dr
(a × r) = ×r+a× =a× ⋅
dt dt dt dt
dr
Therefore ∫  a ×  dt = a × r + c .
 dt 
Here the constant of integration c is a vector quantity.
6. If r = |r | and ^
r is a unit vector in the direction of r, then
d ^ d 1  1 d r 1 dr
(r)=  r = − 2 r.
dt dt  r  r dt r dt
 1 dr 1 dr 
Therefore ∫  − 2 r dt = ^
r + c.
 r dt r dt 
7. If c is a constant scalar and r a vector function of a scalar t ,then obviously
∫ cr dt = c ∫ r dt .
8. If r and s are two vector functions of the scalar t , then obviously
∫ (r + s) dt = ∫ r dt + ∫ s dt.
D-290
2 d2 r
Example 1: Evaluate ∫ r× 2
dt, where r = 2t 2 i + t j − 3t 3 k .
1 dt
(Bundelkhand 2004; Kumaun 08; Kanpur 13)
Solution: Given r = 2t 2 i + t j − 3t 3 k .
dr d2 r
∴ = 4t i + j − 9t 2 k and = 4i + 0 j − 18tk .
dt dt 2
d2 r
∴ r× 2
= (2t 2 i + tj − 3t 3 k ) × (4i + 0 j − 18tk )
dt
 i j k 
= 2t 2 t − 3t 
3
 
4 0 −18t 
= − 18t 2 i − (− 36t 3 + 12t 3 ) j − 4t k

= − 18 t 2 i + 24t 3 j − 4tk.
2 d2 r 2
∴ ∫1 r× 2
dt = ∫1 (−18t 2 i + 24t 3 j − 4tk ) dt
dt
2 2 2
= − 18i ∫1 t 2 dt + 24 j ∫ t 3 dt − 4k ∫ t dt
1 1
2 2 2
 t3   t4   t2 
= − 18i   + 24 j   − 4k  
 3 1  4 1  2 1
= − 6 (8 − 1) i + 6 (16 − 1) j − 2 (4 − 1) k
= − 42i + 90 j − 6k .
d2 r
Example 2: Find the value of r satisfying the equation = ta + b, where a and b are
dt 2
constant vectors.
d2 r dr 1 2
Solution: Integrating the equation 2
= ta + b, we get = t a + tb + c, where
dt dt 2
c is constant.
Again integrating, we get
1 1
r = t 3 + t 2 b + tc + d , where d is constant.
6 2
Example 3: If r (t) = 5t 2 i + tj − t 3 k , prove that
2  d2 r 
∫1  r × 2  dt = − 14i + 75 j − 15k .
 dt  (Kumaun 2000; Meerut 01, 04, 05, 07, 10, 10B;
Kanpur 10; Rohilkhand 13)
D-291
 d2 r  dr
Solution: We have ∫  r × 2  dt = r × + c.
 dt  dt
2  d2 r  dr 2
∴ ∫1  r × 2  dt = r ×  ⋅
 dt   dt 1
dr dr
Let us now find r × ⋅ We have = 10 t i + j − 3t 2 k .
dt dt
dr
∴ r× = (5t 2 i + t j − t 3 k ) × (10 t i + j − 3t 2 k )
dt
 i j k 
 
= 5t 2 t − t 3 = − 2t 3 i + 5t 4 j − 5t 2 k .

 

10 t 1 − 3 t 2
 
2  d2 r  2
∴ ∫1  dt 
[
 r × 2  dt = −2t 3 i + 5t 4 j − 5t 2 k ]1
2 2 2
[
= − 2t 3 ]1 [ ]
i + 5t 4
1 [ ]
j − 5t 2
1
k
= − 14i + 75 j − 15k.
 2i − j + 2k , when t = 2
Example 4: Given that r (t) = 
 4i − 2 j + 3k , when t = 3,
show that ∫
3  r • dr  dt = 10.
 
2  dt  (Meerut 2003, 13B; Bundelkhand 08;
Kanpur 09, 11; Agra 06; Avadh 10; Purvanchal 13)
3
Solution: We have  r • dr  dt =  1 r 2  ⋅
∫ 


dt  2  2
When t = 3, r = 4i − 2 j + 3k .
∴ when t = 3, r 2 = (4i − 2 j + 3k ) • (4i − 2 j + 3k ) = 16 + 4 + 9 = 29.
When t = 2 , r = 2i − j + 2k .
∴ when t = 2 , r 2 = 4 + 1 + 4 = 9.
∴
3  r • dr  dt = 1 [29 − 9] = 10.
∫2 


dt  2
Example 5: The acceleration of a particle at any time t ≥ 0 is given by
dv
a= = 12 cos 2t i − 8 sin 2t j + 16 t k .
dt
If the velocity v and displacement r are zero at t = 0, find v and r at any time.
(Agra 2007)
dv
Solution: We have = 12 cos 2t i − 8 sin 2t j + 16t k .
dt
Integrating, we get
D-292
v = i ∫ 12 cos 2t dt + j ∫ −8 sin 2t dt + k ∫ 16t dt
or v = 6 sin 2t i + 4 cos 2t j + 8t 2 k + c.
When t = 0, v = 0.
∴ 0 = 0 i + 4 j + 0 k + c or c = − 4 j.
dr
∴ v= = 6 sin 2t i + (4 cos 2t − 4) j + 8t 2 k .
dt
Integrating, we get
2
r = i ∫ 6 sin 2t dt + j ∫ (4 cos 2t − 4) dt + k ∫ 8t dt
8 3
= − 3 cos 2t i + (2 sin 2t − 4t) j + t k + d,
3
where d is constant.
When t = 0, r = 0.
∴ 0 = − 3i + 0 j + 0 k + d . ∴ d = 3i .
8 3
∴ r = − 3 cos 2t i + (2 sin 2t − 4t) j + t k + 3i
3
8
= (3 − 3 cos 2t) i + (2 sin 2t − 4t) j + t 3 k .
3
1. If f (t) = (t − t 2 ) i + 2 t 3 j − 3k , find
2
(i) ∫ f (t) dt (ii) ∫1 f (t) dt.
1
2. Evaluate ∫ ( e t i + e −2 t j + t k ) dt . (Garhwal 2001, 02)
0
1
3. If f (t) = t i + (t 2 − 2 t) j + (3t 2 + 3t 3 ) k , find ∫ f (t) dt.
0
4. If r = t i − t 2 j + (t − 1) k and s = 2t 2 i + 6t k , evaluate
2 2
(i) ∫0 r • s dt, (ii) ∫0 r × s dt
(Rohilkhand 2008)
d2 r
5. (i) Find the value of r satisfying the equation = a , where a is a
dt 2
dr
constant vector. Also it is given that when t = 0, r = 0 and = u.
dt
D-293
d2 r
(ii) Solve the equation = a where a is a constant vector ; given that
dt 2
dr
r =0 and = 0 when t = 0.
dt (Bundelkhand 2008)
d2 r
6. Find the value of r satisfying the equation 2
= 6ti − 24t 2 j + 4 sin t k ,
dt
given that r = 2i + j and dr / dt = − i − 3 k at t = 0.
(Agra 2001; Meerut 11)
7. The acceleration of a particle at any time t is e t i + e2 t j + k .

Find v, given that v = i + j at t = 0.
2
8. Evaluate ∫ (a • b × c) dt , where
1
a = ti − 3 j + 2tk , b = i − 2 j + 2k , c = 3i + t j − k . (Meerut 2013)
d2 r
9. Integrate = − nr 2 .
dt 2 (Kumaun 2009)
A nswers 1
t 2 t 3  t4
1. (i)  −  i+ j − 3t k + c
2 3 2
5 15
(ii) − i+ j−3k
6 2
1 1
2. (e − 1) i − (e −2 − 1) j + k
2 2
1 2 7
3. i − j+ k
2 3 4
4. (i) 12
40 64
(ii) − 24 i − j+ k
3 5
1 2
5. (i) t a + tu
2
1 2
(ii) t a
2
6. r = (t 3 − t + 2) i + (1 − 2t 4 ) j + (t − 4 sin t) k
1 2t
7. et i + (e + 1) j + t k
2
8. 0
9. − n2 r 2 + c
D-294

(a), (b), (c) and (d).
1
1. If F (t) = t i + (t 2 − 2 t) j + (3t 2 + 3t 3 ) k , then the value of ∫ F (t) dt is
0
1 2 7 1 2 7
(a) i + j+ k (b) i − j+ k
2 3 4 2 3 4
1 2 7
(c) − i − j + k (d) None of these
2 3 4
(Bundelkhand 2001)
1
2. If r = t i − t 2 j + (t − 1) k , and s = 2t 2 i + 6t k , then the value of ∫ r • s dt is
0
(a) 10 (b) 12
(c) 15 (d) None of these
(Kumaun 2007, 10)
1
3. The value of ∫ (e t i + e − 2 t j + t k ) dt is
0
1 −2 1 1 −2 1
(a) e i − e j+ k (b) (e − 1) i − (e + 1) j + k
2 2 2 2
1 −2 1
(c) (e − 1) i − (e − 1) j + k (d) none of these
2 2
(Garhwal 2008)
2
d r
4. The value of r satisfying the equation = ta + b, where a and b are
dt 2
constant vectors is
t3 t2 t3 t2
(a) + b + tc + d (b) − b + tc + d
6 2 6 2
t3 t2 t3 t2
(c) − b − tc + d (d) − − b − tc + d
6 2 6 2
5. The acceleration of a particle at any time t ≥ 0 is given by

dv
a= = 12 cos 2t i − 8 sin 2t j + 16t k , the velocity v at any time is
dt
(a) 6 sin 2t i − 4 cos 2t j + 8t 2 k + c (b) 6 sin 2t i + 4 cos 2t j + 8t 2 k + c
(c) 6 sin 2t i − 4 cos 2t j − 8t 2 k + c (d) none of these
D-295
6. If f (t) = (t − t 2 ) i + 2 t 3 j − 3 k , the value of ∫ f (t) dt is
 t2 t3  t4  t2 t3  t4
(a)  + i+ j − 3t k + c (b)  − i+ j − 3t k + c
2 3 2 2 3 2
 t2 t3  t4  t2 t3  t4
(c)  − i− j − 3t k + c (d)  + i− j − 3t k + c
2 3 2 2 3 2
7. The acceleration of a particle at any time t is e t i + e2 t j + k , the velocity v

at any time is
1 2t
(a) e t i + 2e2 t j + k + c (b) e t i + e j+ c
2
1 2t
(c) e t i + e j + tk + c (d) none of these
2

1
1. ∫0 [t i + (t 2 − 2t) j ] dt = …… .
2. If F (t) = 3t 2 i + t j + 2 k and G (t) = 6t 2 i + (t − 1) j + 3t k , then

1  dF • G + F • dG dt = …… .
∫0  dt dt  (Meerut 2011)
2
2 d r
3. If r (t) = 5t 2 i + t j − t 3 k , then ∫ r× dt = …… .
1 dt 2 (Kumaun 2013)
1 dr
4. If r = t 2 i + j − t k , then ∫ r• dt = …… .
0 dt (Kumaun 2014)
True or False
 dr d2 r  dr 2
1. ∫ 2 • 2  dt =   + c .
 dt dt   dt 
1
2. The value of ∫ (e t i + e − 2 t j + t k ) dt is (e − 1) i − (e − 2 − 1) j + k .
0
A nswers

1. (b) 2. (d) 3. (c) 4. (a) 5. (b)
6. (b) 7. (c)
D-296

1 2
1. i− j 2. 24
2 3
3. −14 i + 75 j − 15 k 4. 1
True or False
1. T 2. F
¨
Line Integrals
D-297
11
Line Integrals
1 Some Preliminary Concepts

riented curve. Suppose C is a
O curve in space. Let us orient C by
taking one of the two directions
along C as the positive direction; the
opposite direction along C is then called
the negative direction . Suppose A is the
initial point and B the terminal point of
C under the chosen orientation. In case these two points coincide, the curve C is
called a closed curve.
Smooth curve. Let r (t) = x (t) i + y (t) j + z (t) k , where r (t) is the position
vector of ( x, y, z ), be the parametric representation of a curve C joining the points A
and B, where t = t1 and t = t2 respectively. We know that dr / dt is a tangent vector
D-298
to this curve at the point r. Suppose the function r (t) is continuous and has a
continuous first derivative not equal to zero vector for all values of t under
consideration. Then the curve C possesses a unique tangent at each of its points. A
curve satisfying these assumptions is called a smooth curve.
A curve C is said to be
piecewise smooth if it is
composed of a finite number
of smooth curves. The curve
C in the adjoining figure is
piecewise smooth as it is
composed of three smooth curves C1 , C2 and C3 . The circle is a smooth closed
curve while the curve consisting of the four sides of a rectangle is a piecewise
smooth closed curve.
Smooth surface. Suppose S is a surface which has a unique normal at each of its
points and the direction of this normal depends continuously on the points of S.
Then S is called a smooth surface.
If a surface S is not smooth but can be subdivided into a finite number of smooth
surfaces, then it is called a piecewise smooth surface. The surface of a sphere is
smooth while the surface of a cube is piecewise
smooth.
Classification of regions. A region R in which
every closed curve can be contracted to a point
without passing out of the region is called a
simply connected region. Otherwise the region
R is multiply-connected. The region interior to a
circle is a simply-connected plane region. The
region interior to a sphere is a simply-connected
region in space. The region between two
concentric circles lying in the same plane is a
multiply-connected plane region.
If we take a closed curve in this region surrounding the inner circle, then it cannot
be contracted to a point without passing out of the region. Therefore the region is
not simply-connected. However the region between two concentric spheres is a
simply-connected region in space. The region between two infinitely long coaxial
cylinders is a multiply-connected region in space.
2 Line Integrals
(Avadh 2014)
Any integral which is to be evaluated along a curve is called a line integral.
D-299
Suppose r (t) = x (t) i + y (t) j + z (t) k ,where r (t) is the position vector of ( x, y, z )
i. e., r (t) = x i + y j + z k , defines a piecewise smooth curve joining two points A
and B. Let t = t1 at A and t = t2 at B. Suppose F( x, y, z ) = F1 i + F2 j + F3 k is a vector
point function defined and continuous along C. If s denotes the arc length of the
dr
curve C, then = t is a unit vector along the tangent to the curve C at the point r.
ds
dr dr
The component of the vector F along this tangent is F • . The integral of F •
ds ds
along C from A to B written as
B  dr  B
∫ A F • ds  ds = ∫ A F • dr = ∫ C F • dr
is an example of a line integral. It is called the tangent line integral of F along C.
Since r = x i + y j + z k , therefore, dr = dx i + dy j + dz k .
∴ F • dr = ( F1 i + F2 j + F3 k ) • (dx i + dy j + dz k )
= F1 dx + F2 dy + F3 dz .
Therefore in components form the above line integral is written as
∫C F • dr = ∫C ( F1 dx + F2 dy + F3 dz ).
The parametric equations of the curve C are x = x(t), y = y(t) and z = z (t).
Therefore we may write
t2  dx dy dz 
∫C F • dr = ∫t 1  F1 dt + F2 dt + F3 dt  dt.
 
Circulation: If C is a simple closed curve ( i.e. a curve which does not intersect itself
anywhere), then the tangent line integral of F around C is called the circulation of F about C.
It is often denoted by
∫ F • dr = ∫ ( F1 dx + F2 dy + F3 dz ).
Work done by a Force. Suppose a force F acts upon a particle. Let the particle be
displaced along a given path C in space. If r denotes the position vector of a point
dr
on C,then is a unit vector along the tangent to C at the point r in the direction of s
ds
dr
increasing. The component of force F along tangent to C is F • . Therefore the
ds
dr
work done by F during a small displacement ds of the particle along C is F •  ds
 ds 
i. e., F • dr. The total work W done by F in this displacement along C, is given by the
line integral
W= ∫C F • dr,
the integration being taken in the sense of the displacement.

D-300
Example 1: Evaluate ∫C F • dr, where F = x 2 i + y 3 j and curve C is the arc of the
parabola y = x 2 in the x-y plane from (0, 0) to (1, 1).
Solution: We shall illustrate two methods for the solution of such a problem.
Method 1. The curve C is the parabola y = x 2 from (0, 0) to (1, 1).
Let x = t ; then y = t 2 . If r is the position vector of any point ( x, y) on C, then
dr
r (t) = xi + yj = ti + t 2 j . ∴ = i + 2tj.
dt
Also in terms of t, F = t 2 i + t 6 j.
At the point (0, 0), t = x = 0. At the point (1, 1), t = 1.
 dr  1 2 6
∴ ∫ C F • dr = ∫ C  F • dt  dt = ∫0 (t i + t j) • (i + 2t j) dt
1
1 2  t3 2t 8  1 1 7
= ∫ t + 2t 7 ) dt =  +  = + = .
0 3 8 3 4 12
 0
Method 2: In the xy-plane we have r = xi + yj .
∴ dr = dxi + dyj.
Therefore, F • dr = ( x 2 i + y 3 j) • (dx i + dy j) = x 2 dx + y 3 dy.

∴ ∫C F • dr = ∫C ( x 2 dx + y 3 dy).
Now along the curve C, y = x 2 . Therefore dy = 2 x dx.

1
∴ ∫C F • dr = ∫ x =0 [ x 2 dx + x 6 (2 x) dx]
1
1 2 7  x3 2 x8  7
= ∫0 (x + 2 x ) dx =  +  = .
3 8  0 12
Example 2: If F = 3 xy i − y 2 j , evaluate ∫ F • dr, where C is the curve in the x y-plane,

C
y = 2 x 2 , from (0, 0) to (1, 2). (Garhwal 2001, 02; Kumaun 07; Rohilkhand 12)
Solution: The parametric equations of the parabola y = 2 x 2 can be taken as
x = t, y = 2t 2 .
At the point (0, 0), x = 0 and so t = 0.Again at the point (1, 2), x = 1and so t = 1.
Now ∫C F • dr = ∫C (3 xyi − y 2 j) • (dx i + dy j)
[∵ r = xi + yj, so that dr = dxi + d y j]

D-301
1  dx dy 
= ∫C (3 xy dx − y 2 dy) = ∫t =0 3 xy − y2  dt
 dt dt 
1
= ∫0 (3 . t . 2 t 2 . 1 − 4 t 4 . 4t) dt
[∵ x = t, y = 2t 2 so that dx / dt = 1 and dy / dt = 4t]

1
1 3 5  t4 t6 
=∫0 (6 t − 16 t ) dt = 6 . − 16 . 
 4 6 0
6 16 3 8 7
= − = − =− .
4 6 2 3 6
Example 3: Evaluate ∫ F • dr along the curve x 2 + y 2 = 1, z = 1in the positive direction
from (0, 1, 1) to (1, 0, 1) where F = (2 x + yz ) i + xzj + ( xy + 2z ) k .
Solution: Let the given curve be denoted by C and let A and B be points (0, 1, 1) and
(1, 0, 1) respectively.
Along the given curve C, we have r = xi + yj + zk .
∴ dr = dx i + dy j + dz k .
B
∴ ∫C F • dr = ∫A [(2 x + yz ) i + xz j + ( xy + 2z ) k ] • (dx i + dy j + dz k )
B
= ∫A [(2 x + yz ) dx + xz dy + ( xy + 2z ) dz ]. ...(1)
In moving from A to B, x varies from 0 to 1 , y varies from 1 to 0 and z remains

constant. We have z = 1 and so dz = 0.
Hence from (1)
1 0
∫C F • dr = ∫0 (2 x + y) dx + ∫ x dy + 0
1
1 1 1
= ∫0 [2 x + √ (1 − x 2 )] dx − ∫ √ (1 − y 2 ) dy = [ x 2 ] 0 = 1,
0
the last two integrals cancel by a property of definite integrals.

Example 4: Evaluate ∫ ( x dy − y dx) around the circle x 2 + y 2 = 1.
(Meerut 2002)
2 2
Solution: Let C denote the circle x + y = 1. The parametric equations of this
circle are x = cos t, y = sin t.
To integrate around the circle C we should vary t from 0 to 2π.
2π  dy dx 
∴ ∫C ( x dy − y dx) = ∫0 x − y  dt
 dt dt 
2π 2π
= ∫0 (cos 2 t + sin2 t) dt = ∫0 dt = 2π.
Example 5: If F = (2 x + y) i + (3 y − x) j , evaluate ∫ F • dr where C is the curve in
C
the xy-plane consisting of the straight lines from (0, 0) to (2, 0) and then to (3, 2).
(Agra 2007; Meerut 11; Purvanchal 14)
D-302
Solution: The path of integration C has been shown in the figure.

It consists of the straight lines OA and AB.
We have ∫C F • dr
= ∫C [(2 x + y) i + (3 y − x) j] • (dx i + dy j)
= ∫C [(2 x + y) dx + (3 y − x) dy].
Now along the straight line OA, y = 0, dy = 0 and x

varies from 0 to 2. The equation of the straight line AB is
2−0
y−0 = ( x − 2) i. e., y = 2 x − 4.
3−2
∴ along AB, y = 2 x − 4, dy = 2 dx and x varies from 2 to 3.
2 3
∴ ∫C F • dr = ∫0 [(2 x + 0) dx + 0] + ∫ [(2 x + 2 x − 4) dx
2
+ (6 x − 12 − x) 2 dx]
2 3
= [x ] 2
0
+∫
2
(14 x − 28) dx
3
3 ( x − 2)2 
= 4 + 14∫ ( x − 2) dx = 4 + 14   = 4 + 7 = 11.
2 2
  2
Example 6: Evaluate ∫ F • dr where F = ( x 2 + y 2 ) i − 2 xyj , curve C is the rectangle in
C
the x y-plane bounded by y = 0, x = a, y = b, x = 0.

(Meerut 2000, 06B, 07B, 09B, 12, 13; Kanpur 10; Bundelkhand 09;
Purvanchal 09)
Solution: In the xy-plane z = 0.
Therefore
r = x i + y j and dr = dx i + dy j .
The path of integration C has been shown in
the figure. It consists of the straight lines
OA, AB, BD and DO.
We have
∫C F • dr = ∫C [( x 2 + y 2 ) i − 2 xy j] • (dx i + dy j)
= ∫C [( x 2 + y 2 ) dx − 2 xy dy].
Now on OA, y = 0, dy = 0 and x varies from 0 to a ; on AB, x = a, dx = 0 and y

varies from 0 to b ; on BD, y = b, dy = 0 and x varies from a to 0; on
DO, x = 0, dx = 0 and y varies from b to 0.
a b 0 0
∴ ∫C F • dr = ∫0 x 2 dx − ∫ 2a y dy + ∫ ( x 2 + b 2 ) dx + ∫ 0 dy
0 a b
D-303
a b 0
 x3   y2   x3 
=   − 2a   + + b 2 x + 0 = − 2ab 2 .
 3 0  2  0  3 a
Example 7: Find the total work done in moving a particle in a force field given by
F = 3 xyi − 5zj + 10 xk
along the curve x = t 2 + 1, y = 2t 2 , z = t 3 from t = 1 to t = 2.
Solution: Let C denote the arc of the given curve from t = 1to t = 2. Then the total
work done
=∫C F • dr = ∫C (3 x yi − 5z j + 10 x k ) • (dx i + dy j + dz k )
=∫ (3 xy dx − 5z dy + 10 x dz )
C
2  dx dy dz 
= ∫1 3 x y − 5z + 10 x  dt
 dt dt dt 
2
= ∫1 [3 (t 2 + 1) (2t)2 (2t) − (5t 3 )(4t) + 10 (t 2 + 1)(3t 2 )] dt
2
= ∫1 (12t 5 + 12t 3 − 20 t 4 + 30 t 4 + 30 t 2 ) dt
2
= ∫1 (12t 5 + 10 t 4 + 12t 3 + 30 t 2 ) dt = 303.
Example 8: Evaluate ∫ F • dr, where F = yz i + zx j + x y k and C is the portion of the

C
curve r = a cos t i + b sin t j + ct k , from t = 0 to t = π / 2. (Avadh 2010)

Solution: Along the curve C,
r = x i + y j + z k = a cos t i + b sin t j + ct k .
∴ x = a cos t, y = b sin t, z = ct.
Now ∫C F • dr = ∫C ( yz i + z x j + x y k ) • (dx i + dy j + dz k )
= ∫C ( yz dx + z x dy + x y dz ) = ∫C d ( x yz )
t = π /2 π /2
= [ x yz ] t =0 = [(a cos t) . (b sin t) . (ct)] 0
= abc [t cos t sin t ] 0π /2 = abc (0 − 0) = 0.
1. Find ∫ t • dr where t is the unit tangent vector and C is the unit circle,
C
in xy-plane, with centre at the origin. (Bundelkhand 2008)

2
2. (i) Integrate the function F = x i − xyj from the point (0, 0) to (1, 1) along
parabola y 2 = x.
D-304
(ii) Evaluate ∫ x y 3 ds, where C is the segment of the line y = 2 x in the

C
xy-plane from (−1, − 2) to (1, 2).

3. Evaluate ∫ F • dr, where F = i cos y − j x sin y and C is the curve
C
2
y = √ (1 − x ) in the xy-plane from (1, 0) to (0, 1). (Agra 2001)
2 2 2
4. (i) Evaluate ∫C F • dr where F is x y i + yj and C is y = 4 x in the
xy-plane from (0, 0) to (4, 4).

(Garhwal 2003; Meerut 04, 06;
Kanpur 09, 11, 12; Rohilkhand 11)
(ii) Evaluate ∫C F • dr, where F = x y i + yz j + z x k and C is the curve
r = ti + t 2 j + t 3 k , t varying from −1 to + 1.
(iii) Evaluate ∫C F • dr, where F = (2 x + y) i + (3 y − x) j + yzx k and C is
the curve x = 2t 2 , y = t, z = t 3 from t = 0 to t = 1. (Kumaun 2011)

5. Evaluate ∫ F • dr where
C
F = c [−3a sin2 t cos t i + a (2 sin t − 3 sin3 t) j + b sin 2t k ]

and C is given by r = a cos t i + a sin t j + bt k from t = π / 4 to π / 2.
6. Evaluate ∫ F • dr where F = z i + x j + y k and C is the arc of the curve
C
r = cos t i + sin t j + t k from t = 0 to t = 2π .

7. Evaluate ∫ F • dr where F = x y i + ( x 2 + y 2 ) j and C is the x-axis from
C
x = 2 to x = 4 and the straight line x = 4 from y = 0 to y = 12.

8. Find the work done in moving a particle in a force field
F = 3 x 2 i + (2 xz − y) j + z k along the line joining (0, 0, 0) to (2, 1, 3).
9. Calculate ∫ [( x 2 + y 2 ) i + ( x 2 − y 2 ) j] • dr where C is the curve :
C
(i) y 2 = x joining (0, 0) to (1, 1). (Meerut 2005B)

(ii) x 7 = y joining (0, 0) to (1, 1).
(iii) consisting of two straight lines joining (0, 0) to (1, 0) and (1, 0) to (1, 1).
(iv) consisting of three straight lines joining (0, 0) to (2, − 2), (2, − 2) to
(0, − 1) and (0, − 1) to (1, 1).
10. Find the circulation of F round the curve C, where
x
F=e sin y i + e x cos y j and C is the rectangle whose vertices are
1 1
(0, 0), (1, 0), (1, π), (0, π).
2 2
11. Find the circulation of F round the curve C, where F = ( x − y) i + ( x + y) j
and C is the circle x 2 + y 2 = 4, z = 0.
D-305
12. (i) If F = (2 x 2 + y 2 ) i + (3 y − 4 x) j , evaluate ∫ F • dr around the triangle
ABC whose vertices are A (0, 0), B (2, 0) and C (2, 1).
(Kumaun 2008)
2 2
(ii) If F = (3 x + 6 y) i − 14 yz j + 20 xz k , then evaluate ∫ F • dr from
(0, 0, 0) to (1, 1, 1) along the curve x = t, y = t 2 , z = t 3 .

(Kumaun 2012)
A nswers 1
1 16
1. 2π 2. (i) (ii) 3. − 1
12 √5
10 277 1
4. (i) 264 (ii) (iii) 5. c (a2 + b 2 )
7 42 2
6. 2 π + π = 3π 7. 768 8. 16
7 38 7
9. (i) (ii) (iii) 1 (iv) − 10. 0
10 45 3
11. 8π 12. (i) −14 / 3 (ii) 5
3 Surface Integrals
(Avadh 2014)
Any integral which is to be evaluated over a surface is called a surface integral.
Suppose S is a surface of finite area. Suppose

f ( x, y, z ) is a single valued function of position
defined over S. Subdivide the area S into n elements
of areas δS1 , δS2 , …, δS n . In each part δS k we
choose an arbitrary point Pk whose coordinates are
( x k , y k , z k ).
We define f ( Pk ) = f ( x k , y k , z k ).
n
Form the sum ∑ f ( Pk ) δS k .
k =1
Now take the limit of this sum as n → ∞ in such a way that the largest of the areas
δS k approaches zero. This limit if it exists, is called the surface integral of f ( x, y, z )
over S and is denoted by ∫ ∫ f ( x, y, z ) dS.
S
It can be shown that if the surface S is piecewise smooth and the function f ( x, y, z )
is continuous over S, then the above limit exists i. e., is independent of the choice of
sub-divisions and points Pk .
D-306
Flux: Suppose S is a piecewise smooth surface

and F( x, y, z ) is a vector function of position
defined and continuous over S. Let P be any point
on the surface S and let n be the unit vector at P in
the direction of outward drawn normal to the
surface S at P. Then F • n is the normal
component of F at P. The integral of F • n over S is
∫ ∫ F • n dS.
S
It is called the flux of F over S.

Let us associate with the differential of surface area dS a vector dS (called vector area)
whose magnitude is dS and whose direction is that of n. Then dS = n dS. Therefore
we can write ∫ ∫ F • n dS = ∫ ∫ F • dS.
S S
Suppose the outward drawn normal to the surface S at P makes angles α, β, γ with
the positive directions of x, y and z axes respectively. If l, m, n are the direction
cosines of the outward drawn normal, then
l = cos α, m = cos β, n = cos γ .
Also n = cos α i + cos β j + cos γ k = l i + m j + n k .
Let F( x, y, z ) = F1 i + F2 j + F3 k . Then
F • n = F1 cos α + F2 cos β + F3 cos γ = F1 l + F2 m + F3 n.
Therefore, we can write
∫ ∫S F • n dS = ∫ ∫S ( F1 cos α + F2 cos β + F3 cos γ ) dS
= ∫ ∫S ( F1 dy dz + F2 dz dx + F3 dx dy),
if we define ∫ ∫S F1 cos α dS = ∫ ∫S F1 dy dz ,
∫ ∫S F2 cos β dS = ∫ ∫S F2 dz dx, ∫ ∫S F3 cos γ dS = ∫ ∫ F3 dx dy.

S
Note 1: Other examples of surface integrals are ∫ ∫ f n dS, ∫ ∫ F × dS

S S
where f ( x, y, z ) is a scalar function of position.
Note 2: Important. In order to evaluate surface
integrals it is convenient to express them as
double integrals taken over the orthogonal
projection of the surface S on one of the
coordinate planes. But this is possible only if any
line perpendicular to the co-ordinate plane
chosen meets the surface S in no more than one
point. If the surface S does not satisfy this
condition, then it can be sub-divided into surfaces which do satisfy this
condition.
D-307
Suppose the surface S is such that any line perpendicular to the xy-plane meets S in
no more than one point. Then the equation of the surface S can be written in the
form z = h ( x, y).
Let R be the orthogonal projection of S on the x y-plane. If γ is the acute angle
which the undirected normal n at P( x, y, z ) to the surface S makes with
z-axis, then it can be shown that cos γ dS = dx dy,
where dS is the small element of area of surface S at the point P.
dx dy dx dy
Therefore dS = = , where k is the unit vector along z-axis.
cos γ |n • k |
dx dy
Hence ∫ ∫S F • n dS = ∫ ∫R F •n .
|n • k|
Thus the surface integral on S can be evaluated with the help of a double integral
integrated over R.
Example 9: Evaluate ∫ ∫ F • n dS, where F = yz i + zx j + xy k and S is that part of the

S
surface of the sphere x + y 2 + z 2 = 1 which lies in the first octant.
2
(Garhwal 2003; kumaun 14))

Solution: A vector normal to the surface S is given by
∇ ( x 2 + y 2 + z 2 ) = 2 x i + 2 y j + 2z k .
Therefore n = a unit normal to any point ( x, y, z ) of S
2 x i + 2 y j + 2z k
= = xi + yj + zk ,
√ (4 x 2 + 4 y 2 + 4z 2 )
since x 2 + y 2 + z 2 = 1 on the surface S.
dx dy
We have ∫ ∫S F • n dS = ∫ ∫R F•n ,
|n • k |
where R is the projection of S on the xy-plane.
The region R is bounded by x-axis, y-axis and the circle x 2 + y 2 = 1, z = 0.
We have F • n = ( yz i + zx j + x y k ) • ( x i + y j + z k ) = 3 x yz .
Also n • k = ( x i + y j + z k ) • k = z . ∴ |n • k | = z .
3 x yz
Hence ∫ ∫ S F • n dS = ∫ ∫ R z dx dy = 3∫ ∫ R x y dx dy
π/2 1
= 3∫ ∫r =0 (r cos θ) (r sin θ) r dθ dr, on changing to polars
θ =0
1
π/2 r4  3  1 3
= 3∫   cos θ sin θ dθ =   = .
0 4 4 2 8
 0
D-308
Example 10: Evaluate ∫ ∫ F • n dS,where F = z i + x j − 3 y 2 zk and S is the surface of

S
the cylinder x 2 + y 2 = 16 included in the first octant between z = 0 and z = 5.
Solution: A vector normal to the surface S is given by

∇ ( x 2 + y 2 ) = 2 x i + 2 y j.
Therefore n = a unit normal to any point of S
2x i + 2 y j xi+ y j
= 2 2
= ,
√ (4 x + 4 y ) 4
since x 2 + y 2 = 16, on the surface S.
dx dz
We have ∫ ∫ F • n dS = ∫ ∫R F•n ,where R is the projection of S on the x-z
S | n • j|
plane. It should be noted that in this case we cannot take the projection of S on the
x-y plane as the surface S is perpendicular to the x-y plane.
 x i + y j 1
Now F • n = ( z i + x j − 3 y2 z k) •   = ( xz + x y ),
 4  4
 x i + y j y
n•j=   •j= ⋅
 4  4
Therefore the required surface integral is
xz + x y dx dz
= ∫∫
R 4 y /4
5 4  xz 
=∫  + x  dx dz , since y = √ (16 − x 2 ) on S
z = 0 ∫ x = 0  √ (16 − x 2 )
 
5
= ∫0 (4z + 8) dz = 90.
4 Volume Integrals
(Avadh 2014)
Suppose V is a volume bounded by a
surface S. Suppose f ( x, y, z ) is a single
valued function of position defined over V.
Subdivide the volume V into n elements of
volumes δV1 , δV2 , …, δVn . In each part δVk
we choose an arbitrary point Pk whose
co-ordinates are ( x k , y k , z k ). We define
f ( Pk ) = f ( x k , y k , z k ).
Form the sum
n
∑ f ( Pk ) δVk .
k =1
D-309
Now take the limit of this sum as n → ∞ in such a way that the largest of the
volumes δVk approaches zero. This limit, if it exists, is called the volume integral of
f ( x, y, z ) over V and is denoted by ∫ ∫ ∫ f ( x, y, z ) dV .
V
It can be shown that if the surface is piecewise smooth and the function f ( x, y, z )
is continuous over V, then the above limit exists i. e., is independent of the choice of
sub-divisions and points Pk .
If we subdivide the volume V into small cuboids by drawing lines parallel to the
three co-ordinates axes, then dV = dx dy dz and the above volume integral
becomes ∫ ∫ ∫ f ( x, y, z) dx dy dz.
V
If F ( x, y, z ) is a vector function, then ∫ ∫ ∫V F dV

is also an example of a volume integral.
Example 11: If F = (2 x 2 − 3z ) i − 2 x y j − 4 x k , then evaluate ∫ ∫ ∫V ∇ • F dV
where V is the closed region bounded by the planes x = 0, y = 0, z = 0 and

2 x + 2 y + z = 4. Also Evaluate ∫ ∫ ∫ ∇ × F dV .
V
Solution: We have F = (2 x 2 − 3z ) i − 2 x y j − 4 x k .
 ∂ ∂ ∂
∴ ∇ • F = i + j +k  • [(2 x 2 − 3z ) i − 2 x y j − 4 x k ]
 ∂x ∂y ∂z 
∂ ∂ ∂
= (2 x 2 − 3z ) + (− 2 x y ) + (− 4 x ) = 4 x − 2 x = 2 x.
∂x ∂y ∂z
∴ ∫ ∫ ∫V ∇ • F dV = ∫ ∫ ∫V 2 x dx dy dz [∵ dV = dx dy dz ]
2 2−x 4 −2 x −2 y
= 2∫ ∫ y =0 ∫z =0 x dx dy dz.
x =0
[Note that we have taken a thin column parallel to z-axis as the elementary volume.
It cuts the boundary at z = 0 and z = 4 − 2 x − 2 y. Also the projection of the plane
2 x + 2 y + z = 4 on the xy-plane is bounded by the axes y = 0, x = 0 and the line
x + y = 2.Hence the limits for y are from 0 to 2 − x and those for x are from 0 to 2]
2 2− x 4 −2 x −2 y
∴ ∫ ∫ ∫V ∇ • F dV = 2∫
x =0 ∫ y =0 x [z ] z =0
dx dy
2 2− x
= 2∫ ∫ y= 0 x (4 − 2 x − 2 y ) dx dy
x =0
2 2− x
= 2∫
x =0 [4 x y − 2 x 2
y − x y2 ] y =0
dx
2
= 2∫ [4 x (2 − x ) − 2 x 2 (2 − x ) − x (2 − x )2 ] dx
0
2
= 2∫ [ x 3 − 4 x 2 + 4 x ] dx, on simplifying
0
2
1 4 32 8
= 2  x 4 − x 3 + 2 x 2  = 2 4 − + 8 = ⋅
4 3  0  3  3
D-310
Second part: We have

 i j k 
 ∂ ∂ ∂ 
∇×F= 
 ∂x ∂y ∂z 
 2 
2 x − 3z − 2x y − 4x
 
∂ ∂  ∂ ∂ 
=  (− 4 x ) − (− 2 x y ) i −  (− 4 x ) − (2 x 2 − 3z ) j
 ∂ y ∂ z   ∂ x ∂ z 
∂ ∂ 
+  (− 2 x y ) − (2 x 2 − 3z ) k
 ∂ x ∂ y 
= 0 i − (− 4 + 3) j + (− 2 y ) k = j − 2 y k .
∴ ∫ ∫ ∫V ∇ × F dV = ∫ ∫ ∫V ( j − 2 y k ) dx dy dz
2 2− x 4 −2 x −2 y
= ∫ x =0 ∫ y =0 ∫z =0 ( j − 2 y k ) dx dy dz
2 2−x
= ∫ x =0 ∫ y =0 ( j − 2 y k ) (4 − 2 x − 2 y ) dx dy
2−x
=
2  j (4 y − 2 x y − y 2 ) − 2 k (2 y 2 − x y 2 − 2 y 3 ) dx
∫ x =0  3  y = 0
2
= ∫ x = 0 [j (2 − x ) (4 − 2 x − 2 + x )
2 
− 2 k (2 − x )2 2 − x − (2 − x )  dx
 3 
=
2 (2 − x )2 j − 2 (2 − x )3 k  dx
∫0  3 
=
2 ( x − 2)2 j + 2 ( x − 2)3 k  dx
∫0  3 
2 2
( x − 2)3  2 ( x − 2)4 
=  j+   k
 3  0 3 4  0
8 8 8
= j − k = ( j − k ).
3 3 3
1. (i) Evaluate ∫ ∫ F • n dS, where F = 18z i − 12 j + 3 y k and S is the

S
surface of the plane 2 x + 3 y + 6z = 12 in the first octant.

(Bundelkhand 2005)
D-311
(ii) Evaluate ∫ ∫ ∫ φ dV , where φ = 45 x 2 y and V is the closed region

V
bounded by the planes 4 x + 2 y + z = 8, x = 0, y = 0, z = 0.

2. Evaluate ∫ ∫ F • n dS, where F = ( x + y 2 ) i − 2 x j + 2 yz k and S is the
S
surface of the plane 2 x + y + 2z = 6 in the first octant.

3. Evaluate ∫ ∫ A • n dS, where A = x y i − x 2 j + ( x + z ) k , S is the portion
S
of the plane 2 x + 2 y + z = 6 included in the first octant and n is a unit
normal to S.
4. If F = 2 yi − 3 j + x 2 k and S is the surface of the parabolic cylinder y 2 = 8 x
in the first octant bounded by the planes y = 4 and z = 6, then evaluate
∫ ∫ F • n dS.
S
5. Evaluate ∫ ∫ F • n dS, where F = yi + 2 xj − z k and S is the surface of the

S
plane 2 x + y = 6 in the first octant cut off by the plane z = 4.
6. Evaluate ∫ ∫ ∫ F dV where F = x i + y j + z k and V is the region bounded
V
by the surfaces x = 0, x = 2, y = 0, y = 6, z = 4 and z = x 2 .
A nswers 2
1. (i) 24 (ii) 128
27
2. 81 3._ 4._132
4
384
5. 108 6. 24 i + 96 j + k
5

Indicate the correct answer for each question by writing the corresponding letter
from (a), (b), (c) and (d).
1. If F = x 2 i + y 3 j and curve C is the arc of the parabola y = x 2 in the
x y-plane from (0, 0) to (1, 1,
) then ∫ F • d r is
C
(a) 7 / 12 (b) 5 / 12
(c) 11 / 12 (d) none of these
D-312
2. The work done in moving a particle in a force field

F = 3 x 2 i + (2 x z − y) j + 3k
along the line joining (0, 0, 0) to (2 , 1, 3) is
(a) 12 (b) 16
(c) 0 (d) 20
3. For a closed surface S, the value of ∫ ∫S r • n dS is
(a) V (b) 2 V
(c) 3V (d) 0
where V is the volume enclosed by S. (Kumaun 2013, 15)
4. The value of ∫ ( x dy − ydx) around the circle x 2 + y 2 = 1 is
(a) 4π (b) 2π
(Kanpur 2006, 12)
5. If F = 3 xy i − y 2 j, the value of ∫C F • dr is ......, where C is the curve in the
xy-plane, y = 2 x 2 , from (0, 0) to (1, 2)

(a) 1 (b) −1
(c) 0 (d) −7 / 6 (Kanpur 2009)
6. The value of ∫ t • dr is ......, where t is the unit tangent vector and C is

C
the unit circle, in xy-plane, with centre at the origin

(a) 2π (b) −2π
(c) π (d) − π
7. The value of ∫C F • dr is ......, where F = i cos y − j x sin y and C is the
curve y = (1 − x 2 ) in the xy-plane from (1, 0) to (0 , 1)

(a) 1 (b) −1
8. The value of ∫ F • dr is ......, where F = x 2 y 2 i + y j and C is y 2 = 4 x in

C
the xy-plane from (0, 0) to (4 , 4)

(a) 268 (b) 265
Line Integrals
D-313
9. If F = (3 x 2 + 6 y) i − 14 yz j + 20 xz 2 k , then the value of ∫ F • dr from

C
(0, 0, 0) to (1, 1, 1) along the curve x = t, y = t 2 , z = t 3 is

(a) 7 (b) 6
10. The value of ∫ [( x 2 + y 2 ) i + ( x 2 − y 2 ) j] • dr where C is the curve

C
x 7 = y joining (0, 0) to (1, 1) is

38 30
(a) (b)
45 45
35
(c) (d) none of these
45

1. Any integral which is to be evaluated along a curve is called a …… .

2. Any integral which is to be evaluated over a surface is called a …… .
dr
3. If r denotes the position vector of a point on a curve C , then is a unit
ds
vector along the …… to C at the point r in the direction of s increasing.
4. ∫ ∫S F • n dS is called the …… of F over S.
5. If t is the unit tangent vector and C is the unit circle in x y-plane, with centre
at the origin, then ∫C t • dr = …… .
6. The value of ∫ ( x dy − y dx) around the circle x 2 + y 2 = 1 is …… .
7. If F = (3 x 2 + 6 y) i − 14 y z j + 20 x z 2
k and C is a straight line joining
(0, 0, 0) to (1, 1, 1,
) then ∫ F • d r = …… .
C
True or False
1. If C is a simple closed curve, then ∫ F • d r is called the circulation of F
C
about C .
2. If F = ax i + by j + cz k , a, b, c are constants, then

2
∫ ∫ S F • n dS = 3 π (a + b + c), where S is the surface of a unit sphere.
D-314
A nswers
1. (a) 2. (b) 3. (c) 4. (b) 5. (d)
6. (a) 7. (a) 8. (c) 9. (c) 10. (a)

1. line integral 2. surface integral 3. tangent
4. flux 5. 2π 6. 2π 7. 13 / 3
True or False
1. T 2. F
¨
D-315
12
G reen's, G auss's and
S toke's T heorems
1 Green’s Theorem in the Plane

et R be a closed bounded region in the x-y plane whose boundary C consists of finitely
L many smooth curves. Let M and N be continuous functions of x and y having continuous
∂M ∂N
partial derivatives and in R. Then
∂y ∂x
 ∂N ∂M 
∫ ∫R  −  dx dy = ∫C ( M dx + N dy ),
 ∂x ∂y 
the line integral being taken along the entire boundary C of R such that R is on the left as one
advances in the direction of integration.
(Meerut 2004, 05, 07, 10; Avadh 10; Purvanchal 08;
Rohilkhand 09B, 12; Kashi 13)
Proof: We shall first prove the theorem for a special region R bounded by a closed
curve C and having the property that any straight line parallel to any one of the
coordinate axes and intersecting R has only one segment (or a single point) in
common with R. This means that R can be represented in both of the forms
a ≤ x ≤ b, f ( x) ≤ y ≤ g ( x)
and c ≤ y ≤ d, p( y) ≤ x ≤ q ( y).
D-316
In the adjoining figure, the equations of the curves AEB and BFA are y = f ( x) and
y = g ( x) respectively. Similarly the
equations of the curves FAE and EBF are
x = p ( y) and x = q ( y) respectively.
We have
∂M
∫ ∫R dx dy
∂y
b  g ( x) ∂M 
= ∫ x = a ∫ y = f ( x) dy  dx
∂y 
y = g ( x)
b  
= ∫x=a  M ( x, y )  dx
  y = f ( x)
b
= ∫ x = a [M [ x, g ( x )] − M [ x, f ( x)]] dx
b a
=− ∫a M [ x, f ( x)] dx − ∫b M [ x, g ( x)] dx
=− ∫C M ( x, y) dx, since y = f ( x) represents the curve AEB
and y = g ( x) represents the curve BFA.

If portions of C are segments parallel to y-axis
such as GH and PQ in the adjoining figure,
then the above result is not affected. The line
integral ∫ M dx over GH is zero because on
GH, we have x = constant implies

dx = 0. Similarly the line integral over PQ is
zero. The equations of QG and HP are y = f ( x)
and y = g ( x) respectively. Hence we have
∂M
−∫ ∫ dx dy = ∫C M ( x, y ) dx. ... (1)
R ∂y
∂N d  q ( y) ∂N 
Similarly, ∫ ∫R dx dy = ∫ y =c ∫ x = p ( y) dx  dy
∂x ∂x 
x = q ( y)
d  
= ∫y=c  N ( x, y)  dy
  x = p ( y)
d
= ∫ y=c [N [ q ( y ), y ] − N [ p ( y ), y]] dy
d c
= ∫ c N [ q ( y ), y ] dy + ∫ d N [ p ( y ), y ] dy
= ∫ N ( x, y ) dy. ... (2)

C
D-317
 ∂N ∂M 
From (1) and (2), we get on adding ∫ ∫R  −  dx dy = ∫C ( M dx + N dy) .
 ∂x ∂y 
The proof of the theorem can now be extended to a region R which can be
subdivided into finitely many special
regions of the above type by drawing
lines (TS in the adjoining figure). In this
case we apply the theorem to each
subregion (R1 and R2 in the figure) and
then add the results. The sum of the left
hand members will be equal to the
integral over R. The sum of the right
hand members will be equal to the
integral over C plus the line integrals over
the curves introduced for subdividing R.
Each of the latter integrals comes twice,
taken once in each direction (as ST and TS in the figure). Therefore these two
integrals cancel each other and thus the sum of the right hand members will be
equal to the line integral over C.
Note: Extension of Green’s theorem in plane to multiply
connected regions.
Green’s theorem in the plane is also
valid for a multiply-connected
region R such as shown in the
adjoining figure. Here the boundary
C of R consists of two parts; the
exterior boundary C1 is traversed in
the anticlockwise sense so that R is
on the left, while the interior
boundary C2 is traversed in the
clockwise sense so that R is on the
left.
In order to establish the theorem, we construct a line such as AD (called a cross cut)
connecting the exterior and interior boundaries. The region bounded by
ADEFGDAPQLHA is simply-connected and so Green’s theorem is valid for it.
Therefore
 ∂N ∂M 
∫C M dx + N dy = ∫ ∫  −  dx dy.
R  ∂x ∂y 
ADEFGDAPQLHA
The integral on the left hand side leaving out the integrand is equal to
∫AD + ∫ C2 + ∫ DA + ∫ C1 = ∫ C2 +∫ , since ∫ =−∫
C1 AD DA
= ∫C ( M dx + N dy ).
Hence the theorem.

D-318
2 Green’s Theorem in the Plane in Vector Notation

We have r = x i + y j so that dr = dx i + dy j.
Let F = Mi + Nj .
Then M dx + N dy = ( M i + N j) • (dx i + dy j) = F • dr.
Also curlF = ∇ × F
i j k
∂ ∂ ∂ ∂N ∂M  ∂N ∂M 
= = − i+ j+  −  k.
 ∂x ∂y ∂z ∂z ∂z  ∂x ∂y 
 
M N 0
∂N ∂M
∴ (∇ × F) • k = − ⋅
∂x ∂y
Hence Green’s theorem in plane can be written as
∫ ∫R (∇ × F) • k dR = ∫C F • dr
where dR = dx dy and k is unit vector perpendicular to the x y-plane.

If s denotes the arc length of C and t is the unit tangent vector to C, then
dr
dr = ds = t ds. Therefore the above result can also be written as
ds
∫ ∫R (∇ × F) • k dR = ∫C F • t ds.
Example 1: Verify Green’s theorem in the plane for ∫C ( xy + y 2 ) dx + x 2 dy , where C
is the closed curve of the region bounded by y = x and y = x 2 . (Rohilkhand 2011)

Solution: By Green’s theorem in plane, we have
 ∂N ∂M 
∫ ∫ R  ∂x − ∂y  dx dy = ∫ C ( M dx + N dy ).
Here M = x y + y 2 , N = x 2 .
The curves y = x and y = x 2 intersect at (0, 0) and
(1, 1). The positive direction in traversing C is as
shown in the figure.
 ∂N ∂M 
We have ∫ ∫R  ∂x − ∂y  dx dy
∂ 2 ∂ 
= ∫ ∫R  (x )− ( xy + y 2 ) dx dy
 ∂x ∂y 
= ∫ ∫R (2 x − x − 2 y ) dx dy = ∫ ∫R ( x − 2 y ) dx dy
D-319
x
1 x 1  2
= ∫ x = 0 ∫ y = x2 ( x − 2 y) dy dx = ∫x =0  xy − y  dx
  y= x 2
1 1
= ∫0 [ x 2 − x 2 − x 3 + x 4 ] dx = ∫0 ( x 4 − x 3 ) dx
1
 x5 x4  1 1 1
= −  = − =− ⋅
 5 4 0 5 4 20
Now let us evaluate the line integral along C. Along y = x 2 , dy = 2 x dx. Therefore
along y = x 2 , the line integral equals
1 1 19
∫0 [{( x )( x 2 ) + x 4 } dx + x 2 (2 x) dx ] = ∫0 (3 x 3 + x 4 ) dx = ⋅
20
Along y = x, dy = dx. Therefore along y = x, the line integral equals
0 0
∫1 [{( x) ( x) + x 2 } dx + x 2 dx ] = ∫1 3 x 4 dx = − 1.
19 1
Therefore the required line integral = −1= − ⋅ Hence the theorem is
20 20
verified.
Example 2: Evaluate by Green’s theorem ∫ ( x 2 − cosh y ) dx + ( y + sin x) dy, where

C
C is the rectangle with vertices (0, 0), (π, 0), (π, 1), (0, 1).
(Meerut 2002, 05B, 06, 13B; Rohilkhand 14)
Solution: By Green’s theorem in plane, we have
 ∂N ∂M 
∫ ∫R  ∂x − ∂y  dx dy = ∫ C ( M dx + N dy ).
Here M = x 2 − cosh y, N = y + sin x.
∂N ∂M
∴ = cos x, = − sinh y.
∂x ∂y
Hence the given line integral is equal to
∫ ∫R (cos x + sinh y ) dx dy
π 1
= ∫ x =0 ∫ y =0 (cos x + sinh y ) dy dx
π 1
=  y cos x + cosh y  dx
∫ x =0   y =0
π
π  
= ∫ x =0 [cos x + cosh 1 − 1] dx =  sin x + x cosh 1 − x  = (cosh 1 − 1).
  0
Example 3: Show that the area bounded by a simple closed curve C is given by
1
( x dy − y dx). Hence find the area of the ellipse x = a cos θ, y = b sin θ.
2 ∫C
D-320
Solution: By Green’s theorem in plane, if R is a plane region bounded by a simple

closed curve C, then
 ∂N ∂M 
∫ ∫R  −  dx dy = ∫C M dx + N dy.
 ∂x ∂y 
Putting M = − y, N = x, we get
∂ ∂ 
∫C ( x dy − y dx) = ∫ ∫R  ( x) − (− y) dx dy
 ∂x ∂y 
=2 ∫ ∫R dx dy
= 2 A, where A is the area bounded by C.

1
Hence A= ∫C ( x dy − y dx).
2
The area of the ellipse
1 1 2π  dy dx 
= ∫ ( x dy − y dx) = ∫  a cos θ − b sin θ  dθ
2 C 2 θ = 0  dθ dθ
1 2π 1 2π
= ∫ (ab cos 2 θ + ab sin2 θ) dθ = ab ∫ dθ = πab.
2 0 2 0
1. Verify Green’s theorem in the plane for

∫C [(2 xy − x 2 ) dx + ( x 2 + y 2 ) dy ],
where C is the boundary of the region enclosed by y = x 2 and y 2 = x
described in the positive sense.
2 2
∫ [(3 x − 8 y ) dx + (4 y − 6 xy) dy],
C
where C is the boundary of the region defined by y = √ x, y = x 2 .
3. Apply Green’s theorem in the plane to evaluate
∫C {( y − sin x ) dx + cos x dy},
where C is the triangle enclosed by the lines y = 0, x = 2π, πy = 2 x.
(Avadh 2010)
−x −x
4. Evaluate by Green’s theorem in plane ∫C (e sin y dx + e cos y dy ),
1 1
where C is the rectangle with vertices (0, 0), (π , 0),  π , π , 0, π .
 2   2 
5. Evaluate by Green’s theorem ∫C (cos x sin y − xy) dx + sin x cos y dy,
where C is the circle x 2 + y 2 = 1. (Kumaun 2012)
6. If F = ( x 2 − y 2 ) i + 2 x y j and r = xi + yj , find the value of ∫ F • dr
around the rectangular boundary x = 0, x = a, y = 0, y = b.
D-321
∫C ( x 2 − xy 3 ) dx + ( y 2 − 2 xy) dy,
where C is the square with vertices (0, 0), (2, 0), (2, 2), (0, 2).
(Meerut 2001)
8. Apply Green’s theorem in the plane to evaluate
∫C [(2 x 2 − y 2 ) dx + ( x 2 + y 2 ) dy], where C is the boundary of the
surface enclosed by the x-axis and the semi-circle y = (1 − x 2 )1 /2 .
9. If C is the simple closed curve in the xy-plane not enclosing the origin,
−i y+ jx
show that ∫ F • dr = 0, where F = .
C x2 + y2
A nswers 1
π 2
3. − − 4. 2 (e −π
− 1) 5. 0 6. 2ab 2
4 π
3 The Divergence Theorem of Gauss

Suppose V is the volume bounded by a closed piecewise smooth surface S. Suppose F ( x, y, z ) is
a vector function of position which is continuous and has continuous first partial derivatives in
V. Then ∫ ∫ ∫ ∇ • F dV = ∫ ∫ F • n dS,
V S
where n is the outward drawn unit normal vector to S.

(Meerut 2000, 01, 06, 10B, 12, 12B;
Bundelkhand 09, 11; Avadh 14; Kashi 14)
Since F • n is the normal component of vector F, therefore divergence theorem may
also be stated as follows :
The surface integral of the normal component of a vector F taken over a closed surface is equal to
the integral of the divergence of F taken over the volume enclosed by the surface.
Cartesian equivalent of Divergence Theorem:
Let F = F1 i + F2 j + F3 k .
∂F ∂F ∂F
Then ∇ • F = div F = 1 + 2 + 3 .
∂x ∂y ∂z
If α, β, γ are the angles which outward drawn unit normal n makes with positive
directions of x, y, z -axes, then cos α, cos β, cos γ are direction cosines of n and we
have n = cos α i + cos β j + cos γ k .
∴ F • n = ( F1 i + F2 j + F3 k ) • (cos α i + cos β j + cos γ k )
= F1 cos α + F2 cos β + F3 cos γ .
D-322
Therefore the divergence theorem can be written as

 ∂F1 ∂F ∂F 
∫ ∫ ∫V  + 2 + 3  dx dy dz
 ∂x ∂y ∂z 
= ∫ ∫S ( F1 cos α + F2 cos β + F3 cos γ ) dS
= ∫∫ F1 dy dz + F2 dz dx + F3 dx dy).
The significance of divergence theorem lies in the fact that a surface integral
may be expressed as a volume integral and vice versa.
Proof of the divergence theorem:
We shall first prove the theorem for a special
region V which is bounded by a piecewise
smooth closed surface S and has the property
that any straight line parallel to any one of the
coordinate axes and intersecting V has only
one segment (or a single point) in common
with V. If R is the orthogonal projection of S
on the xy-plane, then V can be represented in
the form f ( x, y) ≤ z ≤ g ( x, y) where ( x, y)
varies in R.
Obviously z = g ( x, y) represents the upper portion S1 of S, z = f ( x, y) represents
the lower portion S2 of S and there may be a remaining vertical portion S3 of S.
We have
∂F3 ∂F3
∫ ∫ ∫V dV = ∫ ∫ ∫V dx dy dz
∂z ∂z
 g ( x, y) ∂F3 
= ∫ ∫ R ∫ z = f ( x, y) dz  dx dy
∂z 
g ( x, y)
= ∫ ∫ R [F3 ( x, y, z)] z = f ( x, y) dx dy
= ∫ ∫ [ F3 [ x, y, g ( x, y)] − F3 [ x, y, f ( x, y)]] dx dy
R
= ∫ ∫R F3 [ x, y, g ( x, y)] dx dy
− ∫ ∫R F3 [ x, y, f ( x, y)] dx dy ...(1)
Now for the vertical portion S3 of S, the normal n 3 to S3 makes a right angle γ with
k. Therefore ∫ ∫ S3 F3 k • n 3 dS3 = 0, since k • n 3 = 0.
For the upper portion S1 of S, the normal n1 to S1 makes an acute angle γ 1 with k.
Therefore k • n1 dS1 = cos γ 1 dS1 = dx dy.
Hence ∫ ∫ S1 F3 k • n1 dS1 = ∫ ∫R F3 [ x, y, g ( x, y)] dx dy.
D-323
For the lower portion S2 of S,the normal n 2 to S2 makes an obtuse angle γ 2 with k.
Therefore k • n 2 dS2 = cos γ 2 dS2 = − dx dy.
Hence ∫ ∫ S1 F3 k • n 2 dS2 = − ∫ ∫R F3 [ x y, f ( x, y)] dx dy.
∴ ∫ ∫S F3 k • n 3 dS3 + ∫ ∫ S1 F3 k • n1 dS1 + ∫ ∫ S2 F3 k • n 2 dS2
=0 + ∫ ∫R F3 [ x, y, g ( x, y)] dx dy
− ∫ ∫R F3 [ x, y, f ( x, y)] dx dy
or with the help of (1), we get

∂F3
∫ ∫S F3 k • n dS = ∫ ∫ ∫V dV . ...(2)
∂z
Similarly, by projecting S on the other co-ordinate planes, we get
∂F2
∫ ∫S F2 j • n dS = ∫ ∫ ∫V dV ...(3)
∂y
∂F1
and ∫ ∫S F1 i • n dS = ∫ ∫ ∫V dV ...(4)
∂x
Adding (2), (3) and (4), we get
 ∂F1 ∂F ∂F 
∫ ∫S ( F1 i + F2 j + F3 k ) • n dS = ∫ ∫ ∫V  + 2 + 3  dV
 ∂x ∂y ∂z 
or ∫ ∫ ∫V ∇ • F dV = ∫ ∫S F • n dS.
The proof of the theorem can now be extended to a region V which can be
subdivided into finitely many special regions of the above type by drawing
auxiliary surfaces. In this case we apply the theorem to each sub-region and then
add the results. The sum of the volume integrals over parts of V will be equal to the
volume integral over V. The surface integrals over auxiliary surfaces cancel in pairs,
while the sum of the remaining surface integrals is equal to the surface integral over
the whole boundary S of V.
Note: The divergence theorem is applicable for a

region V if it is bounded by two closed surfaces
S1 and S2 one of which lies within the other.
Here outward drawn normals will have the
directions as shown in the figure.
D-324
4 Some Deductions from Divergence Theorem

1. Green’s theorem: Let φ and ψ be scalar point functions which together with their
derivatives in any direction are uniform and continuous within the region V bounded by a
closed surface S, then
∫ ∫ ∫V (φ∇ 2 ψ − ψ∇ 2 φ) dV = ∫ ∫S (φ∇ ψ − ψ∇ φ) • n dS.
Proof: By divergence theorem, we have ∫ ∫ ∫V ∇ • F dV = ∫ ∫S F • n dS.
Putting F = φ∇ψ, we get ∇ • F = ∇ • (φ∇ψ)

= φ (∇ • ∇ψ) + (∇φ) • (∇ψ) = φ∇ 2 ψ + (∇φ) • (∇ψ).
Also F • n = (φ∇ψ) • n.
∴ divergence theorem gives
∫ ∫ ∫V [φ∇ 2 ψ + (∇φ) • (∇ψ)] dV = ∫ ∫S (φ∇ψ) • n dS ...(1)
This is called Green’s first identity or theorem.

Interchanging φ and ψ in (1), we get
∫ ∫ ∫V [ψ∇ 2 φ + (∇ψ) • (∇φ)] dV = ∫ ∫S [ψ∇φ] • n dS ...(2)
Subtracting (2) from (1), we get

∫ ∫ ∫V (φ∇ 2 ψ − ψ∇ 2 φ) dV = ∫ ∫S (φ∇ ψ − ψ∇φ) • n dS ...(3)
This is called Green’s second identity or Green’s theorem in symmetrical form.

∂ψ ∂φ
Since ∇ψ = n and ∇φ = n, therefore
∂n ∂n
 ∂ψ ∂φ 
(φ∇ψ − ψ∇φ) • n =  φ n−ψ n • n
 ∂n ∂n 
∂ψ ∂φ
=φ −ψ .
∂n ∂n
Hence (3) can also be written as
 ∂ψ ∂φ
∫ ∫ ∫V (φ∇ 2 ψ − ψ∇ 2 φ) dV = ∫ ∫S φ − ψ  dS.
 ∂n ∂n
Note: Harmonic function: If a scalar point function φ satisfies Laplace’s equation
∇ 2 φ = 0, then φ is called harmonic function. If φ and ψ are both harmonic functions,
then ∇ 2 φ = 0, ∇ 2 ψ = 0.
 ∂ψ ∂φ
Hence from Green’s second identity, we get ∫ ∫S φ − ψ  dS = 0.
 ∂n ∂n
2. Prove that ∫ ∫ ∫ ∇φ dV = ∫ ∫S φn dS.
V
Proof: By divergence theorem, we have ∫ ∫ ∫ ∇ • F dV = ∫ ∫S F • n dS.

V
D-325
Taking F = φ C where C is an arbitrary constant non-zero vector, we get

∫ ∫ ∫V ∇ • (φC) dV = ∫ ∫S (φC) • n dS. ...(1)
Now ∇ • (φ C) = (∇φ) • C + φ(∇ • C) = (∇φ) • C, since ∇ • C = 0.

Also (φ C) • n = C • (φ n).
∴ (1) becomes
∫ ∫ ∫V C • (∇φ) dV = ∫ ∫S C • (φn) dS
or C•∫∫∫ ∇φ dV = C • ∫ (φn) dS
V S
or C • ∫ ∫ ∫ ∇φ dV − ∫ ∫S φn dS = 0.
 V 
Since C is an arbitrary vector, therefore we must have
∫ ∫ ∫V ∇φ dV = ∫ ∫S φn dS.
3. Prove that ∫ ∫ ∫ ∇ × B dV = ∫ ∫S n × B dS.

V
Proof: In divergence theorem taking F = B × C, where C is an arbitrary constant

vector, we get
∫ ∫ ∫V ∇ • (B × C) dV = ∫ ∫S (B × C) • n dS. ...(1)
Now ∇ • (B × C) = C • curl B − B • curl C = C • curl B, since curl C = 0.

Also (B × C) • n = [B, C, n] = [C, n, B] = C • (n × B).
∴ (1) becomes
∫ ∫ ∫V (C • curl B) dV = ∫ ∫ S C • (n × B) dS
or C • ∫ ∫ ∫ (∇ × B) dV = C • ∫ ∫ (n × B) dS
V S
or C • ∫ ∫ ∫ (∇ × B) dV − ∫ ∫S (n × B) dS = 0.
 V 
Since C is an arbitrary vector therefore we can take C as a non-zero vector which is
not perpendicular to the vector
∫ ∫ ∫V (∇ × B) dV − ∫ ∫S (n × B) dS.
Hence we must have
∫ ∫ ∫V (∇ × B) dV − ∫ ∫S (n × B) dS = 0
or ∫ ∫ ∫V (∇ × B) dV = ∫ ∫ (n × B) dS.
S
Example 4: (i) For any closed surface S, prove that ∫ ∫ curl F • n dS = 0.

S
(Meerut 2009B; Purvanchal 14; Avadh 14)
D-326
(ii) Evaluate ∫ ∫ r • n dS, where S is a closed surface.

S
(Agra 2006; Kumaun 07; Bundelkhand 09; Purvanchal 14)
(iii) If F = ax i + by j + cz k , a, b, c are constants show that
4
∫ ∫S F • n dS = 3 π (a + b + c),
where S is the surface of a unit sphere.
(Bundelkhand 2001, 07, 08; Rohilkhand 07; Kashi 13)
Solution: (i) By divergence theorem, we have
∫ ∫S curl F • n dS = ∫ ∫ ∫V (div curl F) dV ,
where V is the volume enclosed by S

= 0, since div curl F = 0.
(ii) By the divergence theorem, we have
∫ ∫S r • n dS = ∫ ∫ ∫V ∇ • r dV = ∫ ∫ ∫V 3 dV ,
since ∇ • r = div r = 3
= 3V, where V is the volume enclosed by S.
(iii) By the divergence theorem, we have
∫ ∫S F • n dS = ∫ ∫ ∫V (∇ • F) dV ,
where V is the volume enclosed by S
= ∫ ∫ ∫V [∇ • (a x i + by j + cz k )] dV
∂ ∂ ∂ 
= ∫ ∫ ∫V  (a x) + (by) + (cz ) dV
 ∂ x ∂ y ∂ z 
= ∫ ∫ ∫V (a + b + c ) dV
4
= (a + b + c ) V = (a + b + c ) π,
3
4 4
since the volume V enclosed by a sphere of unit radius is equal to π(1)3 i. e., π.
3 3
Example 5(i): Show that ∫ ∫ n dS = 0 for any closed surface S.
S (Purvanchal 14)
(ii) Prove that ∫ ∫ r × n dS = 0 for any closed surface S.
S (Agra 2007)
(iii) Prove that ∫ ∫ n × (a × r) dS = 2Va, where a is a constant vector and V is the volume
S
enclosed by the closed surface S. (Avadh 2011, 12)

Solution: (i) Let C be any arbitrary constant vector.
Then C• ∫ ∫ n dS = ∫ ∫S C • n dS
S
= ∫ ∫ ∫V (∇ • C) dV , by divergence theorem
= 0, since div C = 0.
D-327
Thus C•∫ ∫ n dS = 0, where C is an arbitrary vector.

S
Therefore we must have ∫ ∫ n dS = 0.

S
(ii) Let C be any arbitrary constant vector. Then
C•∫∫ r × n dS = ∫ ∫S C • [(r × n)] dS = ∫ ∫S (C × r) • n dS
S
= ∫ ∫ ∫V [∇ • (C × r)] dV , by divergence theorem
= ∫ ∫ ∫V [r • curl C − C • curl r] dV = 0,
since curl C = 0 and curl r = 0.
Thus C•∫∫ r × n dS = 0, where C is an arbitrary vector.
S
Therefore, we must have ∫ ∫ r × n dS = 0.

S
(iii) We know that

∫ ∫ ∫V ∇ × B dV = ∫ ∫S n × B dS. [See article 4, part 3]
Putting B = a × r, we get
∫ ∫S n × (a × r) dS = ∫ ∫ ∫V ∇ × (a × r) dV
= ∫ ∫ ∫V curl (a × r) dV
= ∫ ∫ ∫V 2a dV , since curl (a × r) = 2a
= 2a ∫ ∫ ∫ dV = 2Va .
V
Example 6: Using the divergence theorem, show that the volume V of a region T bounded by
a surface S is
V = ∫∫ x dy dz = ∫ ∫ y dz dx = ∫ ∫ z dx dy
S S S
1
= ( x dy dz + y dz dx + z dx dy).
3 ∫ ∫S (Meerut 2010)
Solution: By divergence theorem, we have
∂ 
∫ ∫ S x dy dz = ∫ ∫ ∫V  ∂x ( x) dV = ∫ ∫ ∫V dV = V
∂ 
∫ ∫S y dz dx = ∫ ∫ ∫V  ( y) dV = ∫ ∫ ∫V dV = V
 ∂y 
∂ 
∫ ∫S z dx dy = ∫ ∫ ∫V  (z ) dV = ∫ ∫ ∫V dV = V .
 ∂z 
Adding these results, we get
3V = ∫ ∫ ( x dy dz + y dz dx + z dx dy )
S
1
or V = ∫ ∫ ( x dy dz + y dz dx + z dx dy).
3 S
D-328
Example 7: Verify divergence theorem for

F = ( x 2 − yz ) i + ( y 2 − zx ) j + ( z 2 − x y ) k
taken over the rectangular parallelopiped 0 ≤ x ≤ a, 0 ≤ y ≤ b, 0 ≤ z ≤ c .
(Meerut 2006B; Avadh 09; Rohilkhand 13)
Solution: We have
div F = ∇ • F
∂ ∂ ∂
= ( x 2 − yz ) + ( y 2 − zx ) + ( z2 − x y )
∂x ∂y ∂z
= 2 x + 2 y + 2z.
∴ volume integer = ∫ ∫ ∫ ∇ • F dV = ∫ ∫ ∫V 2( x + y + z ) dV
V
c b a
= 2∫ ∫ y =0 ∫ x =0 ( x + y + z ) dx dy dz
z= 0
a
c b  x2 
=2 ∫z =0 ∫ y =0  + yx + zx dy dz
2  x =0
c b  a2 
=2 ∫z =0 ∫ y =0  + ay + az  dy dz
2 
b
c  a2 y2 
= 2∫  y+a + azy dz
z =0 2 2 
  y =0
c 2
 a b ab 2
=2 ∫z =0  + + abz  dz
 2 2 
c
 a2 b ab 2 z 2
=2 z + z + ab 
 2 2 2 0
= [a2 bc + ab 2 c + abc 2 ] = abc (a + b + c ).
Surface Integral: We shall now calculate ∫ ∫ F • n dS over the six faces of the
S
rectangular parallelopiped.
Over the face DEFG, n = i, x = a.
Therefore, ∫ ∫ DEFG F • n dS
c b
= ∫z =0 ∫ y =0 [(a2 − y z ) i
+ ( y 2 − za) j + (z 2 − ay) k ] • i dy dz
c b
= ∫z =0 ∫ y =0 (a2 − yz ) dy dz
b
c  2 y2 
= ∫z =0 a y − z  dz
 2  y =0
D-329
c
c  2 zb 2   2 z2 2  2 c 2 b2
= ∫z =0 a b −  dz = a bz − b  = a bc − ⋅
 2   4 0 4
Over the face ABCO, n = − i, x = 0. Therefore
∫ ∫ABCO F • n dS = ∫∫ [(0 − yz ) i + ... + ... ] • (− i) dy dz
b
c b c  y2 
= ∫z =0 ∫ y =0 y z dy dz = ∫z =0  z dz
 2  y =0
2 2 2
c b b c
= ∫z =0 z dz = ⋅
2 4
Over the face ABEF, n = j , y = b. Therefore
c a
∫ ∫ ABEF F • n dS = ∫z =0 ∫ x =0 [( x 2 − bz ) i + (b 2 − zx) j
+ ( z 2 − bx ) k ] • j dx dz
c a a2 c 2
= ∫z =0 ∫ x =0 (b 2 − zx) dx dz = b 2 ca − ⋅
4
Over the face OGDC, n = − j, y = 0. Therefore
c a c 2 a2
∫ ∫ OGDC F • n dS = ∫z=0 ∫ x =0 zx dx dz = ⋅
4
Over the face BCDE, n = k , z = c . Therefore
b a a2 b 2
∫ ∫ BCDE F • n dS = ∫ y =0 ∫ x =0 (c 2 − xy ) dx dy = c 2 ab − ⋅
4
Over the face AFGO, n = − k , z = 0. Therefore
b a a2 b 2
∫ ∫ AFGO F • n dS = ∫ y =0 ∫ x =0 x y dx dy = ⋅
4
Adding the six surface integrals, we get
 c 2 b2 c 2 b2   2 a2 c 2 a2 c 2 
∫ ∫S F • n dS =  a2 bc − +  +  b ca − + 
 4 4   4 4 
 a2 b 2 a2 b 2 
+  c 2 ab − + 
 4 4 
= abc (a + b + c ).
Hence the theorem is verified.
Example 8: If F = x i − y j + (z 2 − 1)k , find the value of ∫ ∫S F • n dS where S is the
closed surface bounded by the planes z = 0, z = 1 and the cylinder x 2 + y 2 = 4.

(Garhwal 2000; Kanpur 05; Avadh 13; Kumaun 15)
Solution: By divergence theorem, we have ∫ ∫S F • n dS = ∫ ∫ ∫V div F dV .

D-330
∂ ∂ ∂ 2
Here div F = ( x) + (− y) + (z − 1) = 1 − 1 + 2z = 2 z.
∂x ∂y ∂z
1 2 √(4 − y 2 )
∴ ∫ ∫ ∫V div F dV = ∫ z = 0 ∫ y =−2 ∫ x = −√(4 − y 2 ) 2z dx dy dz
1 2 √(4 − y 2 )
= ∫ z = 0 ∫ y = −2 [2zx] dy dz
x = −√(4 − y 2 )
1 2
= ∫z = 0 ∫ y = −2 4z √ (4 − y 2 ) dy dz
1
2  z2 
= ∫ y = −2 4 √ (4 − y 2 ) dy
 2  z =0
2 2
= 2∫ √ (4 − y 2 ) dy = 4∫ √ (4 − y 2 ) dy
y = −2 0
y y 2
=4 √ (4 − y 2 ) + 2 sin −1 
2 2 0
π
= 4 [2 sin −1 1] = 4 (2) = 4π.
2
1. (i) Verify divergence theorem for F = (2 x − z )i + x 2 yj − xz 2 k taken over

the region bounded by x = 0, x = 2, y = 0, y = 2, z = 0, z = 2.
(ii) Verify divergence theorem for F = (2 x − z ) i + x 2 y j − xz 2 k taken

over the region bounded by x = 0, x = 1, y = 0, y = 1, z = 0, z = 1.
(Garhwal 2001; Kumaun 14)
2
2. (i) If F = 4 xz i − y j + yz k and S is the surface bounded by
x = 0, y = 0, z = 0, x = 1, y = 1, z = 1, evaluate ∫ ∫S F • n dS.
(ii) Evaluate ∫ ∫ x 2 dy dz + y 2 dz dx + 2z ( xy − x − y) dx dy
S
where S is the surface of the cube 0 ≤ x ≤ 1, 0 ≤ y ≤ 1, 0 ≤ z ≤ 1.

(Kumaun 2015)
3. (i) Evaluate ∫ ∫S [4 xz dy dz − y 2 dz dx + yz dx dy] where S is the
surface of the cube bounded by the planes x = 0, y = 0, z = 0, x = 1,

y = 1 and z = 1. (Meerut 2005, 06B, 10B, 11)
D-331
(ii) Apply Gauss’s divergence theorem to evaluate

∫ ∫S [( x 3 − yz ) dy dz − 2 x 2 y dz dx + z dx dy ]
over the surface of a cube bounded by the coordinate planes and the
planes x = y = z = a. (Rohilkhand 2011)
4. (i) State divergence theorem of Gauss.
(ii) Use Gauss divergence theorem to show that
1 5
∫ ∫S {( x 3 − yz ) i − 2 x 2 y j + 2 k } • n dS = a ,
3
where S denotes the surface of the cube bounded by the planes
x = 0, x = a, y = 0, y = a, z = 0, z = a. (Bundelkhand 2005, 06)
5. Evaluate ∫ ∫ ( x i + y j + z k ) • n dS where S denotes the surface of the cube
S
bounded by the planes x = 0, y = 0, z = 0, x = a, y = a, z = a by the

application of Gauss divergence theorem. Verify your answer by evaluating the
integral directly. (Garhwal 2003)
6. (i) Evaluate by divergence theorem the integral
2 2 3 2
∫ ∫ xz dy dz + ( x y − z ) dz dx + (2 xy + y z) dx dy,
S
where S is the entire surface of the hemispherical region bounded by
z = √ (a2 − x 2 − y 2 ) and z = 0.
(ii) Evaluate ∫ ∫S ( y 2 z 2 i + z 2 x 2 j + z 2 y 2 k ) • n dS
where S is the part of the sphere x 2 + y 2 + z 2 = 1 above the xy-plane

and bounded by this plane. (Bundelkhand 2006)
7. (i) If F = ax i + by j + cz k , where a,b,c are constants, show that
4π
∫ ∫ S (n • F) dS = 3 (a + b + c),
S being the surface of the sphere ( x − 1)2 + ( y − 2)2 + ( z − 3)2 = 1.
(ii) If S is any closed surface enclosing a volume V and
F = x i + 2 y j + 3 z k , prove that ∫ ∫ F • n dS = 6V .
S
(Rohilkhand 2009B)
8. Verify the divergence theorem for F = 4 xi − 2 y j + z 2 k 2
taken over the region bounded by the surfaces x 2 + y 2 = 4, z = 0, z = 3.

9. Use Gauss divergence theorem to find ∫ ∫ F • n dS, where
S
F = 2 x 2 y i − y 2 j + 4 xz 2 k and S is the closed surface in the first octant
bounded by y 2 + z 2 = 9 and x = 2.
10. If F = y i + ( x − 2 xz ) j − x y k , evaluate ∫ ∫S (∇ × F) • n dS where S is the
surface of the sphere x 2 + y 2 + z 2 = a2 above the xy-plane.

D-332
11. Evaluate ∫ ∫S (∇ × F) • n dS,

where F = ( x 2 + y − 4) i + 3 x y j + (2 xz + z 2 ) k and S is the surface of the
paraboloid z = 4 − ( x 2 + y 2 ) above the xy-plane.
12. Compute
(i) ∫ ∫S (a2 x 2 + b 2 y 2 + c 2 z 2 )1 /2 dS, and
(ii) ∫ ∫S (a2 x 2 + b 2 y 2 + c 2 z 2 ) −1 /2 dS
over the ellipsoid ax 2 + by 2 + cz 2 = 1.
13. Evaluate ∫ ∫S ( x 2 + y 2 ) dS, where S is the surface of the cone
z 2 = 3 ( x 2 + y 2 ) bounded by z = 0 and z = 3.
14. Show that ∫ ∫S ( x 2 i + y 2 j + z 2 k ) • n dS vanishes where S denotes the
x2 y2 z2
surface of the ellipsoid + + = 1.
a2 b2 c2
(Meerut 2005, 07; Kumaun 11, 13)
15. If n is the unit outward drawn normal to any closed surface S, show that
∫ ∫ ∫V div n dV = S.
A nswers 2
3 1 3  a3 
2. (i) (ii) 3. (i) (ii) a2  + a
2 2 2  3 
2πa5 π
5. 3a 3 6. (i) (ii)
5 12
9. 180 10. 0
4 4π
11. − 4π 12.(i) π abc. (ii)
3 √ (abc )
13. 9π
5 Stoke’s Theorem
Let S be a piecewise smooth open surface bounded by a piecewise smooth simple closed curve C.
Let F ( x, y, z ) be a continuous vector function which has continuous first partial derivatives
in a region of space which contains S in its interior. Then
∫C F • dr = ∫ ∫S (∇ × F) • n dS = ∫ ∫S (curl F) • dS
D-333
where C is traversed in the positive direction. The direction of C is called positive if an observer,
walking on the boundary of S in this direction, with his head pointing in the direction of
outward drawn normal n to S, has the surface on the left.
(Meerut 2009; Bundelkhand 10)
Note: F • dr =  F • dr  ds = (F • t) ds, where t is unit tangent vector

∫C ∫C 


ds  ∫C
to C. Therefore F • t is the component of F in the direction of the tangent vector of
C. Also (∇ × F) • n is the component of curl F in the direction of outward drawn
normal vector n of S.Therefore in words Stoke’s theorem may be stated as follows:
The line integral of the tangential component of vector F taken around a simple closed curve C is
equal to the surface integral of the normal component of the curl of F taken over any surface S
having C as its boundary.
Cartesian equivalent of Stoke’s theorem:

Let F = F1 i + F2 j + F3 k . Let outward drawn normal vector n of S make angles
α, β, γ with positive directions of x, y, z axes.
Then n = cos α i + cos β j + cos γ k .
 i j k 
 ∂ ∂ ∂ 
Also ∇×F=
 ∂z 
 ∂x ∂y 
 F1 F2 F3 
 ∂F ∂F   ∂F ∂F   ∂F ∂F 
=  3 − 2  i +  1 − 3  j +  2 − 1  k.
 ∂y ∂z   ∂z ∂x   ∂x ∂y 
 ∂F ∂F   ∂F ∂F 
∴ (∇ × F) • n =  3 − 2  cos α +  1 − 3  cos β
 ∂ y ∂ z   ∂ z ∂x 
 ∂F ∂F 
+  2 − 1  cos γ.
 ∂x ∂y 
Also F • dr = ( F1 i + F2 j + F3 k ) • (dx i + dy j + dz k )
= F1 dx + F2 dy + F3 dz .
∴ Stoke’s theorem can be written as
∫C F1 dx + F2 dy + F3 dz
 ∂F3 ∂F   ∂F ∂F 
= ∫ ∫S  − 2  cos α +  1 − 3  cos β
 ∂y ∂z   ∂z ∂x 
 ∂F ∂F  
+  2 − 1  cos γ  dS.
 ∂x ∂y  
D-334
Proof of Stoke’s theorem: Let S be a

surface which is such that its
projections on the xy, yz and zx planes
are regions bounded by simple closed
curves. Suppose S can be represented
simultaneously in the forms
z = f ( x, y ), y = g ( x, z ),
x = h ( z, y ) where f , g, h are
continuous functions and have
continuous first partial derivatives.
Consider the integral
∫ ∫S [∇ × ( F1 i)] • n dS.
 i j k 
 ∂ ∂ ∂  ∂F1 ∂F
We have ∇ × ( F1 i) = = j − 1 k.
 
 ∂x ∂y ∂z  ∂z ∂y
 F1 0 0 
 ∂F ∂F 
∴ [∇ × ( F1 i)] • n =  1 j • n − 1 k • n
 ∂z ∂y 
∂F1 ∂F1
= cos β − cos γ .
∂z ∂y
 ∂F1 ∂F 
∴ ∫ ∫S [∇ × ( F1 i)] • n dS = ∫∫ S  cos β − 1 cos γ  dS.
 ∂z ∂y 
We shall prove that
 ∂F1 ∂F 
∫ ∫S  cos β − 1 cos γ  dS = ∫C F1 dx.
 ∂z ∂y 
Let R be the orthogonal projection of S on the xy-plane and let Γ be its boundary
which is oriented as shown in the figure. Using the representation z = f ( x, y ) of S,
we may write the line integral over C as a line integral over Γ. Thus
∫C F1 ( x, y, z ) dx = ∫ Γ F1 [ x, y, f ( x, y )] dx
= ∫ Γ {F1 [ x, y, f ( x, y )] dx + 0 dy}
∂F1
= −∫∫ dx dy,
R
∂y
by Green’s theorem in plane for the region R.
∂F1 [ x, y, f ( x, y )] ∂F1 ( x, y, z ) ∂F1 ( x, y, z ) ∂f

But = + ⋅
∂y ∂y ∂z ∂y
[∵ z = f ( x, y)]
D-335
 ∂F1 ∂F ∂f 
∴ ∫C F1 ( x, y, z ) dx = − ∫ ∫R  + 1  dx dy ...(1)
 ∂y ∂z ∂y 
Now the equation z = f ( x, y ) of the surface S can be written as
φ ( x, y, z ) ≡ z − f ( x, y ) = 0.
∂f ∂f
We have grad φ = − i− j + k.
∂x ∂y
Let | grad φ | = a.
grad φ
Since grad φ is normal to S, therefore, we get n = ± ⋅
a
But the components of both n and grad φ in positive direction of z-axis are positive.
Therefore
grad φ
n=+
a
1 ∂f 1 ∂f 1
or cos α i + cos β j + cos γ k = − i− j + k.
a ∂x a ∂y a
1 ∂f 1 ∂f 1
∴ cos α = − , cos β = − , cos γ = ⋅
a ∂x a ∂y a
dx dy
Now dS = = a dx dy.
cos γ
 ∂F1 ∂F 
∴ ∫ ∫S  cos β − 1 cos γ  dS
 ∂z ∂y 
 ∂F1  1 ∂f  ∂F1 1
= ∫ ∫R  −  −  a dx dy
 ∂z  a ∂y  ∂y a 
 ∂F1 ∂F ∂f 
=− ∫ ∫R  + 1  dx dy. ...(2)
 ∂y ∂z ∂y 
From (1) and (2), we get
 ∂F1 ∂F 
∫C F1 dx = ∫ ∫S  cos β − 1 cos γ  dS
 ∂z ∂y 
= ∫ ∫S [∇ × ( F1 i)] • n dS ...(3)
Similarly, by projections on the other coordinate planes, we get

∫C F2 dy = ∫ ∫S [∇ × ( F2 j)] • n dS ...(4)
∫C F3 dz = ∫ ∫S [∇ × ( F3 k )] • n dS ...(5)
Adding (3), (4), (5), we get

∫C ( F1 dx + F2 dy + F3 dz ) = ∫ ∫S [∇ × ( F1 i + F2 j + F3 k )] • n dS
D-336
or ∫C F • dr = ∫ ∫S (∇ × F) • n dS.
If the surface S does not satisfy the restrictions imposed above, even then Stoke’s
theorem will be true provided S can be subdivided into surfaces S1 , S2 , ... , S k
with boundaries C1 , C2 , ... , Ck which do satisfy the restrictions. Stoke’s theorem
holds for each such surface. The sum of surface integrals over S1 , S2 , ... , S k will
give us surface integral over S while the sum of the integrals over C1 , C2 , ... , Ck
will give us line integral over C.
Note: Green’s theorem in plane is a special case of Stoke’s theorem. If R is a
region in the xy-plane bounded by a closed curve C, then in vector form Green’s
theorem in plane can be written as
∫ ∫R (∇ × F) • k dR = ∫C F • dr.
This is nothing but a special case of Stoke’s theorem because here k = n = outward
drawn unit normal to the surface of region R.
Example 9: Prove that ∫ r • dr = 0. (Meerut 2010)

C
Solution: By Stoke’s theorem ∫C r • dr = ∫ ∫S (curl r) • n dS = 0,since curl r = 0.
Example 10: By Stoke’s theorem prove that div curl F = 0.
Solution: Let V be any volume enclosed by a closed surface.

Then by divergence theorem
∫ ∫ ∫V ∇ • (curl F) dV = ∫ ∫S (curl F) • n dS.
Divide the surface S into two portions S1 and S2
by a closed curve C. Then
∫ ∫S (curl F) • n dS = ∫ ∫ S1 (curl F) • n dS1
+ ∫ ∫ S2 (curl F) • n dS2 . ... (1)
By Stoke’s theorem right hand side of (1) is

= ∫C F • dr − ∫C F • dr = 0.
Negative sign has been taken in the second integral because the positive directions
about the boundaries of the two surfaces are opposite.
∴ ∫ ∫ ∫V ∇ • (curl F) dV = 0.
Now this equation is true for all volume elements V. Therefore we have
∇ • (curl F) = 0 or div curl F = 0.
D-337
Example 11: Verify Stoke’s theorem for F = y i + z j + x k where S is the upper half
surface of the sphere x 2 + y 2 + z 2 = 1 and C is its boundary.
(Agra 2000, 06; Kanpur 09; Kumaun 07, 10, 13)
Solution: The boundary C of S is a circle in the xy-plane of radius unity and centre
origin. The equations of the curve C are x 2 + y 2 = 1, z = 0. Suppose
x = cos t, y = sin t, z = 0, 0 ≤ t < 2 π are parametric equation of C. Then
∫C F • dr = ∫C ( y i + z j + x k ) • (dx i + dy j + dz k )
= ∫C ( y dx + z dy + x dz )
= ∫C y dx, since on C, z = 0 and dz = 0

2π dx
= ∫0 sin t dt
dt
2π
= ∫0 − sin2 t dt
1 2π
=− (1 − cos 2t ) dt
2 ∫0
1  sin 2t  2 π
=− t −
2 2  0
= − π. ...(1)
Now let us evaluate ∫ ∫ curl F • n dS. We have
S
 i j k 
 ∂ ∂ ∂ 
curl F = ∇ × F = = – i – j – k.
 ∂z 
 ∂x ∂y 
 y z x 
If S1 is the plane region bounded by the circle C, then by an application of
divergence theorem, we have
∫ ∫S curl F • n dS = ∫ ∫ S1 curl F • k dS
[See example 4 after article 4]

= ∫ ∫ S1 (− i − j − k ) • k dS
= ∫ ∫S1 (− 1) dS
=− ∫ ∫S1 dS = − S1 .
But S1 = area of a circle of radius 1 = π (1)2 = π.

∴ ∫ ∫S curl F • n dS = − π. ...(2)
Hence from (1) and (2), the theorem is verified.

D-338
Example 12: Verify Stoke’s theorem for F = (2 x − y) i − yz 2 j − y 2 z k , where S is the

upper half surface of the sphere x 2 + y 2 + z 2 = 1 and C is its boundary.
(Kumaun 2003; Kanpur 10, 14; Avadh 09)
Solution: The boundary C of S is a circle in the xy-plane of radius unity and centre
origin. Suppose x = cos t, y = sin t, z = 0,0 ≤ t < 2 π are parametric equations of
C. Then
2 2
∫ F • dr = ∫ [(2 x − y )i − yz j − y z k ] • (dx i + dy j + dz k )
C C
=∫C [(2 x − y) dx − yz 2 dy − y 2 z dz ]
=∫ (2 x − y) dx, since z = 0 and dz = 0

C
2π dx
= ∫0 (2 cos t − sin t) dt
dt
2π
=−∫ (2 cos t − sin t) sin t dt
0
2π 1
=−∫ [sin 2t −
(1 − cos 2t)] dt
0 2
2π
 cos 2t 1 1 sin 2t 
= − − − t+
 2 2 2 2  0
1 1 1 1
= − [(− + ) − (π − 0) + (0 − 0)]
2 2 2 4
= π. ...(1)
 i j k 
 ∂ ∂ ∂ 
And (∇ × F) =  
∂x ∂y ∂z
 2

 2 x − y − yz − y2 z 
= (− 2 yz + 2 yz ) i − (0 − 0) j + (0 + 1) k
= k.
Let S1 be the plane region bounded by the circle C. If S′ is the surface consisting of
the surfaces S and S1 , then S′ is a closed surface.
∴ by an application of Gauss divergence theorem, we have
∫ ∫ S′ curl F • n dS = 0 [See example 4(i) after article 4 ]
or ∫ ∫S curl F • n dS + ∫ ∫ S1 curl F • n dS = 0
[∵ S ′ consists of S and S1 ]
or ∫ ∫S curl F • n dS − ∫ ∫ S1 curl F • k dS = 0
[∵ on S1 , n = − k ]
or ∫ ∫S curl F • n dS = ∫ ∫ S1 curl F • k dS.
D-339
∴ ∫ ∫S curl F • n dS = ∫ ∫ S1 curl F • k dS
= ∫ ∫ S1 k • k dS
= ∫ ∫ S1 dS = S1 = π . ...(2)
Note that S1 = area of a circle of radius 1 = π (1)2 = π .

Hence from (1) and (2) Stoke’s theorem is verified.
Example 13: Verify Stoke’s theorem for F = ( x 2 + y 2 ) i − 2 x yj taken round the

rectangle bounded by x = ± a, y = 0, y = b. (Bundelkhand 2007; Agra 08;
Kumaun 15)
 i j k 
 ∂ ∂ ∂ 
Solution: We have curl F =  
∂x ∂y ∂z
 2 
x + y2 −2 xy 0 
= (− 2 y − 2 y ) k = − 4 y k .
Also n = k.
∴ ∫ ∫S (curl F) • n dS
b a
= ∫ y = 0 ∫ x = −a (− 4 y k ) • k dx dy
b a
=−4 ∫ y = 0 ∫ x = −a y dx dy
b a
= − 4∫ [ xy] x = − a dy
y= 0
b
= − 4∫ 2ay dy
y=0
b
= − 4 [ay 2 ] 0 = − 4ab 2 .
2 2
Also o∫ F • dr = ∫ C [( x + y ) i − 2 xy j ] • (dx i + dy j)
C
= o∫ [( x 2 + y 2 ) dx − 2 xy dy ]
C
=∫ [( x 2 + y 2 ) dx − 2 xy dy ] + ∫ +∫ +∫ .
DA AB BE ED
Along DA , y = 0 and dy = 0. Along AB, x = a and dx = 0.

Along BE, y = b and dy = 0. Along ED, x = − a and dx = 0.
a b
∴ ∫C F • dr = ∫ x = −a x 2 dx + ∫ − 2ay dy
y =0
−a 0
+ ∫x=a ( x 2 + b 2 ) dx + ∫ y=b 2ay dy
a a b
= ∫ −a x 2 dx − ∫ ( x 2 + b 2 ) dx − 4a ∫ y dy
−a 0
D-340
a b
=− ∫ −a x 2 dx − 4a ∫0 y dy
b
 y2 
2
= − 2ab − 4a   = − 4ab 2 .
 2 
0
Thus ∫C F • dr = ∫ ∫S (curl F) • n dS.
Hence the theorem is verified.
Example 14: Evaluate ∫C F • dr by Stoke’s theorem where
F = y 2 i + x 2 j − (x + z) k
and C is the boundary of the triangle with vertices at (0, 0, 0), (1, 0, 0), (1, 1, 0).
(Avadh 2013)
Solution: We have
i j k
Y
∂ ∂ ∂
Curl F =
∂x ∂y ∂z B (1, 1)
y2 x2 (x + z)
= 0 i + j + 2 ( x − y ) k.
Also we note that z co-ordinate of each vertex
of the triangle is zero. Therefore the triangle lies O
A (1, 0) X
in the xy-plane. So n = k .
∴ Curl F • n = [ j + 2 ( x − y)k ] • k
= 2 ( x − y).
In the figure, we have only considered the x y plane.
The equation of the line OB is y = x.
By Stoke’s theorem
∫C F • dr = ∫ ∫S (curl F ) • n dS
1 x
= ∫ x =0 ∫ y =0
2 ( x − y ) dx dy
x
1
 y2 
=2 ∫ x =0 x y −  dx
 2 
 y =0
 2 2
1 x
= 2∫ x −  dx
0 2

1 x2
= 2∫ dx
0 2
1 1
= ∫0 x 2 dx = ⋅
3
D-341
1. (i) State Stoke’s theorem.

(ii) By Stoke’s theorem prove that curl grad φ = 0. (Kumaun 2014)
(iii) Verify Stoke’s theorem for the function F = zi + xj + y k where curve
is the unit circle in the xy-plane bounding the hemisphere
z = √ (1 − x 2 − y 2 ). (Garhwal 2003; Agra 07)
2. Verify Stoke’s theorem for the vector F = z i + x j + y k taken over the half
of the sphere x 2 + y 2 + z 2 = a2 lying above the xy-plane.
3. (i) Verify Stoke’s theorem for the function F = x 2 i + xy j integrated along
the rectangle, in the plane z = 0, whose sides are along the lines
x = 0, y = 0, x = a and y = b.
(ii) Verify Stoke’s theorem for the function F = ( x 2 − y 2 ) i + 2xy j in
the rectangular region in the xy-plane bounded by the lines
x = 0, x = a, y = 0 and y = b . (Kanpur 2008)
2
(iii) Verify Stoke’s theorem for the function F = x i + xy j, integrated
round the square, in the plane z = 0, whose sides are along the lines
x = 0, y = 0, x = a, y = a. (Agra 2002)
4. Verify Stoke’s theorem for the vector A = 3 y i − xz j + yz 2 k , where S is
the surface of the paraboloid 2z = x 2 + y 2 bounded by z = 2 and C is its
boundary.
5. (i) By converting into a line integral evaluate
∫ ∫S (∇ × A) • n dS, where A = ( x − z ) i + ( x 3 + yz ) j − 3 xy 2 k
and S is the surface of the cone z = 2 − √ ( x 2 + y 2 ) above the

xy-plane.
(ii) By converting into a line integral evaluate ∫ ∫S (∇ × F) • n dS
where F = ( x 2 + y − 4) i + 3 xy j + (2 xy + z 2 ) k and S is the surface of

the paraboloid z = 4 − ( x 2 + y 2 ) above the xy-plane.
6. (i) Evaluate by Stoke’s theorem ∫C (e x dx + 2 y dy − dz )
where C is the curve x 2 + y 2 = 4, z = 2. (Garhwal 2001; Meerut 09B)

(ii) Evaluate by Stoke’s theorem ∫C ( yz dx + xz dy + xy dz )
where C is the curve x 2 + y 2 = 1, z = y 2 . (Meerut 2006B)

D-342
7. (i) Evaluate ∫ ∫ (∇ × F) • n dS, where F = ( y − z + 2) i +

S
( yz + 4) j − xz k and S is the surface of the cube x = y = z = 0,

x = y = z = 2 above the xy-plane.
(ii) Evaluate by Stoke’s theorem ∫ (sin z dx − cos x dy + sin y dz ) where
C
C is the boundary of the rectangle 0 ≤ x ≤ π, 0 ≤ y ≤ 1, z = 3.

(Meerut 2009)
8. If f = ∇ φ and g = ∇ψ are two vector point functions, such that ∇ 2 φ = 0,
∇ 2 ψ = 0, show that
∫ ∫S (g • ∇) f • dS = ∫C (f × g) • dr + ∫ ∫ (f • ∇)g • dS.
S
9. Prove that a necessary and sufficient condition that ∫ F • dr = 0 for every

C
closed curve C lying in a simply connected region R is that ∇ × F = 0

identically.
10. Apply Stoke’s theorem to prove that
∫C ( y dx + z dy + x dz ) = − 2 √ 2 πa2
where C is the curve given by x 2 + y 2 + z 2 − 2ax − 2ay = 0, x + y = 2a
and begins at the point (2 a, 0, 0) and goes at first below the z-plane.
(Meerut 2005, 06B)
11. Use Stoke’s theorem to evaluate ∫ ∫ (∇ × F) • n dS,
S
where F = y i + ( x − 2 xz ) j − x y k and S is the surface of sphere

x 2 + y 2 + z 2 = a2 , above the xy-plane. (Kumaun 2012)
A nswers 3
5. (i) 12π (ii) − 4π 6. (i) 0 (ii) 0
7. (i) − 4 (ii) 2
11. 0
6 Line Integrals Independent of Path

Let F ( x, y, z ) = f ( x, y, z ) i + g ( x, y, z ) j + h ( x, y, z ) k be a vector point function
defined and continuous in a region R of space. Let P and Q be two points in R and
let C be a path joining P to Q. Then
∫C F • dr = ∫ ( f dx + g dy + h dz ) ... (1)
D-343
is called the line integral of F along C. In general the value of this line integral
depends not only on the end points P and Q of the path C but also on C.
In other words, if we integrate from P to Q along different paths, we shall, in
general, get different values of the integral. The line integral (1) is said to be
independent of path in R, if for every pair of end points P and Q in R the value of the
integral is the same for all paths C in R starting from P and ending at Q.
In this case the value of this line integral will depend on the choice of P and Q and
not on the choice of the path joining P to Q.
Definition: The expression f dx + g dy + h dz is said to be an exact differential if there

exists a single valued scalar point function φ ( x, y, z ), having continuous first partial
derivatives such that dφ = f dx + g dy + h dz .
It can be easily seen that f dx + g dy + h dz is an exact differential if and only if the
vector function F = f i + g j + h k is the gradient of a single valued scalar function
φ ( x, y, z ).
Because F = grad φ
∂φ ∂φ ∂φ
if, and only if f i + g j+ hk = i+ j+ k
∂x ∂y ∂z
∂φ ∂φ ∂φ
if, and only if f = , g= , h=
∂x ∂y ∂z
∂φ ∂φ ∂φ
if, and only if f dx + g dy + h dz = dx + dy + dz
∂x ∂y ∂z
if, and only if f dx + g dy + h dz = d φ.
Thus F = grad φ if, and only if f dx + g dy + h dz is an exact differential dφ.

Theorem 1: Let f (x, y, z), g (x, y, z) and h (x, y, z) be continuous in a region R of space.
Then the line integral
∫ ( f dx + g dy + h dz )
is independent of path in R if and only if the differential form under the integral sign is exact in
R.
Or
Let F ( x, y, z ) be continuous in region R of space. Then the line integral ∫C F • dr
is independent of the path C in R joining P and Q if and only if F = grad φ where φ ( x, y, z )

is a single-valued scalar function having continuous first partial derivatives in R.
Proof: Suppose F = grad φ in R. Let P and Q be any two points in R and let C be
any path from P to Q in R.
Then ∫C F • dr = ∫C ∇φ • dr
D-344
 ∂φ ∂φ ∂φ 
= ∫C  i+ j+ k  • (dx i + dy j + dz k)
 ∂x ∂y ∂z 
 ∂φ ∂φ ∂φ 
= ∫C  dx + dy + dz 
 ∂ x ∂ y ∂ z 
= ∫C dφ
Q
= ∫P dφ
Q
= [ φ ]P
= φ (Q) − φ ( P).
Thus the line integral depends only on points P and Q and not on the path joining
them. This is true, of course, only if φ ( x, y, z ) is single valued at all points P and Q.
Conversely, suppose the line integral ∫C F • dr is independent of the path C
joining any two points P and Q in R. Let P be a fixed point ( x0 , y0 , z 0 ) in R and

let Q be any point ( x, y, z ) in R.
( x, y, z)
Let φ ( x, y, z ) = ∫( x0 , y0 , z0 ) F • dr
( x, y, z)  dr 
= ∫ x0 , y0 , z0  F • ds  ds.
Differentiating both sides with respect to s, we get
dφ dr
=F• .
ds ds
dφ ∂φ dx ∂φ dy ∂φ dz
But = + +
ds ∂x ds ∂y ds ∂z ds
 ∂φ ∂φ ∂φ   dx dy dz 
= i+ j+ k  • i + j+ k 
 ∂ x ∂ y ∂ z   ds ds ds 
dr
=∇φ• .
ds
dr dr
∴ F• =∇φ•
ds ds
dr
or (∇ φ – F) • = 0.
ds
Now this result is true irrespective of the path joining P to Q i. e. this result is true
dr
irrespective of the direction of which is tangent vector to C. Therefore we must
ds
have
∇φ – F = 0
i.e., ∇ φ = F.
This completes the proof of the theorem.
D-345
Definition: A vector field F ( x, y, z ) defined and continuous in a region R of space is said to

be a conservative vector field if the line integral ∫ F • dr is independent of the path C in R
C
joining P and Q where P and Q are any two points in R.

By theorem 1, vector field F ( x, y, z ) is conservative if and only if F = ∇ φ where
φ ( x, y, z ) is a single valued scalar function having continuous first partial
derivatives in R. The function φ ( x, y, z ) is called the scalar potential of the vector
field F.
Theorem 2: Let F ( x, y, z ) be a vector function defined and continuous in a region R of

Q
space. Then the line integral ∫ F • dr is independent of the path joining any two points P
P
and Q in R if and only if ∫C F • dr = 0 for every simple closed path in R.
Proof: Let C be any simple closed path in R and

let the line integral be independent of path in
R. Take two points P and Q on C and subdivide
C into two arcs PBQ and QAP. Then
∫C F • dr = ∫ PBQAP F • dr
= ∫ PBQ F • dr + ∫ F • dr
QAP
= ∫ PBQ F • dr − ∫ F • dr
PAQ
= 0,
since the integral from P to Q along a path through B is equal to the integral from P
to Q along a path through A.
Conversely, suppose that the integral under consideration is zero on every simple
closed path in R. Let P and Q be any two points in R which join P to Q and do not
cross. Then
∫ PBQAP F • dr = ∫ PBQ F • dr + ∫ QAP F • dr
= ∫ PBQ F • dr − ∫ F • dr.
PAQ
But as given, we have

∫ PBQAP F • dr = 0.
∴ ∫ PBQ F • dr − ∫ PAQ F • dr = 0
or ∫ PBQ F • dr = ∫ PAQ F • dr.
This completes the proof of the theorem.

D-346
Theorem 3: Let F ( x, y, z ) = f i + g j + h k be a continuous vector function having

continuous first partial derivatives in a region R of space. If ∫ f dx + g dy + h dz is
independent of path in R and consequently f dx + g dy + h dz is an exact differential in R,

then curl F = 0 everywhere in R. Conversely, if R is simply connected and curl F = 0
everywhere in R, then f dx + g dy + h dz is an exact differential in R or
∫ f dx + g dy + h dz is independent of path in R.
Proof: Suppose ∫ ( f dx + g dy + h dz ) is independent of path in R. Then

f dx + g dy + h dz is an exact differential in R. Therefore
F = f i + g j + h k = grad φ.
∴ curl F = curl ( grad φ ) = 0.
Converse. Suppose R is simply connected and curl F = 0 everywhere in R. Let C be
any simple closed path in R. Since R is simply connected, therefore we can find a
surface S in R having C as its boundary. Therefore by Stoke’s theorem
∫C F • dr = ∫ ∫S (curl F ) • n dS = 0.
Thus ∫C F • dr is zero for every simple closed path C in R.
Therefore ∫ F • dr is independent of path in R.
Therefore F = ∇ φ and consequently f dx + g dy + h dz is an exact differential dφ.
Note: The assumption that R be simply connected is essential and cannot be

omitted. It is obvious from the following illustration.
Illustration: Let
y x
F=− 2 2
i+ 2
j.
x + y x + y2
Here F is not defined at origin. In every region R of the xy-plane not containing the
origin, we have
 
 i j k 
 
∂ ∂ ∂ 
Curl F = 
 ∂x ∂y ∂z 
 y x 
− 2 2 2
0 
 x + y x + y2 
 ∂  x  ∂  y  
= 0i + 0 j +     
 x 2 + y 2  + ∂y  x2 + y2   k
 ∂x     
D-347
 x2 + y2 − 2 x2 x2 + y2 − 2 y2 
= 2 2 2
+  k
 (x + y ) ( x 2 + y 2 )2 
=0k
= 0.
Suppose R is simply connected. For example let R be the region enclosed by a
simple closed curve C not enclosing the origin. Then
 y x 
∫C F • dr = ∫C  − x 2 + y 2 dx + x 2 + y 2 dy 
∂  x  ∂  y 
  − 
= ∫ ∫ R   x 2 + y 2  − ∂y  x 2 + y 2   dx dy,
 ∂x     
by Green’s theorem in plane
= 0.
Suppose R is not simply connected. Let R be the region of the xy-plane contained
1 3
between concentric circles of radii and and having centre at origin. Obviously
2 2
R is not simply connected. We have z = 0, everywhere in R. Let C be a closed curve
in R. The parametric equations of C can be taken as x = cos t, y = sin t, z = 0,
0 ≤ t < 2π.
 y y 
We have ∫C F • dr = ∫C − dx + dy
 x2 + y2 x2 + y2 
 
2π  sin t dx cos t dy 
= ∫t=0 − 2 2 dt
+ 2 2  dt
 cos t + sin t cos t + sin t dt 
2π
= ∫0 (sin2 t + cos 2 t) dt
= 2 π.
Thus we see that
∫C F • dr ≠ 0.
Definition: Irrotational vector field: A vector field F is said to be irrotational if curl

F = 0.
We see that an irrotational field F is characterized by any one of the three
conditions :
(i) F = ∇ φ,
(ii) ∇ × F = 0,
(iii) ∫C F • dr = 0 for every closed path.
Any one of these conditions implies the other two.

D-348
Example 15: Are the following forms exact ?

(i)__e dx + e dy + e z dz .________
y x
(ii) yz dx + xz dy + xy dz .
Solution:
y
(i) Here F=e i + e x j + e z k.
We have
i j k
∂ ∂ ∂
Curl F =
∂x ∂y ∂z
e y
ex ez
= 0 i + 0 j + (e x − e y ) k .
Since curl F ≠ 0, therefore the given form is not exact.
(ii) Here F = yz i + xz j + xy k .
We have
i j k
∂ ∂ ∂
Curl F =
∂x ∂y ∂z
yz xz xy
= ( x − x)i − ( y − y) j + (z − z )k
= 0.
Since curl F = 0, therefore the given form is exact.
Example 16: In each of following cases show that the given differential form is exact and
find a function φ such that the form equals dφ :
(i) cos x dx − 2 yz dy − y 2 dz .
(ii) (z 2 − 2 xy) dx − x 2 dy + 2 xz dz .
Solution: (i) Here F = cos x i − 2 yz j − y 2 k .

We have
 i j k 
 ∂ ∂ ∂ 
Curl F =  
 ∂x ∂y ∂z 
cos x − 2 yz − y 2

= (− 2 y + 2 y) i + 0 j + 0 k = 0.
∴ the given form is exact.
D-349
Let F = ∇ φ,
∂φ ∂φ ∂φ
or cos x i − 2 yz j − y 2 k = i+ j+ k.
∂x ∂y ∂z
∂φ
Then = cos x whence φ = sin x + f1 ( y, z ) ... (1)
∂x
∂φ
= − 2 yz whence φ = − y 2 z + f 2 ( x, z ) ... (2)
∂y
∂φ
= − y 2 whence φ = − y 2 z + f 3 ( x, y). ... (3)
∂z
(1), (2), (3) each represents φ. These agree if we choose
f1 ( y, z ) = − y 2 z , f 2 ( x, z ) = sin x, f 3 ( x, y) = sin x.
∴ φ = sin x − y 2 z to which may be added any constant.
∴ φ = sin x − y 2 z + C.
(ii) Here F = (z 2 − 2 x y) i − x 2 j + 2 xz k . We have
i j k
∂ ∂ ∂
Curl F =
∂x ∂y ∂z
z2 − 2 xy − x 2 2 xz
= 0 i + 0 j + 0 k = 0.
∴ the given form is exact.
Let F =∇ φ
∂φ ∂φ ∂φ
or (z 2 − 2 xy) i − x 2 j + 2 xz k = i+ j+ k.
∂x ∂y ∂z
∂φ
Then = z 2 − 2 xy whence φ = z 2 x − x 2 y + f1 ( y, z ) ... (1)
∂x
∂φ
= − x2 whence φ = − x 2 y + f 2 ( x, z ) ... (2)
∂y
∂φ
= 2 xz whence φ = xz 2 + f 3 ( x, y). ... (3)
∂z
f1 ( y, z ) = 0, f 2 ( x, z ) = xz 2 , f 3 ( x, y) = − x 2 y.
∴ φ = z 2 x − x 2 y to which may be added any constant.
∴ φ = z 2 x − x 2 y + C.
Example 17: Show that the vector field F given by

F = ( x 2 − yz ) i + ( y 2 − zx) j + (z 2 − xy) k
is irrotational. Find a scalar φ such that F = ∇φ.
D-350
Solution: We have
i j k
∂ ∂ ∂
Curl F =
∂x ∂y ∂z
2
x − yz y 2 − zx z 2 − xy
= (− x + x) i − (− y + y) j + (− z + z ) k = 0.
∴ The vector field F is irrotational.
Let F = ∇φ
∂φ ∂φ ∂φ
or ( x 2 − yz ) i + ( y 2 − zx) j + (z 2
− xy) k = i+ j+ k.
∂x ∂y ∂z
∂φ x3
Then = x 2 − yz whence φ = − xyz + f1 ( y, z ) ... (1)
∂x 3
∂φ y3
= y 2 − zx whence φ = − xyz + f 2 ( x, z ) ... (2)
∂y 3
∂φ z3
= z 2 − x y whence φ = − x y z + f 3 ( x, y). ... (3)
∂z 3
y3 z3
f1 ( y, z ) = + ,
3 3
3
x + z3
f 2 ( x, z ) = ,
3
x3 + y3
f 3 ( x, y) = .
3
x3 + y3 + z 3
Therefore φ= − xyz + C.
3
7 Physical Interpretation of Divergence and Curl

Physical interpretation of divergence: Suppose that there is a fluid motion whose
velocity at any point is v ( x, y, z). Then the loss of fluid per unit volume per unit time in a small
parallelopiped having centre at P (x, y, z) and edges parallel to the co-ordinate axes and having
lengths δx, δy, δz respectively, is given approximately by
div v = ∇ • v.
Let v = v1 i + v2 j + v3 k .
x-component of velocity v at P = v1 ( x, y, z ).
∴ x-component of v at centre of face AFED which is perpendicular to x-axis and is

nearer to origin
D-351
 δx 
= v1  x − , y, z 
 2 
δx ∂v1
= v1 ( x, y, z ) − +…
2 ∂x
by Taylor’s theorem
δx ∂v1
= v1 ( x, y, z ) −
2 ∂x
approximately.
Similarly x-component of v at centre of opposite

face GHCB
δx ∂v1
= v1 + approximately.
2 ∂x
∴ volume of fluid entering the parallelopiped across AFED per unit time
 δx ∂v1 
=  v1 −  δy δz .
 2 ∂x 
Also volume of fluid going out the parallelopiped across GHCB per unit time
 δx ∂v1 
=  v1 +  δy δz .
 2 ∂x 
∴ loss in volume per unit time in the direction of x-axis
 δx ∂v1   δx ∂v1 
=  v1 +  δy δz −  v1 −  δy δz
 2 ∂x   2 ∂x 
∂v1
= δx δy δz .
∂x
Similarly, loss in volume per unit time in y direction
∂v
= 2 δx δy δz ,
∂y
and loss in volume per unit time in z direction
∂v
= 3 δx δy δz .
∂z
∴ total loss of the fluid per unit volume per unit time symbol ·
 ∂v1 ∂v ∂v 
 + 2 + 3  δx δy δz
 ∂x ∂y ∂z 
=
δx δy δz
∂v1 ∂v2 ∂v3
= + +
∂x ∂y ∂z
= ∇ • v = div v.
D-352
Physical interpretation of curl: Let S be a circular disc of small radius r and centre
P bounded by the circle C. Let F ( x, y, z ) be a continuously differentiable vector
function in S. Then by Stoke’s theorem
∫C F • dr = ∫ ∫S (curl F ) • n dS
= (curl F ) • n ∫ ∫S dS,
by mean value theorem of integral calculus where (curl F ) • n is some value

intermediate between the maximum and minimum values of (curl F ) • n over S.
∴ ∫C F • dr = (curl F ) • n S.
 F • dr
∫ C 
∴ (curl F ) • n = .
S
Taking limit as r → 0, we get at P,
∫ C F • dr
(curl F ) • n = lim .
r→0 S
Now (curl F ) • n is normal component of curl F at P and ∫ F • dr is circulation of
C
F about C. Therefore the normal component of the curl can be interpreted

physically as the limit of the circulation per unit area.
1. Are the following forms exact ?

(i) x dx − y dy + z dz .
(ii) y 2 z 3 dx + 2 xyz 3 dy + 3 xy 2 z 2 dz .
2. In each of following cases show that the given differential form is exact
and find a function φ such that the form equals dφ :
(i) x dx − y dy − z dz .
(ii) dx + z dy + y dz .
3. (i) Show that
( y 2 z 3 cos x − 4 x 3 z ) dx + 2z 3 y sin x dy + (3 y 2 z 2 sin x − x 4 ) dz
is an exact differential of some function φ and find this function.
(ii) Show that F = (2 x y + z 3 ) i + x 2 j + 3 xz 2 k is a conservative force
field. Find the scalar potential. Find also the work done in moving an
object in this field from (1, − 2, 1) to (3, 1, 4).
D-353
4 (i) Show that the vector field

F = (2 xy 2 + yz ) i + (2 x 2 y + xz + 2 yz 2 ) j + (2 y 2 z + xy) k
is conservative.
(ii) Show that F = x i + yj + z k is conservative and find φ such that
F = ∇φ.
5. (i) Show that F = (sin y + z ) i + ( x cos y − z ) j + ( x − y) k is a
conservative vector field and find a function φ such that F = ∇φ.
(ii) Evaluate ∫ 2 xyz 2 dx + ( x 2 z 2 + z cos yz ) dy + (2 x 2 yz + y cos yz ) dz

C
where C is any path from (0, 0, 1) to (1, π / 4 , 2).

6. Show that the following vector functions F are irrotational and find the
corresponding scalar φ such that F = ∇φ .
(i) F = (sin y + z cos x) i + ( x cos y + sin z ) j + ( y cos z + sin x) k .
(ii) F = ( y sin z − sin x) i + ( x sin z + 2 yz ) j + ( xy cos z + y 2 ) k .
(iii) F = x 3 i + y 3 j + z 3 k .
7. Find a, b, c if F = (3 x − 3 y + az ) i + (bx + 2 y − 4z ) j + (2x + cy + z ) k
is irrotational.
8. Evaluate ∫C yz dx + ( xz + 1) dy + x y dz , where C is any path from
(1, 0, 0) to (2, 1, 4).

9. Show that the form under the integral sign is exact and evaluate
(2 , 0,1)
∫(0, 2 ,1) [ze x dx + 3 yz dy + (e x + y 2 ) dz ].
A nswers 4
1. (i) Exact (ii) Exact
2
x − y2 − z 2
2. (i) Exact; φ = +C
2
(ii) Exact; φ = x + yz + C
3. (i) φ = y 2 z 3 sin x − x 4 z + C
(ii) φ = x 2 y − xz 3 + C; 202
1
4. (ii) φ = ( x 2 + y 2 + z 2 ) + C.
2
5. (i) φ = x sin y + x z − yz + C.
(ii) π +1
D-354
6. (i) φ = x sin y + z sin x + y sin z + C

(ii) φ = xy sin z + cos x + y 2 z + C
1
(iii) φ = ( x 4 + y 4 + z 4 ) + C
4
7. a = 2, b = − 3, c = − 4.
8. 9
9. e2 − 5

(a), (b), (c) and (d).
1. If C is the curve x 2 + y 2 = 1, z = y 2 , then by Stoke’s theorem

∫ ( yz dx + zx dy + xy dz) is
C
(a) 0 (b) 3
2. If S denotes the surface of the cube bounded by the planes x = 0, x = a, y = 0,

y = a , z = 0, z = a then by the application of Gauss divergence theorem the
value of ∫ ∫ ( x i + y j + z k ) • n dS is
S
(a) a3 (b) 2a3

3
(c) 3a (d) 0
3. The value of ∫ r • dr is
C
(a) 1 (b) 3
(Kumaun 2007)
4. Gauss divergence theorem relates to
(a) surface and volume integral (b) line and volume integral
(c) line and surface integral (d) all of these (Kanpur 2011)
5. By Stoke's theorem the value of div curl F is

(a) 1 (b) 2
(c) 3 (d) 0 (Kanpur 2013)
D-355
6. By Green's theorem the value of ∫ ( x 2 − cosh y) dx + ( y + sin x) dy,

C
where C is the rectangle with vertices (0, 0),(π, 0),(π, 1),(0, 1), is
(a) 1 − cosh1 (b) 1 + cosh1
(c) cosh1 − 1 (d) cosh1 + 1
7. By Green's theorem the value of ∫ (cos x sin y − x y) dx + sin x cos y dy,
C
where C is the circle x 2 + y 2 = 1, is

(a) 1 (b) −1
8. For any closed surface S, ∫ ∫ curl F • n dS is
S
(a) 0 (b) 1
(c) −1 (d) 3
9. The value of ∫ ∫ r • n dS, where S is a closed surface, is
S
(a) V (b) − V
(c) 3V (d) −3V
10. If F = ax i + byj + cz k ; a, b, c are constants, then the value of ∫ ∫S F • n dS is
1 1
(a) π (a + b + c ) (b) − π (a + b + c )
3 3
4 4
(c) π (a + b + c ) (d) − π (a + b + c )
3 3
11. The value of ∫ ∫ n dS, for any closed surface S is
S
(a) 0 (b) 1
(c) −1 (d) 2
x
12. By Stoke's theorem the value of ∫ (e dx + 2 y dy − dz ), where C is the
C
curve x 2 + y 2 = 4, z = 2, is
(a) 1 (b) −1
(c) 0 (d) 2

1. For any closed surface S, ∫ ∫ curl F • n dS = …… .

S
2. If n is the unit outward drawn normal to any closed surface S, then
∫ ∫ ∫V div n dV = …… .
(Bundelkhand 2008)
D-356
3. The value of ∫C r • d r = …… .
(Agra 2008)
4. A necessary and sufficient condition that ∫ F • d r = 0 for every closed
C
curve C lying in a simply connected region R is that ∇ × F = ……
identically.
5. By Stoke’s theorem, ∫C F • d r = …… .
6. ∫ ∫S r • n dS = …… .
(Kumaun 2009)
True or False
1. ∫ ∫S n dS = 0 for any surface S.

2. Green’s theorem in plane is a special case of Stoke’s theorem.
3. Green’s theorem states that “the surface integral of the normal component of
a vector F taken over a closed surface is equal to the integral of the divergence
of F taken over the volume enclosed by the surface”.
A nswers
1. (a) 2. (c) 3. (c) 4. (a) 5. (d)
6. (c) 7. (c) 8. (a) 9. (c) 10. (c)
11. (a) 12. (c)

1. 0 2. S 3. 0 4. 0
5. ∫ ∫ (∇ × F) • n dS 6. 3V
S
True or False
1. F 2. T 3. F

Krishnas - Differential Equations Vector Calculus by

Uploaded by

Document Information

Original Title

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

Krishnas - Differential Equations Vector Calculus by

Uploaded by

Copyright:

Available Formats

h n a's

Kris TEXT BOOK on

Shilpi Singh Manoj Kumar Mishra

KRISHNA Prakashan Media (P) Ltd.

Unit-1: Differential equation, formation of a differential equation, order and degree,

Unit-2: Linear differential equations with constant, coefficients, homogeneous linear

Unit-3: Simultaneous linear differential equations with constant coefficients, general

Chap-1: Differential Equations of First Order and First Degree..............D-01—D-54

Chap-2: Differential Equations of the First Order but not of the

Chap-3: Orthogonal Trajectories........................................................D-83—D-94

Chap-4: Linear Differential Equations with Constant Coefficients........D-95—D-136

Chap-5: Homogeneous Linear Differential Equations.......................D-137—D-150

Chap-6: Ordinary Simultaneous Differential Equations.....................D-151—D-172

Chap-7: Linear Equations of Second Order with Variable

Chap-8: Differentiation of Vectors...................................................D-215—D-240

Chap-9: Gradient, Divergence and Curl...........................................D-241—D-286

Chap-10: Integration of Vectors.........................................................D-287—D-296

Chap-11: Line Integrals....................................................................D-297—D-314

Chap-12: Green's, Gauss's and Stoke's Theorems...............................D-315—D-356

Differential Equations of First Order

Differential Equations of The First

Linear Differential Equations with

Ordinary Simultaneous Differential

Linear Equations of Second Order

1. Gradient, Divergence and Curl

1. Green's, Gauss's and Stoke's Theorems

A function y = f ( x) is called a Solution (or the primitive) of a differential equation if,

A solution of a differential equation, containing independent arbitrary constants

Solution: The equation of all circles of radius a is given by

h and k being parameters (i. e., arbitrary constants).

which is the required differential equation.

1. Find the differential equation of the family of curves

2. Find the differential equation corresponding to

5. Find the differential equation corresponding to the family of curves

6. Show that Ax2 + By2 = 1 is the solution of

7. Show that v = (a / r) + B is a solution of

2 Differential Equations of First Order and First Degree

2.1 Variables Separable

Example 3: Solve (1 + x2 ) dy = (1 + y2 ) dx (Meerut 2003)

Example 5: Solve dy / dx = ( x + y)2 . (Avadh 2011; Kanpur 15)

Thus the variables being separated and so integrating, we get

or x + y = tan ( x + c ) as the required solution where c is an arbitrary constant.

Solve the following differential equations :

5. (i) (ds / dx) + x2 = x2 e3 s .

6. (i) log (dy / dx) = ax + by.

10. (i) 4 x + y + 1 = 2 tan (2 x + k )

2.2 Homogeneous Equations

To solve such an equation, put

Now replacing v by ( y / x) after integration, we get the required solution.

Example 7: Solve ( x2 − y2 ) dx + 2 xy dy = 0 . (Gorakhpur 2008; Kumaun 13)

Solution: The given equation can be written as

Example 8: Solve y2 dx + ( xy + x2 ) dy = 0 . (Avadh 2010)

Solution: The given equation can be written as

∴ putting y = vx , this equation reduces to

in which the variables are separable.

Example 10: Solve ( x3 − 3 xy2 ) dx = ( y3 − 3 x2 y) dy.

(Meerut 2007; Kumaun 09; Garhwal 13)

or log (cx) = log [{(v + 1)1 /2 (v − 1)1 /2 } /(v2 + 1)]

Solve the following differential equations :

2. y − x (dy / dx) = x + y (dy / dx). (Gorakhpur 2010; Garhwal 08)

4. ( x2 + y2 ) (dy / dx) = xy. (Kanpur 2002; Kumaun 12)

10. (1 + e x / y) dx + e x / y{1 − ( x / y)} dy = 0 . (Meerut 2006)

11. sin ( y / x) = cx 12. tan ( y / x) = log (c / x)